{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 6.623101377605087,
  "eval_steps": 3000,
  "global_step": 75000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 7.74092288017273,
      "epoch": 0.0004415400918403391,
      "grad_norm": 0.9375,
      "learning_rate": 2e-06,
      "loss": 7.5069,
      "mean_token_accuracy": 0.0897733099758625,
      "num_tokens": 9390.0,
      "step": 5
    },
    {
      "entropy": 7.730613660812378,
      "epoch": 0.0008830801836806782,
      "grad_norm": 0.953125,
      "learning_rate": 4.5e-06,
      "loss": 7.6006,
      "mean_token_accuracy": 0.09082280173897743,
      "num_tokens": 18671.0,
      "step": 10
    },
    {
      "entropy": 7.711504316329956,
      "epoch": 0.0013246202755210173,
      "grad_norm": 0.9375,
      "learning_rate": 7e-06,
      "loss": 7.5458,
      "mean_token_accuracy": 0.09541130512952804,
      "num_tokens": 27614.0,
      "step": 15
    },
    {
      "entropy": 7.760569763183594,
      "epoch": 0.0017661603673613563,
      "grad_norm": 0.94140625,
      "learning_rate": 9.5e-06,
      "loss": 7.6561,
      "mean_token_accuracy": 0.08549572601914406,
      "num_tokens": 37850.0,
      "step": 20
    },
    {
      "entropy": 7.768425226211548,
      "epoch": 0.0022077004592016957,
      "grad_norm": 0.9375,
      "learning_rate": 1.2e-05,
      "loss": 7.5482,
      "mean_token_accuracy": 0.10141252279281616,
      "num_tokens": 47166.0,
      "step": 25
    },
    {
      "entropy": 7.844929361343384,
      "epoch": 0.0026492405510420347,
      "grad_norm": 0.88671875,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 7.5478,
      "mean_token_accuracy": 0.0926934964954853,
      "num_tokens": 55500.0,
      "step": 30
    },
    {
      "entropy": 7.811486482620239,
      "epoch": 0.0030907806428823736,
      "grad_norm": 0.88671875,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 7.5563,
      "mean_token_accuracy": 0.08848847448825836,
      "num_tokens": 63851.0,
      "step": 35
    },
    {
      "entropy": 7.876116132736206,
      "epoch": 0.0035323207347227126,
      "grad_norm": 0.9453125,
      "learning_rate": 1.95e-05,
      "loss": 7.5805,
      "mean_token_accuracy": 0.09331603050231933,
      "num_tokens": 73697.0,
      "step": 40
    },
    {
      "entropy": 7.89044713973999,
      "epoch": 0.003973860826563052,
      "grad_norm": 1.0390625,
      "learning_rate": 2.2e-05,
      "loss": 7.5408,
      "mean_token_accuracy": 0.09619837477803231,
      "num_tokens": 83000.0,
      "step": 45
    },
    {
      "entropy": 7.889915847778321,
      "epoch": 0.004415400918403391,
      "grad_norm": 0.80859375,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 7.5514,
      "mean_token_accuracy": 0.09451765418052674,
      "num_tokens": 92982.0,
      "step": 50
    },
    {
      "entropy": 7.9315505027771,
      "epoch": 0.00485694101024373,
      "grad_norm": 1.0078125,
      "learning_rate": 2.7e-05,
      "loss": 7.5183,
      "mean_token_accuracy": 0.09080160632729531,
      "num_tokens": 101455.0,
      "step": 55
    },
    {
      "entropy": 7.839016962051391,
      "epoch": 0.005298481102084069,
      "grad_norm": 0.8359375,
      "learning_rate": 2.95e-05,
      "loss": 7.5322,
      "mean_token_accuracy": 0.09154317900538445,
      "num_tokens": 110782.0,
      "step": 60
    },
    {
      "entropy": 7.82890944480896,
      "epoch": 0.005740021193924408,
      "grad_norm": 0.89453125,
      "learning_rate": 3.2e-05,
      "loss": 7.4277,
      "mean_token_accuracy": 0.09484038800001145,
      "num_tokens": 119241.0,
      "step": 65
    },
    {
      "entropy": 7.828151416778565,
      "epoch": 0.006181561285764747,
      "grad_norm": 0.94140625,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 7.5235,
      "mean_token_accuracy": 0.08991453871130943,
      "num_tokens": 127903.0,
      "step": 70
    },
    {
      "entropy": 7.8103076934814455,
      "epoch": 0.006623101377605086,
      "grad_norm": 0.90234375,
      "learning_rate": 3.7e-05,
      "loss": 7.5,
      "mean_token_accuracy": 0.09334810897707939,
      "num_tokens": 137370.0,
      "step": 75
    },
    {
      "entropy": 7.8186195373535154,
      "epoch": 0.007064641469445425,
      "grad_norm": 0.84375,
      "learning_rate": 3.95e-05,
      "loss": 7.4449,
      "mean_token_accuracy": 0.09487180560827255,
      "num_tokens": 146582.0,
      "step": 80
    },
    {
      "entropy": 7.844836282730102,
      "epoch": 0.007506181561285765,
      "grad_norm": 0.9296875,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 7.4263,
      "mean_token_accuracy": 0.10111527815461159,
      "num_tokens": 154933.0,
      "step": 85
    },
    {
      "entropy": 7.830266809463501,
      "epoch": 0.007947721653126103,
      "grad_norm": 0.875,
      "learning_rate": 4.45e-05,
      "loss": 7.5282,
      "mean_token_accuracy": 0.09698798581957817,
      "num_tokens": 165157.0,
      "step": 90
    },
    {
      "entropy": 7.830750465393066,
      "epoch": 0.008389261744966443,
      "grad_norm": 0.84375,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 7.4588,
      "mean_token_accuracy": 0.0967378407716751,
      "num_tokens": 174958.0,
      "step": 95
    },
    {
      "entropy": 7.780857229232788,
      "epoch": 0.008830801836806783,
      "grad_norm": 0.93359375,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 7.5012,
      "mean_token_accuracy": 0.09829336479306221,
      "num_tokens": 184256.0,
      "step": 100
    },
    {
      "entropy": 7.829332780838013,
      "epoch": 0.009272341928647121,
      "grad_norm": 0.8984375,
      "learning_rate": 5.2e-05,
      "loss": 7.4522,
      "mean_token_accuracy": 0.09847906306385994,
      "num_tokens": 192894.0,
      "step": 105
    },
    {
      "entropy": 7.851593732833862,
      "epoch": 0.00971388202048746,
      "grad_norm": 0.96875,
      "learning_rate": 5.45e-05,
      "loss": 7.4989,
      "mean_token_accuracy": 0.097539983689785,
      "num_tokens": 202675.0,
      "step": 110
    },
    {
      "entropy": 7.8246654033660885,
      "epoch": 0.010155422112327799,
      "grad_norm": 0.90625,
      "learning_rate": 5.7e-05,
      "loss": 7.5051,
      "mean_token_accuracy": 0.09736146405339241,
      "num_tokens": 212261.0,
      "step": 115
    },
    {
      "entropy": 7.855020570755005,
      "epoch": 0.010596962204168139,
      "grad_norm": 0.90234375,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 7.582,
      "mean_token_accuracy": 0.09346140772104264,
      "num_tokens": 222329.0,
      "step": 120
    },
    {
      "entropy": 7.888062286376953,
      "epoch": 0.011038502296008477,
      "grad_norm": 0.984375,
      "learning_rate": 6.2e-05,
      "loss": 7.5076,
      "mean_token_accuracy": 0.08999184891581535,
      "num_tokens": 231247.0,
      "step": 125
    },
    {
      "entropy": 7.855972003936768,
      "epoch": 0.011480042387848817,
      "grad_norm": 1.0859375,
      "learning_rate": 6.450000000000001e-05,
      "loss": 7.4591,
      "mean_token_accuracy": 0.10294558182358741,
      "num_tokens": 239978.0,
      "step": 130
    },
    {
      "entropy": 7.773235321044922,
      "epoch": 0.011921582479689156,
      "grad_norm": 0.9375,
      "learning_rate": 6.7e-05,
      "loss": 7.4769,
      "mean_token_accuracy": 0.09530732408165932,
      "num_tokens": 249735.0,
      "step": 135
    },
    {
      "entropy": 7.8279619216918945,
      "epoch": 0.012363122571529495,
      "grad_norm": 1.15625,
      "learning_rate": 6.950000000000001e-05,
      "loss": 7.5184,
      "mean_token_accuracy": 0.0965347282588482,
      "num_tokens": 259369.0,
      "step": 140
    },
    {
      "entropy": 7.942261266708374,
      "epoch": 0.012804662663369834,
      "grad_norm": 0.91015625,
      "learning_rate": 7.2e-05,
      "loss": 7.5286,
      "mean_token_accuracy": 0.09435783103108406,
      "num_tokens": 268645.0,
      "step": 145
    },
    {
      "entropy": 7.800592231750488,
      "epoch": 0.013246202755210172,
      "grad_norm": 0.96875,
      "learning_rate": 7.45e-05,
      "loss": 7.4194,
      "mean_token_accuracy": 0.10034767463803292,
      "num_tokens": 276667.0,
      "step": 150
    },
    {
      "entropy": 7.797780895233155,
      "epoch": 0.013687742847050512,
      "grad_norm": 0.87890625,
      "learning_rate": 7.7e-05,
      "loss": 7.4485,
      "mean_token_accuracy": 0.1016088955104351,
      "num_tokens": 286017.0,
      "step": 155
    },
    {
      "entropy": 7.838827800750733,
      "epoch": 0.01412928293889085,
      "grad_norm": 0.8984375,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.4773,
      "mean_token_accuracy": 0.0957870528101921,
      "num_tokens": 295631.0,
      "step": 160
    },
    {
      "entropy": 7.811876916885376,
      "epoch": 0.01457082303073119,
      "grad_norm": 0.91015625,
      "learning_rate": 8.2e-05,
      "loss": 7.4769,
      "mean_token_accuracy": 0.10056489408016205,
      "num_tokens": 304704.0,
      "step": 165
    },
    {
      "entropy": 7.836099576950073,
      "epoch": 0.01501236312257153,
      "grad_norm": 0.921875,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.506,
      "mean_token_accuracy": 0.09576743617653846,
      "num_tokens": 314195.0,
      "step": 170
    },
    {
      "entropy": 7.863115644454956,
      "epoch": 0.015453903214411868,
      "grad_norm": 1.078125,
      "learning_rate": 8.7e-05,
      "loss": 7.5172,
      "mean_token_accuracy": 0.09137798845767975,
      "num_tokens": 323379.0,
      "step": 175
    },
    {
      "entropy": 7.841438961029053,
      "epoch": 0.015895443306252206,
      "grad_norm": 0.9140625,
      "learning_rate": 8.95e-05,
      "loss": 7.4337,
      "mean_token_accuracy": 0.10475980639457702,
      "num_tokens": 332322.0,
      "step": 180
    },
    {
      "entropy": 7.761782026290893,
      "epoch": 0.016336983398092548,
      "grad_norm": 1.0,
      "learning_rate": 9.2e-05,
      "loss": 7.4066,
      "mean_token_accuracy": 0.10371800139546394,
      "num_tokens": 341735.0,
      "step": 185
    },
    {
      "entropy": 7.892005348205567,
      "epoch": 0.016778523489932886,
      "grad_norm": 0.96875,
      "learning_rate": 9.45e-05,
      "loss": 7.5071,
      "mean_token_accuracy": 0.09807236269116401,
      "num_tokens": 351209.0,
      "step": 190
    },
    {
      "entropy": 7.800058889389038,
      "epoch": 0.017220063581773224,
      "grad_norm": 0.91796875,
      "learning_rate": 9.7e-05,
      "loss": 7.4442,
      "mean_token_accuracy": 0.10099716112017632,
      "num_tokens": 360467.0,
      "step": 195
    },
    {
      "entropy": 7.8116179466247555,
      "epoch": 0.017661603673613566,
      "grad_norm": 0.8828125,
      "learning_rate": 9.95e-05,
      "loss": 7.4501,
      "mean_token_accuracy": 0.10495151057839394,
      "num_tokens": 370361.0,
      "step": 200
    },
    {
      "entropy": 7.931649780273437,
      "epoch": 0.018103143765453904,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000102,
      "loss": 7.496,
      "mean_token_accuracy": 0.0929229848086834,
      "num_tokens": 380366.0,
      "step": 205
    },
    {
      "entropy": 7.765449905395508,
      "epoch": 0.018544683857294242,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.4157,
      "mean_token_accuracy": 0.1013551875948906,
      "num_tokens": 390690.0,
      "step": 210
    },
    {
      "entropy": 7.802529335021973,
      "epoch": 0.01898622394913458,
      "grad_norm": 0.953125,
      "learning_rate": 0.000107,
      "loss": 7.4996,
      "mean_token_accuracy": 0.09649925082921981,
      "num_tokens": 400722.0,
      "step": 215
    },
    {
      "entropy": 7.8250895023345945,
      "epoch": 0.01942776404097492,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001095,
      "loss": 7.3837,
      "mean_token_accuracy": 0.1022213265299797,
      "num_tokens": 410223.0,
      "step": 220
    },
    {
      "entropy": 7.719676780700683,
      "epoch": 0.01986930413281526,
      "grad_norm": 0.953125,
      "learning_rate": 0.000112,
      "loss": 7.4551,
      "mean_token_accuracy": 0.10060659795999527,
      "num_tokens": 420214.0,
      "step": 225
    },
    {
      "entropy": 7.863075590133667,
      "epoch": 0.020310844224655598,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0001145,
      "loss": 7.4203,
      "mean_token_accuracy": 0.10097097754478454,
      "num_tokens": 429407.0,
      "step": 230
    },
    {
      "entropy": 7.792139530181885,
      "epoch": 0.02075238431649594,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.3518,
      "mean_token_accuracy": 0.0997156098484993,
      "num_tokens": 438403.0,
      "step": 235
    },
    {
      "entropy": 7.722495126724243,
      "epoch": 0.021193924408336277,
      "grad_norm": 1.0625,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.404,
      "mean_token_accuracy": 0.10603894293308258,
      "num_tokens": 447466.0,
      "step": 240
    },
    {
      "entropy": 7.889057874679565,
      "epoch": 0.021635464500176615,
      "grad_norm": 1.09375,
      "learning_rate": 0.000122,
      "loss": 7.5207,
      "mean_token_accuracy": 0.09140023961663246,
      "num_tokens": 457141.0,
      "step": 245
    },
    {
      "entropy": 7.798206424713134,
      "epoch": 0.022077004592016954,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0001245,
      "loss": 7.3654,
      "mean_token_accuracy": 0.10516906753182412,
      "num_tokens": 465708.0,
      "step": 250
    },
    {
      "entropy": 7.7851112365722654,
      "epoch": 0.022518544683857295,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000127,
      "loss": 7.4083,
      "mean_token_accuracy": 0.10015119016170501,
      "num_tokens": 475369.0,
      "step": 255
    },
    {
      "entropy": 7.747976684570313,
      "epoch": 0.022960084775697633,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0001295,
      "loss": 7.3678,
      "mean_token_accuracy": 0.10004653111100197,
      "num_tokens": 484249.0,
      "step": 260
    },
    {
      "entropy": 7.840033864974975,
      "epoch": 0.02340162486753797,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000132,
      "loss": 7.4932,
      "mean_token_accuracy": 0.09563242942094803,
      "num_tokens": 493303.0,
      "step": 265
    },
    {
      "entropy": 7.812006616592408,
      "epoch": 0.023843164959378313,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.449,
      "mean_token_accuracy": 0.10385405197739601,
      "num_tokens": 501503.0,
      "step": 270
    },
    {
      "entropy": 7.883935546875,
      "epoch": 0.02428470505121865,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.3669,
      "mean_token_accuracy": 0.10403206944465637,
      "num_tokens": 509661.0,
      "step": 275
    },
    {
      "entropy": 7.698855304718018,
      "epoch": 0.02472624514305899,
      "grad_norm": 1.0,
      "learning_rate": 0.0001395,
      "loss": 7.3695,
      "mean_token_accuracy": 0.10018852427601814,
      "num_tokens": 519464.0,
      "step": 280
    },
    {
      "entropy": 7.810435962677002,
      "epoch": 0.025167785234899327,
      "grad_norm": 1.21875,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.3442,
      "mean_token_accuracy": 0.10638531595468521,
      "num_tokens": 527968.0,
      "step": 285
    },
    {
      "entropy": 7.726656579971314,
      "epoch": 0.02560932532673967,
      "grad_norm": 1.015625,
      "learning_rate": 0.0001445,
      "loss": 7.4085,
      "mean_token_accuracy": 0.10700022876262665,
      "num_tokens": 537234.0,
      "step": 290
    },
    {
      "entropy": 7.8001549243927,
      "epoch": 0.026050865418580007,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000147,
      "loss": 7.3728,
      "mean_token_accuracy": 0.10392982810735703,
      "num_tokens": 546398.0,
      "step": 295
    },
    {
      "entropy": 7.735007047653198,
      "epoch": 0.026492405510420345,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0001495,
      "loss": 7.3857,
      "mean_token_accuracy": 0.10330862030386925,
      "num_tokens": 555362.0,
      "step": 300
    },
    {
      "entropy": 7.846773290634156,
      "epoch": 0.026933945602260687,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000152,
      "loss": 7.4235,
      "mean_token_accuracy": 0.09694105237722397,
      "num_tokens": 564575.0,
      "step": 305
    },
    {
      "entropy": 7.824136352539062,
      "epoch": 0.027375485694101025,
      "grad_norm": 1.09375,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.4528,
      "mean_token_accuracy": 0.10274629592895508,
      "num_tokens": 573915.0,
      "step": 310
    },
    {
      "entropy": 7.747010231018066,
      "epoch": 0.027817025785941363,
      "grad_norm": 1.109375,
      "learning_rate": 0.000157,
      "loss": 7.3458,
      "mean_token_accuracy": 0.10164012908935546,
      "num_tokens": 583216.0,
      "step": 315
    },
    {
      "entropy": 7.775792646408081,
      "epoch": 0.0282585658777817,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001595,
      "loss": 7.3946,
      "mean_token_accuracy": 0.10511558204889297,
      "num_tokens": 591955.0,
      "step": 320
    },
    {
      "entropy": 7.6797911643981935,
      "epoch": 0.028700105969622042,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000162,
      "loss": 7.3672,
      "mean_token_accuracy": 0.103659038990736,
      "num_tokens": 600999.0,
      "step": 325
    },
    {
      "entropy": 7.804133558273316,
      "epoch": 0.02914164606146238,
      "grad_norm": 1.0,
      "learning_rate": 0.00016450000000000001,
      "loss": 7.2868,
      "mean_token_accuracy": 0.11126004680991172,
      "num_tokens": 609478.0,
      "step": 330
    },
    {
      "entropy": 7.633207130432129,
      "epoch": 0.02958318615330272,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00016700000000000002,
      "loss": 7.3672,
      "mean_token_accuracy": 0.10404063463211059,
      "num_tokens": 618348.0,
      "step": 335
    },
    {
      "entropy": 7.855530452728272,
      "epoch": 0.03002472624514306,
      "grad_norm": 1.25,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.4232,
      "mean_token_accuracy": 0.10137763321399688,
      "num_tokens": 628548.0,
      "step": 340
    },
    {
      "entropy": 7.745834493637085,
      "epoch": 0.0304662663369834,
      "grad_norm": 1.09375,
      "learning_rate": 0.00017199999999999998,
      "loss": 7.3651,
      "mean_token_accuracy": 0.10630879923701286,
      "num_tokens": 637489.0,
      "step": 345
    },
    {
      "entropy": 7.684433507919311,
      "epoch": 0.030907806428823736,
      "grad_norm": 1.125,
      "learning_rate": 0.00017449999999999999,
      "loss": 7.3503,
      "mean_token_accuracy": 0.10281694158911706,
      "num_tokens": 646715.0,
      "step": 350
    },
    {
      "entropy": 7.803950309753418,
      "epoch": 0.031349346520664075,
      "grad_norm": 1.015625,
      "learning_rate": 0.000177,
      "loss": 7.3499,
      "mean_token_accuracy": 0.10527043938636779,
      "num_tokens": 656858.0,
      "step": 355
    },
    {
      "entropy": 7.707482957839966,
      "epoch": 0.03179088661250441,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001795,
      "loss": 7.3208,
      "mean_token_accuracy": 0.10721890702843666,
      "num_tokens": 665968.0,
      "step": 360
    },
    {
      "entropy": 7.764640760421753,
      "epoch": 0.03223242670434476,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000182,
      "loss": 7.315,
      "mean_token_accuracy": 0.10370069816708564,
      "num_tokens": 674295.0,
      "step": 365
    },
    {
      "entropy": 7.695739364624023,
      "epoch": 0.032673966796185096,
      "grad_norm": 1.0,
      "learning_rate": 0.0001845,
      "loss": 7.2613,
      "mean_token_accuracy": 0.10737213939428329,
      "num_tokens": 683559.0,
      "step": 370
    },
    {
      "entropy": 7.558203125,
      "epoch": 0.033115506888025434,
      "grad_norm": 1.328125,
      "learning_rate": 0.000187,
      "loss": 7.2997,
      "mean_token_accuracy": 0.10474916398525239,
      "num_tokens": 692402.0,
      "step": 375
    },
    {
      "entropy": 7.85990538597107,
      "epoch": 0.03355704697986577,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001895,
      "loss": 7.4051,
      "mean_token_accuracy": 0.09768549725413322,
      "num_tokens": 702052.0,
      "step": 380
    },
    {
      "entropy": 7.703890705108643,
      "epoch": 0.03399858707170611,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000192,
      "loss": 7.3332,
      "mean_token_accuracy": 0.10592006072402001,
      "num_tokens": 711926.0,
      "step": 385
    },
    {
      "entropy": 7.764595699310303,
      "epoch": 0.03444012716354645,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0001945,
      "loss": 7.2823,
      "mean_token_accuracy": 0.09970764666795731,
      "num_tokens": 720948.0,
      "step": 390
    },
    {
      "entropy": 7.651279592514038,
      "epoch": 0.034881667255386786,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00019700000000000002,
      "loss": 7.3228,
      "mean_token_accuracy": 0.11015688627958298,
      "num_tokens": 729611.0,
      "step": 395
    },
    {
      "entropy": 7.79816370010376,
      "epoch": 0.03532320734722713,
      "grad_norm": 1.171875,
      "learning_rate": 0.00019950000000000002,
      "loss": 7.3073,
      "mean_token_accuracy": 0.09714326411485671,
      "num_tokens": 738433.0,
      "step": 400
    },
    {
      "entropy": 7.687422800064087,
      "epoch": 0.03576474743906747,
      "grad_norm": 1.078125,
      "learning_rate": 0.000202,
      "loss": 7.3504,
      "mean_token_accuracy": 0.103574138879776,
      "num_tokens": 747310.0,
      "step": 405
    },
    {
      "entropy": 7.690566205978394,
      "epoch": 0.03620628753090781,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00020449999999999998,
      "loss": 7.3092,
      "mean_token_accuracy": 0.10628200694918633,
      "num_tokens": 756362.0,
      "step": 410
    },
    {
      "entropy": 7.803203916549682,
      "epoch": 0.036647827622748146,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000207,
      "loss": 7.3251,
      "mean_token_accuracy": 0.09963449612259864,
      "num_tokens": 764978.0,
      "step": 415
    },
    {
      "entropy": 7.557148265838623,
      "epoch": 0.037089367714588484,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002095,
      "loss": 7.299,
      "mean_token_accuracy": 0.10668339729309081,
      "num_tokens": 774058.0,
      "step": 420
    },
    {
      "entropy": 7.685351037979126,
      "epoch": 0.03753090780642882,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000212,
      "loss": 7.2881,
      "mean_token_accuracy": 0.10115964189171792,
      "num_tokens": 783332.0,
      "step": 425
    },
    {
      "entropy": 7.726617670059204,
      "epoch": 0.03797244789826916,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002145,
      "loss": 7.3501,
      "mean_token_accuracy": 0.10238557234406472,
      "num_tokens": 792481.0,
      "step": 430
    },
    {
      "entropy": 7.598264503479004,
      "epoch": 0.038413987990109505,
      "grad_norm": 1.203125,
      "learning_rate": 0.00021700000000000002,
      "loss": 7.2437,
      "mean_token_accuracy": 0.11149554774165153,
      "num_tokens": 801396.0,
      "step": 435
    },
    {
      "entropy": 7.771391916275024,
      "epoch": 0.03885552808194984,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002195,
      "loss": 7.3027,
      "mean_token_accuracy": 0.10071089640259742,
      "num_tokens": 809903.0,
      "step": 440
    },
    {
      "entropy": 7.631546688079834,
      "epoch": 0.03929706817379018,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000222,
      "loss": 7.2639,
      "mean_token_accuracy": 0.10943293869495392,
      "num_tokens": 819144.0,
      "step": 445
    },
    {
      "entropy": 7.66679310798645,
      "epoch": 0.03973860826563052,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002245,
      "loss": 7.2976,
      "mean_token_accuracy": 0.10613659396767616,
      "num_tokens": 828881.0,
      "step": 450
    },
    {
      "entropy": 7.639922952651977,
      "epoch": 0.04018014835747086,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00022700000000000002,
      "loss": 7.224,
      "mean_token_accuracy": 0.09946361035108567,
      "num_tokens": 837588.0,
      "step": 455
    },
    {
      "entropy": 7.582935905456543,
      "epoch": 0.040621688449311195,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00022950000000000002,
      "loss": 7.2197,
      "mean_token_accuracy": 0.1077489398419857,
      "num_tokens": 847002.0,
      "step": 460
    },
    {
      "entropy": 7.601295232772827,
      "epoch": 0.041063228541151534,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00023200000000000003,
      "loss": 7.2374,
      "mean_token_accuracy": 0.10669372826814652,
      "num_tokens": 855791.0,
      "step": 465
    },
    {
      "entropy": 7.683427000045777,
      "epoch": 0.04150476863299188,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023449999999999998,
      "loss": 7.2737,
      "mean_token_accuracy": 0.10692826434969901,
      "num_tokens": 865392.0,
      "step": 470
    },
    {
      "entropy": 7.632460880279541,
      "epoch": 0.04194630872483222,
      "grad_norm": 1.140625,
      "learning_rate": 0.000237,
      "loss": 7.3431,
      "mean_token_accuracy": 0.10185316428542138,
      "num_tokens": 874807.0,
      "step": 475
    },
    {
      "entropy": 7.7053466796875,
      "epoch": 0.042387848816672555,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002395,
      "loss": 7.3151,
      "mean_token_accuracy": 0.107633588463068,
      "num_tokens": 883928.0,
      "step": 480
    },
    {
      "entropy": 7.573175430297852,
      "epoch": 0.04282938890851289,
      "grad_norm": 1.046875,
      "learning_rate": 0.000242,
      "loss": 7.2153,
      "mean_token_accuracy": 0.10783804580569267,
      "num_tokens": 893047.0,
      "step": 485
    },
    {
      "entropy": 7.575657176971435,
      "epoch": 0.04327092900035323,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002445,
      "loss": 7.2168,
      "mean_token_accuracy": 0.1058282122015953,
      "num_tokens": 901645.0,
      "step": 490
    },
    {
      "entropy": 7.620817947387695,
      "epoch": 0.04371246909219357,
      "grad_norm": 1.234375,
      "learning_rate": 0.000247,
      "loss": 7.2325,
      "mean_token_accuracy": 0.10362469628453255,
      "num_tokens": 911169.0,
      "step": 495
    },
    {
      "entropy": 7.675788545608521,
      "epoch": 0.04415400918403391,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002495,
      "loss": 7.239,
      "mean_token_accuracy": 0.09903389513492585,
      "num_tokens": 921382.0,
      "step": 500
    },
    {
      "entropy": 7.509087562561035,
      "epoch": 0.04459554927587425,
      "grad_norm": 1.15625,
      "learning_rate": 0.000252,
      "loss": 7.1784,
      "mean_token_accuracy": 0.10705869123339654,
      "num_tokens": 930409.0,
      "step": 505
    },
    {
      "entropy": 7.563630390167236,
      "epoch": 0.04503708936771459,
      "grad_norm": 1.125,
      "learning_rate": 0.0002545,
      "loss": 7.1437,
      "mean_token_accuracy": 0.11103534623980522,
      "num_tokens": 939049.0,
      "step": 510
    },
    {
      "entropy": 7.656319618225098,
      "epoch": 0.04547862945955493,
      "grad_norm": 1.140625,
      "learning_rate": 0.000257,
      "loss": 7.1669,
      "mean_token_accuracy": 0.10674394220113755,
      "num_tokens": 947575.0,
      "step": 515
    },
    {
      "entropy": 7.617980527877807,
      "epoch": 0.045920169551395267,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002595,
      "loss": 7.3263,
      "mean_token_accuracy": 0.10124524608254433,
      "num_tokens": 957848.0,
      "step": 520
    },
    {
      "entropy": 7.594845724105835,
      "epoch": 0.046361709643235605,
      "grad_norm": 1.21875,
      "learning_rate": 0.000262,
      "loss": 7.1638,
      "mean_token_accuracy": 0.1080893911421299,
      "num_tokens": 966521.0,
      "step": 525
    },
    {
      "entropy": 7.610377073287964,
      "epoch": 0.04680324973507594,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00026450000000000003,
      "loss": 7.2181,
      "mean_token_accuracy": 0.10909054726362229,
      "num_tokens": 975827.0,
      "step": 530
    },
    {
      "entropy": 7.499869298934937,
      "epoch": 0.04724478982691628,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00026700000000000004,
      "loss": 7.1793,
      "mean_token_accuracy": 0.11137327626347542,
      "num_tokens": 985292.0,
      "step": 535
    },
    {
      "entropy": 7.5960760593414305,
      "epoch": 0.047686329918756626,
      "grad_norm": 1.15625,
      "learning_rate": 0.00026950000000000005,
      "loss": 7.154,
      "mean_token_accuracy": 0.10950936675071717,
      "num_tokens": 994791.0,
      "step": 540
    },
    {
      "entropy": 7.469157934188843,
      "epoch": 0.048127870010596964,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027200000000000005,
      "loss": 7.1317,
      "mean_token_accuracy": 0.11670516580343246,
      "num_tokens": 1003700.0,
      "step": 545
    },
    {
      "entropy": 7.523545932769776,
      "epoch": 0.0485694101024373,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002745,
      "loss": 7.1673,
      "mean_token_accuracy": 0.11030299812555314,
      "num_tokens": 1012682.0,
      "step": 550
    },
    {
      "entropy": 7.583120203018188,
      "epoch": 0.04901095019427764,
      "grad_norm": 1.453125,
      "learning_rate": 0.000277,
      "loss": 7.1875,
      "mean_token_accuracy": 0.1056957058608532,
      "num_tokens": 1021018.0,
      "step": 555
    },
    {
      "entropy": 7.558827114105225,
      "epoch": 0.04945249028611798,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002795,
      "loss": 7.0584,
      "mean_token_accuracy": 0.1091762512922287,
      "num_tokens": 1029744.0,
      "step": 560
    },
    {
      "entropy": 7.384686899185181,
      "epoch": 0.049894030377958316,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028199999999999997,
      "loss": 7.1435,
      "mean_token_accuracy": 0.11267905756831169,
      "num_tokens": 1038805.0,
      "step": 565
    },
    {
      "entropy": 7.607424783706665,
      "epoch": 0.050335570469798654,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002845,
      "loss": 7.1996,
      "mean_token_accuracy": 0.10786581486463546,
      "num_tokens": 1047656.0,
      "step": 570
    },
    {
      "entropy": 7.606092405319214,
      "epoch": 0.050777110561639,
      "grad_norm": 1.15625,
      "learning_rate": 0.000287,
      "loss": 7.1792,
      "mean_token_accuracy": 0.11288331896066665,
      "num_tokens": 1056598.0,
      "step": 575
    },
    {
      "entropy": 7.5449169158935545,
      "epoch": 0.05121865065347934,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002895,
      "loss": 7.1883,
      "mean_token_accuracy": 0.10918098837137222,
      "num_tokens": 1065226.0,
      "step": 580
    },
    {
      "entropy": 7.479248285293579,
      "epoch": 0.051660190745319676,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000292,
      "loss": 7.1217,
      "mean_token_accuracy": 0.11170855611562729,
      "num_tokens": 1074661.0,
      "step": 585
    },
    {
      "entropy": 7.5761857509613035,
      "epoch": 0.052101730837160014,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002945,
      "loss": 7.2093,
      "mean_token_accuracy": 0.1055817686021328,
      "num_tokens": 1083921.0,
      "step": 590
    },
    {
      "entropy": 7.475902557373047,
      "epoch": 0.05254327092900035,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000297,
      "loss": 7.1024,
      "mean_token_accuracy": 0.12069160491228104,
      "num_tokens": 1093399.0,
      "step": 595
    },
    {
      "entropy": 7.477249240875244,
      "epoch": 0.05298481102084069,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002995,
      "loss": 7.1806,
      "mean_token_accuracy": 0.1091334879398346,
      "num_tokens": 1104065.0,
      "step": 600
    },
    {
      "entropy": 7.454396152496338,
      "epoch": 0.05342635111268103,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000302,
      "loss": 7.0759,
      "mean_token_accuracy": 0.11466379538178444,
      "num_tokens": 1112995.0,
      "step": 605
    },
    {
      "entropy": 7.4687096118927006,
      "epoch": 0.05386789120452137,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003045,
      "loss": 7.1202,
      "mean_token_accuracy": 0.10947456136345864,
      "num_tokens": 1121637.0,
      "step": 610
    },
    {
      "entropy": 7.428070116043091,
      "epoch": 0.05430943129636171,
      "grad_norm": 1.140625,
      "learning_rate": 0.000307,
      "loss": 7.061,
      "mean_token_accuracy": 0.1107931189239025,
      "num_tokens": 1131166.0,
      "step": 615
    },
    {
      "entropy": 7.479091644287109,
      "epoch": 0.05475097138820205,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003095,
      "loss": 7.0598,
      "mean_token_accuracy": 0.11267657652497291,
      "num_tokens": 1140888.0,
      "step": 620
    },
    {
      "entropy": 7.516183996200562,
      "epoch": 0.05519251148004239,
      "grad_norm": 1.140625,
      "learning_rate": 0.000312,
      "loss": 7.2209,
      "mean_token_accuracy": 0.10797442942857742,
      "num_tokens": 1150278.0,
      "step": 625
    },
    {
      "entropy": 7.450537157058716,
      "epoch": 0.055634051571882726,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003145,
      "loss": 7.1601,
      "mean_token_accuracy": 0.10592598244547843,
      "num_tokens": 1159348.0,
      "step": 630
    },
    {
      "entropy": 7.588924980163574,
      "epoch": 0.056075591663723064,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000317,
      "loss": 7.1788,
      "mean_token_accuracy": 0.10933139771223069,
      "num_tokens": 1168883.0,
      "step": 635
    },
    {
      "entropy": 7.51312780380249,
      "epoch": 0.0565171317555634,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003195,
      "loss": 7.0915,
      "mean_token_accuracy": 0.1159443736076355,
      "num_tokens": 1178572.0,
      "step": 640
    },
    {
      "entropy": 7.437539958953858,
      "epoch": 0.05695867184740375,
      "grad_norm": 1.0625,
      "learning_rate": 0.000322,
      "loss": 7.17,
      "mean_token_accuracy": 0.10953533723950386,
      "num_tokens": 1188909.0,
      "step": 645
    },
    {
      "entropy": 7.465111255645752,
      "epoch": 0.057400211939244085,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032450000000000003,
      "loss": 7.0672,
      "mean_token_accuracy": 0.11020829901099205,
      "num_tokens": 1197705.0,
      "step": 650
    },
    {
      "entropy": 7.397136974334717,
      "epoch": 0.05784175203108442,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.9218,
      "mean_token_accuracy": 0.11375543102622032,
      "num_tokens": 1206351.0,
      "step": 655
    },
    {
      "entropy": 7.38561692237854,
      "epoch": 0.05828329212292476,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032950000000000004,
      "loss": 7.0438,
      "mean_token_accuracy": 0.11089925691485406,
      "num_tokens": 1214984.0,
      "step": 660
    },
    {
      "entropy": 7.4726481437683105,
      "epoch": 0.0587248322147651,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033200000000000005,
      "loss": 7.1247,
      "mean_token_accuracy": 0.11059069335460663,
      "num_tokens": 1224485.0,
      "step": 665
    },
    {
      "entropy": 7.461330795288086,
      "epoch": 0.05916637230660544,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033450000000000005,
      "loss": 7.1386,
      "mean_token_accuracy": 0.10253455564379692,
      "num_tokens": 1233560.0,
      "step": 670
    },
    {
      "entropy": 7.473884534835816,
      "epoch": 0.05960791239844578,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000337,
      "loss": 7.0729,
      "mean_token_accuracy": 0.10994852259755135,
      "num_tokens": 1242628.0,
      "step": 675
    },
    {
      "entropy": 7.392604112625122,
      "epoch": 0.06004945249028612,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003395,
      "loss": 6.9809,
      "mean_token_accuracy": 0.1146723486483097,
      "num_tokens": 1251004.0,
      "step": 680
    },
    {
      "entropy": 7.428996276855469,
      "epoch": 0.06049099258212646,
      "grad_norm": 1.03125,
      "learning_rate": 0.000342,
      "loss": 7.1634,
      "mean_token_accuracy": 0.11115597635507583,
      "num_tokens": 1260344.0,
      "step": 685
    },
    {
      "entropy": 7.401163053512573,
      "epoch": 0.0609325326739668,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034449999999999997,
      "loss": 7.0696,
      "mean_token_accuracy": 0.11442951783537865,
      "num_tokens": 1269988.0,
      "step": 690
    },
    {
      "entropy": 7.402554750442505,
      "epoch": 0.061374072765807135,
      "grad_norm": 1.21875,
      "learning_rate": 0.000347,
      "loss": 7.0428,
      "mean_token_accuracy": 0.11391758620738983,
      "num_tokens": 1280912.0,
      "step": 695
    },
    {
      "entropy": 7.406301021575928,
      "epoch": 0.06181561285764747,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003495,
      "loss": 7.0467,
      "mean_token_accuracy": 0.11682257801294327,
      "num_tokens": 1289684.0,
      "step": 700
    },
    {
      "entropy": 7.3641218662261965,
      "epoch": 0.06225715294948781,
      "grad_norm": 1.265625,
      "learning_rate": 0.000352,
      "loss": 7.0803,
      "mean_token_accuracy": 0.10842933505773544,
      "num_tokens": 1298853.0,
      "step": 705
    },
    {
      "entropy": 7.3719359874725345,
      "epoch": 0.06269869304132815,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003545,
      "loss": 7.0386,
      "mean_token_accuracy": 0.11791565641760826,
      "num_tokens": 1309112.0,
      "step": 710
    },
    {
      "entropy": 7.369606351852417,
      "epoch": 0.0631402331331685,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000357,
      "loss": 7.0135,
      "mean_token_accuracy": 0.11014628857374191,
      "num_tokens": 1319064.0,
      "step": 715
    },
    {
      "entropy": 7.296480321884156,
      "epoch": 0.06358177322500883,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003595,
      "loss": 6.9938,
      "mean_token_accuracy": 0.11697245836257934,
      "num_tokens": 1327889.0,
      "step": 720
    },
    {
      "entropy": 7.403489398956299,
      "epoch": 0.06402331331684917,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000362,
      "loss": 7.0083,
      "mean_token_accuracy": 0.1104284293949604,
      "num_tokens": 1337241.0,
      "step": 725
    },
    {
      "entropy": 7.347339487075805,
      "epoch": 0.06446485340868952,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003645,
      "loss": 7.0718,
      "mean_token_accuracy": 0.11292454376816749,
      "num_tokens": 1346527.0,
      "step": 730
    },
    {
      "entropy": 7.347748756408691,
      "epoch": 0.06490639350052985,
      "grad_norm": 1.09375,
      "learning_rate": 0.000367,
      "loss": 7.0639,
      "mean_token_accuracy": 0.11358218491077424,
      "num_tokens": 1355677.0,
      "step": 735
    },
    {
      "entropy": 7.3271379470825195,
      "epoch": 0.06534793359237019,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003695,
      "loss": 7.0468,
      "mean_token_accuracy": 0.1157820999622345,
      "num_tokens": 1364874.0,
      "step": 740
    },
    {
      "entropy": 7.400119829177856,
      "epoch": 0.06578947368421052,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000372,
      "loss": 6.9839,
      "mean_token_accuracy": 0.12082682698965072,
      "num_tokens": 1373717.0,
      "step": 745
    },
    {
      "entropy": 7.417098903656006,
      "epoch": 0.06623101377605087,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003745,
      "loss": 6.9526,
      "mean_token_accuracy": 0.11417729035019875,
      "num_tokens": 1382767.0,
      "step": 750
    },
    {
      "entropy": 7.194312381744385,
      "epoch": 0.0666725538678912,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000377,
      "loss": 6.9799,
      "mean_token_accuracy": 0.11586436182260514,
      "num_tokens": 1391190.0,
      "step": 755
    },
    {
      "entropy": 7.339599084854126,
      "epoch": 0.06711409395973154,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003795,
      "loss": 6.9817,
      "mean_token_accuracy": 0.1170587882399559,
      "num_tokens": 1400722.0,
      "step": 760
    },
    {
      "entropy": 7.400132036209106,
      "epoch": 0.06755563405157189,
      "grad_norm": 1.140625,
      "learning_rate": 0.000382,
      "loss": 7.0091,
      "mean_token_accuracy": 0.10940207540988922,
      "num_tokens": 1409328.0,
      "step": 765
    },
    {
      "entropy": 7.115442705154419,
      "epoch": 0.06799717414341222,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003845,
      "loss": 6.8837,
      "mean_token_accuracy": 0.12403405010700226,
      "num_tokens": 1418504.0,
      "step": 770
    },
    {
      "entropy": 7.437279510498047,
      "epoch": 0.06843871423525257,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038700000000000003,
      "loss": 7.0359,
      "mean_token_accuracy": 0.10912163555622101,
      "num_tokens": 1427690.0,
      "step": 775
    },
    {
      "entropy": 7.26606240272522,
      "epoch": 0.0688802543270929,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038950000000000003,
      "loss": 7.0132,
      "mean_token_accuracy": 0.11617867797613143,
      "num_tokens": 1436798.0,
      "step": 780
    },
    {
      "entropy": 7.26966495513916,
      "epoch": 0.06932179441893324,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039200000000000004,
      "loss": 7.01,
      "mean_token_accuracy": 0.11528639197349548,
      "num_tokens": 1446357.0,
      "step": 785
    },
    {
      "entropy": 7.302881193161011,
      "epoch": 0.06976333451077357,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.9315,
      "mean_token_accuracy": 0.11601672917604447,
      "num_tokens": 1455998.0,
      "step": 790
    },
    {
      "entropy": 7.307020282745361,
      "epoch": 0.07020487460261392,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.9813,
      "mean_token_accuracy": 0.11031874865293503,
      "num_tokens": 1465237.0,
      "step": 795
    },
    {
      "entropy": 7.243889856338501,
      "epoch": 0.07064641469445426,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003995,
      "loss": 6.9248,
      "mean_token_accuracy": 0.12101015448570251,
      "num_tokens": 1474363.0,
      "step": 800
    },
    {
      "entropy": 7.248915958404541,
      "epoch": 0.0710879547862946,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000402,
      "loss": 6.9906,
      "mean_token_accuracy": 0.12257542759180069,
      "num_tokens": 1483379.0,
      "step": 805
    },
    {
      "entropy": 7.34154782295227,
      "epoch": 0.07152949487813494,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004045,
      "loss": 6.9828,
      "mean_token_accuracy": 0.11638684496283531,
      "num_tokens": 1492507.0,
      "step": 810
    },
    {
      "entropy": 7.212753200531006,
      "epoch": 0.07197103496997527,
      "grad_norm": 1.25,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.8604,
      "mean_token_accuracy": 0.12233690097928047,
      "num_tokens": 1500888.0,
      "step": 815
    },
    {
      "entropy": 7.181662464141846,
      "epoch": 0.07241257506181561,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004095,
      "loss": 6.9234,
      "mean_token_accuracy": 0.11673298925161361,
      "num_tokens": 1510310.0,
      "step": 820
    },
    {
      "entropy": 7.263974905014038,
      "epoch": 0.07285411515365595,
      "grad_norm": 1.171875,
      "learning_rate": 0.000412,
      "loss": 6.8977,
      "mean_token_accuracy": 0.12379544451832772,
      "num_tokens": 1519427.0,
      "step": 825
    },
    {
      "entropy": 7.25563235282898,
      "epoch": 0.07329565524549629,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004145,
      "loss": 6.9332,
      "mean_token_accuracy": 0.11141304299235344,
      "num_tokens": 1529456.0,
      "step": 830
    },
    {
      "entropy": 7.141549444198608,
      "epoch": 0.07373719533733664,
      "grad_norm": 1.15625,
      "learning_rate": 0.000417,
      "loss": 6.8617,
      "mean_token_accuracy": 0.11971206367015838,
      "num_tokens": 1537695.0,
      "step": 835
    },
    {
      "entropy": 7.242198467254639,
      "epoch": 0.07417873542917697,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004195,
      "loss": 7.0075,
      "mean_token_accuracy": 0.11687812730669975,
      "num_tokens": 1547511.0,
      "step": 840
    },
    {
      "entropy": 7.309858417510986,
      "epoch": 0.07462027552101731,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000422,
      "loss": 6.9839,
      "mean_token_accuracy": 0.11879770755767823,
      "num_tokens": 1557035.0,
      "step": 845
    },
    {
      "entropy": 7.211091947555542,
      "epoch": 0.07506181561285764,
      "grad_norm": 1.0,
      "learning_rate": 0.0004245,
      "loss": 6.9346,
      "mean_token_accuracy": 0.12483425140380859,
      "num_tokens": 1566773.0,
      "step": 850
    },
    {
      "entropy": 7.208451414108277,
      "epoch": 0.07550335570469799,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000427,
      "loss": 6.927,
      "mean_token_accuracy": 0.11267467886209488,
      "num_tokens": 1576873.0,
      "step": 855
    },
    {
      "entropy": 7.186734676361084,
      "epoch": 0.07594489579653832,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004295,
      "loss": 6.9684,
      "mean_token_accuracy": 0.11325102299451828,
      "num_tokens": 1586170.0,
      "step": 860
    },
    {
      "entropy": 7.32284631729126,
      "epoch": 0.07638643588837866,
      "grad_norm": 1.328125,
      "learning_rate": 0.000432,
      "loss": 7.0153,
      "mean_token_accuracy": 0.11636607497930526,
      "num_tokens": 1596054.0,
      "step": 865
    },
    {
      "entropy": 7.200122261047364,
      "epoch": 0.07682797598021901,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004345,
      "loss": 6.8751,
      "mean_token_accuracy": 0.1230311393737793,
      "num_tokens": 1604544.0,
      "step": 870
    },
    {
      "entropy": 7.244127464294434,
      "epoch": 0.07726951607205934,
      "grad_norm": 1.171875,
      "learning_rate": 0.000437,
      "loss": 6.9023,
      "mean_token_accuracy": 0.12475202232599258,
      "num_tokens": 1614580.0,
      "step": 875
    },
    {
      "entropy": 7.1907594203948975,
      "epoch": 0.07771105616389969,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004395,
      "loss": 6.9948,
      "mean_token_accuracy": 0.115571129322052,
      "num_tokens": 1624701.0,
      "step": 880
    },
    {
      "entropy": 7.136907148361206,
      "epoch": 0.07815259625574002,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000442,
      "loss": 6.9193,
      "mean_token_accuracy": 0.11373112797737121,
      "num_tokens": 1634085.0,
      "step": 885
    },
    {
      "entropy": 7.195565986633301,
      "epoch": 0.07859413634758036,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004445,
      "loss": 6.8968,
      "mean_token_accuracy": 0.11859833672642708,
      "num_tokens": 1643190.0,
      "step": 890
    },
    {
      "entropy": 7.176092052459717,
      "epoch": 0.0790356764394207,
      "grad_norm": 1.03125,
      "learning_rate": 0.000447,
      "loss": 6.9171,
      "mean_token_accuracy": 0.11816032081842423,
      "num_tokens": 1652705.0,
      "step": 895
    },
    {
      "entropy": 7.225932455062866,
      "epoch": 0.07947721653126104,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.9033,
      "mean_token_accuracy": 0.11247398778796196,
      "num_tokens": 1662210.0,
      "step": 900
    },
    {
      "entropy": 7.222397232055664,
      "epoch": 0.07991875662310138,
      "grad_norm": 0.984375,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.931,
      "mean_token_accuracy": 0.11510396897792816,
      "num_tokens": 1671893.0,
      "step": 905
    },
    {
      "entropy": 7.168801355361938,
      "epoch": 0.08036029671494171,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.8842,
      "mean_token_accuracy": 0.12047431021928787,
      "num_tokens": 1681217.0,
      "step": 910
    },
    {
      "entropy": 7.189701175689697,
      "epoch": 0.08080183680678206,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.9145,
      "mean_token_accuracy": 0.11493082717061043,
      "num_tokens": 1690447.0,
      "step": 915
    },
    {
      "entropy": 7.148307180404663,
      "epoch": 0.08124337689862239,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045950000000000006,
      "loss": 7.001,
      "mean_token_accuracy": 0.10863983184099198,
      "num_tokens": 1700355.0,
      "step": 920
    },
    {
      "entropy": 7.2164154052734375,
      "epoch": 0.08168491699046274,
      "grad_norm": 1.015625,
      "learning_rate": 0.000462,
      "loss": 6.906,
      "mean_token_accuracy": 0.11343243196606637,
      "num_tokens": 1709449.0,
      "step": 925
    },
    {
      "entropy": 7.191054010391236,
      "epoch": 0.08212645708230307,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004645,
      "loss": 6.9311,
      "mean_token_accuracy": 0.11408023163676262,
      "num_tokens": 1718838.0,
      "step": 930
    },
    {
      "entropy": 7.111620044708252,
      "epoch": 0.08256799717414341,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000467,
      "loss": 6.8844,
      "mean_token_accuracy": 0.12342581376433373,
      "num_tokens": 1728594.0,
      "step": 935
    },
    {
      "entropy": 7.233243608474732,
      "epoch": 0.08300953726598376,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004695,
      "loss": 6.9305,
      "mean_token_accuracy": 0.12092314288020134,
      "num_tokens": 1738814.0,
      "step": 940
    },
    {
      "entropy": 7.168387937545776,
      "epoch": 0.08345107735782409,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000472,
      "loss": 6.8546,
      "mean_token_accuracy": 0.12510684356093407,
      "num_tokens": 1747644.0,
      "step": 945
    },
    {
      "entropy": 7.201950550079346,
      "epoch": 0.08389261744966443,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004745,
      "loss": 6.9079,
      "mean_token_accuracy": 0.11906211525201797,
      "num_tokens": 1757489.0,
      "step": 950
    },
    {
      "entropy": 7.130580997467041,
      "epoch": 0.08433415754150476,
      "grad_norm": 1.03125,
      "learning_rate": 0.000477,
      "loss": 6.8296,
      "mean_token_accuracy": 0.12410564720630646,
      "num_tokens": 1767546.0,
      "step": 955
    },
    {
      "entropy": 7.093601322174072,
      "epoch": 0.08477569763334511,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004795,
      "loss": 6.8928,
      "mean_token_accuracy": 0.10743151605129242,
      "num_tokens": 1776035.0,
      "step": 960
    },
    {
      "entropy": 7.145628499984741,
      "epoch": 0.08521723772518544,
      "grad_norm": 0.953125,
      "learning_rate": 0.000482,
      "loss": 6.953,
      "mean_token_accuracy": 0.11847842559218406,
      "num_tokens": 1786161.0,
      "step": 965
    },
    {
      "entropy": 7.274262189865112,
      "epoch": 0.08565877781702579,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004845,
      "loss": 7.007,
      "mean_token_accuracy": 0.11388442665338516,
      "num_tokens": 1796093.0,
      "step": 970
    },
    {
      "entropy": 7.080674505233764,
      "epoch": 0.08610031790886613,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000487,
      "loss": 6.8546,
      "mean_token_accuracy": 0.11734381765127182,
      "num_tokens": 1805574.0,
      "step": 975
    },
    {
      "entropy": 7.141940927505493,
      "epoch": 0.08654185800070646,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004895,
      "loss": 6.8924,
      "mean_token_accuracy": 0.11525135561823845,
      "num_tokens": 1815175.0,
      "step": 980
    },
    {
      "entropy": 7.089342260360718,
      "epoch": 0.08698339809254681,
      "grad_norm": 1.03125,
      "learning_rate": 0.000492,
      "loss": 6.8082,
      "mean_token_accuracy": 0.12424175813794136,
      "num_tokens": 1824683.0,
      "step": 985
    },
    {
      "entropy": 7.1276007175445555,
      "epoch": 0.08742493818438714,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004945,
      "loss": 6.7918,
      "mean_token_accuracy": 0.1199628159403801,
      "num_tokens": 1833174.0,
      "step": 990
    },
    {
      "entropy": 7.066866636276245,
      "epoch": 0.08786647827622748,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000497,
      "loss": 6.8796,
      "mean_token_accuracy": 0.11747556179761887,
      "num_tokens": 1842409.0,
      "step": 995
    },
    {
      "entropy": 7.007741022109985,
      "epoch": 0.08830801836806781,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995,
      "loss": 6.7861,
      "mean_token_accuracy": 0.12650212571024894,
      "num_tokens": 1851441.0,
      "step": 1000
    },
    {
      "entropy": 7.240553998947144,
      "epoch": 0.08874955845990816,
      "grad_norm": 1.140625,
      "learning_rate": 0.000499999998589561,
      "loss": 6.9076,
      "mean_token_accuracy": 0.11784921512007714,
      "num_tokens": 1861188.0,
      "step": 1005
    },
    {
      "entropy": 7.046397066116333,
      "epoch": 0.0891910985517485,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999999928596523,
      "loss": 6.8583,
      "mean_token_accuracy": 0.12223245799541474,
      "num_tokens": 1870284.0,
      "step": 1010
    },
    {
      "entropy": 7.0789531707763675,
      "epoch": 0.08963263864358884,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999999827221219,
      "loss": 6.8283,
      "mean_token_accuracy": 0.12814989313483238,
      "num_tokens": 1879744.0,
      "step": 1015
    },
    {
      "entropy": 7.0941243171691895,
      "epoch": 0.09007417873542918,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999999681769696,
      "loss": 6.8228,
      "mean_token_accuracy": 0.12476407438516617,
      "num_tokens": 1889241.0,
      "step": 1020
    },
    {
      "entropy": 7.043097877502442,
      "epoch": 0.09051571882726951,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000499999949224196,
      "loss": 6.8413,
      "mean_token_accuracy": 0.12211177572607994,
      "num_tokens": 1898247.0,
      "step": 1025
    },
    {
      "entropy": 7.068183851242066,
      "epoch": 0.09095725891910986,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999999258638013,
      "loss": 6.7999,
      "mean_token_accuracy": 0.122597586363554,
      "num_tokens": 1907559.0,
      "step": 1030
    },
    {
      "entropy": 7.042861270904541,
      "epoch": 0.09139879901095019,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999998980957861,
      "loss": 6.8116,
      "mean_token_accuracy": 0.12296328842639923,
      "num_tokens": 1917072.0,
      "step": 1035
    },
    {
      "entropy": 7.02711238861084,
      "epoch": 0.09184033910279053,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999998659201508,
      "loss": 6.7577,
      "mean_token_accuracy": 0.1265648201107979,
      "num_tokens": 1926597.0,
      "step": 1040
    },
    {
      "entropy": 7.008166885375976,
      "epoch": 0.09228187919463088,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999998293368961,
      "loss": 6.8148,
      "mean_token_accuracy": 0.11956329345703125,
      "num_tokens": 1935978.0,
      "step": 1045
    },
    {
      "entropy": 7.03216814994812,
      "epoch": 0.09272341928647121,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999997883460227,
      "loss": 6.7551,
      "mean_token_accuracy": 0.12221006751060486,
      "num_tokens": 1944424.0,
      "step": 1050
    },
    {
      "entropy": 7.016500616073609,
      "epoch": 0.09316495937831155,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999997429475314,
      "loss": 6.7199,
      "mean_token_accuracy": 0.1299702636897564,
      "num_tokens": 1953844.0,
      "step": 1055
    },
    {
      "entropy": 7.052721452713013,
      "epoch": 0.09360649947015189,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999996931414232,
      "loss": 6.8195,
      "mean_token_accuracy": 0.12436865568161011,
      "num_tokens": 1963974.0,
      "step": 1060
    },
    {
      "entropy": 7.030517196655273,
      "epoch": 0.09404803956199223,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999996389276988,
      "loss": 6.7732,
      "mean_token_accuracy": 0.13136502131819724,
      "num_tokens": 1973466.0,
      "step": 1065
    },
    {
      "entropy": 6.936943769454956,
      "epoch": 0.09448957965383256,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999995803063596,
      "loss": 6.7879,
      "mean_token_accuracy": 0.1239612028002739,
      "num_tokens": 1983478.0,
      "step": 1070
    },
    {
      "entropy": 7.092060518264771,
      "epoch": 0.0949311197456729,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999995172774065,
      "loss": 6.8703,
      "mean_token_accuracy": 0.12069909051060676,
      "num_tokens": 1992775.0,
      "step": 1075
    },
    {
      "entropy": 7.053188705444336,
      "epoch": 0.09537265983751325,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999994498408408,
      "loss": 6.8269,
      "mean_token_accuracy": 0.11928088068962098,
      "num_tokens": 2002526.0,
      "step": 1080
    },
    {
      "entropy": 7.058347511291504,
      "epoch": 0.09581419992935358,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999993779966639,
      "loss": 6.8484,
      "mean_token_accuracy": 0.11959519013762474,
      "num_tokens": 2012476.0,
      "step": 1085
    },
    {
      "entropy": 6.85009355545044,
      "epoch": 0.09625574002119393,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999993017448771,
      "loss": 6.6854,
      "mean_token_accuracy": 0.1360420987010002,
      "num_tokens": 2021252.0,
      "step": 1090
    },
    {
      "entropy": 7.145733261108399,
      "epoch": 0.09669728011303426,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999992210854821,
      "loss": 6.8438,
      "mean_token_accuracy": 0.1258312664926052,
      "num_tokens": 2031438.0,
      "step": 1095
    },
    {
      "entropy": 7.019267272949219,
      "epoch": 0.0971388202048746,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999991360184801,
      "loss": 6.883,
      "mean_token_accuracy": 0.12097008153796196,
      "num_tokens": 2041319.0,
      "step": 1100
    },
    {
      "entropy": 7.073525238037109,
      "epoch": 0.09758036029671494,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999990465438731,
      "loss": 6.862,
      "mean_token_accuracy": 0.1219029612839222,
      "num_tokens": 2052060.0,
      "step": 1105
    },
    {
      "entropy": 6.95486044883728,
      "epoch": 0.09802190038855528,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999989526616628,
      "loss": 6.7495,
      "mean_token_accuracy": 0.12949764505028724,
      "num_tokens": 2061331.0,
      "step": 1110
    },
    {
      "entropy": 6.9770708084106445,
      "epoch": 0.09846344048039563,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999988543718509,
      "loss": 6.7566,
      "mean_token_accuracy": 0.12212359756231309,
      "num_tokens": 2070006.0,
      "step": 1115
    },
    {
      "entropy": 7.0353333950042725,
      "epoch": 0.09890498057223596,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999987516744394,
      "loss": 6.7158,
      "mean_token_accuracy": 0.13030195459723473,
      "num_tokens": 2079089.0,
      "step": 1120
    },
    {
      "entropy": 6.971569681167603,
      "epoch": 0.0993465206640763,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999986445694303,
      "loss": 6.7236,
      "mean_token_accuracy": 0.12542433887720109,
      "num_tokens": 2087237.0,
      "step": 1125
    },
    {
      "entropy": 6.899944829940796,
      "epoch": 0.09978806075591663,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999985330568258,
      "loss": 6.6822,
      "mean_token_accuracy": 0.13330597206950187,
      "num_tokens": 2096598.0,
      "step": 1130
    },
    {
      "entropy": 7.01187915802002,
      "epoch": 0.10022960084775698,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999984171366278,
      "loss": 6.7788,
      "mean_token_accuracy": 0.11807065233588218,
      "num_tokens": 2106106.0,
      "step": 1135
    },
    {
      "entropy": 6.9801586151123045,
      "epoch": 0.10067114093959731,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499998296808839,
      "loss": 6.7582,
      "mean_token_accuracy": 0.11982692703604698,
      "num_tokens": 2115478.0,
      "step": 1140
    },
    {
      "entropy": 7.019436502456665,
      "epoch": 0.10111268103143765,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004999981720734615,
      "loss": 6.7836,
      "mean_token_accuracy": 0.12526557743549346,
      "num_tokens": 2124439.0,
      "step": 1145
    },
    {
      "entropy": 7.01093807220459,
      "epoch": 0.101554221123278,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004999980429304977,
      "loss": 6.796,
      "mean_token_accuracy": 0.11945406049489975,
      "num_tokens": 2133718.0,
      "step": 1150
    },
    {
      "entropy": 6.833093214035034,
      "epoch": 0.10199576121511833,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999979093799502,
      "loss": 6.6939,
      "mean_token_accuracy": 0.12498606964945794,
      "num_tokens": 2142978.0,
      "step": 1155
    },
    {
      "entropy": 6.946839380264282,
      "epoch": 0.10243730130695868,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999977714218217,
      "loss": 6.7688,
      "mean_token_accuracy": 0.12323360666632652,
      "num_tokens": 2152250.0,
      "step": 1160
    },
    {
      "entropy": 6.99408917427063,
      "epoch": 0.102878841398799,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999976290561147,
      "loss": 6.7613,
      "mean_token_accuracy": 0.12313725724816323,
      "num_tokens": 2161620.0,
      "step": 1165
    },
    {
      "entropy": 6.988214206695557,
      "epoch": 0.10332038149063935,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999974822828322,
      "loss": 6.7814,
      "mean_token_accuracy": 0.12046999335289002,
      "num_tokens": 2170856.0,
      "step": 1170
    },
    {
      "entropy": 6.9620466232299805,
      "epoch": 0.10376192158247968,
      "grad_norm": 0.984375,
      "learning_rate": 0.000499997331101977,
      "loss": 6.7547,
      "mean_token_accuracy": 0.12075523287057877,
      "num_tokens": 2180926.0,
      "step": 1175
    },
    {
      "entropy": 6.9654961109161375,
      "epoch": 0.10420346167432003,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499997175513552,
      "loss": 6.7462,
      "mean_token_accuracy": 0.12285592630505562,
      "num_tokens": 2190248.0,
      "step": 1180
    },
    {
      "entropy": 6.9309080123901365,
      "epoch": 0.10464500176616037,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999970155175603,
      "loss": 6.7514,
      "mean_token_accuracy": 0.12348200678825379,
      "num_tokens": 2199833.0,
      "step": 1185
    },
    {
      "entropy": 6.959126377105713,
      "epoch": 0.1050865418580007,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499996851114005,
      "loss": 6.7184,
      "mean_token_accuracy": 0.12861018776893615,
      "num_tokens": 2208240.0,
      "step": 1190
    },
    {
      "entropy": 6.932129383087158,
      "epoch": 0.10552808194984105,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999966823028894,
      "loss": 6.7618,
      "mean_token_accuracy": 0.11978181153535843,
      "num_tokens": 2218758.0,
      "step": 1195
    },
    {
      "entropy": 6.9820332527160645,
      "epoch": 0.10596962204168138,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999965090842168,
      "loss": 6.7626,
      "mean_token_accuracy": 0.1300593949854374,
      "num_tokens": 2228443.0,
      "step": 1200
    },
    {
      "entropy": 6.858966636657715,
      "epoch": 0.10641116213352173,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999963314579905,
      "loss": 6.6107,
      "mean_token_accuracy": 0.1355985902249813,
      "num_tokens": 2236787.0,
      "step": 1205
    },
    {
      "entropy": 6.947892761230468,
      "epoch": 0.10685270222536206,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999961494242139,
      "loss": 6.7746,
      "mean_token_accuracy": 0.12228678241372108,
      "num_tokens": 2247089.0,
      "step": 1210
    },
    {
      "entropy": 6.849541854858399,
      "epoch": 0.1072942423172024,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999959629828908,
      "loss": 6.7035,
      "mean_token_accuracy": 0.12353255078196526,
      "num_tokens": 2256045.0,
      "step": 1215
    },
    {
      "entropy": 7.002861881256104,
      "epoch": 0.10773578240904275,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999957721340248,
      "loss": 6.7841,
      "mean_token_accuracy": 0.12372039780020713,
      "num_tokens": 2265835.0,
      "step": 1220
    },
    {
      "entropy": 6.845533561706543,
      "epoch": 0.10817732250088308,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999955768776195,
      "loss": 6.7134,
      "mean_token_accuracy": 0.12301751002669334,
      "num_tokens": 2275318.0,
      "step": 1225
    },
    {
      "entropy": 6.958423137664795,
      "epoch": 0.10861886259272342,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999953772136788,
      "loss": 6.7049,
      "mean_token_accuracy": 0.12736163064837455,
      "num_tokens": 2284821.0,
      "step": 1230
    },
    {
      "entropy": 6.942495965957642,
      "epoch": 0.10906040268456375,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999951731422068,
      "loss": 6.6712,
      "mean_token_accuracy": 0.12646759897470475,
      "num_tokens": 2294013.0,
      "step": 1235
    },
    {
      "entropy": 6.917871713638306,
      "epoch": 0.1095019427764041,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999949646632072,
      "loss": 6.6694,
      "mean_token_accuracy": 0.12900351211428643,
      "num_tokens": 2302727.0,
      "step": 1240
    },
    {
      "entropy": 6.849976634979248,
      "epoch": 0.10994348286824443,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999947517766842,
      "loss": 6.6906,
      "mean_token_accuracy": 0.12817103341221808,
      "num_tokens": 2312032.0,
      "step": 1245
    },
    {
      "entropy": 7.057008552551269,
      "epoch": 0.11038502296008477,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000499994534482642,
      "loss": 6.7819,
      "mean_token_accuracy": 0.12531876042485238,
      "num_tokens": 2321839.0,
      "step": 1250
    },
    {
      "entropy": 6.899663734436035,
      "epoch": 0.11082656305192512,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999943127810847,
      "loss": 6.7555,
      "mean_token_accuracy": 0.1212481401860714,
      "num_tokens": 2331255.0,
      "step": 1255
    },
    {
      "entropy": 6.816301393508911,
      "epoch": 0.11126810314376545,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999940866720169,
      "loss": 6.5529,
      "mean_token_accuracy": 0.13424451053142547,
      "num_tokens": 2340038.0,
      "step": 1260
    },
    {
      "entropy": 6.8964927196502686,
      "epoch": 0.1117096432356058,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999938561554429,
      "loss": 6.6959,
      "mean_token_accuracy": 0.12449696063995361,
      "num_tokens": 2348901.0,
      "step": 1265
    },
    {
      "entropy": 6.875371694564819,
      "epoch": 0.11215118332744613,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999936212313672,
      "loss": 6.7697,
      "mean_token_accuracy": 0.119696693867445,
      "num_tokens": 2358842.0,
      "step": 1270
    },
    {
      "entropy": 6.907081985473633,
      "epoch": 0.11259272341928647,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999933818997943,
      "loss": 6.6654,
      "mean_token_accuracy": 0.12900050207972527,
      "num_tokens": 2368650.0,
      "step": 1275
    },
    {
      "entropy": 6.866778802871704,
      "epoch": 0.1130342635111268,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999931381607292,
      "loss": 6.7041,
      "mean_token_accuracy": 0.1312815934419632,
      "num_tokens": 2377916.0,
      "step": 1280
    },
    {
      "entropy": 6.936831331253051,
      "epoch": 0.11347580360296715,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999928900141764,
      "loss": 6.6683,
      "mean_token_accuracy": 0.12833984717726707,
      "num_tokens": 2387507.0,
      "step": 1285
    },
    {
      "entropy": 6.833027648925781,
      "epoch": 0.1139173436948075,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499992637460141,
      "loss": 6.7166,
      "mean_token_accuracy": 0.12722958698868753,
      "num_tokens": 2396148.0,
      "step": 1290
    },
    {
      "entropy": 6.854029512405395,
      "epoch": 0.11435888378664782,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999923804986275,
      "loss": 6.5844,
      "mean_token_accuracy": 0.12824990823864937,
      "num_tokens": 2404891.0,
      "step": 1295
    },
    {
      "entropy": 6.922542858123779,
      "epoch": 0.11480042387848817,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999921191296415,
      "loss": 6.6288,
      "mean_token_accuracy": 0.12811262235045434,
      "num_tokens": 2414406.0,
      "step": 1300
    },
    {
      "entropy": 6.860514211654663,
      "epoch": 0.1152419639703285,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999918533531877,
      "loss": 6.7126,
      "mean_token_accuracy": 0.12714724242687225,
      "num_tokens": 2424363.0,
      "step": 1305
    },
    {
      "entropy": 6.917534399032593,
      "epoch": 0.11568350406216885,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999915831692714,
      "loss": 6.6534,
      "mean_token_accuracy": 0.12873802334070206,
      "num_tokens": 2433753.0,
      "step": 1310
    },
    {
      "entropy": 6.790404272079468,
      "epoch": 0.11612504415400918,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999913085778981,
      "loss": 6.6769,
      "mean_token_accuracy": 0.12445072457194328,
      "num_tokens": 2443275.0,
      "step": 1315
    },
    {
      "entropy": 6.909478187561035,
      "epoch": 0.11656658424584952,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999910295790729,
      "loss": 6.6931,
      "mean_token_accuracy": 0.12268302440643311,
      "num_tokens": 2452510.0,
      "step": 1320
    },
    {
      "entropy": 6.915535879135132,
      "epoch": 0.11700812433768987,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999907461728014,
      "loss": 6.7807,
      "mean_token_accuracy": 0.1234104610979557,
      "num_tokens": 2462742.0,
      "step": 1325
    },
    {
      "entropy": 6.835855340957641,
      "epoch": 0.1174496644295302,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999904583590893,
      "loss": 6.6549,
      "mean_token_accuracy": 0.12718312442302704,
      "num_tokens": 2471409.0,
      "step": 1330
    },
    {
      "entropy": 6.876492547988891,
      "epoch": 0.11789120452137054,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999901661379418,
      "loss": 6.6053,
      "mean_token_accuracy": 0.1328142687678337,
      "num_tokens": 2481011.0,
      "step": 1335
    },
    {
      "entropy": 6.79636025428772,
      "epoch": 0.11833274461321087,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999898695093652,
      "loss": 6.6995,
      "mean_token_accuracy": 0.12458944991230965,
      "num_tokens": 2490664.0,
      "step": 1340
    },
    {
      "entropy": 6.888507127761841,
      "epoch": 0.11877428470505122,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999895684733648,
      "loss": 6.6235,
      "mean_token_accuracy": 0.13585169985890388,
      "num_tokens": 2499799.0,
      "step": 1345
    },
    {
      "entropy": 6.845640707015991,
      "epoch": 0.11921582479689156,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004999892630299467,
      "loss": 6.6025,
      "mean_token_accuracy": 0.12698834463953973,
      "num_tokens": 2508780.0,
      "step": 1350
    },
    {
      "entropy": 6.898168849945068,
      "epoch": 0.1196573648887319,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999889531791171,
      "loss": 6.6341,
      "mean_token_accuracy": 0.12324970662593841,
      "num_tokens": 2517741.0,
      "step": 1355
    },
    {
      "entropy": 6.750218534469605,
      "epoch": 0.12009890498057224,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999886389208817,
      "loss": 6.7105,
      "mean_token_accuracy": 0.12327531054615974,
      "num_tokens": 2528742.0,
      "step": 1360
    },
    {
      "entropy": 6.880941247940063,
      "epoch": 0.12054044507241257,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999883202552468,
      "loss": 6.6444,
      "mean_token_accuracy": 0.1296772688627243,
      "num_tokens": 2538609.0,
      "step": 1365
    },
    {
      "entropy": 6.886623859405518,
      "epoch": 0.12098198516425292,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999879971822189,
      "loss": 6.6274,
      "mean_token_accuracy": 0.12891870513558387,
      "num_tokens": 2547772.0,
      "step": 1370
    },
    {
      "entropy": 6.6913806915283205,
      "epoch": 0.12142352525609325,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999876697018038,
      "loss": 6.6008,
      "mean_token_accuracy": 0.13324561789631845,
      "num_tokens": 2556114.0,
      "step": 1375
    },
    {
      "entropy": 6.90076265335083,
      "epoch": 0.1218650653479336,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999873378140085,
      "loss": 6.6339,
      "mean_token_accuracy": 0.12988200038671494,
      "num_tokens": 2566814.0,
      "step": 1380
    },
    {
      "entropy": 6.784169101715088,
      "epoch": 0.12230660543977394,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004999870015188389,
      "loss": 6.6031,
      "mean_token_accuracy": 0.13186766728758811,
      "num_tokens": 2576030.0,
      "step": 1385
    },
    {
      "entropy": 6.762492084503174,
      "epoch": 0.12274814553161427,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999866608163021,
      "loss": 6.6231,
      "mean_token_accuracy": 0.13228361681103706,
      "num_tokens": 2585756.0,
      "step": 1390
    },
    {
      "entropy": 6.82820086479187,
      "epoch": 0.12318968562345461,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999863157064045,
      "loss": 6.6779,
      "mean_token_accuracy": 0.1298144727945328,
      "num_tokens": 2595676.0,
      "step": 1395
    },
    {
      "entropy": 6.848227310180664,
      "epoch": 0.12363122571529495,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999859661891529,
      "loss": 6.6748,
      "mean_token_accuracy": 0.1304032675921917,
      "num_tokens": 2606197.0,
      "step": 1400
    },
    {
      "entropy": 6.879591560363769,
      "epoch": 0.12407276580713529,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999856122645543,
      "loss": 6.5967,
      "mean_token_accuracy": 0.12973913177847862,
      "num_tokens": 2615311.0,
      "step": 1405
    },
    {
      "entropy": 6.734095478057862,
      "epoch": 0.12451430589897562,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999852539326154,
      "loss": 6.5243,
      "mean_token_accuracy": 0.1325017362833023,
      "num_tokens": 2624074.0,
      "step": 1410
    },
    {
      "entropy": 6.758558845520019,
      "epoch": 0.12495584599081597,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999848911933434,
      "loss": 6.6522,
      "mean_token_accuracy": 0.13497036173939705,
      "num_tokens": 2633877.0,
      "step": 1415
    },
    {
      "entropy": 6.745906019210816,
      "epoch": 0.1253973860826563,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999845240467453,
      "loss": 6.575,
      "mean_token_accuracy": 0.12966776564717292,
      "num_tokens": 2643330.0,
      "step": 1420
    },
    {
      "entropy": 6.863449001312256,
      "epoch": 0.12583892617449666,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999841524928282,
      "loss": 6.5664,
      "mean_token_accuracy": 0.13387957438826562,
      "num_tokens": 2652070.0,
      "step": 1425
    },
    {
      "entropy": 6.756294584274292,
      "epoch": 0.126280466266337,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999837765315997,
      "loss": 6.6242,
      "mean_token_accuracy": 0.12831888422369958,
      "num_tokens": 2660546.0,
      "step": 1430
    },
    {
      "entropy": 6.676768016815186,
      "epoch": 0.12672200635817732,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999833961630669,
      "loss": 6.507,
      "mean_token_accuracy": 0.13364252746105193,
      "num_tokens": 2669938.0,
      "step": 1435
    },
    {
      "entropy": 6.824239492416382,
      "epoch": 0.12716354645001765,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999830113872374,
      "loss": 6.637,
      "mean_token_accuracy": 0.13026153072714805,
      "num_tokens": 2679814.0,
      "step": 1440
    },
    {
      "entropy": 6.675518751144409,
      "epoch": 0.127605086541858,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999826222041186,
      "loss": 6.5447,
      "mean_token_accuracy": 0.13804391250014306,
      "num_tokens": 2688733.0,
      "step": 1445
    },
    {
      "entropy": 6.745492744445801,
      "epoch": 0.12804662663369834,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999822286137182,
      "loss": 6.5346,
      "mean_token_accuracy": 0.13509366363286973,
      "num_tokens": 2697744.0,
      "step": 1450
    },
    {
      "entropy": 6.658346796035767,
      "epoch": 0.12848816672553867,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999818306160439,
      "loss": 6.4971,
      "mean_token_accuracy": 0.13757269233465194,
      "num_tokens": 2707037.0,
      "step": 1455
    },
    {
      "entropy": 6.859276628494262,
      "epoch": 0.12892970681737903,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999814282111034,
      "loss": 6.6552,
      "mean_token_accuracy": 0.13125866502523423,
      "num_tokens": 2717345.0,
      "step": 1460
    },
    {
      "entropy": 6.818531513214111,
      "epoch": 0.12937124690921936,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999810213989047,
      "loss": 6.5966,
      "mean_token_accuracy": 0.13112419992685317,
      "num_tokens": 2726892.0,
      "step": 1465
    },
    {
      "entropy": 6.735688495635986,
      "epoch": 0.1298127870010597,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999806101794558,
      "loss": 6.572,
      "mean_token_accuracy": 0.13323748409748076,
      "num_tokens": 2736479.0,
      "step": 1470
    },
    {
      "entropy": 6.835882139205933,
      "epoch": 0.13025432709290002,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999801945527648,
      "loss": 6.6304,
      "mean_token_accuracy": 0.12691461741924287,
      "num_tokens": 2745998.0,
      "step": 1475
    },
    {
      "entropy": 6.753359127044678,
      "epoch": 0.13069586718474038,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999797745188395,
      "loss": 6.6159,
      "mean_token_accuracy": 0.1317651830613613,
      "num_tokens": 2754346.0,
      "step": 1480
    },
    {
      "entropy": 6.801856279373169,
      "epoch": 0.13113740727658071,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999793500776886,
      "loss": 6.5416,
      "mean_token_accuracy": 0.13924193307757377,
      "num_tokens": 2763391.0,
      "step": 1485
    },
    {
      "entropy": 6.706637144088745,
      "epoch": 0.13157894736842105,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999789212293201,
      "loss": 6.5867,
      "mean_token_accuracy": 0.13430250883102418,
      "num_tokens": 2772764.0,
      "step": 1490
    },
    {
      "entropy": 6.875431776046753,
      "epoch": 0.1320204874602614,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999784879737423,
      "loss": 6.6781,
      "mean_token_accuracy": 0.1260899782180786,
      "num_tokens": 2782312.0,
      "step": 1495
    },
    {
      "entropy": 6.746471691131592,
      "epoch": 0.13246202755210174,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999780503109642,
      "loss": 6.5979,
      "mean_token_accuracy": 0.1290306605398655,
      "num_tokens": 2791159.0,
      "step": 1500
    },
    {
      "entropy": 6.704893779754639,
      "epoch": 0.13290356764394207,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999776082409939,
      "loss": 6.408,
      "mean_token_accuracy": 0.13926204219460486,
      "num_tokens": 2799319.0,
      "step": 1505
    },
    {
      "entropy": 6.6507806301116945,
      "epoch": 0.1333451077357824,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999771617638401,
      "loss": 6.5335,
      "mean_token_accuracy": 0.13334701284766198,
      "num_tokens": 2807401.0,
      "step": 1510
    },
    {
      "entropy": 6.7505804061889645,
      "epoch": 0.13378664782762276,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999767108795118,
      "loss": 6.6075,
      "mean_token_accuracy": 0.1309545248746872,
      "num_tokens": 2817734.0,
      "step": 1515
    },
    {
      "entropy": 6.790192747116089,
      "epoch": 0.1342281879194631,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999762555880176,
      "loss": 6.5906,
      "mean_token_accuracy": 0.12997578456997871,
      "num_tokens": 2828235.0,
      "step": 1520
    },
    {
      "entropy": 6.771645879745483,
      "epoch": 0.13466972801130342,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999757958893666,
      "loss": 6.6309,
      "mean_token_accuracy": 0.1289198786020279,
      "num_tokens": 2837453.0,
      "step": 1525
    },
    {
      "entropy": 6.712240839004517,
      "epoch": 0.13511126810314378,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999753317835677,
      "loss": 6.587,
      "mean_token_accuracy": 0.12751804441213607,
      "num_tokens": 2847055.0,
      "step": 1530
    },
    {
      "entropy": 6.758566999435425,
      "epoch": 0.1355528081949841,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999748632706299,
      "loss": 6.5597,
      "mean_token_accuracy": 0.13620577305555343,
      "num_tokens": 2857101.0,
      "step": 1535
    },
    {
      "entropy": 6.699764680862427,
      "epoch": 0.13599434828682444,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999743903505626,
      "loss": 6.4845,
      "mean_token_accuracy": 0.13134869784116746,
      "num_tokens": 2866685.0,
      "step": 1540
    },
    {
      "entropy": 6.813828182220459,
      "epoch": 0.13643588837866477,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999739130233749,
      "loss": 6.5821,
      "mean_token_accuracy": 0.1299674078822136,
      "num_tokens": 2876022.0,
      "step": 1545
    },
    {
      "entropy": 6.708414220809937,
      "epoch": 0.13687742847050513,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999734312890761,
      "loss": 6.5192,
      "mean_token_accuracy": 0.12855476662516593,
      "num_tokens": 2885560.0,
      "step": 1550
    },
    {
      "entropy": 6.690548992156982,
      "epoch": 0.13731896856234546,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999729451476757,
      "loss": 6.5568,
      "mean_token_accuracy": 0.13441056087613107,
      "num_tokens": 2894686.0,
      "step": 1555
    },
    {
      "entropy": 6.749054193496704,
      "epoch": 0.1377605086541858,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999724545991835,
      "loss": 6.5599,
      "mean_token_accuracy": 0.1379253476858139,
      "num_tokens": 2904390.0,
      "step": 1560
    },
    {
      "entropy": 6.715701723098755,
      "epoch": 0.13820204874602615,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999719596436086,
      "loss": 6.6185,
      "mean_token_accuracy": 0.13158311247825621,
      "num_tokens": 2913311.0,
      "step": 1565
    },
    {
      "entropy": 6.777320003509521,
      "epoch": 0.13864358883786648,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999714602809611,
      "loss": 6.4916,
      "mean_token_accuracy": 0.13447055071592331,
      "num_tokens": 2923196.0,
      "step": 1570
    },
    {
      "entropy": 6.681281852722168,
      "epoch": 0.1390851289297068,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999709565112506,
      "loss": 6.5513,
      "mean_token_accuracy": 0.13382177129387857,
      "num_tokens": 2932813.0,
      "step": 1575
    },
    {
      "entropy": 6.7223258972167965,
      "epoch": 0.13952666902154715,
      "grad_norm": 0.9375,
      "learning_rate": 0.000499970448334487,
      "loss": 6.4762,
      "mean_token_accuracy": 0.12845111936330794,
      "num_tokens": 2942694.0,
      "step": 1580
    },
    {
      "entropy": 6.633160924911499,
      "epoch": 0.1399682091133875,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999699357506803,
      "loss": 6.4203,
      "mean_token_accuracy": 0.14310574680566787,
      "num_tokens": 2950932.0,
      "step": 1585
    },
    {
      "entropy": 6.733539772033692,
      "epoch": 0.14040974920522784,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999694187598406,
      "loss": 6.5208,
      "mean_token_accuracy": 0.1363385871052742,
      "num_tokens": 2960102.0,
      "step": 1590
    },
    {
      "entropy": 6.716689872741699,
      "epoch": 0.14085128929706817,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004999688973619777,
      "loss": 6.5161,
      "mean_token_accuracy": 0.1320077233016491,
      "num_tokens": 2969968.0,
      "step": 1595
    },
    {
      "entropy": 6.624239683151245,
      "epoch": 0.14129282938890853,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999683715571022,
      "loss": 6.4732,
      "mean_token_accuracy": 0.14136387258768082,
      "num_tokens": 2978880.0,
      "step": 1600
    },
    {
      "entropy": 6.666266059875488,
      "epoch": 0.14173436948074886,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999678413452242,
      "loss": 6.5087,
      "mean_token_accuracy": 0.1332849942147732,
      "num_tokens": 2988369.0,
      "step": 1605
    },
    {
      "entropy": 6.669563817977905,
      "epoch": 0.1421759095725892,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999673067263542,
      "loss": 6.5549,
      "mean_token_accuracy": 0.1297246351838112,
      "num_tokens": 2997070.0,
      "step": 1610
    },
    {
      "entropy": 6.7209991931915285,
      "epoch": 0.14261744966442952,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999667677005026,
      "loss": 6.4881,
      "mean_token_accuracy": 0.1361941583454609,
      "num_tokens": 3006547.0,
      "step": 1615
    },
    {
      "entropy": 6.670996856689453,
      "epoch": 0.14305898975626988,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999662242676799,
      "loss": 6.5007,
      "mean_token_accuracy": 0.13505242466926576,
      "num_tokens": 3015821.0,
      "step": 1620
    },
    {
      "entropy": 6.667819404602051,
      "epoch": 0.1435005298481102,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999656764278968,
      "loss": 6.4863,
      "mean_token_accuracy": 0.13409159779548646,
      "num_tokens": 3024750.0,
      "step": 1625
    },
    {
      "entropy": 6.699454641342163,
      "epoch": 0.14394206993995054,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999651241811642,
      "loss": 6.4543,
      "mean_token_accuracy": 0.13378062024712561,
      "num_tokens": 3033345.0,
      "step": 1630
    },
    {
      "entropy": 6.63138108253479,
      "epoch": 0.1443836100317909,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999645675274925,
      "loss": 6.4314,
      "mean_token_accuracy": 0.1369638502597809,
      "num_tokens": 3042060.0,
      "step": 1635
    },
    {
      "entropy": 6.6189075946807865,
      "epoch": 0.14482515012363123,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999640064668931,
      "loss": 6.584,
      "mean_token_accuracy": 0.1317050538957119,
      "num_tokens": 3052490.0,
      "step": 1640
    },
    {
      "entropy": 6.820568704605103,
      "epoch": 0.14526669021547156,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999634409993766,
      "loss": 6.5606,
      "mean_token_accuracy": 0.1320354349911213,
      "num_tokens": 3061934.0,
      "step": 1645
    },
    {
      "entropy": 6.609713792800903,
      "epoch": 0.1457082303073119,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999628711249544,
      "loss": 6.4512,
      "mean_token_accuracy": 0.14322613999247552,
      "num_tokens": 3070890.0,
      "step": 1650
    },
    {
      "entropy": 6.7396402835845945,
      "epoch": 0.14614977039915225,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999622968436373,
      "loss": 6.4715,
      "mean_token_accuracy": 0.13257212415337563,
      "num_tokens": 3079933.0,
      "step": 1655
    },
    {
      "entropy": 6.579079866409302,
      "epoch": 0.14659131049099258,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999617181554369,
      "loss": 6.5248,
      "mean_token_accuracy": 0.13561063706874849,
      "num_tokens": 3089910.0,
      "step": 1660
    },
    {
      "entropy": 6.757257795333862,
      "epoch": 0.1470328505828329,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004999611350603643,
      "loss": 6.5275,
      "mean_token_accuracy": 0.13599081262946128,
      "num_tokens": 3098676.0,
      "step": 1665
    },
    {
      "entropy": 6.642300748825074,
      "epoch": 0.14747439067467327,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000499960547558431,
      "loss": 6.6165,
      "mean_token_accuracy": 0.13092149198055267,
      "num_tokens": 3109055.0,
      "step": 1670
    },
    {
      "entropy": 6.597188997268677,
      "epoch": 0.1479159307665136,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999599556496486,
      "loss": 6.4706,
      "mean_token_accuracy": 0.13985515162348747,
      "num_tokens": 3117517.0,
      "step": 1675
    },
    {
      "entropy": 6.640227365493774,
      "epoch": 0.14835747085835393,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999593593340286,
      "loss": 6.463,
      "mean_token_accuracy": 0.14289223924279212,
      "num_tokens": 3126606.0,
      "step": 1680
    },
    {
      "entropy": 6.657995748519897,
      "epoch": 0.14879901095019427,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999587586115826,
      "loss": 6.4902,
      "mean_token_accuracy": 0.1365136206150055,
      "num_tokens": 3135444.0,
      "step": 1685
    },
    {
      "entropy": 6.682824659347534,
      "epoch": 0.14924055104203462,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999581534823226,
      "loss": 6.4214,
      "mean_token_accuracy": 0.14298791959881782,
      "num_tokens": 3144967.0,
      "step": 1690
    },
    {
      "entropy": 6.649335050582886,
      "epoch": 0.14968209113387496,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999575439462601,
      "loss": 6.4563,
      "mean_token_accuracy": 0.13346426337957382,
      "num_tokens": 3153898.0,
      "step": 1695
    },
    {
      "entropy": 6.639866065979004,
      "epoch": 0.1501236312257153,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999569300034075,
      "loss": 6.5099,
      "mean_token_accuracy": 0.13122038245201112,
      "num_tokens": 3162341.0,
      "step": 1700
    },
    {
      "entropy": 6.621887493133545,
      "epoch": 0.15056517131755565,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999563116537764,
      "loss": 6.3038,
      "mean_token_accuracy": 0.1461081884801388,
      "num_tokens": 3171420.0,
      "step": 1705
    },
    {
      "entropy": 6.58403959274292,
      "epoch": 0.15100671140939598,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999556888973792,
      "loss": 6.3453,
      "mean_token_accuracy": 0.14070172980427742,
      "num_tokens": 3180221.0,
      "step": 1710
    },
    {
      "entropy": 6.7422089099884035,
      "epoch": 0.1514482515012363,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999550617342279,
      "loss": 6.5005,
      "mean_token_accuracy": 0.1325171984732151,
      "num_tokens": 3189156.0,
      "step": 1715
    },
    {
      "entropy": 6.5561034202575685,
      "epoch": 0.15188979159307664,
      "grad_norm": 1.0,
      "learning_rate": 0.000499954430164335,
      "loss": 6.4253,
      "mean_token_accuracy": 0.13670367896556854,
      "num_tokens": 3199870.0,
      "step": 1720
    },
    {
      "entropy": 6.655413389205933,
      "epoch": 0.152331331684917,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999537941877127,
      "loss": 6.4231,
      "mean_token_accuracy": 0.14037118926644326,
      "num_tokens": 3208815.0,
      "step": 1725
    },
    {
      "entropy": 6.575525569915771,
      "epoch": 0.15277287177675733,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999531538043735,
      "loss": 6.4845,
      "mean_token_accuracy": 0.137246023863554,
      "num_tokens": 3218692.0,
      "step": 1730
    },
    {
      "entropy": 6.627863502502441,
      "epoch": 0.15321441186859766,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999525090143298,
      "loss": 6.4482,
      "mean_token_accuracy": 0.14504025802016257,
      "num_tokens": 3227604.0,
      "step": 1735
    },
    {
      "entropy": 6.625754451751709,
      "epoch": 0.15365595196043802,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999518598175946,
      "loss": 6.4675,
      "mean_token_accuracy": 0.14471371322870255,
      "num_tokens": 3237912.0,
      "step": 1740
    },
    {
      "entropy": 6.527244615554809,
      "epoch": 0.15409749205227835,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999512062141805,
      "loss": 6.3302,
      "mean_token_accuracy": 0.14219417944550514,
      "num_tokens": 3246003.0,
      "step": 1745
    },
    {
      "entropy": 6.632448387145996,
      "epoch": 0.15453903214411868,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999505482040999,
      "loss": 6.4719,
      "mean_token_accuracy": 0.13460535928606987,
      "num_tokens": 3256363.0,
      "step": 1750
    },
    {
      "entropy": 6.610141611099243,
      "epoch": 0.154980572235959,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999498857873662,
      "loss": 6.4661,
      "mean_token_accuracy": 0.13534337058663368,
      "num_tokens": 3265822.0,
      "step": 1755
    },
    {
      "entropy": 6.538852977752685,
      "epoch": 0.15542211232779937,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999492189639921,
      "loss": 6.3683,
      "mean_token_accuracy": 0.14053185135126114,
      "num_tokens": 3274614.0,
      "step": 1760
    },
    {
      "entropy": 6.577081727981567,
      "epoch": 0.1558636524196397,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999485477339907,
      "loss": 6.4384,
      "mean_token_accuracy": 0.13609417378902436,
      "num_tokens": 3283800.0,
      "step": 1765
    },
    {
      "entropy": 6.623507022857666,
      "epoch": 0.15630519251148003,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999478720973753,
      "loss": 6.3582,
      "mean_token_accuracy": 0.14612039923667908,
      "num_tokens": 3293221.0,
      "step": 1770
    },
    {
      "entropy": 6.631183099746704,
      "epoch": 0.1567467326033204,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000499947192054159,
      "loss": 6.5087,
      "mean_token_accuracy": 0.13212059661746026,
      "num_tokens": 3302852.0,
      "step": 1775
    },
    {
      "entropy": 6.668994379043579,
      "epoch": 0.15718827269516072,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499946507604355,
      "loss": 6.3761,
      "mean_token_accuracy": 0.14372102692723274,
      "num_tokens": 3311752.0,
      "step": 1780
    },
    {
      "entropy": 6.522418546676636,
      "epoch": 0.15762981278700106,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499945818747977,
      "loss": 6.4622,
      "mean_token_accuracy": 0.14355819597840308,
      "num_tokens": 3321339.0,
      "step": 1785
    },
    {
      "entropy": 6.667537021636963,
      "epoch": 0.1580713528788414,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999451254850383,
      "loss": 6.4756,
      "mean_token_accuracy": 0.13703957572579384,
      "num_tokens": 3330269.0,
      "step": 1790
    },
    {
      "entropy": 6.634947395324707,
      "epoch": 0.15851289297068175,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999444278155525,
      "loss": 6.4032,
      "mean_token_accuracy": 0.14509668350219726,
      "num_tokens": 3340770.0,
      "step": 1795
    },
    {
      "entropy": 6.548423147201538,
      "epoch": 0.15895443306252208,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999437257395333,
      "loss": 6.4725,
      "mean_token_accuracy": 0.1425343744456768,
      "num_tokens": 3349976.0,
      "step": 1800
    },
    {
      "entropy": 6.667039012908935,
      "epoch": 0.1593959731543624,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999430192569944,
      "loss": 6.5242,
      "mean_token_accuracy": 0.13105336651206018,
      "num_tokens": 3359764.0,
      "step": 1805
    },
    {
      "entropy": 6.581896781921387,
      "epoch": 0.15983751324620277,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004999423083679498,
      "loss": 6.405,
      "mean_token_accuracy": 0.13858939930796624,
      "num_tokens": 3369939.0,
      "step": 1810
    },
    {
      "entropy": 6.633963489532471,
      "epoch": 0.1602790533380431,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999415930724133,
      "loss": 6.5557,
      "mean_token_accuracy": 0.1334507644176483,
      "num_tokens": 3381326.0,
      "step": 1815
    },
    {
      "entropy": 6.706169319152832,
      "epoch": 0.16072059342988343,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999408733703988,
      "loss": 6.5135,
      "mean_token_accuracy": 0.13054400235414504,
      "num_tokens": 3391016.0,
      "step": 1820
    },
    {
      "entropy": 6.57712345123291,
      "epoch": 0.16116213352172376,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999401492619207,
      "loss": 6.2754,
      "mean_token_accuracy": 0.14139460623264313,
      "num_tokens": 3400294.0,
      "step": 1825
    },
    {
      "entropy": 6.533513784408569,
      "epoch": 0.16160367361356412,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999394207469928,
      "loss": 6.4737,
      "mean_token_accuracy": 0.13292715698480606,
      "num_tokens": 3409685.0,
      "step": 1830
    },
    {
      "entropy": 6.695759248733521,
      "epoch": 0.16204521370540445,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999386878256297,
      "loss": 6.3887,
      "mean_token_accuracy": 0.13601208254694938,
      "num_tokens": 3418946.0,
      "step": 1835
    },
    {
      "entropy": 6.4282214641571045,
      "epoch": 0.16248675379724478,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999379504978457,
      "loss": 6.3763,
      "mean_token_accuracy": 0.13835799768567086,
      "num_tokens": 3428245.0,
      "step": 1840
    },
    {
      "entropy": 6.614362144470215,
      "epoch": 0.16292829388908514,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499937208763655,
      "loss": 6.3958,
      "mean_token_accuracy": 0.14421852231025695,
      "num_tokens": 3437580.0,
      "step": 1845
    },
    {
      "entropy": 6.64262170791626,
      "epoch": 0.16336983398092547,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999364626230724,
      "loss": 6.3701,
      "mean_token_accuracy": 0.14203060865402223,
      "num_tokens": 3446393.0,
      "step": 1850
    },
    {
      "entropy": 6.477195358276367,
      "epoch": 0.1638113740727658,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999357120761124,
      "loss": 6.4748,
      "mean_token_accuracy": 0.1370338521897793,
      "num_tokens": 3455956.0,
      "step": 1855
    },
    {
      "entropy": 6.65583610534668,
      "epoch": 0.16425291416460613,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999349571227898,
      "loss": 6.4634,
      "mean_token_accuracy": 0.1417573131620884,
      "num_tokens": 3465722.0,
      "step": 1860
    },
    {
      "entropy": 6.579311370849609,
      "epoch": 0.1646944542564465,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999341977631193,
      "loss": 6.4233,
      "mean_token_accuracy": 0.13521963283419608,
      "num_tokens": 3475467.0,
      "step": 1865
    },
    {
      "entropy": 6.563374280929565,
      "epoch": 0.16513599434828682,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999334339971157,
      "loss": 6.3294,
      "mean_token_accuracy": 0.140533210337162,
      "num_tokens": 3484931.0,
      "step": 1870
    },
    {
      "entropy": 6.542880487442017,
      "epoch": 0.16557753444012716,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999326658247942,
      "loss": 6.4374,
      "mean_token_accuracy": 0.14069321602582932,
      "num_tokens": 3494001.0,
      "step": 1875
    },
    {
      "entropy": 6.596772766113281,
      "epoch": 0.16601907453196751,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999318932461696,
      "loss": 6.3622,
      "mean_token_accuracy": 0.14485045671463012,
      "num_tokens": 3503021.0,
      "step": 1880
    },
    {
      "entropy": 6.477844142913819,
      "epoch": 0.16646061462380785,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999311162612571,
      "loss": 6.4034,
      "mean_token_accuracy": 0.1405440613627434,
      "num_tokens": 3513005.0,
      "step": 1885
    },
    {
      "entropy": 6.664361810684204,
      "epoch": 0.16690215471564818,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499930334870072,
      "loss": 6.468,
      "mean_token_accuracy": 0.14140931516885757,
      "num_tokens": 3523219.0,
      "step": 1890
    },
    {
      "entropy": 6.607090282440185,
      "epoch": 0.1673436948074885,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999295490726296,
      "loss": 6.4475,
      "mean_token_accuracy": 0.13736767545342446,
      "num_tokens": 3532917.0,
      "step": 1895
    },
    {
      "entropy": 6.548594951629639,
      "epoch": 0.16778523489932887,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999287588689453,
      "loss": 6.4314,
      "mean_token_accuracy": 0.14416549280285834,
      "num_tokens": 3542845.0,
      "step": 1900
    },
    {
      "entropy": 6.615040254592896,
      "epoch": 0.1682267749911692,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999279642590344,
      "loss": 6.4644,
      "mean_token_accuracy": 0.14099383279681205,
      "num_tokens": 3552126.0,
      "step": 1905
    },
    {
      "entropy": 6.59620246887207,
      "epoch": 0.16866831508300953,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999271652429127,
      "loss": 6.4683,
      "mean_token_accuracy": 0.13365241289138793,
      "num_tokens": 3561254.0,
      "step": 1910
    },
    {
      "entropy": 6.603018665313721,
      "epoch": 0.1691098551748499,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999263618205958,
      "loss": 6.3064,
      "mean_token_accuracy": 0.14801083952188493,
      "num_tokens": 3569781.0,
      "step": 1915
    },
    {
      "entropy": 6.507657670974732,
      "epoch": 0.16955139526669022,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999255539920993,
      "loss": 6.4154,
      "mean_token_accuracy": 0.13945700079202653,
      "num_tokens": 3579664.0,
      "step": 1920
    },
    {
      "entropy": 6.607545375823975,
      "epoch": 0.16999293535853055,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999247417574391,
      "loss": 6.4491,
      "mean_token_accuracy": 0.14012959450483323,
      "num_tokens": 3588671.0,
      "step": 1925
    },
    {
      "entropy": 6.5801286697387695,
      "epoch": 0.17043447545037088,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999239251166312,
      "loss": 6.3639,
      "mean_token_accuracy": 0.14608858227729798,
      "num_tokens": 3597656.0,
      "step": 1930
    },
    {
      "entropy": 6.465410947799683,
      "epoch": 0.17087601554221124,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999231040696914,
      "loss": 6.373,
      "mean_token_accuracy": 0.14378879442811013,
      "num_tokens": 3608017.0,
      "step": 1935
    },
    {
      "entropy": 6.607225227355957,
      "epoch": 0.17131755563405157,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999222786166361,
      "loss": 6.4412,
      "mean_token_accuracy": 0.14090094193816186,
      "num_tokens": 3618189.0,
      "step": 1940
    },
    {
      "entropy": 6.596541976928711,
      "epoch": 0.1717590957258919,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999214487574812,
      "loss": 6.3977,
      "mean_token_accuracy": 0.1351962171494961,
      "num_tokens": 3627211.0,
      "step": 1945
    },
    {
      "entropy": 6.564425277709961,
      "epoch": 0.17220063581773226,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999206144922431,
      "loss": 6.331,
      "mean_token_accuracy": 0.1392312839627266,
      "num_tokens": 3636781.0,
      "step": 1950
    },
    {
      "entropy": 6.521996641159058,
      "epoch": 0.1726421759095726,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499919775820938,
      "loss": 6.4204,
      "mean_token_accuracy": 0.1457922250032425,
      "num_tokens": 3644891.0,
      "step": 1955
    },
    {
      "entropy": 6.611178350448609,
      "epoch": 0.17308371600141292,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999189327435825,
      "loss": 6.4558,
      "mean_token_accuracy": 0.13934372514486312,
      "num_tokens": 3655477.0,
      "step": 1960
    },
    {
      "entropy": 6.615030193328858,
      "epoch": 0.17352525609325326,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999180852601929,
      "loss": 6.4555,
      "mean_token_accuracy": 0.1424385577440262,
      "num_tokens": 3664542.0,
      "step": 1965
    },
    {
      "entropy": 6.556003856658935,
      "epoch": 0.17396679618509361,
      "grad_norm": 1.140625,
      "learning_rate": 0.000499917233370786,
      "loss": 6.3637,
      "mean_token_accuracy": 0.13885778263211251,
      "num_tokens": 3673806.0,
      "step": 1970
    },
    {
      "entropy": 6.537166738510132,
      "epoch": 0.17440833627693395,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999163770753784,
      "loss": 6.3706,
      "mean_token_accuracy": 0.14099680185317992,
      "num_tokens": 3683238.0,
      "step": 1975
    },
    {
      "entropy": 6.504315948486328,
      "epoch": 0.17484987636877428,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999155163739869,
      "loss": 6.3407,
      "mean_token_accuracy": 0.14536889716982843,
      "num_tokens": 3692161.0,
      "step": 1980
    },
    {
      "entropy": 6.581552124023437,
      "epoch": 0.17529141646061464,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999146512666284,
      "loss": 6.367,
      "mean_token_accuracy": 0.1379326641559601,
      "num_tokens": 3701431.0,
      "step": 1985
    },
    {
      "entropy": 6.487573671340942,
      "epoch": 0.17573295655245497,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999137817533197,
      "loss": 6.3461,
      "mean_token_accuracy": 0.14381761997938156,
      "num_tokens": 3710963.0,
      "step": 1990
    },
    {
      "entropy": 6.506614542007446,
      "epoch": 0.1761744966442953,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999129078340779,
      "loss": 6.3353,
      "mean_token_accuracy": 0.14846789166331292,
      "num_tokens": 3720177.0,
      "step": 1995
    },
    {
      "entropy": 6.530578517913819,
      "epoch": 0.17661603673613563,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999120295089202,
      "loss": 6.2971,
      "mean_token_accuracy": 0.14959900006651877,
      "num_tokens": 3728123.0,
      "step": 2000
    },
    {
      "entropy": 6.542397546768188,
      "epoch": 0.177057576827976,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999111467778639,
      "loss": 6.4341,
      "mean_token_accuracy": 0.1391107402741909,
      "num_tokens": 3736869.0,
      "step": 2005
    },
    {
      "entropy": 6.53966269493103,
      "epoch": 0.17749911691981632,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499910259640926,
      "loss": 6.3237,
      "mean_token_accuracy": 0.14294100552797318,
      "num_tokens": 3745830.0,
      "step": 2010
    },
    {
      "entropy": 6.51638708114624,
      "epoch": 0.17794065701165665,
      "grad_norm": 0.984375,
      "learning_rate": 0.000499909368098124,
      "loss": 6.3564,
      "mean_token_accuracy": 0.13725989684462547,
      "num_tokens": 3755019.0,
      "step": 2015
    },
    {
      "entropy": 6.511772108078003,
      "epoch": 0.178382197103497,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999084721494754,
      "loss": 6.331,
      "mean_token_accuracy": 0.14094301015138627,
      "num_tokens": 3764814.0,
      "step": 2020
    },
    {
      "entropy": 6.452651882171631,
      "epoch": 0.17882373719533734,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999075717949978,
      "loss": 6.2835,
      "mean_token_accuracy": 0.14623287171125413,
      "num_tokens": 3774258.0,
      "step": 2025
    },
    {
      "entropy": 6.427514886856079,
      "epoch": 0.17926527728717767,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999066670347089,
      "loss": 6.3158,
      "mean_token_accuracy": 0.1432214617729187,
      "num_tokens": 3783961.0,
      "step": 2030
    },
    {
      "entropy": 6.497758913040161,
      "epoch": 0.179706817379018,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999057578686261,
      "loss": 6.2899,
      "mean_token_accuracy": 0.14475978389382363,
      "num_tokens": 3792417.0,
      "step": 2035
    },
    {
      "entropy": 6.472495174407959,
      "epoch": 0.18014835747085836,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999048442967675,
      "loss": 6.2812,
      "mean_token_accuracy": 0.14342265501618384,
      "num_tokens": 3801479.0,
      "step": 2040
    },
    {
      "entropy": 6.572735977172852,
      "epoch": 0.1805898975626987,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999039263191508,
      "loss": 6.3581,
      "mean_token_accuracy": 0.13960720375180244,
      "num_tokens": 3810799.0,
      "step": 2045
    },
    {
      "entropy": 6.510642623901367,
      "epoch": 0.18103143765453902,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999030039357943,
      "loss": 6.3896,
      "mean_token_accuracy": 0.13761023208498954,
      "num_tokens": 3820966.0,
      "step": 2050
    },
    {
      "entropy": 6.4944600582122805,
      "epoch": 0.18147297774637938,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999020771467158,
      "loss": 6.3647,
      "mean_token_accuracy": 0.1407288655638695,
      "num_tokens": 3829247.0,
      "step": 2055
    },
    {
      "entropy": 6.609596920013428,
      "epoch": 0.1819145178382197,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999011459519335,
      "loss": 6.3036,
      "mean_token_accuracy": 0.14792885035276412,
      "num_tokens": 3838114.0,
      "step": 2060
    },
    {
      "entropy": 6.411144733428955,
      "epoch": 0.18235605793006004,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999002103514655,
      "loss": 6.3792,
      "mean_token_accuracy": 0.142097420245409,
      "num_tokens": 3848075.0,
      "step": 2065
    },
    {
      "entropy": 6.5771557807922365,
      "epoch": 0.18279759802190038,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004998992703453304,
      "loss": 6.4163,
      "mean_token_accuracy": 0.14099472612142563,
      "num_tokens": 3857934.0,
      "step": 2070
    },
    {
      "entropy": 6.490412330627441,
      "epoch": 0.18323913811374073,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998983259335466,
      "loss": 6.2769,
      "mean_token_accuracy": 0.15167167782783508,
      "num_tokens": 3866707.0,
      "step": 2075
    },
    {
      "entropy": 6.511580276489258,
      "epoch": 0.18368067820558107,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998973771161324,
      "loss": 6.3342,
      "mean_token_accuracy": 0.14083316251635553,
      "num_tokens": 3875233.0,
      "step": 2080
    },
    {
      "entropy": 6.495502281188965,
      "epoch": 0.1841222182974214,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004998964238931065,
      "loss": 6.3311,
      "mean_token_accuracy": 0.1447795733809471,
      "num_tokens": 3885173.0,
      "step": 2085
    },
    {
      "entropy": 6.511646032333374,
      "epoch": 0.18456375838926176,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998954662644876,
      "loss": 6.3103,
      "mean_token_accuracy": 0.14369324892759322,
      "num_tokens": 3894198.0,
      "step": 2090
    },
    {
      "entropy": 6.465791845321656,
      "epoch": 0.1850052984811021,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998945042302943,
      "loss": 6.3043,
      "mean_token_accuracy": 0.14370952174067497,
      "num_tokens": 3904076.0,
      "step": 2095
    },
    {
      "entropy": 6.529692268371582,
      "epoch": 0.18544683857294242,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998935377905457,
      "loss": 6.414,
      "mean_token_accuracy": 0.13739102557301522,
      "num_tokens": 3913204.0,
      "step": 2100
    },
    {
      "entropy": 6.59678544998169,
      "epoch": 0.18588837866478275,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998925669452605,
      "loss": 6.3665,
      "mean_token_accuracy": 0.14387816935777664,
      "num_tokens": 3922148.0,
      "step": 2105
    },
    {
      "entropy": 6.451368236541748,
      "epoch": 0.1863299187566231,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004998915916944579,
      "loss": 6.336,
      "mean_token_accuracy": 0.14496995508670807,
      "num_tokens": 3931333.0,
      "step": 2110
    },
    {
      "entropy": 6.508979606628418,
      "epoch": 0.18677145884846344,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998906120381568,
      "loss": 6.2869,
      "mean_token_accuracy": 0.14773214906454085,
      "num_tokens": 3941061.0,
      "step": 2115
    },
    {
      "entropy": 6.536288022994995,
      "epoch": 0.18721299894030377,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998896279763766,
      "loss": 6.3841,
      "mean_token_accuracy": 0.13904453590512275,
      "num_tokens": 3950075.0,
      "step": 2120
    },
    {
      "entropy": 6.4915399074554445,
      "epoch": 0.18765453903214413,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998886395091365,
      "loss": 6.2552,
      "mean_token_accuracy": 0.1421836346387863,
      "num_tokens": 3958885.0,
      "step": 2125
    },
    {
      "entropy": 6.379503393173218,
      "epoch": 0.18809607912398446,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998876466364559,
      "loss": 6.3307,
      "mean_token_accuracy": 0.1476896308362484,
      "num_tokens": 3968218.0,
      "step": 2130
    },
    {
      "entropy": 6.450522613525391,
      "epoch": 0.1885376192158248,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998866493583541,
      "loss": 6.2918,
      "mean_token_accuracy": 0.14630952328443528,
      "num_tokens": 3977435.0,
      "step": 2135
    },
    {
      "entropy": 6.4365739822387695,
      "epoch": 0.18897915930766512,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004998856476748509,
      "loss": 6.2738,
      "mean_token_accuracy": 0.14805593118071556,
      "num_tokens": 3986608.0,
      "step": 2140
    },
    {
      "entropy": 6.498980379104614,
      "epoch": 0.18942069939950548,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998846415859656,
      "loss": 6.2719,
      "mean_token_accuracy": 0.14355492740869522,
      "num_tokens": 3996087.0,
      "step": 2145
    },
    {
      "entropy": 6.4271481990814205,
      "epoch": 0.1898622394913458,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998836310917182,
      "loss": 6.3259,
      "mean_token_accuracy": 0.13990697860717774,
      "num_tokens": 4006257.0,
      "step": 2150
    },
    {
      "entropy": 6.432379961013794,
      "epoch": 0.19030377958318614,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004998826161921282,
      "loss": 6.3248,
      "mean_token_accuracy": 0.1525232471525669,
      "num_tokens": 4015904.0,
      "step": 2155
    },
    {
      "entropy": 6.395722532272339,
      "epoch": 0.1907453196750265,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004998815968872157,
      "loss": 6.3357,
      "mean_token_accuracy": 0.14217145144939422,
      "num_tokens": 4025417.0,
      "step": 2160
    },
    {
      "entropy": 6.434484338760376,
      "epoch": 0.19118685976686683,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998805731770007,
      "loss": 6.194,
      "mean_token_accuracy": 0.16098422557115555,
      "num_tokens": 4035181.0,
      "step": 2165
    },
    {
      "entropy": 6.466515350341797,
      "epoch": 0.19162839985870717,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000499879545061503,
      "loss": 6.3656,
      "mean_token_accuracy": 0.14232496544718742,
      "num_tokens": 4045112.0,
      "step": 2170
    },
    {
      "entropy": 6.544062900543213,
      "epoch": 0.1920699399505475,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004998785125407432,
      "loss": 6.468,
      "mean_token_accuracy": 0.13045382052659987,
      "num_tokens": 4054566.0,
      "step": 2175
    },
    {
      "entropy": 6.538053798675537,
      "epoch": 0.19251148004238786,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000499877475614741,
      "loss": 6.2829,
      "mean_token_accuracy": 0.14255435317754744,
      "num_tokens": 4063960.0,
      "step": 2180
    },
    {
      "entropy": 6.318663692474365,
      "epoch": 0.1929530201342282,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004998764342835169,
      "loss": 6.2248,
      "mean_token_accuracy": 0.14946237206459045,
      "num_tokens": 4072620.0,
      "step": 2185
    },
    {
      "entropy": 6.361247396469116,
      "epoch": 0.19339456022606852,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998753885470915,
      "loss": 6.1939,
      "mean_token_accuracy": 0.15080117732286452,
      "num_tokens": 4081590.0,
      "step": 2190
    },
    {
      "entropy": 6.448186588287354,
      "epoch": 0.19383610031790888,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998743384054851,
      "loss": 6.3243,
      "mean_token_accuracy": 0.15017441809177398,
      "num_tokens": 4090758.0,
      "step": 2195
    },
    {
      "entropy": 6.359209203720093,
      "epoch": 0.1942776404097492,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004998732838587183,
      "loss": 6.1452,
      "mean_token_accuracy": 0.1547905460000038,
      "num_tokens": 4099281.0,
      "step": 2200
    },
    {
      "entropy": 6.484091281890869,
      "epoch": 0.19471918050158954,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004998722249068118,
      "loss": 6.2579,
      "mean_token_accuracy": 0.14498598128557205,
      "num_tokens": 4108953.0,
      "step": 2205
    },
    {
      "entropy": 6.412591505050659,
      "epoch": 0.19516072059342987,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004998711615497863,
      "loss": 6.365,
      "mean_token_accuracy": 0.1421401210129261,
      "num_tokens": 4118799.0,
      "step": 2210
    },
    {
      "entropy": 6.515138959884643,
      "epoch": 0.19560226068527023,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004998700937876626,
      "loss": 6.3215,
      "mean_token_accuracy": 0.14281034767627715,
      "num_tokens": 4127862.0,
      "step": 2215
    },
    {
      "entropy": 6.504071760177612,
      "epoch": 0.19604380077711056,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998690216204615,
      "loss": 6.4025,
      "mean_token_accuracy": 0.13254041969776154,
      "num_tokens": 4139029.0,
      "step": 2220
    },
    {
      "entropy": 6.468188190460205,
      "epoch": 0.1964853408689509,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998679450482043,
      "loss": 6.2211,
      "mean_token_accuracy": 0.15794688761234282,
      "num_tokens": 4148257.0,
      "step": 2225
    },
    {
      "entropy": 6.399339389801026,
      "epoch": 0.19692688096079125,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000499866864070912,
      "loss": 6.1777,
      "mean_token_accuracy": 0.14184310287237167,
      "num_tokens": 4157626.0,
      "step": 2230
    },
    {
      "entropy": 6.394889450073242,
      "epoch": 0.19736842105263158,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998657786886056,
      "loss": 6.4048,
      "mean_token_accuracy": 0.1460250087082386,
      "num_tokens": 4166804.0,
      "step": 2235
    },
    {
      "entropy": 6.561651229858398,
      "epoch": 0.1978099611444719,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998646889013066,
      "loss": 6.2711,
      "mean_token_accuracy": 0.14743182435631752,
      "num_tokens": 4175701.0,
      "step": 2240
    },
    {
      "entropy": 6.471790933609009,
      "epoch": 0.19825150123631224,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004998635947090362,
      "loss": 6.3589,
      "mean_token_accuracy": 0.1456327997148037,
      "num_tokens": 4184711.0,
      "step": 2245
    },
    {
      "entropy": 6.440358591079712,
      "epoch": 0.1986930413281526,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004998624961118158,
      "loss": 6.3148,
      "mean_token_accuracy": 0.1504688784480095,
      "num_tokens": 4193931.0,
      "step": 2250
    },
    {
      "entropy": 6.443251514434815,
      "epoch": 0.19913458141999293,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499861393109667,
      "loss": 6.1933,
      "mean_token_accuracy": 0.15764214992523193,
      "num_tokens": 4203000.0,
      "step": 2255
    },
    {
      "entropy": 6.3125227928161625,
      "epoch": 0.19957612151183327,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998602857026114,
      "loss": 6.214,
      "mean_token_accuracy": 0.1494602829217911,
      "num_tokens": 4211977.0,
      "step": 2260
    },
    {
      "entropy": 6.4371459007263185,
      "epoch": 0.20001766160367362,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998591738906708,
      "loss": 6.2855,
      "mean_token_accuracy": 0.15524563640356065,
      "num_tokens": 4220375.0,
      "step": 2265
    },
    {
      "entropy": 6.506232404708863,
      "epoch": 0.20045920169551396,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998580576738668,
      "loss": 6.3022,
      "mean_token_accuracy": 0.14764061123132705,
      "num_tokens": 4230506.0,
      "step": 2270
    },
    {
      "entropy": 6.4268731594085695,
      "epoch": 0.2009007417873543,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998569370522213,
      "loss": 6.2711,
      "mean_token_accuracy": 0.1487172946333885,
      "num_tokens": 4240235.0,
      "step": 2275
    },
    {
      "entropy": 6.3987833023071286,
      "epoch": 0.20134228187919462,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004998558120257563,
      "loss": 6.3105,
      "mean_token_accuracy": 0.1501203641295433,
      "num_tokens": 4249775.0,
      "step": 2280
    },
    {
      "entropy": 6.428017330169678,
      "epoch": 0.20178382197103498,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998546825944938,
      "loss": 6.2596,
      "mean_token_accuracy": 0.14586075246334076,
      "num_tokens": 4258523.0,
      "step": 2285
    },
    {
      "entropy": 6.456875276565552,
      "epoch": 0.2022253620628753,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499853548758456,
      "loss": 6.2549,
      "mean_token_accuracy": 0.14920931309461594,
      "num_tokens": 4267683.0,
      "step": 2290
    },
    {
      "entropy": 6.502974796295166,
      "epoch": 0.20266690215471564,
      "grad_norm": 1.5,
      "learning_rate": 0.000499852410517665,
      "loss": 6.3396,
      "mean_token_accuracy": 0.1415552243590355,
      "num_tokens": 4276903.0,
      "step": 2295
    },
    {
      "entropy": 6.390913295745849,
      "epoch": 0.203108442246556,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004998512678721431,
      "loss": 6.2765,
      "mean_token_accuracy": 0.14179632887244226,
      "num_tokens": 4287257.0,
      "step": 2300
    },
    {
      "entropy": 6.467247819900512,
      "epoch": 0.20354998233839633,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499850120821913,
      "loss": 6.2994,
      "mean_token_accuracy": 0.1429171495139599,
      "num_tokens": 4297345.0,
      "step": 2305
    },
    {
      "entropy": 6.504390478134155,
      "epoch": 0.20399152243023666,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998489693669967,
      "loss": 6.2774,
      "mean_token_accuracy": 0.14350235760211943,
      "num_tokens": 4306533.0,
      "step": 2310
    },
    {
      "entropy": 6.430126476287842,
      "epoch": 0.204433062522077,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000499847813507417,
      "loss": 6.3142,
      "mean_token_accuracy": 0.14808136820793152,
      "num_tokens": 4316338.0,
      "step": 2315
    },
    {
      "entropy": 6.418506526947022,
      "epoch": 0.20487460261391735,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004998466532431966,
      "loss": 6.2461,
      "mean_token_accuracy": 0.14408156126737595,
      "num_tokens": 4326585.0,
      "step": 2320
    },
    {
      "entropy": 6.502306318283081,
      "epoch": 0.20531614270575768,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004998454885743581,
      "loss": 6.4021,
      "mean_token_accuracy": 0.1425124689936638,
      "num_tokens": 4336490.0,
      "step": 2325
    },
    {
      "entropy": 6.4134602546691895,
      "epoch": 0.205757682797598,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004998443195009242,
      "loss": 6.266,
      "mean_token_accuracy": 0.14605539664626122,
      "num_tokens": 4346264.0,
      "step": 2330
    },
    {
      "entropy": 6.444290590286255,
      "epoch": 0.20619922288943837,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998431460229182,
      "loss": 6.261,
      "mean_token_accuracy": 0.14759050458669662,
      "num_tokens": 4355102.0,
      "step": 2335
    },
    {
      "entropy": 6.518517303466797,
      "epoch": 0.2066407629812787,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998419681403627,
      "loss": 6.4309,
      "mean_token_accuracy": 0.139105836302042,
      "num_tokens": 4365569.0,
      "step": 2340
    },
    {
      "entropy": 6.428114986419677,
      "epoch": 0.20708230307311903,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004998407858532809,
      "loss": 6.2633,
      "mean_token_accuracy": 0.1455647498369217,
      "num_tokens": 4375437.0,
      "step": 2345
    },
    {
      "entropy": 6.348156213760376,
      "epoch": 0.20752384316495937,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499839599161696,
      "loss": 6.2444,
      "mean_token_accuracy": 0.15116954892873763,
      "num_tokens": 4384420.0,
      "step": 2350
    },
    {
      "entropy": 6.430051231384278,
      "epoch": 0.20796538325679972,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998384080656314,
      "loss": 6.1993,
      "mean_token_accuracy": 0.15055150538682938,
      "num_tokens": 4393730.0,
      "step": 2355
    },
    {
      "entropy": 6.349585485458374,
      "epoch": 0.20840692334864006,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00049983721256511,
      "loss": 6.1842,
      "mean_token_accuracy": 0.14877034425735475,
      "num_tokens": 4402731.0,
      "step": 2360
    },
    {
      "entropy": 6.3974512100219725,
      "epoch": 0.2088484634404804,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998360126601556,
      "loss": 6.2405,
      "mean_token_accuracy": 0.1482672870159149,
      "num_tokens": 4411606.0,
      "step": 2365
    },
    {
      "entropy": 6.3687732219696045,
      "epoch": 0.20929000353232075,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998348083507916,
      "loss": 6.3177,
      "mean_token_accuracy": 0.14799434095621108,
      "num_tokens": 4421685.0,
      "step": 2370
    },
    {
      "entropy": 6.44785623550415,
      "epoch": 0.20973154362416108,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998335996370416,
      "loss": 6.2883,
      "mean_token_accuracy": 0.14761801213026046,
      "num_tokens": 4431765.0,
      "step": 2375
    },
    {
      "entropy": 6.392046165466309,
      "epoch": 0.2101730837160014,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004998323865189291,
      "loss": 6.2439,
      "mean_token_accuracy": 0.14551597088575363,
      "num_tokens": 4441191.0,
      "step": 2380
    },
    {
      "entropy": 6.444688892364502,
      "epoch": 0.21061462380784174,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998311689964781,
      "loss": 6.3302,
      "mean_token_accuracy": 0.138855642080307,
      "num_tokens": 4450156.0,
      "step": 2385
    },
    {
      "entropy": 6.510595750808716,
      "epoch": 0.2110561638996821,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998299470697125,
      "loss": 6.3199,
      "mean_token_accuracy": 0.15451483130455018,
      "num_tokens": 4459466.0,
      "step": 2390
    },
    {
      "entropy": 6.423422145843506,
      "epoch": 0.21149770399152243,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004998287207386559,
      "loss": 6.3232,
      "mean_token_accuracy": 0.14224404096603394,
      "num_tokens": 4468539.0,
      "step": 2395
    },
    {
      "entropy": 6.416548156738282,
      "epoch": 0.21193924408336276,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998274900033326,
      "loss": 6.2395,
      "mean_token_accuracy": 0.15238402634859086,
      "num_tokens": 4477579.0,
      "step": 2400
    },
    {
      "entropy": 6.291043853759765,
      "epoch": 0.21238078417520312,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004998262548637667,
      "loss": 6.1989,
      "mean_token_accuracy": 0.15245457291603087,
      "num_tokens": 4486800.0,
      "step": 2405
    },
    {
      "entropy": 6.308730173110962,
      "epoch": 0.21282232426704345,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998250153199822,
      "loss": 6.1484,
      "mean_token_accuracy": 0.15403092205524443,
      "num_tokens": 4495985.0,
      "step": 2410
    },
    {
      "entropy": 6.466254568099975,
      "epoch": 0.21326386435888378,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004998237713720036,
      "loss": 6.3239,
      "mean_token_accuracy": 0.14338542893528938,
      "num_tokens": 4504944.0,
      "step": 2415
    },
    {
      "entropy": 6.381775999069214,
      "epoch": 0.2137054044507241,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004998225230198552,
      "loss": 6.1987,
      "mean_token_accuracy": 0.1538494348526001,
      "num_tokens": 4515402.0,
      "step": 2420
    },
    {
      "entropy": 6.330594635009765,
      "epoch": 0.21414694454256447,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004998212702635614,
      "loss": 6.2685,
      "mean_token_accuracy": 0.14354601204395295,
      "num_tokens": 4525009.0,
      "step": 2425
    },
    {
      "entropy": 6.5298998832702635,
      "epoch": 0.2145884846344048,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998200131031469,
      "loss": 6.3256,
      "mean_token_accuracy": 0.14529723376035691,
      "num_tokens": 4534460.0,
      "step": 2430
    },
    {
      "entropy": 6.389695167541504,
      "epoch": 0.21503002472624513,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004998187515386361,
      "loss": 6.1518,
      "mean_token_accuracy": 0.1601933628320694,
      "num_tokens": 4543748.0,
      "step": 2435
    },
    {
      "entropy": 6.282786703109741,
      "epoch": 0.2154715648180855,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004998174855700538,
      "loss": 6.2753,
      "mean_token_accuracy": 0.15319369435310365,
      "num_tokens": 4552722.0,
      "step": 2440
    },
    {
      "entropy": 6.487105512619019,
      "epoch": 0.21591310490992582,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998162151974248,
      "loss": 6.1183,
      "mean_token_accuracy": 0.15194425731897354,
      "num_tokens": 4561607.0,
      "step": 2445
    },
    {
      "entropy": 6.372227382659912,
      "epoch": 0.21635464500176615,
      "grad_norm": 0.859375,
      "learning_rate": 0.000499814940420774,
      "loss": 6.4008,
      "mean_token_accuracy": 0.13625949174165725,
      "num_tokens": 4572524.0,
      "step": 2450
    },
    {
      "entropy": 6.387830066680908,
      "epoch": 0.21679618509360649,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004998136612401266,
      "loss": 6.2112,
      "mean_token_accuracy": 0.1517648920416832,
      "num_tokens": 4581601.0,
      "step": 2455
    },
    {
      "entropy": 6.404532241821289,
      "epoch": 0.21723772518544684,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004998123776555071,
      "loss": 6.2609,
      "mean_token_accuracy": 0.1447027012705803,
      "num_tokens": 4591795.0,
      "step": 2460
    },
    {
      "entropy": 6.37783408164978,
      "epoch": 0.21767926527728718,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004998110896669412,
      "loss": 6.2334,
      "mean_token_accuracy": 0.14586359560489653,
      "num_tokens": 4600745.0,
      "step": 2465
    },
    {
      "entropy": 6.4204151153564455,
      "epoch": 0.2181208053691275,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004998097972744539,
      "loss": 6.2744,
      "mean_token_accuracy": 0.1438829004764557,
      "num_tokens": 4610490.0,
      "step": 2470
    },
    {
      "entropy": 6.344004535675049,
      "epoch": 0.21856234546096787,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998085004780705,
      "loss": 6.1917,
      "mean_token_accuracy": 0.1555175483226776,
      "num_tokens": 4619511.0,
      "step": 2475
    },
    {
      "entropy": 6.415195465087891,
      "epoch": 0.2190038855528082,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998071992778164,
      "loss": 6.2515,
      "mean_token_accuracy": 0.14207181110978126,
      "num_tokens": 4628186.0,
      "step": 2480
    },
    {
      "entropy": 6.438273286819458,
      "epoch": 0.21944542564464853,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499805893673717,
      "loss": 6.2691,
      "mean_token_accuracy": 0.14731226563453675,
      "num_tokens": 4637431.0,
      "step": 2485
    },
    {
      "entropy": 6.356544446945191,
      "epoch": 0.21988696573648886,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004998045836657982,
      "loss": 6.1291,
      "mean_token_accuracy": 0.15215693414211273,
      "num_tokens": 4646627.0,
      "step": 2490
    },
    {
      "entropy": 6.363220739364624,
      "epoch": 0.22032850582832922,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998032692540853,
      "loss": 6.2228,
      "mean_token_accuracy": 0.14926744401454925,
      "num_tokens": 4656095.0,
      "step": 2495
    },
    {
      "entropy": 6.444636392593384,
      "epoch": 0.22077004592016955,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004998019504386044,
      "loss": 6.2983,
      "mean_token_accuracy": 0.1443503238260746,
      "num_tokens": 4665807.0,
      "step": 2500
    },
    {
      "entropy": 6.39345760345459,
      "epoch": 0.22121158601200988,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998006272193809,
      "loss": 6.2515,
      "mean_token_accuracy": 0.14276793897151946,
      "num_tokens": 4674459.0,
      "step": 2505
    },
    {
      "entropy": 6.420822715759277,
      "epoch": 0.22165312610385024,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004997992995964412,
      "loss": 6.3335,
      "mean_token_accuracy": 0.14485810697078705,
      "num_tokens": 4684063.0,
      "step": 2510
    },
    {
      "entropy": 6.430441427230835,
      "epoch": 0.22209466619569057,
      "grad_norm": 1.125,
      "learning_rate": 0.0004997979675698109,
      "loss": 6.2237,
      "mean_token_accuracy": 0.1505151480436325,
      "num_tokens": 4692807.0,
      "step": 2515
    },
    {
      "entropy": 6.493818712234497,
      "epoch": 0.2225362062875309,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004997966311395164,
      "loss": 6.1509,
      "mean_token_accuracy": 0.15266127437353133,
      "num_tokens": 4701100.0,
      "step": 2520
    },
    {
      "entropy": 6.261106777191162,
      "epoch": 0.22297774637937123,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004997952903055836,
      "loss": 6.2014,
      "mean_token_accuracy": 0.1452627256512642,
      "num_tokens": 4710697.0,
      "step": 2525
    },
    {
      "entropy": 6.3538493633270265,
      "epoch": 0.2234192864712116,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499793945068039,
      "loss": 6.1951,
      "mean_token_accuracy": 0.14937613308429717,
      "num_tokens": 4718745.0,
      "step": 2530
    },
    {
      "entropy": 6.333649778366089,
      "epoch": 0.22386082656305192,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004997925954269088,
      "loss": 6.173,
      "mean_token_accuracy": 0.15181030780076982,
      "num_tokens": 4728056.0,
      "step": 2535
    },
    {
      "entropy": 6.434943294525146,
      "epoch": 0.22430236665489225,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004997912413822196,
      "loss": 6.2935,
      "mean_token_accuracy": 0.15063572451472282,
      "num_tokens": 4737605.0,
      "step": 2540
    },
    {
      "entropy": 6.293275022506714,
      "epoch": 0.2247439067467326,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004997898829339979,
      "loss": 6.114,
      "mean_token_accuracy": 0.16079342663288115,
      "num_tokens": 4746168.0,
      "step": 2545
    },
    {
      "entropy": 6.297460889816284,
      "epoch": 0.22518544683857294,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00049978852008227,
      "loss": 6.1821,
      "mean_token_accuracy": 0.1518242985010147,
      "num_tokens": 4755072.0,
      "step": 2550
    },
    {
      "entropy": 6.327752351760864,
      "epoch": 0.22562698693041328,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000499787152827063,
      "loss": 6.192,
      "mean_token_accuracy": 0.15824129432439804,
      "num_tokens": 4764580.0,
      "step": 2555
    },
    {
      "entropy": 6.290549898147583,
      "epoch": 0.2260685270222536,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997857811684035,
      "loss": 6.2145,
      "mean_token_accuracy": 0.15303274542093276,
      "num_tokens": 4774135.0,
      "step": 2560
    },
    {
      "entropy": 6.410637426376343,
      "epoch": 0.22651006711409397,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004997844051063183,
      "loss": 6.3043,
      "mean_token_accuracy": 0.14438919797539712,
      "num_tokens": 4784733.0,
      "step": 2565
    },
    {
      "entropy": 6.41192512512207,
      "epoch": 0.2269516072059343,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004997830246408346,
      "loss": 6.2387,
      "mean_token_accuracy": 0.15428498163819312,
      "num_tokens": 4795327.0,
      "step": 2570
    },
    {
      "entropy": 6.278723859786988,
      "epoch": 0.22739314729777463,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004997816397719791,
      "loss": 6.2157,
      "mean_token_accuracy": 0.14635045304894448,
      "num_tokens": 4804314.0,
      "step": 2575
    },
    {
      "entropy": 6.441279935836792,
      "epoch": 0.227834687389615,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004997802504997792,
      "loss": 6.303,
      "mean_token_accuracy": 0.14009749442338942,
      "num_tokens": 4813637.0,
      "step": 2580
    },
    {
      "entropy": 6.366417980194091,
      "epoch": 0.22827622748145532,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004997788568242621,
      "loss": 6.1695,
      "mean_token_accuracy": 0.15038952380418777,
      "num_tokens": 4823094.0,
      "step": 2585
    },
    {
      "entropy": 6.292829847335815,
      "epoch": 0.22871776757329565,
      "grad_norm": 0.83984375,
      "learning_rate": 0.000499777458745455,
      "loss": 6.1123,
      "mean_token_accuracy": 0.1579282909631729,
      "num_tokens": 4833199.0,
      "step": 2590
    },
    {
      "entropy": 6.3866489887237545,
      "epoch": 0.22915930766513598,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004997760562633853,
      "loss": 6.2102,
      "mean_token_accuracy": 0.15025776177644729,
      "num_tokens": 4842970.0,
      "step": 2595
    },
    {
      "entropy": 6.37343578338623,
      "epoch": 0.22960084775697634,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004997746493780804,
      "loss": 6.2792,
      "mean_token_accuracy": 0.14459849298000335,
      "num_tokens": 4852043.0,
      "step": 2600
    },
    {
      "entropy": 6.356409358978271,
      "epoch": 0.23004238784881667,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000499773238089568,
      "loss": 6.1844,
      "mean_token_accuracy": 0.14483380317687988,
      "num_tokens": 4862232.0,
      "step": 2605
    },
    {
      "entropy": 6.343675518035889,
      "epoch": 0.230483927940657,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997718223978758,
      "loss": 6.1204,
      "mean_token_accuracy": 0.14981301128864288,
      "num_tokens": 4871186.0,
      "step": 2610
    },
    {
      "entropy": 6.302071237564087,
      "epoch": 0.23092546803249736,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004997704023030315,
      "loss": 6.2134,
      "mean_token_accuracy": 0.1568485602736473,
      "num_tokens": 4879974.0,
      "step": 2615
    },
    {
      "entropy": 6.37543830871582,
      "epoch": 0.2313670081243377,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004997689778050627,
      "loss": 6.2599,
      "mean_token_accuracy": 0.15443349331617356,
      "num_tokens": 4890300.0,
      "step": 2620
    },
    {
      "entropy": 6.287172794342041,
      "epoch": 0.23180854821617802,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004997675489039975,
      "loss": 6.2494,
      "mean_token_accuracy": 0.1468710109591484,
      "num_tokens": 4900428.0,
      "step": 2625
    },
    {
      "entropy": 6.366749954223633,
      "epoch": 0.23225008830801835,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004997661155998638,
      "loss": 6.2476,
      "mean_token_accuracy": 0.15147435963153838,
      "num_tokens": 4910092.0,
      "step": 2630
    },
    {
      "entropy": 6.374930906295776,
      "epoch": 0.2326916283998587,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004997646778926898,
      "loss": 6.2313,
      "mean_token_accuracy": 0.14748895019292832,
      "num_tokens": 4919593.0,
      "step": 2635
    },
    {
      "entropy": 6.320575475692749,
      "epoch": 0.23313316849169904,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004997632357825035,
      "loss": 6.1889,
      "mean_token_accuracy": 0.15245203971862792,
      "num_tokens": 4929098.0,
      "step": 2640
    },
    {
      "entropy": 6.37444224357605,
      "epoch": 0.23357470858353938,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997617892693333,
      "loss": 6.2685,
      "mean_token_accuracy": 0.14796690195798873,
      "num_tokens": 4938265.0,
      "step": 2645
    },
    {
      "entropy": 6.381702041625976,
      "epoch": 0.23401624867537973,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004997603383532075,
      "loss": 6.2157,
      "mean_token_accuracy": 0.15212348997592925,
      "num_tokens": 4946694.0,
      "step": 2650
    },
    {
      "entropy": 6.345135974884033,
      "epoch": 0.23445778876722007,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004997588830341545,
      "loss": 6.2007,
      "mean_token_accuracy": 0.1495561882853508,
      "num_tokens": 4955296.0,
      "step": 2655
    },
    {
      "entropy": 6.262336206436157,
      "epoch": 0.2348993288590604,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004997574233122028,
      "loss": 6.1783,
      "mean_token_accuracy": 0.15376680493354797,
      "num_tokens": 4964409.0,
      "step": 2660
    },
    {
      "entropy": 6.344006538391113,
      "epoch": 0.23534086895090076,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004997559591873809,
      "loss": 6.1848,
      "mean_token_accuracy": 0.15347870737314223,
      "num_tokens": 4973449.0,
      "step": 2665
    },
    {
      "entropy": 6.325773239135742,
      "epoch": 0.2357824090427411,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004997544906597178,
      "loss": 6.1774,
      "mean_token_accuracy": 0.1539867863059044,
      "num_tokens": 4983057.0,
      "step": 2670
    },
    {
      "entropy": 6.2636034965515135,
      "epoch": 0.23622394913458142,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004997530177292418,
      "loss": 6.2376,
      "mean_token_accuracy": 0.14284619241952895,
      "num_tokens": 4991950.0,
      "step": 2675
    },
    {
      "entropy": 6.425727081298828,
      "epoch": 0.23666548922642175,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004997515403959823,
      "loss": 6.1908,
      "mean_token_accuracy": 0.15085240155458451,
      "num_tokens": 5001042.0,
      "step": 2680
    },
    {
      "entropy": 6.364514875411987,
      "epoch": 0.2371070293182621,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997500586599677,
      "loss": 6.1239,
      "mean_token_accuracy": 0.15567631870508195,
      "num_tokens": 5009827.0,
      "step": 2685
    },
    {
      "entropy": 6.238472986221313,
      "epoch": 0.23754856941010244,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004997485725212274,
      "loss": 6.1721,
      "mean_token_accuracy": 0.15332650095224382,
      "num_tokens": 5018708.0,
      "step": 2690
    },
    {
      "entropy": 6.277514266967773,
      "epoch": 0.23799010950194277,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004997470819797903,
      "loss": 6.0688,
      "mean_token_accuracy": 0.1563563570380211,
      "num_tokens": 5027522.0,
      "step": 2695
    },
    {
      "entropy": 6.287735462188721,
      "epoch": 0.23843164959378313,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004997455870356857,
      "loss": 6.1948,
      "mean_token_accuracy": 0.15198080092668534,
      "num_tokens": 5035755.0,
      "step": 2700
    },
    {
      "entropy": 6.307868242263794,
      "epoch": 0.23887318968562346,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997440876889429,
      "loss": 6.1377,
      "mean_token_accuracy": 0.15334794968366622,
      "num_tokens": 5045289.0,
      "step": 2705
    },
    {
      "entropy": 6.278635931015015,
      "epoch": 0.2393147297774638,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004997425839395913,
      "loss": 6.1641,
      "mean_token_accuracy": 0.1568767547607422,
      "num_tokens": 5053774.0,
      "step": 2710
    },
    {
      "entropy": 6.4377528667449955,
      "epoch": 0.23975626986930412,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004997410757876602,
      "loss": 6.2444,
      "mean_token_accuracy": 0.14437243491411209,
      "num_tokens": 5062911.0,
      "step": 2715
    },
    {
      "entropy": 6.3234247207641605,
      "epoch": 0.24019780996114448,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004997395632331793,
      "loss": 6.1091,
      "mean_token_accuracy": 0.15637961626052857,
      "num_tokens": 5072107.0,
      "step": 2720
    },
    {
      "entropy": 6.226778221130371,
      "epoch": 0.2406393500529848,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997380462761781,
      "loss": 6.0586,
      "mean_token_accuracy": 0.15822814106941224,
      "num_tokens": 5080588.0,
      "step": 2725
    },
    {
      "entropy": 6.362985134124756,
      "epoch": 0.24108089014482514,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004997365249166864,
      "loss": 6.2812,
      "mean_token_accuracy": 0.148160120844841,
      "num_tokens": 5090262.0,
      "step": 2730
    },
    {
      "entropy": 6.364387369155883,
      "epoch": 0.2415224302366655,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004997349991547342,
      "loss": 6.1672,
      "mean_token_accuracy": 0.15980131253600122,
      "num_tokens": 5099285.0,
      "step": 2735
    },
    {
      "entropy": 6.311175918579101,
      "epoch": 0.24196397032850583,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004997334689903509,
      "loss": 6.2297,
      "mean_token_accuracy": 0.1506420761346817,
      "num_tokens": 5109115.0,
      "step": 2740
    },
    {
      "entropy": 6.381357097625733,
      "epoch": 0.24240551042034617,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997319344235668,
      "loss": 6.2471,
      "mean_token_accuracy": 0.15130761936306952,
      "num_tokens": 5117977.0,
      "step": 2745
    },
    {
      "entropy": 6.371263027191162,
      "epoch": 0.2428470505121865,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000499730395454412,
      "loss": 6.1953,
      "mean_token_accuracy": 0.15422819852828978,
      "num_tokens": 5127457.0,
      "step": 2750
    },
    {
      "entropy": 6.326935291290283,
      "epoch": 0.24328859060402686,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004997288520829166,
      "loss": 6.2614,
      "mean_token_accuracy": 0.14421235620975495,
      "num_tokens": 5137310.0,
      "step": 2755
    },
    {
      "entropy": 6.359060192108155,
      "epoch": 0.2437301306958672,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997273043091107,
      "loss": 6.1923,
      "mean_token_accuracy": 0.1464222900569439,
      "num_tokens": 5146963.0,
      "step": 2760
    },
    {
      "entropy": 6.268268156051636,
      "epoch": 0.24417167078770752,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997257521330248,
      "loss": 6.1424,
      "mean_token_accuracy": 0.14788112491369249,
      "num_tokens": 5155521.0,
      "step": 2765
    },
    {
      "entropy": 6.351244401931763,
      "epoch": 0.24461321087954788,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997241955546892,
      "loss": 6.1424,
      "mean_token_accuracy": 0.1483145385980606,
      "num_tokens": 5165182.0,
      "step": 2770
    },
    {
      "entropy": 6.31810302734375,
      "epoch": 0.2450547509713882,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004997226345741343,
      "loss": 6.1684,
      "mean_token_accuracy": 0.15033391118049622,
      "num_tokens": 5175511.0,
      "step": 2775
    },
    {
      "entropy": 6.3378277778625485,
      "epoch": 0.24549629106322854,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499721069191391,
      "loss": 6.1013,
      "mean_token_accuracy": 0.1578337162733078,
      "num_tokens": 5184772.0,
      "step": 2780
    },
    {
      "entropy": 6.310634183883667,
      "epoch": 0.24593783115506887,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004997194994064896,
      "loss": 6.1866,
      "mean_token_accuracy": 0.1496903583407402,
      "num_tokens": 5195136.0,
      "step": 2785
    },
    {
      "entropy": 6.332416534423828,
      "epoch": 0.24637937124690923,
      "grad_norm": 0.96875,
      "learning_rate": 0.000499717925219461,
      "loss": 6.1167,
      "mean_token_accuracy": 0.15987685918807984,
      "num_tokens": 5203163.0,
      "step": 2790
    },
    {
      "entropy": 6.345263051986694,
      "epoch": 0.24682091133874956,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004997163466303362,
      "loss": 6.2061,
      "mean_token_accuracy": 0.1503751039505005,
      "num_tokens": 5213233.0,
      "step": 2795
    },
    {
      "entropy": 6.292828702926636,
      "epoch": 0.2472624514305899,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000499714763639146,
      "loss": 6.1372,
      "mean_token_accuracy": 0.15086597949266434,
      "num_tokens": 5222940.0,
      "step": 2800
    },
    {
      "entropy": 6.270041036605835,
      "epoch": 0.24770399152243025,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997131762459211,
      "loss": 6.1676,
      "mean_token_accuracy": 0.15103059709072114,
      "num_tokens": 5232263.0,
      "step": 2805
    },
    {
      "entropy": 6.331594228744507,
      "epoch": 0.24814553161427058,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997115844506932,
      "loss": 6.1405,
      "mean_token_accuracy": 0.15087736994028092,
      "num_tokens": 5241536.0,
      "step": 2810
    },
    {
      "entropy": 6.34823522567749,
      "epoch": 0.2485870717061109,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997099882534929,
      "loss": 6.1704,
      "mean_token_accuracy": 0.14839458093047142,
      "num_tokens": 5250702.0,
      "step": 2815
    },
    {
      "entropy": 6.377634048461914,
      "epoch": 0.24902861179795124,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004997083876543519,
      "loss": 6.1759,
      "mean_token_accuracy": 0.1505747467279434,
      "num_tokens": 5259811.0,
      "step": 2820
    },
    {
      "entropy": 6.377023410797119,
      "epoch": 0.2494701518897916,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004997067826533014,
      "loss": 6.2543,
      "mean_token_accuracy": 0.1441645845770836,
      "num_tokens": 5270518.0,
      "step": 2825
    },
    {
      "entropy": 6.403309440612793,
      "epoch": 0.24991169198163193,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004997051732503726,
      "loss": 6.1311,
      "mean_token_accuracy": 0.15377675592899323,
      "num_tokens": 5279538.0,
      "step": 2830
    },
    {
      "entropy": 6.22372031211853,
      "epoch": 0.25035323207347226,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004997035594455975,
      "loss": 6.1674,
      "mean_token_accuracy": 0.1481279417872429,
      "num_tokens": 5289633.0,
      "step": 2835
    },
    {
      "entropy": 6.387079381942749,
      "epoch": 0.2507947721653126,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997019412390074,
      "loss": 6.2575,
      "mean_token_accuracy": 0.15361315459012986,
      "num_tokens": 5299148.0,
      "step": 2840
    },
    {
      "entropy": 6.307563066482544,
      "epoch": 0.2512363122571529,
      "grad_norm": 0.90625,
      "learning_rate": 0.000499700318630634,
      "loss": 6.1357,
      "mean_token_accuracy": 0.15210598707199097,
      "num_tokens": 5309090.0,
      "step": 2845
    },
    {
      "entropy": 6.273115110397339,
      "epoch": 0.2516778523489933,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004996986916205092,
      "loss": 6.2278,
      "mean_token_accuracy": 0.15018689632415771,
      "num_tokens": 5318798.0,
      "step": 2850
    },
    {
      "entropy": 6.345541620254517,
      "epoch": 0.25211939244083365,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996970602086648,
      "loss": 6.0964,
      "mean_token_accuracy": 0.15610066950321197,
      "num_tokens": 5327915.0,
      "step": 2855
    },
    {
      "entropy": 6.211011552810669,
      "epoch": 0.252560932532674,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004996954243951327,
      "loss": 6.1209,
      "mean_token_accuracy": 0.15709759593009948,
      "num_tokens": 5336970.0,
      "step": 2860
    },
    {
      "entropy": 6.323676109313965,
      "epoch": 0.2530024726245143,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004996937841799451,
      "loss": 6.0769,
      "mean_token_accuracy": 0.15997228324413298,
      "num_tokens": 5345167.0,
      "step": 2865
    },
    {
      "entropy": 6.199875545501709,
      "epoch": 0.25344401271635464,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004996921395631342,
      "loss": 6.071,
      "mean_token_accuracy": 0.1565924033522606,
      "num_tokens": 5353399.0,
      "step": 2870
    },
    {
      "entropy": 6.27602071762085,
      "epoch": 0.25388555280819497,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499690490544732,
      "loss": 6.1828,
      "mean_token_accuracy": 0.1513873428106308,
      "num_tokens": 5363224.0,
      "step": 2875
    },
    {
      "entropy": 6.339487314224243,
      "epoch": 0.2543270929000353,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004996888371247707,
      "loss": 6.1923,
      "mean_token_accuracy": 0.14602998048067092,
      "num_tokens": 5372274.0,
      "step": 2880
    },
    {
      "entropy": 6.343375635147095,
      "epoch": 0.2547686329918757,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000499687179303283,
      "loss": 6.1815,
      "mean_token_accuracy": 0.15472851544618607,
      "num_tokens": 5380240.0,
      "step": 2885
    },
    {
      "entropy": 6.228068780899048,
      "epoch": 0.255210173083716,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004996855170803012,
      "loss": 6.0344,
      "mean_token_accuracy": 0.15991738438606262,
      "num_tokens": 5389390.0,
      "step": 2890
    },
    {
      "entropy": 6.28302755355835,
      "epoch": 0.25565171317555635,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004996838504558581,
      "loss": 6.1902,
      "mean_token_accuracy": 0.1521752417087555,
      "num_tokens": 5399425.0,
      "step": 2895
    },
    {
      "entropy": 6.355008220672607,
      "epoch": 0.2560932532673967,
      "grad_norm": 0.96875,
      "learning_rate": 0.000499682179429986,
      "loss": 6.2011,
      "mean_token_accuracy": 0.1508815959095955,
      "num_tokens": 5408717.0,
      "step": 2900
    },
    {
      "entropy": 6.280868196487427,
      "epoch": 0.256534793359237,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004996805040027178,
      "loss": 6.1368,
      "mean_token_accuracy": 0.14726324751973152,
      "num_tokens": 5418475.0,
      "step": 2905
    },
    {
      "entropy": 6.2741169929504395,
      "epoch": 0.25697633345107734,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004996788241740863,
      "loss": 6.1707,
      "mean_token_accuracy": 0.1536627657711506,
      "num_tokens": 5428403.0,
      "step": 2910
    },
    {
      "entropy": 6.269294166564942,
      "epoch": 0.2574178735429177,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004996771399441243,
      "loss": 6.2226,
      "mean_token_accuracy": 0.14502449780702592,
      "num_tokens": 5437347.0,
      "step": 2915
    },
    {
      "entropy": 6.407036876678466,
      "epoch": 0.25785941363475806,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004996754513128652,
      "loss": 6.1309,
      "mean_token_accuracy": 0.15732353776693345,
      "num_tokens": 5446804.0,
      "step": 2920
    },
    {
      "entropy": 6.222485494613648,
      "epoch": 0.2583009537265984,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004996737582803416,
      "loss": 6.0819,
      "mean_token_accuracy": 0.15406210720539093,
      "num_tokens": 5455888.0,
      "step": 2925
    },
    {
      "entropy": 6.3086316108703615,
      "epoch": 0.2587424938184387,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004996720608465868,
      "loss": 6.0772,
      "mean_token_accuracy": 0.15540871918201446,
      "num_tokens": 5463977.0,
      "step": 2930
    },
    {
      "entropy": 6.277329921722412,
      "epoch": 0.25918403391027905,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004996703590116342,
      "loss": 6.1841,
      "mean_token_accuracy": 0.14695459753274917,
      "num_tokens": 5473780.0,
      "step": 2935
    },
    {
      "entropy": 6.239596223831176,
      "epoch": 0.2596255740021194,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996686527755171,
      "loss": 6.0776,
      "mean_token_accuracy": 0.15587782710790635,
      "num_tokens": 5482151.0,
      "step": 2940
    },
    {
      "entropy": 6.316288614273072,
      "epoch": 0.2600671140939597,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004996669421382687,
      "loss": 6.0935,
      "mean_token_accuracy": 0.16187871843576432,
      "num_tokens": 5491103.0,
      "step": 2945
    },
    {
      "entropy": 6.257398414611816,
      "epoch": 0.26050865418580005,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996652270999228,
      "loss": 6.1026,
      "mean_token_accuracy": 0.1524055764079094,
      "num_tokens": 5500367.0,
      "step": 2950
    },
    {
      "entropy": 6.286735200881958,
      "epoch": 0.26095019427764043,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996635076605128,
      "loss": 6.1432,
      "mean_token_accuracy": 0.15162961781024933,
      "num_tokens": 5509631.0,
      "step": 2955
    },
    {
      "entropy": 6.3608544826507565,
      "epoch": 0.26139173436948077,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996617838200725,
      "loss": 6.1488,
      "mean_token_accuracy": 0.1510395437479019,
      "num_tokens": 5518635.0,
      "step": 2960
    },
    {
      "entropy": 6.2053210735321045,
      "epoch": 0.2618332744613211,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004996600555786357,
      "loss": 6.1166,
      "mean_token_accuracy": 0.1505479171872139,
      "num_tokens": 5527696.0,
      "step": 2965
    },
    {
      "entropy": 6.343273687362671,
      "epoch": 0.26227481455316143,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996583229362362,
      "loss": 6.077,
      "mean_token_accuracy": 0.15462881475687026,
      "num_tokens": 5536632.0,
      "step": 2970
    },
    {
      "entropy": 6.356432819366455,
      "epoch": 0.26271635464500176,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004996565858929078,
      "loss": 6.1634,
      "mean_token_accuracy": 0.15574948489665985,
      "num_tokens": 5545825.0,
      "step": 2975
    },
    {
      "entropy": 6.2175110340118405,
      "epoch": 0.2631578947368421,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004996548444486847,
      "loss": 6.0344,
      "mean_token_accuracy": 0.1576728031039238,
      "num_tokens": 5555158.0,
      "step": 2980
    },
    {
      "entropy": 6.174654150009156,
      "epoch": 0.2635994348286824,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004996530986036008,
      "loss": 5.9889,
      "mean_token_accuracy": 0.16068323999643325,
      "num_tokens": 5564218.0,
      "step": 2985
    },
    {
      "entropy": 6.207144069671631,
      "epoch": 0.2640409749205228,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996513483576907,
      "loss": 6.1208,
      "mean_token_accuracy": 0.15933286994695664,
      "num_tokens": 5572760.0,
      "step": 2990
    },
    {
      "entropy": 6.368608522415161,
      "epoch": 0.26448251501236314,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996495937109884,
      "loss": 6.1754,
      "mean_token_accuracy": 0.14925839453935624,
      "num_tokens": 5581660.0,
      "step": 2995
    },
    {
      "entropy": 6.248563766479492,
      "epoch": 0.26492405510420347,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996478346635283,
      "loss": 6.0906,
      "mean_token_accuracy": 0.15518097579479218,
      "num_tokens": 5590664.0,
      "step": 3000
    },
    {
      "epoch": 0.26492405510420347,
      "eval_entropy": 5.978648100514914,
      "eval_loss": 6.153229713439941,
      "eval_mean_token_accuracy": 0.1595988018887533,
      "eval_num_tokens": 5590664.0,
      "eval_runtime": 20.3151,
      "eval_samples_per_second": 1738.363,
      "eval_steps_per_second": 217.326,
      "step": 3000
    },
    {
      "entropy": 6.155827331542969,
      "epoch": 0.2653655951960438,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996460712153448,
      "loss": 5.9704,
      "mean_token_accuracy": 0.16879431158304214,
      "num_tokens": 5598727.0,
      "step": 3005
    },
    {
      "entropy": 6.174825668334961,
      "epoch": 0.26580713528788413,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004996443033664726,
      "loss": 6.1576,
      "mean_token_accuracy": 0.15344538390636445,
      "num_tokens": 5608549.0,
      "step": 3010
    },
    {
      "entropy": 6.312784004211426,
      "epoch": 0.26624867537972446,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996425311169463,
      "loss": 6.0951,
      "mean_token_accuracy": 0.16055944710969924,
      "num_tokens": 5617628.0,
      "step": 3015
    },
    {
      "entropy": 6.304688596725464,
      "epoch": 0.2666902154715648,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004996407544668005,
      "loss": 6.1541,
      "mean_token_accuracy": 0.1486222580075264,
      "num_tokens": 5627182.0,
      "step": 3020
    },
    {
      "entropy": 6.258115196228028,
      "epoch": 0.2671317555634052,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004996389734160701,
      "loss": 6.1827,
      "mean_token_accuracy": 0.15313406884670258,
      "num_tokens": 5636413.0,
      "step": 3025
    },
    {
      "entropy": 6.3286293029785154,
      "epoch": 0.2675732956552455,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00049963718796479,
      "loss": 6.1165,
      "mean_token_accuracy": 0.15353906974196435,
      "num_tokens": 5646025.0,
      "step": 3030
    },
    {
      "entropy": 6.2249290466308596,
      "epoch": 0.26801483574708584,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004996353981129952,
      "loss": 6.1401,
      "mean_token_accuracy": 0.15831294059753417,
      "num_tokens": 5656577.0,
      "step": 3035
    },
    {
      "entropy": 6.266787099838257,
      "epoch": 0.2684563758389262,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004996336038607206,
      "loss": 6.1096,
      "mean_token_accuracy": 0.15671551674604417,
      "num_tokens": 5665163.0,
      "step": 3040
    },
    {
      "entropy": 6.256934928894043,
      "epoch": 0.2688979159307665,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004996318052080015,
      "loss": 6.1384,
      "mean_token_accuracy": 0.1572331577539444,
      "num_tokens": 5675196.0,
      "step": 3045
    },
    {
      "entropy": 6.291875505447388,
      "epoch": 0.26933945602260684,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004996300021548731,
      "loss": 6.0604,
      "mean_token_accuracy": 0.1568915903568268,
      "num_tokens": 5684361.0,
      "step": 3050
    },
    {
      "entropy": 6.193565845489502,
      "epoch": 0.26978099611444717,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004996281947013707,
      "loss": 6.063,
      "mean_token_accuracy": 0.15895331352949144,
      "num_tokens": 5694187.0,
      "step": 3055
    },
    {
      "entropy": 6.237375640869141,
      "epoch": 0.27022253620628756,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996263828475298,
      "loss": 6.1115,
      "mean_token_accuracy": 0.14960144013166427,
      "num_tokens": 5702858.0,
      "step": 3060
    },
    {
      "entropy": 6.268642711639404,
      "epoch": 0.2706640762981279,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004996245665933857,
      "loss": 6.0305,
      "mean_token_accuracy": 0.16427914202213287,
      "num_tokens": 5712727.0,
      "step": 3065
    },
    {
      "entropy": 6.29313907623291,
      "epoch": 0.2711056163899682,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996227459389741,
      "loss": 6.1448,
      "mean_token_accuracy": 0.15978289842605592,
      "num_tokens": 5722904.0,
      "step": 3070
    },
    {
      "entropy": 6.289960145950317,
      "epoch": 0.27154715648180855,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004996209208843307,
      "loss": 6.2333,
      "mean_token_accuracy": 0.15185119658708573,
      "num_tokens": 5732228.0,
      "step": 3075
    },
    {
      "entropy": 6.30490255355835,
      "epoch": 0.2719886965736489,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004996190914294912,
      "loss": 6.1829,
      "mean_token_accuracy": 0.1575074926018715,
      "num_tokens": 5743548.0,
      "step": 3080
    },
    {
      "entropy": 6.295343112945557,
      "epoch": 0.2724302366654892,
      "grad_norm": 0.875,
      "learning_rate": 0.0004996172575744914,
      "loss": 6.1796,
      "mean_token_accuracy": 0.146508327126503,
      "num_tokens": 5754112.0,
      "step": 3085
    },
    {
      "entropy": 6.300486421585083,
      "epoch": 0.27287177675732954,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004996154193193673,
      "loss": 6.0018,
      "mean_token_accuracy": 0.16367141306400299,
      "num_tokens": 5763377.0,
      "step": 3090
    },
    {
      "entropy": 6.170904731750488,
      "epoch": 0.27331331684916993,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004996135766641549,
      "loss": 5.9686,
      "mean_token_accuracy": 0.16288323402404786,
      "num_tokens": 5772000.0,
      "step": 3095
    },
    {
      "entropy": 6.214351940155029,
      "epoch": 0.27375485694101026,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004996117296088903,
      "loss": 6.0917,
      "mean_token_accuracy": 0.1531004711985588,
      "num_tokens": 5782192.0,
      "step": 3100
    },
    {
      "entropy": 6.343245601654052,
      "epoch": 0.2741963970328506,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004996098781536095,
      "loss": 6.1403,
      "mean_token_accuracy": 0.15874934047460557,
      "num_tokens": 5791163.0,
      "step": 3105
    },
    {
      "entropy": 6.304137182235718,
      "epoch": 0.2746379371246909,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004996080222983492,
      "loss": 6.1539,
      "mean_token_accuracy": 0.15088921338319777,
      "num_tokens": 5801815.0,
      "step": 3110
    },
    {
      "entropy": 6.357846736907959,
      "epoch": 0.27507947721653125,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004996061620431453,
      "loss": 6.1611,
      "mean_token_accuracy": 0.152420574426651,
      "num_tokens": 5811545.0,
      "step": 3115
    },
    {
      "entropy": 6.260142993927002,
      "epoch": 0.2755210173083716,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004996042973880344,
      "loss": 6.1806,
      "mean_token_accuracy": 0.15066104382276535,
      "num_tokens": 5820285.0,
      "step": 3120
    },
    {
      "entropy": 6.208466100692749,
      "epoch": 0.2759625574002119,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004996024283330532,
      "loss": 6.0189,
      "mean_token_accuracy": 0.15600398033857346,
      "num_tokens": 5829284.0,
      "step": 3125
    },
    {
      "entropy": 6.209367752075195,
      "epoch": 0.2764040974920523,
      "grad_norm": 0.921875,
      "learning_rate": 0.000499600554878238,
      "loss": 6.1824,
      "mean_token_accuracy": 0.1472199648618698,
      "num_tokens": 5838973.0,
      "step": 3130
    },
    {
      "entropy": 6.292955923080444,
      "epoch": 0.27684563758389263,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995986770236258,
      "loss": 6.0744,
      "mean_token_accuracy": 0.1584090195596218,
      "num_tokens": 5847434.0,
      "step": 3135
    },
    {
      "entropy": 6.26265139579773,
      "epoch": 0.27728717767573297,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004995967947692533,
      "loss": 6.0732,
      "mean_token_accuracy": 0.1618253469467163,
      "num_tokens": 5856464.0,
      "step": 3140
    },
    {
      "entropy": 6.268027257919312,
      "epoch": 0.2777287177675733,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004995949081151571,
      "loss": 6.1406,
      "mean_token_accuracy": 0.15451844334602355,
      "num_tokens": 5865892.0,
      "step": 3145
    },
    {
      "entropy": 6.259841060638427,
      "epoch": 0.2781702578594136,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004995930170613746,
      "loss": 6.107,
      "mean_token_accuracy": 0.16154323220252992,
      "num_tokens": 5874606.0,
      "step": 3150
    },
    {
      "entropy": 6.247693395614624,
      "epoch": 0.27861179795125396,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004995911216079425,
      "loss": 6.0967,
      "mean_token_accuracy": 0.15935523062944412,
      "num_tokens": 5883853.0,
      "step": 3155
    },
    {
      "entropy": 6.314785385131836,
      "epoch": 0.2790533380430943,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004995892217548981,
      "loss": 6.1328,
      "mean_token_accuracy": 0.15520500987768174,
      "num_tokens": 5892841.0,
      "step": 3160
    },
    {
      "entropy": 6.289192247390747,
      "epoch": 0.2794948781349347,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004995873175022786,
      "loss": 6.0993,
      "mean_token_accuracy": 0.15804312825202943,
      "num_tokens": 5901966.0,
      "step": 3165
    },
    {
      "entropy": 6.273761415481568,
      "epoch": 0.279936418226775,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004995854088501213,
      "loss": 6.0034,
      "mean_token_accuracy": 0.1539376214146614,
      "num_tokens": 5910003.0,
      "step": 3170
    },
    {
      "entropy": 6.177750110626221,
      "epoch": 0.28037795831861534,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004995834957984634,
      "loss": 6.1634,
      "mean_token_accuracy": 0.15334712862968444,
      "num_tokens": 5920307.0,
      "step": 3175
    },
    {
      "entropy": 6.249305629730225,
      "epoch": 0.28081949841045567,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004995815783473428,
      "loss": 6.0055,
      "mean_token_accuracy": 0.16042215675115584,
      "num_tokens": 5929875.0,
      "step": 3180
    },
    {
      "entropy": 6.254670238494873,
      "epoch": 0.281261038502296,
      "grad_norm": 1.0,
      "learning_rate": 0.0004995796564967967,
      "loss": 5.9757,
      "mean_token_accuracy": 0.15762123465538025,
      "num_tokens": 5938570.0,
      "step": 3185
    },
    {
      "entropy": 6.167326974868774,
      "epoch": 0.28170257859413633,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004995777302468628,
      "loss": 6.127,
      "mean_token_accuracy": 0.1609175220131874,
      "num_tokens": 5947693.0,
      "step": 3190
    },
    {
      "entropy": 6.24321608543396,
      "epoch": 0.28214411868597666,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004995757995975789,
      "loss": 6.2097,
      "mean_token_accuracy": 0.15263804495334626,
      "num_tokens": 5957377.0,
      "step": 3195
    },
    {
      "entropy": 6.332187747955322,
      "epoch": 0.28258565877781705,
      "grad_norm": 1.0,
      "learning_rate": 0.0004995738645489828,
      "loss": 6.1274,
      "mean_token_accuracy": 0.1612716645002365,
      "num_tokens": 5966443.0,
      "step": 3200
    },
    {
      "entropy": 6.200443315505981,
      "epoch": 0.2830271988696574,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004995719251011124,
      "loss": 6.0412,
      "mean_token_accuracy": 0.15479797273874282,
      "num_tokens": 5975027.0,
      "step": 3205
    },
    {
      "entropy": 6.266233110427857,
      "epoch": 0.2834687389614977,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004995699812540058,
      "loss": 6.1083,
      "mean_token_accuracy": 0.1580372929573059,
      "num_tokens": 5983722.0,
      "step": 3210
    },
    {
      "entropy": 6.205762672424316,
      "epoch": 0.28391027905333804,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000499568033007701,
      "loss": 6.1233,
      "mean_token_accuracy": 0.15123755782842635,
      "num_tokens": 5993358.0,
      "step": 3215
    },
    {
      "entropy": 6.283669900894165,
      "epoch": 0.2843518191451784,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004995660803622361,
      "loss": 6.0542,
      "mean_token_accuracy": 0.16230252236127854,
      "num_tokens": 6002743.0,
      "step": 3220
    },
    {
      "entropy": 6.320738506317139,
      "epoch": 0.2847933592370187,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004995641233176494,
      "loss": 6.2182,
      "mean_token_accuracy": 0.1505182772874832,
      "num_tokens": 6013727.0,
      "step": 3225
    },
    {
      "entropy": 6.313286638259887,
      "epoch": 0.28523489932885904,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004995621618739792,
      "loss": 6.162,
      "mean_token_accuracy": 0.15084221735596656,
      "num_tokens": 6023397.0,
      "step": 3230
    },
    {
      "entropy": 6.2680644512176515,
      "epoch": 0.2856764394206994,
      "grad_norm": 0.953125,
      "learning_rate": 0.000499560196031264,
      "loss": 6.0331,
      "mean_token_accuracy": 0.16098945140838622,
      "num_tokens": 6032453.0,
      "step": 3235
    },
    {
      "entropy": 6.175032663345337,
      "epoch": 0.28611797951253976,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004995582257895423,
      "loss": 6.02,
      "mean_token_accuracy": 0.15336361676454544,
      "num_tokens": 6042441.0,
      "step": 3240
    },
    {
      "entropy": 6.2402129650115965,
      "epoch": 0.2865595196043801,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995562511488528,
      "loss": 6.0108,
      "mean_token_accuracy": 0.1634404867887497,
      "num_tokens": 6050881.0,
      "step": 3245
    },
    {
      "entropy": 6.239614248275757,
      "epoch": 0.2870010596962204,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004995542721092337,
      "loss": 6.0702,
      "mean_token_accuracy": 0.159418722987175,
      "num_tokens": 6060804.0,
      "step": 3250
    },
    {
      "entropy": 6.330715084075928,
      "epoch": 0.28744259978806075,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004995522886707244,
      "loss": 6.1609,
      "mean_token_accuracy": 0.15356300622224808,
      "num_tokens": 6070774.0,
      "step": 3255
    },
    {
      "entropy": 6.253082895278931,
      "epoch": 0.2878841398799011,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004995503008333634,
      "loss": 6.1559,
      "mean_token_accuracy": 0.15287251621484757,
      "num_tokens": 6080158.0,
      "step": 3260
    },
    {
      "entropy": 6.263114023208618,
      "epoch": 0.2883256799717414,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004995483085971897,
      "loss": 6.05,
      "mean_token_accuracy": 0.15794130861759187,
      "num_tokens": 6089183.0,
      "step": 3265
    },
    {
      "entropy": 6.209271335601807,
      "epoch": 0.2887672200635818,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004995463119622424,
      "loss": 6.1544,
      "mean_token_accuracy": 0.15566999167203904,
      "num_tokens": 6098536.0,
      "step": 3270
    },
    {
      "entropy": 6.124404239654541,
      "epoch": 0.28920876015542213,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004995443109285604,
      "loss": 5.9011,
      "mean_token_accuracy": 0.1713046222925186,
      "num_tokens": 6107745.0,
      "step": 3275
    },
    {
      "entropy": 6.262866878509522,
      "epoch": 0.28965030024726246,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004995423054961832,
      "loss": 6.0899,
      "mean_token_accuracy": 0.15983563363552095,
      "num_tokens": 6117512.0,
      "step": 3280
    },
    {
      "entropy": 6.150252914428711,
      "epoch": 0.2900918403391028,
      "grad_norm": 0.96875,
      "learning_rate": 0.00049954029566515,
      "loss": 6.1128,
      "mean_token_accuracy": 0.16109101474285126,
      "num_tokens": 6126030.0,
      "step": 3285
    },
    {
      "entropy": 6.30514931678772,
      "epoch": 0.2905333804309431,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004995382814355,
      "loss": 6.1308,
      "mean_token_accuracy": 0.1520018458366394,
      "num_tokens": 6134888.0,
      "step": 3290
    },
    {
      "entropy": 6.29309344291687,
      "epoch": 0.29097492052278345,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004995362628072728,
      "loss": 6.0945,
      "mean_token_accuracy": 0.16048415303230285,
      "num_tokens": 6144274.0,
      "step": 3295
    },
    {
      "entropy": 6.152520227432251,
      "epoch": 0.2914164606146238,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004995342397805078,
      "loss": 6.109,
      "mean_token_accuracy": 0.1570252701640129,
      "num_tokens": 6153406.0,
      "step": 3300
    },
    {
      "entropy": 6.257608938217163,
      "epoch": 0.29185800070646417,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004995322123552448,
      "loss": 6.0337,
      "mean_token_accuracy": 0.15978619009256362,
      "num_tokens": 6162743.0,
      "step": 3305
    },
    {
      "entropy": 6.219382572174072,
      "epoch": 0.2922995407983045,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004995301805315235,
      "loss": 6.0253,
      "mean_token_accuracy": 0.16270643919706346,
      "num_tokens": 6171997.0,
      "step": 3310
    },
    {
      "entropy": 6.169751071929932,
      "epoch": 0.29274108089014483,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004995281443093837,
      "loss": 6.0379,
      "mean_token_accuracy": 0.16006233543157578,
      "num_tokens": 6181275.0,
      "step": 3315
    },
    {
      "entropy": 6.232631635665894,
      "epoch": 0.29318262098198516,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004995261036888652,
      "loss": 6.1001,
      "mean_token_accuracy": 0.1559043176472187,
      "num_tokens": 6191640.0,
      "step": 3320
    },
    {
      "entropy": 6.329467058181763,
      "epoch": 0.2936241610738255,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004995240586700081,
      "loss": 6.1076,
      "mean_token_accuracy": 0.15200517624616622,
      "num_tokens": 6201508.0,
      "step": 3325
    },
    {
      "entropy": 6.1505632400512695,
      "epoch": 0.2940657011656658,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004995220092528522,
      "loss": 5.9859,
      "mean_token_accuracy": 0.1673257976770401,
      "num_tokens": 6209902.0,
      "step": 3330
    },
    {
      "entropy": 6.255099248886109,
      "epoch": 0.29450724125750616,
      "grad_norm": 0.921875,
      "learning_rate": 0.000499519955437438,
      "loss": 6.1516,
      "mean_token_accuracy": 0.14913926422595977,
      "num_tokens": 6219760.0,
      "step": 3335
    },
    {
      "entropy": 6.2619555473327635,
      "epoch": 0.29494878134934654,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004995178972238054,
      "loss": 6.1247,
      "mean_token_accuracy": 0.15912023782730103,
      "num_tokens": 6228721.0,
      "step": 3340
    },
    {
      "entropy": 6.183184289932251,
      "epoch": 0.2953903214411869,
      "grad_norm": 0.953125,
      "learning_rate": 0.000499515834611995,
      "loss": 5.9324,
      "mean_token_accuracy": 0.1727195903658867,
      "num_tokens": 6237070.0,
      "step": 3345
    },
    {
      "entropy": 6.158340120315552,
      "epoch": 0.2958318615330272,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004995137676020472,
      "loss": 6.0929,
      "mean_token_accuracy": 0.15666336119174956,
      "num_tokens": 6245659.0,
      "step": 3350
    },
    {
      "entropy": 6.225800704956055,
      "epoch": 0.29627340162486754,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004995116961940023,
      "loss": 6.0472,
      "mean_token_accuracy": 0.16301591396331788,
      "num_tokens": 6255175.0,
      "step": 3355
    },
    {
      "entropy": 6.226327657699585,
      "epoch": 0.29671494171670787,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004995096203879009,
      "loss": 6.0769,
      "mean_token_accuracy": 0.15317506715655327,
      "num_tokens": 6264962.0,
      "step": 3360
    },
    {
      "entropy": 6.216853618621826,
      "epoch": 0.2971564818085482,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004995075401837837,
      "loss": 5.9968,
      "mean_token_accuracy": 0.16241736561059952,
      "num_tokens": 6273411.0,
      "step": 3365
    },
    {
      "entropy": 6.083216285705566,
      "epoch": 0.29759802190038853,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004995054555816915,
      "loss": 5.9669,
      "mean_token_accuracy": 0.16453457474708558,
      "num_tokens": 6282618.0,
      "step": 3370
    },
    {
      "entropy": 6.205840110778809,
      "epoch": 0.2980395619922289,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004995033665816651,
      "loss": 5.9706,
      "mean_token_accuracy": 0.1685549646615982,
      "num_tokens": 6292008.0,
      "step": 3375
    },
    {
      "entropy": 6.129777622222901,
      "epoch": 0.29848110208406925,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004995012731837454,
      "loss": 5.9386,
      "mean_token_accuracy": 0.1638485983014107,
      "num_tokens": 6301079.0,
      "step": 3380
    },
    {
      "entropy": 6.172431135177613,
      "epoch": 0.2989226421759096,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004994991753879736,
      "loss": 6.1135,
      "mean_token_accuracy": 0.15206168442964554,
      "num_tokens": 6310543.0,
      "step": 3385
    },
    {
      "entropy": 6.267715644836426,
      "epoch": 0.2993641822677499,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004994970731943904,
      "loss": 6.0568,
      "mean_token_accuracy": 0.15798714011907578,
      "num_tokens": 6320037.0,
      "step": 3390
    },
    {
      "entropy": 6.189213752746582,
      "epoch": 0.29980572235959024,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004994949666030374,
      "loss": 6.0594,
      "mean_token_accuracy": 0.1617031455039978,
      "num_tokens": 6328948.0,
      "step": 3395
    },
    {
      "entropy": 6.132642126083374,
      "epoch": 0.3002472624514306,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004994928556139557,
      "loss": 5.9311,
      "mean_token_accuracy": 0.16827442049980162,
      "num_tokens": 6337449.0,
      "step": 3400
    },
    {
      "entropy": 6.117817974090576,
      "epoch": 0.3006888025432709,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004994907402271865,
      "loss": 5.9745,
      "mean_token_accuracy": 0.1659398391842842,
      "num_tokens": 6346130.0,
      "step": 3405
    },
    {
      "entropy": 6.122855234146118,
      "epoch": 0.3011303426351113,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004994886204427715,
      "loss": 6.0239,
      "mean_token_accuracy": 0.16507700830698013,
      "num_tokens": 6354973.0,
      "step": 3410
    },
    {
      "entropy": 6.271668004989624,
      "epoch": 0.3015718827269516,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004994864962607519,
      "loss": 6.1041,
      "mean_token_accuracy": 0.15557809472084044,
      "num_tokens": 6364107.0,
      "step": 3415
    },
    {
      "entropy": 6.2369715690612795,
      "epoch": 0.30201342281879195,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004994843676811697,
      "loss": 6.0198,
      "mean_token_accuracy": 0.1597614735364914,
      "num_tokens": 6372859.0,
      "step": 3420
    },
    {
      "entropy": 6.088519191741943,
      "epoch": 0.3024549629106323,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004994822347040664,
      "loss": 5.8991,
      "mean_token_accuracy": 0.17260994017124176,
      "num_tokens": 6381818.0,
      "step": 3425
    },
    {
      "entropy": 6.1546448230743405,
      "epoch": 0.3028965030024726,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004994800973294837,
      "loss": 6.0699,
      "mean_token_accuracy": 0.15940487533807754,
      "num_tokens": 6391460.0,
      "step": 3430
    },
    {
      "entropy": 6.290472555160522,
      "epoch": 0.30333804309431295,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004994779555574636,
      "loss": 6.1024,
      "mean_token_accuracy": 0.15706333965063096,
      "num_tokens": 6401461.0,
      "step": 3435
    },
    {
      "entropy": 6.211634826660156,
      "epoch": 0.3037795831861533,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499475809388048,
      "loss": 5.9738,
      "mean_token_accuracy": 0.15956642478704453,
      "num_tokens": 6410435.0,
      "step": 3440
    },
    {
      "entropy": 6.20361361503601,
      "epoch": 0.30422112327799367,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000499473658821279,
      "loss": 6.1684,
      "mean_token_accuracy": 0.15328484475612641,
      "num_tokens": 6420710.0,
      "step": 3445
    },
    {
      "entropy": 6.279194784164429,
      "epoch": 0.304662663369834,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004994715038571986,
      "loss": 6.0188,
      "mean_token_accuracy": 0.1564917415380478,
      "num_tokens": 6429882.0,
      "step": 3450
    },
    {
      "entropy": 6.139882755279541,
      "epoch": 0.30510420346167433,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004994693444958493,
      "loss": 5.9924,
      "mean_token_accuracy": 0.16666608154773713,
      "num_tokens": 6439183.0,
      "step": 3455
    },
    {
      "entropy": 6.22991909980774,
      "epoch": 0.30554574355351466,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000499467180737273,
      "loss": 6.1876,
      "mean_token_accuracy": 0.15393830388784407,
      "num_tokens": 6448460.0,
      "step": 3460
    },
    {
      "entropy": 6.241871309280396,
      "epoch": 0.305987283645355,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004994650125815124,
      "loss": 6.0346,
      "mean_token_accuracy": 0.16356185972690582,
      "num_tokens": 6457687.0,
      "step": 3465
    },
    {
      "entropy": 6.1781104564666744,
      "epoch": 0.3064288237371953,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004994628400286097,
      "loss": 5.9815,
      "mean_token_accuracy": 0.16612761467695236,
      "num_tokens": 6467057.0,
      "step": 3470
    },
    {
      "entropy": 6.136842584609985,
      "epoch": 0.30687036382903565,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004994606630786078,
      "loss": 6.0576,
      "mean_token_accuracy": 0.16337907910346985,
      "num_tokens": 6477484.0,
      "step": 3475
    },
    {
      "entropy": 6.187231159210205,
      "epoch": 0.30731190392087604,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004994584817315492,
      "loss": 6.1329,
      "mean_token_accuracy": 0.15671780258417128,
      "num_tokens": 6488381.0,
      "step": 3480
    },
    {
      "entropy": 6.257808637619019,
      "epoch": 0.30775344401271637,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004994562959874765,
      "loss": 6.0139,
      "mean_token_accuracy": 0.16207417994737625,
      "num_tokens": 6498279.0,
      "step": 3485
    },
    {
      "entropy": 6.21480975151062,
      "epoch": 0.3081949841045567,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004994541058464326,
      "loss": 6.0169,
      "mean_token_accuracy": 0.16068682819604874,
      "num_tokens": 6508008.0,
      "step": 3490
    },
    {
      "entropy": 6.236310052871704,
      "epoch": 0.30863652419639703,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004994519113084605,
      "loss": 6.0915,
      "mean_token_accuracy": 0.15919461101293564,
      "num_tokens": 6517263.0,
      "step": 3495
    },
    {
      "entropy": 6.244526529312134,
      "epoch": 0.30907806428823736,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004994497123736029,
      "loss": 6.1388,
      "mean_token_accuracy": 0.15009350925683976,
      "num_tokens": 6527682.0,
      "step": 3500
    },
    {
      "entropy": 6.20444884300232,
      "epoch": 0.3095196043800777,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004994475090419034,
      "loss": 6.0139,
      "mean_token_accuracy": 0.16034948080778122,
      "num_tokens": 6537143.0,
      "step": 3505
    },
    {
      "entropy": 6.1589967727661135,
      "epoch": 0.309961144471918,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004994453013134047,
      "loss": 6.004,
      "mean_token_accuracy": 0.16135213673114776,
      "num_tokens": 6546561.0,
      "step": 3510
    },
    {
      "entropy": 6.153254127502441,
      "epoch": 0.3104026845637584,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004994430891881502,
      "loss": 6.0215,
      "mean_token_accuracy": 0.1592785008251667,
      "num_tokens": 6555806.0,
      "step": 3515
    },
    {
      "entropy": 6.202929401397705,
      "epoch": 0.31084422465559874,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004994408726661832,
      "loss": 6.05,
      "mean_token_accuracy": 0.1624141365289688,
      "num_tokens": 6564836.0,
      "step": 3520
    },
    {
      "entropy": 6.133888912200928,
      "epoch": 0.3112857647474391,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004994386517475472,
      "loss": 6.0371,
      "mean_token_accuracy": 0.155925951898098,
      "num_tokens": 6574997.0,
      "step": 3525
    },
    {
      "entropy": 6.208050298690796,
      "epoch": 0.3117273048392794,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004994364264322856,
      "loss": 6.0597,
      "mean_token_accuracy": 0.15896133482456207,
      "num_tokens": 6584589.0,
      "step": 3530
    },
    {
      "entropy": 6.167923545837402,
      "epoch": 0.31216884493111974,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004994341967204421,
      "loss": 6.0051,
      "mean_token_accuracy": 0.16373383700847627,
      "num_tokens": 6592898.0,
      "step": 3535
    },
    {
      "entropy": 6.255459213256836,
      "epoch": 0.31261038502296007,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004994319626120603,
      "loss": 6.0143,
      "mean_token_accuracy": 0.1714682549238205,
      "num_tokens": 6602454.0,
      "step": 3540
    },
    {
      "entropy": 6.140329837799072,
      "epoch": 0.3130519251148004,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004994297241071841,
      "loss": 6.0189,
      "mean_token_accuracy": 0.16352686434984207,
      "num_tokens": 6612454.0,
      "step": 3545
    },
    {
      "entropy": 6.176025867462158,
      "epoch": 0.3134934652066408,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000499427481205857,
      "loss": 6.0204,
      "mean_token_accuracy": 0.16263296753168105,
      "num_tokens": 6621286.0,
      "step": 3550
    },
    {
      "entropy": 6.158733367919922,
      "epoch": 0.3139350052984811,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004994252339081234,
      "loss": 5.9611,
      "mean_token_accuracy": 0.16285568177700044,
      "num_tokens": 6629684.0,
      "step": 3555
    },
    {
      "entropy": 6.135843944549561,
      "epoch": 0.31437654539032145,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000499422982214027,
      "loss": 6.0543,
      "mean_token_accuracy": 0.1660478115081787,
      "num_tokens": 6638836.0,
      "step": 3560
    },
    {
      "entropy": 6.2141636371612545,
      "epoch": 0.3148180854821618,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004994207261236121,
      "loss": 5.9942,
      "mean_token_accuracy": 0.16404083073139192,
      "num_tokens": 6647337.0,
      "step": 3565
    },
    {
      "entropy": 6.096640634536743,
      "epoch": 0.3152596255740021,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004994184656369227,
      "loss": 5.9733,
      "mean_token_accuracy": 0.15804241746664047,
      "num_tokens": 6657246.0,
      "step": 3570
    },
    {
      "entropy": 6.225153303146362,
      "epoch": 0.31570116566584244,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004994162007540033,
      "loss": 6.1202,
      "mean_token_accuracy": 0.15196825563907623,
      "num_tokens": 6667981.0,
      "step": 3575
    },
    {
      "entropy": 6.204136848449707,
      "epoch": 0.3161427057576828,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004994139314748981,
      "loss": 5.9703,
      "mean_token_accuracy": 0.16491541713476182,
      "num_tokens": 6677462.0,
      "step": 3580
    },
    {
      "entropy": 6.090105485916138,
      "epoch": 0.31658424584952316,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004994116577996517,
      "loss": 6.007,
      "mean_token_accuracy": 0.17120114117860794,
      "num_tokens": 6687225.0,
      "step": 3585
    },
    {
      "entropy": 6.172569704055786,
      "epoch": 0.3170257859413635,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004994093797283084,
      "loss": 6.086,
      "mean_token_accuracy": 0.15607115626335144,
      "num_tokens": 6696345.0,
      "step": 3590
    },
    {
      "entropy": 6.24049186706543,
      "epoch": 0.3174673260332038,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004994070972609132,
      "loss": 6.0527,
      "mean_token_accuracy": 0.16193179190158843,
      "num_tokens": 6706207.0,
      "step": 3595
    },
    {
      "entropy": 6.173488807678223,
      "epoch": 0.31790886612504415,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004994048103975103,
      "loss": 5.9788,
      "mean_token_accuracy": 0.15975290089845656,
      "num_tokens": 6714430.0,
      "step": 3600
    },
    {
      "entropy": 6.130007123947143,
      "epoch": 0.3183504062168845,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000499402519138145,
      "loss": 5.9923,
      "mean_token_accuracy": 0.1620108649134636,
      "num_tokens": 6724139.0,
      "step": 3605
    },
    {
      "entropy": 6.143885087966919,
      "epoch": 0.3187919463087248,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004994002234828619,
      "loss": 6.0353,
      "mean_token_accuracy": 0.16159311830997466,
      "num_tokens": 6733525.0,
      "step": 3610
    },
    {
      "entropy": 6.262012004852295,
      "epoch": 0.31923348640056515,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000499397923431706,
      "loss": 6.1456,
      "mean_token_accuracy": 0.15688765943050384,
      "num_tokens": 6743738.0,
      "step": 3615
    },
    {
      "entropy": 6.226923990249634,
      "epoch": 0.31967502649240553,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004993956189847226,
      "loss": 5.9809,
      "mean_token_accuracy": 0.1651754342019558,
      "num_tokens": 6752900.0,
      "step": 3620
    },
    {
      "entropy": 6.146596193313599,
      "epoch": 0.32011656658424587,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004993933101419565,
      "loss": 5.9722,
      "mean_token_accuracy": 0.1666112557053566,
      "num_tokens": 6761776.0,
      "step": 3625
    },
    {
      "entropy": 6.1981529712677,
      "epoch": 0.3205581066760862,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004993909969034531,
      "loss": 6.12,
      "mean_token_accuracy": 0.15697512328624724,
      "num_tokens": 6771543.0,
      "step": 3630
    },
    {
      "entropy": 6.192965984344482,
      "epoch": 0.3209996467679265,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004993886792692576,
      "loss": 6.005,
      "mean_token_accuracy": 0.16011940240859984,
      "num_tokens": 6780010.0,
      "step": 3635
    },
    {
      "entropy": 6.1778465747833256,
      "epoch": 0.32144118685976686,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004993863572394156,
      "loss": 6.0593,
      "mean_token_accuracy": 0.1610572248697281,
      "num_tokens": 6788830.0,
      "step": 3640
    },
    {
      "entropy": 6.20711760520935,
      "epoch": 0.3218827269516072,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004993840308139724,
      "loss": 6.0473,
      "mean_token_accuracy": 0.16559676080942154,
      "num_tokens": 6799121.0,
      "step": 3645
    },
    {
      "entropy": 6.192859411239624,
      "epoch": 0.3223242670434475,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004993816999929738,
      "loss": 5.9948,
      "mean_token_accuracy": 0.15838830024003983,
      "num_tokens": 6808519.0,
      "step": 3650
    },
    {
      "entropy": 6.124808979034424,
      "epoch": 0.3227658071352879,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004993793647764651,
      "loss": 6.0069,
      "mean_token_accuracy": 0.17118934392929078,
      "num_tokens": 6817583.0,
      "step": 3655
    },
    {
      "entropy": 6.23282527923584,
      "epoch": 0.32320734722712824,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004993770251644923,
      "loss": 6.0196,
      "mean_token_accuracy": 0.162643201649189,
      "num_tokens": 6827732.0,
      "step": 3660
    },
    {
      "entropy": 6.109570169448853,
      "epoch": 0.32364888731896857,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993746811571013,
      "loss": 5.9775,
      "mean_token_accuracy": 0.16294157058000563,
      "num_tokens": 6836903.0,
      "step": 3665
    },
    {
      "entropy": 6.069127988815308,
      "epoch": 0.3240904274108089,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004993723327543379,
      "loss": 5.986,
      "mean_token_accuracy": 0.17082992941141129,
      "num_tokens": 6845684.0,
      "step": 3670
    },
    {
      "entropy": 6.141482591629028,
      "epoch": 0.32453196750264923,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000499369979956248,
      "loss": 5.9745,
      "mean_token_accuracy": 0.16510514169931412,
      "num_tokens": 6854650.0,
      "step": 3675
    },
    {
      "entropy": 6.195809412002563,
      "epoch": 0.32497350759448956,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004993676227628779,
      "loss": 5.9775,
      "mean_token_accuracy": 0.15800541490316392,
      "num_tokens": 6864340.0,
      "step": 3680
    },
    {
      "entropy": 6.201231145858765,
      "epoch": 0.3254150476863299,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004993652611742736,
      "loss": 6.022,
      "mean_token_accuracy": 0.16170618385076524,
      "num_tokens": 6873244.0,
      "step": 3685
    },
    {
      "entropy": 6.065858983993531,
      "epoch": 0.3258565877781703,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004993628951904815,
      "loss": 5.8965,
      "mean_token_accuracy": 0.16670404374599457,
      "num_tokens": 6882392.0,
      "step": 3690
    },
    {
      "entropy": 6.172443294525147,
      "epoch": 0.3262981278700106,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004993605248115479,
      "loss": 6.1129,
      "mean_token_accuracy": 0.14900100082159043,
      "num_tokens": 6891515.0,
      "step": 3695
    },
    {
      "entropy": 6.265958166122436,
      "epoch": 0.32673966796185094,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004993581500375191,
      "loss": 6.0042,
      "mean_token_accuracy": 0.16194652020931244,
      "num_tokens": 6900104.0,
      "step": 3700
    },
    {
      "entropy": 6.21707010269165,
      "epoch": 0.3271812080536913,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004993557708684417,
      "loss": 6.1181,
      "mean_token_accuracy": 0.15311858505010606,
      "num_tokens": 6910443.0,
      "step": 3705
    },
    {
      "entropy": 6.130673027038574,
      "epoch": 0.3276227481455316,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004993533873043625,
      "loss": 5.983,
      "mean_token_accuracy": 0.16641464680433274,
      "num_tokens": 6919882.0,
      "step": 3710
    },
    {
      "entropy": 6.2687431335449215,
      "epoch": 0.32806428823737194,
      "grad_norm": 0.9375,
      "learning_rate": 0.000499350999345328,
      "loss": 6.0932,
      "mean_token_accuracy": 0.15396894812583922,
      "num_tokens": 6929757.0,
      "step": 3715
    },
    {
      "entropy": 6.1528526782989506,
      "epoch": 0.32850582832921227,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000499348606991385,
      "loss": 6.0822,
      "mean_token_accuracy": 0.15250903815031053,
      "num_tokens": 6938578.0,
      "step": 3720
    },
    {
      "entropy": 6.175461196899414,
      "epoch": 0.32894736842105265,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004993462102425805,
      "loss": 5.9653,
      "mean_token_accuracy": 0.16177183985710145,
      "num_tokens": 6947920.0,
      "step": 3725
    },
    {
      "entropy": 6.154925155639648,
      "epoch": 0.329388908512893,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004993438090989612,
      "loss": 5.9195,
      "mean_token_accuracy": 0.16460449397563934,
      "num_tokens": 6957291.0,
      "step": 3730
    },
    {
      "entropy": 6.145351982116699,
      "epoch": 0.3298304486047333,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004993414035605743,
      "loss": 5.9816,
      "mean_token_accuracy": 0.171938194334507,
      "num_tokens": 6966504.0,
      "step": 3735
    },
    {
      "entropy": 6.169538497924805,
      "epoch": 0.33027198869657365,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004993389936274669,
      "loss": 6.0167,
      "mean_token_accuracy": 0.16437650918960572,
      "num_tokens": 6976234.0,
      "step": 3740
    },
    {
      "entropy": 6.15001654624939,
      "epoch": 0.330713528788414,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004993365792996862,
      "loss": 6.0641,
      "mean_token_accuracy": 0.15483692288398743,
      "num_tokens": 6986372.0,
      "step": 3745
    },
    {
      "entropy": 6.160980892181397,
      "epoch": 0.3311550688802543,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004993341605772795,
      "loss": 5.9526,
      "mean_token_accuracy": 0.167180797457695,
      "num_tokens": 6995553.0,
      "step": 3750
    },
    {
      "entropy": 6.0821184635162355,
      "epoch": 0.33159660897209464,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004993317374602941,
      "loss": 5.8782,
      "mean_token_accuracy": 0.1738182157278061,
      "num_tokens": 7005675.0,
      "step": 3755
    },
    {
      "entropy": 6.085787343978882,
      "epoch": 0.33203814906393503,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993293099487777,
      "loss": 5.9976,
      "mean_token_accuracy": 0.16012882590293884,
      "num_tokens": 7014953.0,
      "step": 3760
    },
    {
      "entropy": 6.188743925094604,
      "epoch": 0.33247968915577536,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004993268780427776,
      "loss": 6.0033,
      "mean_token_accuracy": 0.17121922522783278,
      "num_tokens": 7025075.0,
      "step": 3765
    },
    {
      "entropy": 6.118736314773559,
      "epoch": 0.3329212292476157,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004993244417423416,
      "loss": 6.0144,
      "mean_token_accuracy": 0.16219213157892226,
      "num_tokens": 7034286.0,
      "step": 3770
    },
    {
      "entropy": 6.19470705986023,
      "epoch": 0.333362769339456,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993220010475174,
      "loss": 6.0445,
      "mean_token_accuracy": 0.15993663668632507,
      "num_tokens": 7043957.0,
      "step": 3775
    },
    {
      "entropy": 6.16349458694458,
      "epoch": 0.33380430943129635,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004993195559583526,
      "loss": 5.966,
      "mean_token_accuracy": 0.1597854785621166,
      "num_tokens": 7053170.0,
      "step": 3780
    },
    {
      "entropy": 6.016193008422851,
      "epoch": 0.3342458495231367,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004993171064748954,
      "loss": 5.955,
      "mean_token_accuracy": 0.16348500102758406,
      "num_tokens": 7062469.0,
      "step": 3785
    },
    {
      "entropy": 6.167666959762573,
      "epoch": 0.334687389614977,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004993146525971937,
      "loss": 5.9591,
      "mean_token_accuracy": 0.16554407477378846,
      "num_tokens": 7071463.0,
      "step": 3790
    },
    {
      "entropy": 6.229743194580078,
      "epoch": 0.3351289297068174,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004993121943252955,
      "loss": 6.0586,
      "mean_token_accuracy": 0.1564462423324585,
      "num_tokens": 7081574.0,
      "step": 3795
    },
    {
      "entropy": 6.085719156265259,
      "epoch": 0.33557046979865773,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004993097316592489,
      "loss": 5.9676,
      "mean_token_accuracy": 0.16341703832149507,
      "num_tokens": 7090835.0,
      "step": 3800
    },
    {
      "entropy": 6.035996341705323,
      "epoch": 0.33601200989049806,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004993072645991023,
      "loss": 5.906,
      "mean_token_accuracy": 0.1561555340886116,
      "num_tokens": 7100521.0,
      "step": 3805
    },
    {
      "entropy": 6.160156917572022,
      "epoch": 0.3364535499823384,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499304793144904,
      "loss": 5.9702,
      "mean_token_accuracy": 0.16260155886411667,
      "num_tokens": 7109011.0,
      "step": 3810
    },
    {
      "entropy": 6.203392267227173,
      "epoch": 0.3368950900741787,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004993023172967022,
      "loss": 5.9987,
      "mean_token_accuracy": 0.15889298021793366,
      "num_tokens": 7119128.0,
      "step": 3815
    },
    {
      "entropy": 6.194024133682251,
      "epoch": 0.33733663016601906,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004992998370545458,
      "loss": 6.081,
      "mean_token_accuracy": 0.15163458064198493,
      "num_tokens": 7128313.0,
      "step": 3820
    },
    {
      "entropy": 6.201267099380493,
      "epoch": 0.3377781702578594,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004992973524184831,
      "loss": 6.0549,
      "mean_token_accuracy": 0.15506500452756883,
      "num_tokens": 7137567.0,
      "step": 3825
    },
    {
      "entropy": 6.092210531234741,
      "epoch": 0.3382197103496998,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004992948633885627,
      "loss": 6.0193,
      "mean_token_accuracy": 0.16242285072803497,
      "num_tokens": 7147254.0,
      "step": 3830
    },
    {
      "entropy": 6.069639110565186,
      "epoch": 0.3386612504415401,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004992923699648335,
      "loss": 5.9133,
      "mean_token_accuracy": 0.17447974383831025,
      "num_tokens": 7156215.0,
      "step": 3835
    },
    {
      "entropy": 6.222735929489136,
      "epoch": 0.33910279053338044,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004992898721473445,
      "loss": 6.0054,
      "mean_token_accuracy": 0.15784852206707,
      "num_tokens": 7165653.0,
      "step": 3840
    },
    {
      "entropy": 6.0844096660614015,
      "epoch": 0.33954433062522077,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004992873699361444,
      "loss": 5.9012,
      "mean_token_accuracy": 0.1631088748574257,
      "num_tokens": 7174797.0,
      "step": 3845
    },
    {
      "entropy": 6.192289066314697,
      "epoch": 0.3399858707170611,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004992848633312822,
      "loss": 5.9327,
      "mean_token_accuracy": 0.1653936520218849,
      "num_tokens": 7184040.0,
      "step": 3850
    },
    {
      "entropy": 6.110203790664673,
      "epoch": 0.34042741080890143,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004992823523328071,
      "loss": 5.8661,
      "mean_token_accuracy": 0.17056092023849487,
      "num_tokens": 7193533.0,
      "step": 3855
    },
    {
      "entropy": 6.074834299087525,
      "epoch": 0.34086895090074176,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004992798369407684,
      "loss": 5.9999,
      "mean_token_accuracy": 0.16645675301551818,
      "num_tokens": 7203355.0,
      "step": 3860
    },
    {
      "entropy": 6.240177392959595,
      "epoch": 0.34131049099258215,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004992773171552152,
      "loss": 5.9403,
      "mean_token_accuracy": 0.16228992938995362,
      "num_tokens": 7212569.0,
      "step": 3865
    },
    {
      "entropy": 6.043364143371582,
      "epoch": 0.3417520310844225,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004992747929761968,
      "loss": 5.968,
      "mean_token_accuracy": 0.16697075814008713,
      "num_tokens": 7221886.0,
      "step": 3870
    },
    {
      "entropy": 6.094742727279663,
      "epoch": 0.3421935711762628,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004992722644037628,
      "loss": 6.0245,
      "mean_token_accuracy": 0.16425753831863404,
      "num_tokens": 7232065.0,
      "step": 3875
    },
    {
      "entropy": 6.176693868637085,
      "epoch": 0.34263511126810314,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004992697314379628,
      "loss": 5.9362,
      "mean_token_accuracy": 0.16599779576063156,
      "num_tokens": 7242287.0,
      "step": 3880
    },
    {
      "entropy": 6.141386938095093,
      "epoch": 0.3430766513599435,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004992671940788462,
      "loss": 5.9315,
      "mean_token_accuracy": 0.16158146411180496,
      "num_tokens": 7250638.0,
      "step": 3885
    },
    {
      "entropy": 6.137431240081787,
      "epoch": 0.3435181914517838,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004992646523264628,
      "loss": 5.9567,
      "mean_token_accuracy": 0.1678780198097229,
      "num_tokens": 7259526.0,
      "step": 3890
    },
    {
      "entropy": 6.091816043853759,
      "epoch": 0.34395973154362414,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004992621061808625,
      "loss": 5.9422,
      "mean_token_accuracy": 0.1681588336825371,
      "num_tokens": 7269260.0,
      "step": 3895
    },
    {
      "entropy": 6.137960958480835,
      "epoch": 0.3444012716354645,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000499259555642095,
      "loss": 5.9608,
      "mean_token_accuracy": 0.1592722475528717,
      "num_tokens": 7279061.0,
      "step": 3900
    },
    {
      "entropy": 6.152037239074707,
      "epoch": 0.34484281172730485,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004992570007102104,
      "loss": 5.9434,
      "mean_token_accuracy": 0.16730886548757554,
      "num_tokens": 7289721.0,
      "step": 3905
    },
    {
      "entropy": 6.210768556594848,
      "epoch": 0.3452843518191452,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004992544413852587,
      "loss": 6.0913,
      "mean_token_accuracy": 0.15217040181159974,
      "num_tokens": 7299836.0,
      "step": 3910
    },
    {
      "entropy": 6.154385423660278,
      "epoch": 0.3457258919109855,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00049925187766729,
      "loss": 6.1203,
      "mean_token_accuracy": 0.1559758186340332,
      "num_tokens": 7309306.0,
      "step": 3915
    },
    {
      "entropy": 6.210061311721802,
      "epoch": 0.34616743200282585,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004992493095563545,
      "loss": 5.9828,
      "mean_token_accuracy": 0.16974906474351883,
      "num_tokens": 7318486.0,
      "step": 3920
    },
    {
      "entropy": 6.071922349929809,
      "epoch": 0.3466089720946662,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004992467370525026,
      "loss": 5.94,
      "mean_token_accuracy": 0.16524540185928344,
      "num_tokens": 7327764.0,
      "step": 3925
    },
    {
      "entropy": 6.158257675170899,
      "epoch": 0.3470505121865065,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004992441601557848,
      "loss": 6.0295,
      "mean_token_accuracy": 0.16136583387851716,
      "num_tokens": 7337118.0,
      "step": 3930
    },
    {
      "entropy": 6.115141105651856,
      "epoch": 0.3474920522783469,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004992415788662514,
      "loss": 5.97,
      "mean_token_accuracy": 0.1630101129412651,
      "num_tokens": 7345973.0,
      "step": 3935
    },
    {
      "entropy": 6.12622766494751,
      "epoch": 0.34793359237018723,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004992389931839529,
      "loss": 5.9695,
      "mean_token_accuracy": 0.16391944289207458,
      "num_tokens": 7355776.0,
      "step": 3940
    },
    {
      "entropy": 6.120577955245972,
      "epoch": 0.34837513246202756,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004992364031089401,
      "loss": 5.9826,
      "mean_token_accuracy": 0.16477784961462022,
      "num_tokens": 7364820.0,
      "step": 3945
    },
    {
      "entropy": 6.1659424781799315,
      "epoch": 0.3488166725538679,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004992338086412636,
      "loss": 5.9967,
      "mean_token_accuracy": 0.16261015236377716,
      "num_tokens": 7373881.0,
      "step": 3950
    },
    {
      "entropy": 6.100357103347778,
      "epoch": 0.3492582126457082,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004992312097809744,
      "loss": 5.9439,
      "mean_token_accuracy": 0.1656390517950058,
      "num_tokens": 7383095.0,
      "step": 3955
    },
    {
      "entropy": 6.111528301239014,
      "epoch": 0.34969975273754855,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004992286065281234,
      "loss": 5.9943,
      "mean_token_accuracy": 0.16113595068454742,
      "num_tokens": 7392702.0,
      "step": 3960
    },
    {
      "entropy": 6.163757038116455,
      "epoch": 0.3501412928293889,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004992259988827614,
      "loss": 6.1647,
      "mean_token_accuracy": 0.15409360080957413,
      "num_tokens": 7402560.0,
      "step": 3965
    },
    {
      "entropy": 6.163148641586304,
      "epoch": 0.35058283292122927,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004992233868449397,
      "loss": 5.9105,
      "mean_token_accuracy": 0.17125627696514129,
      "num_tokens": 7412375.0,
      "step": 3970
    },
    {
      "entropy": 6.125844812393188,
      "epoch": 0.3510243730130696,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004992207704147093,
      "loss": 5.9829,
      "mean_token_accuracy": 0.1594072923064232,
      "num_tokens": 7420734.0,
      "step": 3975
    },
    {
      "entropy": 6.08319845199585,
      "epoch": 0.35146591310490993,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004992181495921216,
      "loss": 5.9549,
      "mean_token_accuracy": 0.1678524360060692,
      "num_tokens": 7430156.0,
      "step": 3980
    },
    {
      "entropy": 6.203280305862426,
      "epoch": 0.35190745319675026,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004992155243772277,
      "loss": 5.9672,
      "mean_token_accuracy": 0.17124156951904296,
      "num_tokens": 7439189.0,
      "step": 3985
    },
    {
      "entropy": 6.162478065490722,
      "epoch": 0.3523489932885906,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004992128947700795,
      "loss": 5.9733,
      "mean_token_accuracy": 0.16030417680740355,
      "num_tokens": 7448768.0,
      "step": 3990
    },
    {
      "entropy": 6.085399436950683,
      "epoch": 0.3527905333804309,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000499210260770728,
      "loss": 5.9384,
      "mean_token_accuracy": 0.16325445622205734,
      "num_tokens": 7458468.0,
      "step": 3995
    },
    {
      "entropy": 6.143225765228271,
      "epoch": 0.35323207347227126,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000499207622379225,
      "loss": 5.9964,
      "mean_token_accuracy": 0.1624176785349846,
      "num_tokens": 7467351.0,
      "step": 4000
    },
    {
      "entropy": 6.191954469680786,
      "epoch": 0.35367361356411164,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004992049795956222,
      "loss": 5.9732,
      "mean_token_accuracy": 0.15985945910215377,
      "num_tokens": 7477116.0,
      "step": 4005
    },
    {
      "entropy": 6.096588611602783,
      "epoch": 0.354115153655952,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004992023324199715,
      "loss": 5.9035,
      "mean_token_accuracy": 0.1694460391998291,
      "num_tokens": 7485216.0,
      "step": 4010
    },
    {
      "entropy": 6.05253791809082,
      "epoch": 0.3545566937477923,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004991996808523245,
      "loss": 6.0587,
      "mean_token_accuracy": 0.15506483614444733,
      "num_tokens": 7495084.0,
      "step": 4015
    },
    {
      "entropy": 6.274289941787719,
      "epoch": 0.35499823383963264,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004991970248927332,
      "loss": 6.0609,
      "mean_token_accuracy": 0.16138618141412736,
      "num_tokens": 7503219.0,
      "step": 4020
    },
    {
      "entropy": 6.107459831237793,
      "epoch": 0.35543977393147297,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004991943645412498,
      "loss": 5.8498,
      "mean_token_accuracy": 0.17398339360952378,
      "num_tokens": 7511961.0,
      "step": 4025
    },
    {
      "entropy": 6.018071126937866,
      "epoch": 0.3558813140233133,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004991916997979263,
      "loss": 5.9675,
      "mean_token_accuracy": 0.16527593061327933,
      "num_tokens": 7521927.0,
      "step": 4030
    },
    {
      "entropy": 6.0901021480560305,
      "epoch": 0.35632285411515363,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004991890306628149,
      "loss": 5.9106,
      "mean_token_accuracy": 0.16498993933200837,
      "num_tokens": 7531093.0,
      "step": 4035
    },
    {
      "entropy": 6.112301921844482,
      "epoch": 0.356764394206994,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004991863571359678,
      "loss": 5.8843,
      "mean_token_accuracy": 0.1641499012708664,
      "num_tokens": 7539813.0,
      "step": 4040
    },
    {
      "entropy": 6.191416263580322,
      "epoch": 0.35720593429883435,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004991836792174376,
      "loss": 6.0347,
      "mean_token_accuracy": 0.15737137496471404,
      "num_tokens": 7548954.0,
      "step": 4045
    },
    {
      "entropy": 6.059986782073975,
      "epoch": 0.3576474743906747,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004991809969072765,
      "loss": 5.8989,
      "mean_token_accuracy": 0.1670374259352684,
      "num_tokens": 7558275.0,
      "step": 4050
    },
    {
      "entropy": 6.059090328216553,
      "epoch": 0.358089014482515,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004991783102055371,
      "loss": 5.9351,
      "mean_token_accuracy": 0.17020276039838791,
      "num_tokens": 7567870.0,
      "step": 4055
    },
    {
      "entropy": 6.14141616821289,
      "epoch": 0.35853055457435534,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004991756191122723,
      "loss": 5.8914,
      "mean_token_accuracy": 0.17240921556949615,
      "num_tokens": 7577485.0,
      "step": 4060
    },
    {
      "entropy": 6.102657413482666,
      "epoch": 0.3589720946661957,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004991729236275346,
      "loss": 5.9435,
      "mean_token_accuracy": 0.16436953246593475,
      "num_tokens": 7587398.0,
      "step": 4065
    },
    {
      "entropy": 6.044604969024658,
      "epoch": 0.359413634758036,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004991702237513768,
      "loss": 5.9639,
      "mean_token_accuracy": 0.16127175837755203,
      "num_tokens": 7596310.0,
      "step": 4070
    },
    {
      "entropy": 6.149261903762818,
      "epoch": 0.3598551748498764,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004991675194838517,
      "loss": 5.9934,
      "mean_token_accuracy": 0.15606556236743926,
      "num_tokens": 7606971.0,
      "step": 4075
    },
    {
      "entropy": 6.130791330337525,
      "epoch": 0.3602967149417167,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004991648108250125,
      "loss": 5.9699,
      "mean_token_accuracy": 0.162691630423069,
      "num_tokens": 7616636.0,
      "step": 4080
    },
    {
      "entropy": 6.138277244567871,
      "epoch": 0.36073825503355705,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000499162097774912,
      "loss": 5.9991,
      "mean_token_accuracy": 0.16216389685869217,
      "num_tokens": 7625128.0,
      "step": 4085
    },
    {
      "entropy": 6.10425295829773,
      "epoch": 0.3611797951253974,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004991593803336037,
      "loss": 5.9044,
      "mean_token_accuracy": 0.1682908684015274,
      "num_tokens": 7634224.0,
      "step": 4090
    },
    {
      "entropy": 6.075386428833008,
      "epoch": 0.3616213352172377,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004991566585011405,
      "loss": 5.8935,
      "mean_token_accuracy": 0.1718817874789238,
      "num_tokens": 7643392.0,
      "step": 4095
    },
    {
      "entropy": 6.03798017501831,
      "epoch": 0.36206287530907805,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004991539322775758,
      "loss": 5.9139,
      "mean_token_accuracy": 0.16705035418272018,
      "num_tokens": 7652491.0,
      "step": 4100
    },
    {
      "entropy": 6.292382764816284,
      "epoch": 0.3625044154009184,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004991512016629632,
      "loss": 6.1198,
      "mean_token_accuracy": 0.14770500659942626,
      "num_tokens": 7663192.0,
      "step": 4105
    },
    {
      "entropy": 6.111117219924926,
      "epoch": 0.36294595549275877,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004991484666573558,
      "loss": 5.9498,
      "mean_token_accuracy": 0.16446800976991655,
      "num_tokens": 7672801.0,
      "step": 4110
    },
    {
      "entropy": 6.096119546890259,
      "epoch": 0.3633874955845991,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004991457272608077,
      "loss": 5.9759,
      "mean_token_accuracy": 0.16090452373027803,
      "num_tokens": 7683013.0,
      "step": 4115
    },
    {
      "entropy": 6.168970012664795,
      "epoch": 0.3638290356764394,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004991429834733721,
      "loss": 6.0052,
      "mean_token_accuracy": 0.1668308272957802,
      "num_tokens": 7692760.0,
      "step": 4120
    },
    {
      "entropy": 6.156150960922242,
      "epoch": 0.36427057576827976,
      "grad_norm": 1.1875,
      "learning_rate": 0.000499140235295103,
      "loss": 5.9444,
      "mean_token_accuracy": 0.16200983226299287,
      "num_tokens": 7702005.0,
      "step": 4125
    },
    {
      "entropy": 6.12065167427063,
      "epoch": 0.3647121158601201,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004991374827260542,
      "loss": 5.9973,
      "mean_token_accuracy": 0.16509999632835387,
      "num_tokens": 7711604.0,
      "step": 4130
    },
    {
      "entropy": 6.022671031951904,
      "epoch": 0.3651536559519604,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004991347257662795,
      "loss": 5.8832,
      "mean_token_accuracy": 0.16578579545021058,
      "num_tokens": 7720434.0,
      "step": 4135
    },
    {
      "entropy": 6.131404399871826,
      "epoch": 0.36559519604380075,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000499131964415833,
      "loss": 5.9468,
      "mean_token_accuracy": 0.167755126953125,
      "num_tokens": 7730733.0,
      "step": 4140
    },
    {
      "entropy": 6.139104986190796,
      "epoch": 0.36603673613564114,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004991291986747689,
      "loss": 5.9964,
      "mean_token_accuracy": 0.16309818848967553,
      "num_tokens": 7740394.0,
      "step": 4145
    },
    {
      "entropy": 6.125427103042602,
      "epoch": 0.36647827622748147,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004991264285431412,
      "loss": 5.962,
      "mean_token_accuracy": 0.1680387258529663,
      "num_tokens": 7748919.0,
      "step": 4150
    },
    {
      "entropy": 6.123518419265747,
      "epoch": 0.3669198163193218,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004991236540210041,
      "loss": 5.974,
      "mean_token_accuracy": 0.16233558654785157,
      "num_tokens": 7758591.0,
      "step": 4155
    },
    {
      "entropy": 6.1502306938171385,
      "epoch": 0.36736135641116213,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004991208751084122,
      "loss": 6.0083,
      "mean_token_accuracy": 0.15538841038942336,
      "num_tokens": 7768269.0,
      "step": 4160
    },
    {
      "entropy": 6.204276371002197,
      "epoch": 0.36780289650300246,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004991180918054199,
      "loss": 6.0402,
      "mean_token_accuracy": 0.15984537899494172,
      "num_tokens": 7776980.0,
      "step": 4165
    },
    {
      "entropy": 6.127771854400635,
      "epoch": 0.3682444365948428,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004991153041120815,
      "loss": 5.941,
      "mean_token_accuracy": 0.1670564740896225,
      "num_tokens": 7786664.0,
      "step": 4170
    },
    {
      "entropy": 6.0249744892120365,
      "epoch": 0.3686859766866831,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004991125120284519,
      "loss": 5.8206,
      "mean_token_accuracy": 0.17915644198656083,
      "num_tokens": 7794981.0,
      "step": 4175
    },
    {
      "entropy": 6.056008577346802,
      "epoch": 0.3691275167785235,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004991097155545856,
      "loss": 5.9086,
      "mean_token_accuracy": 0.16700488328933716,
      "num_tokens": 7804850.0,
      "step": 4180
    },
    {
      "entropy": 6.102103614807129,
      "epoch": 0.36956905687036384,
      "grad_norm": 1.0,
      "learning_rate": 0.0004991069146905374,
      "loss": 5.9142,
      "mean_token_accuracy": 0.1711506649851799,
      "num_tokens": 7814117.0,
      "step": 4185
    },
    {
      "entropy": 6.098058605194092,
      "epoch": 0.3700105969622042,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004991041094363621,
      "loss": 5.9473,
      "mean_token_accuracy": 0.16947417855262756,
      "num_tokens": 7823384.0,
      "step": 4190
    },
    {
      "entropy": 6.091346216201782,
      "epoch": 0.3704521370540445,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004991012997921149,
      "loss": 5.9529,
      "mean_token_accuracy": 0.16536522805690765,
      "num_tokens": 7834072.0,
      "step": 4195
    },
    {
      "entropy": 6.12990288734436,
      "epoch": 0.37089367714588484,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004990984857578506,
      "loss": 5.9442,
      "mean_token_accuracy": 0.16497403979301453,
      "num_tokens": 7843334.0,
      "step": 4200
    },
    {
      "entropy": 5.994000387191773,
      "epoch": 0.37133521723772517,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004990956673336245,
      "loss": 5.865,
      "mean_token_accuracy": 0.17073259353637696,
      "num_tokens": 7853187.0,
      "step": 4205
    },
    {
      "entropy": 6.0792876243591305,
      "epoch": 0.3717767573295655,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004990928445194917,
      "loss": 5.9617,
      "mean_token_accuracy": 0.1619952842593193,
      "num_tokens": 7862262.0,
      "step": 4210
    },
    {
      "entropy": 6.071585988998413,
      "epoch": 0.3722182974214059,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004990900173155074,
      "loss": 5.8975,
      "mean_token_accuracy": 0.16614274084568023,
      "num_tokens": 7870725.0,
      "step": 4215
    },
    {
      "entropy": 6.031388998031616,
      "epoch": 0.3726598375132462,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004990871857217273,
      "loss": 5.8055,
      "mean_token_accuracy": 0.1678528070449829,
      "num_tokens": 7879492.0,
      "step": 4220
    },
    {
      "entropy": 6.110537576675415,
      "epoch": 0.37310137760508655,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004990843497382066,
      "loss": 5.9115,
      "mean_token_accuracy": 0.16305805295705794,
      "num_tokens": 7888710.0,
      "step": 4225
    },
    {
      "entropy": 6.0610274314880375,
      "epoch": 0.3735429176969269,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004990815093650009,
      "loss": 5.9409,
      "mean_token_accuracy": 0.16872991621494293,
      "num_tokens": 7898313.0,
      "step": 4230
    },
    {
      "entropy": 6.102960062026978,
      "epoch": 0.3739844577887672,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004990786646021659,
      "loss": 5.9634,
      "mean_token_accuracy": 0.16352523416280745,
      "num_tokens": 7906726.0,
      "step": 4235
    },
    {
      "entropy": 6.1046594142913815,
      "epoch": 0.37442599788060754,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004990758154497573,
      "loss": 5.8937,
      "mean_token_accuracy": 0.17032238245010375,
      "num_tokens": 7915554.0,
      "step": 4240
    },
    {
      "entropy": 6.101940011978149,
      "epoch": 0.3748675379724479,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004990729619078309,
      "loss": 6.0112,
      "mean_token_accuracy": 0.16436887234449388,
      "num_tokens": 7924732.0,
      "step": 4245
    },
    {
      "entropy": 6.106853342056274,
      "epoch": 0.37530907806428826,
      "grad_norm": 0.875,
      "learning_rate": 0.0004990701039764427,
      "loss": 5.9303,
      "mean_token_accuracy": 0.1686663657426834,
      "num_tokens": 7933808.0,
      "step": 4250
    },
    {
      "entropy": 6.086512899398803,
      "epoch": 0.3757506181561286,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004990672416556485,
      "loss": 5.8807,
      "mean_token_accuracy": 0.1658458963036537,
      "num_tokens": 7942558.0,
      "step": 4255
    },
    {
      "entropy": 6.064576911926269,
      "epoch": 0.3761921582479689,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004990643749455045,
      "loss": 5.9793,
      "mean_token_accuracy": 0.16173869967460633,
      "num_tokens": 7951951.0,
      "step": 4260
    },
    {
      "entropy": 6.199822521209716,
      "epoch": 0.37663369833980925,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004990615038460667,
      "loss": 5.8825,
      "mean_token_accuracy": 0.16723221093416213,
      "num_tokens": 7960543.0,
      "step": 4265
    },
    {
      "entropy": 6.080804681777954,
      "epoch": 0.3770752384316496,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004990586283573916,
      "loss": 5.8682,
      "mean_token_accuracy": 0.1681565523147583,
      "num_tokens": 7970273.0,
      "step": 4270
    },
    {
      "entropy": 6.062951707839966,
      "epoch": 0.3775167785234899,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004990557484795355,
      "loss": 5.9628,
      "mean_token_accuracy": 0.167630435526371,
      "num_tokens": 7978888.0,
      "step": 4275
    },
    {
      "entropy": 6.145836114883423,
      "epoch": 0.37795831861533025,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004990528642125545,
      "loss": 5.9531,
      "mean_token_accuracy": 0.15775933712720872,
      "num_tokens": 7988843.0,
      "step": 4280
    },
    {
      "entropy": 6.107465028762817,
      "epoch": 0.37839985870717063,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004990499755565055,
      "loss": 5.9345,
      "mean_token_accuracy": 0.16518889516592025,
      "num_tokens": 7998403.0,
      "step": 4285
    },
    {
      "entropy": 6.123253870010376,
      "epoch": 0.37884139879901096,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004990470825114448,
      "loss": 5.9763,
      "mean_token_accuracy": 0.1672004759311676,
      "num_tokens": 8007493.0,
      "step": 4290
    },
    {
      "entropy": 6.10452790260315,
      "epoch": 0.3792829388908513,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004990441850774292,
      "loss": 5.9169,
      "mean_token_accuracy": 0.1715584471821785,
      "num_tokens": 8016399.0,
      "step": 4295
    },
    {
      "entropy": 6.0872495651245115,
      "epoch": 0.3797244789826916,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004990412832545155,
      "loss": 5.9401,
      "mean_token_accuracy": 0.1636321425437927,
      "num_tokens": 8025891.0,
      "step": 4300
    },
    {
      "entropy": 6.022824382781982,
      "epoch": 0.38016601907453196,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004990383770427603,
      "loss": 5.8622,
      "mean_token_accuracy": 0.1687710165977478,
      "num_tokens": 8034733.0,
      "step": 4305
    },
    {
      "entropy": 6.052893352508545,
      "epoch": 0.3806075591663723,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004990354664422209,
      "loss": 5.9131,
      "mean_token_accuracy": 0.17057955414056777,
      "num_tokens": 8044443.0,
      "step": 4310
    },
    {
      "entropy": 6.0984782695770265,
      "epoch": 0.3810490992582126,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004990325514529541,
      "loss": 5.9031,
      "mean_token_accuracy": 0.17352878600358962,
      "num_tokens": 8053920.0,
      "step": 4315
    },
    {
      "entropy": 6.143695259094239,
      "epoch": 0.381490639350053,
      "grad_norm": 0.875,
      "learning_rate": 0.0004990296320750169,
      "loss": 5.8922,
      "mean_token_accuracy": 0.16723661124706268,
      "num_tokens": 8063455.0,
      "step": 4320
    },
    {
      "entropy": 6.028673410415649,
      "epoch": 0.38193217944189334,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004990267083084667,
      "loss": 5.9811,
      "mean_token_accuracy": 0.16257134675979615,
      "num_tokens": 8073353.0,
      "step": 4325
    },
    {
      "entropy": 6.11106915473938,
      "epoch": 0.38237371953373367,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004990237801533607,
      "loss": 5.9733,
      "mean_token_accuracy": 0.17016141563653947,
      "num_tokens": 8082164.0,
      "step": 4330
    },
    {
      "entropy": 6.187211561203003,
      "epoch": 0.382815259625574,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004990208476097562,
      "loss": 5.993,
      "mean_token_accuracy": 0.16434455662965775,
      "num_tokens": 8090790.0,
      "step": 4335
    },
    {
      "entropy": 6.228871631622314,
      "epoch": 0.38325679971741433,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004990179106777109,
      "loss": 6.0236,
      "mean_token_accuracy": 0.1600963294506073,
      "num_tokens": 8100412.0,
      "step": 4340
    },
    {
      "entropy": 6.002315425872803,
      "epoch": 0.38369833980925466,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004990149693572819,
      "loss": 5.982,
      "mean_token_accuracy": 0.16522495746612548,
      "num_tokens": 8110002.0,
      "step": 4345
    },
    {
      "entropy": 6.066628885269165,
      "epoch": 0.384139879901095,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004990120236485271,
      "loss": 5.8771,
      "mean_token_accuracy": 0.16987511217594148,
      "num_tokens": 8118898.0,
      "step": 4350
    },
    {
      "entropy": 6.056738805770874,
      "epoch": 0.3845814199929354,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004990090735515043,
      "loss": 5.9422,
      "mean_token_accuracy": 0.15970152467489243,
      "num_tokens": 8128207.0,
      "step": 4355
    },
    {
      "entropy": 6.094995737075806,
      "epoch": 0.3850229600847757,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499006119066271,
      "loss": 5.9157,
      "mean_token_accuracy": 0.16543019860982894,
      "num_tokens": 8137317.0,
      "step": 4360
    },
    {
      "entropy": 6.137070083618164,
      "epoch": 0.38546450017661604,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004990031601928854,
      "loss": 5.9609,
      "mean_token_accuracy": 0.16855712682008744,
      "num_tokens": 8146519.0,
      "step": 4365
    },
    {
      "entropy": 6.0023377418518065,
      "epoch": 0.3859060402684564,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004990001969314051,
      "loss": 5.9722,
      "mean_token_accuracy": 0.16350623667240144,
      "num_tokens": 8155916.0,
      "step": 4370
    },
    {
      "entropy": 6.148402261734009,
      "epoch": 0.3863475803602967,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004989972292818884,
      "loss": 5.9791,
      "mean_token_accuracy": 0.16953045129776,
      "num_tokens": 8165251.0,
      "step": 4375
    },
    {
      "entropy": 6.093685007095337,
      "epoch": 0.38678912045213704,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004989942572443934,
      "loss": 5.973,
      "mean_token_accuracy": 0.16652768552303315,
      "num_tokens": 8174761.0,
      "step": 4380
    },
    {
      "entropy": 6.088238048553467,
      "epoch": 0.38723066054397737,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004989912808189784,
      "loss": 5.9255,
      "mean_token_accuracy": 0.1713149756193161,
      "num_tokens": 8182980.0,
      "step": 4385
    },
    {
      "entropy": 6.1683008670806885,
      "epoch": 0.38767220063581775,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004989883000057013,
      "loss": 5.9693,
      "mean_token_accuracy": 0.15927564948797227,
      "num_tokens": 8193306.0,
      "step": 4390
    },
    {
      "entropy": 6.058181524276733,
      "epoch": 0.3881137407276581,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000498985314804621,
      "loss": 5.8593,
      "mean_token_accuracy": 0.16985781043767928,
      "num_tokens": 8203446.0,
      "step": 4395
    },
    {
      "entropy": 6.02800440788269,
      "epoch": 0.3885552808194984,
      "grad_norm": 0.875,
      "learning_rate": 0.0004989823252157958,
      "loss": 5.9169,
      "mean_token_accuracy": 0.16310855597257615,
      "num_tokens": 8212414.0,
      "step": 4400
    },
    {
      "entropy": 6.018846893310547,
      "epoch": 0.38899682091133875,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004989793312392841,
      "loss": 5.8359,
      "mean_token_accuracy": 0.17101139277219773,
      "num_tokens": 8221509.0,
      "step": 4405
    },
    {
      "entropy": 6.062700700759888,
      "epoch": 0.3894383610031791,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004989763328751448,
      "loss": 5.9423,
      "mean_token_accuracy": 0.1717799872159958,
      "num_tokens": 8230017.0,
      "step": 4410
    },
    {
      "entropy": 6.0845201969146725,
      "epoch": 0.3898799010950194,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004989733301234365,
      "loss": 5.8743,
      "mean_token_accuracy": 0.168140971660614,
      "num_tokens": 8238853.0,
      "step": 4415
    },
    {
      "entropy": 6.119135093688965,
      "epoch": 0.39032144118685974,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000498970322984218,
      "loss": 5.9573,
      "mean_token_accuracy": 0.1616509214043617,
      "num_tokens": 8248112.0,
      "step": 4420
    },
    {
      "entropy": 6.067855501174927,
      "epoch": 0.39076298127870013,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004989673114575483,
      "loss": 5.9268,
      "mean_token_accuracy": 0.16173332184553146,
      "num_tokens": 8257166.0,
      "step": 4425
    },
    {
      "entropy": 6.099668502807617,
      "epoch": 0.39120452137054046,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004989642955434863,
      "loss": 5.9336,
      "mean_token_accuracy": 0.1655343070626259,
      "num_tokens": 8266229.0,
      "step": 4430
    },
    {
      "entropy": 6.116667413711548,
      "epoch": 0.3916460614623808,
      "grad_norm": 0.875,
      "learning_rate": 0.0004989612752420912,
      "loss": 5.9014,
      "mean_token_accuracy": 0.16495247632265092,
      "num_tokens": 8275102.0,
      "step": 4435
    },
    {
      "entropy": 5.985392999649048,
      "epoch": 0.3920876015542211,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000498958250553422,
      "loss": 5.8257,
      "mean_token_accuracy": 0.17586065083742142,
      "num_tokens": 8283847.0,
      "step": 4440
    },
    {
      "entropy": 6.055251932144165,
      "epoch": 0.39252914164606145,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004989552214775381,
      "loss": 5.9606,
      "mean_token_accuracy": 0.16416369527578353,
      "num_tokens": 8292622.0,
      "step": 4445
    },
    {
      "entropy": 6.125879526138306,
      "epoch": 0.3929706817379018,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004989521880144988,
      "loss": 5.8634,
      "mean_token_accuracy": 0.17751139402389526,
      "num_tokens": 8301026.0,
      "step": 4450
    },
    {
      "entropy": 6.148255681991577,
      "epoch": 0.3934122218297421,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004989491501643635,
      "loss": 6.1177,
      "mean_token_accuracy": 0.15533054769039153,
      "num_tokens": 8310977.0,
      "step": 4455
    },
    {
      "entropy": 6.070802021026611,
      "epoch": 0.3938537619215825,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004989461079271916,
      "loss": 5.8891,
      "mean_token_accuracy": 0.17560941129922866,
      "num_tokens": 8319391.0,
      "step": 4460
    },
    {
      "entropy": 6.047619771957398,
      "epoch": 0.39429530201342283,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004989430613030429,
      "loss": 5.8413,
      "mean_token_accuracy": 0.16868884712457657,
      "num_tokens": 8328639.0,
      "step": 4465
    },
    {
      "entropy": 6.128910255432129,
      "epoch": 0.39473684210526316,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000498940010291977,
      "loss": 5.9599,
      "mean_token_accuracy": 0.16220096349716187,
      "num_tokens": 8338190.0,
      "step": 4470
    },
    {
      "entropy": 6.021871757507324,
      "epoch": 0.3951783821971035,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004989369548940536,
      "loss": 5.8946,
      "mean_token_accuracy": 0.16686442196369172,
      "num_tokens": 8346874.0,
      "step": 4475
    },
    {
      "entropy": 6.107971143722534,
      "epoch": 0.3956199222889438,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004989338951093327,
      "loss": 5.8898,
      "mean_token_accuracy": 0.17452280074357987,
      "num_tokens": 8356446.0,
      "step": 4480
    },
    {
      "entropy": 6.077869462966919,
      "epoch": 0.39606146238078416,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004989308309378741,
      "loss": 5.8862,
      "mean_token_accuracy": 0.17257023006677627,
      "num_tokens": 8365694.0,
      "step": 4485
    },
    {
      "entropy": 6.063984346389771,
      "epoch": 0.3965030024726245,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004989277623797379,
      "loss": 5.9406,
      "mean_token_accuracy": 0.16608838438987733,
      "num_tokens": 8374282.0,
      "step": 4490
    },
    {
      "entropy": 6.068101119995117,
      "epoch": 0.3969445425644649,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004989246894349841,
      "loss": 5.908,
      "mean_token_accuracy": 0.16799129843711852,
      "num_tokens": 8383315.0,
      "step": 4495
    },
    {
      "entropy": 6.085828971862793,
      "epoch": 0.3973860826563052,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004989216121036732,
      "loss": 5.8716,
      "mean_token_accuracy": 0.16725074946880342,
      "num_tokens": 8392263.0,
      "step": 4500
    },
    {
      "entropy": 6.094727611541748,
      "epoch": 0.39782762274814554,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004989185303858651,
      "loss": 5.9052,
      "mean_token_accuracy": 0.17103357762098312,
      "num_tokens": 8400734.0,
      "step": 4505
    },
    {
      "entropy": 6.054353046417236,
      "epoch": 0.39826916283998587,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004989154442816203,
      "loss": 5.9632,
      "mean_token_accuracy": 0.16649849265813826,
      "num_tokens": 8410635.0,
      "step": 4510
    },
    {
      "entropy": 6.089747953414917,
      "epoch": 0.3987107029318262,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004989123537909994,
      "loss": 5.9367,
      "mean_token_accuracy": 0.1698456108570099,
      "num_tokens": 8420111.0,
      "step": 4515
    },
    {
      "entropy": 6.013007402420044,
      "epoch": 0.39915224302366653,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004989092589140629,
      "loss": 5.8902,
      "mean_token_accuracy": 0.1611982986330986,
      "num_tokens": 8429459.0,
      "step": 4520
    },
    {
      "entropy": 6.087127161026001,
      "epoch": 0.39959378311550686,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004989061596508712,
      "loss": 5.8813,
      "mean_token_accuracy": 0.17483796626329423,
      "num_tokens": 8438083.0,
      "step": 4525
    },
    {
      "entropy": 6.033258438110352,
      "epoch": 0.40003532320734725,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004989030560014853,
      "loss": 5.9093,
      "mean_token_accuracy": 0.17043748795986174,
      "num_tokens": 8447713.0,
      "step": 4530
    },
    {
      "entropy": 6.111389303207398,
      "epoch": 0.4004768632991876,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004988999479659657,
      "loss": 5.88,
      "mean_token_accuracy": 0.16914622038602828,
      "num_tokens": 8457394.0,
      "step": 4535
    },
    {
      "entropy": 6.061598920822144,
      "epoch": 0.4009184033910279,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004988968355443737,
      "loss": 5.8502,
      "mean_token_accuracy": 0.17563896626234055,
      "num_tokens": 8467333.0,
      "step": 4540
    },
    {
      "entropy": 6.019485664367676,
      "epoch": 0.40135994348286824,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004988937187367699,
      "loss": 5.9683,
      "mean_token_accuracy": 0.16143568158149718,
      "num_tokens": 8477530.0,
      "step": 4545
    },
    {
      "entropy": 6.126996564865112,
      "epoch": 0.4018014835747086,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004988905975432154,
      "loss": 6.0199,
      "mean_token_accuracy": 0.16645699143409728,
      "num_tokens": 8486861.0,
      "step": 4550
    },
    {
      "entropy": 6.076916742324829,
      "epoch": 0.4022430236665489,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004988874719637715,
      "loss": 5.8415,
      "mean_token_accuracy": 0.17304785847663878,
      "num_tokens": 8496541.0,
      "step": 4555
    },
    {
      "entropy": 6.047940254211426,
      "epoch": 0.40268456375838924,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004988843419984994,
      "loss": 5.9254,
      "mean_token_accuracy": 0.1652942880988121,
      "num_tokens": 8505667.0,
      "step": 4560
    },
    {
      "entropy": 6.153884840011597,
      "epoch": 0.4031261038502296,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004988812076474604,
      "loss": 5.8971,
      "mean_token_accuracy": 0.16712863743305206,
      "num_tokens": 8515133.0,
      "step": 4565
    },
    {
      "entropy": 6.094613790512085,
      "epoch": 0.40356764394206995,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004988780689107158,
      "loss": 5.87,
      "mean_token_accuracy": 0.17001326382160187,
      "num_tokens": 8524797.0,
      "step": 4570
    },
    {
      "entropy": 5.936724328994751,
      "epoch": 0.4040091840339103,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004988749257883271,
      "loss": 5.893,
      "mean_token_accuracy": 0.1669435977935791,
      "num_tokens": 8534667.0,
      "step": 4575
    },
    {
      "entropy": 6.046252727508545,
      "epoch": 0.4044507241257506,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000498871778280356,
      "loss": 5.9061,
      "mean_token_accuracy": 0.16667770594358444,
      "num_tokens": 8543874.0,
      "step": 4580
    },
    {
      "entropy": 6.062067222595215,
      "epoch": 0.40489226421759095,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004988686263868641,
      "loss": 5.8976,
      "mean_token_accuracy": 0.1650629997253418,
      "num_tokens": 8553620.0,
      "step": 4585
    },
    {
      "entropy": 6.07551589012146,
      "epoch": 0.4053338043094313,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004988654701079131,
      "loss": 5.9763,
      "mean_token_accuracy": 0.16740324944257737,
      "num_tokens": 8563857.0,
      "step": 4590
    },
    {
      "entropy": 6.113534784317016,
      "epoch": 0.4057753444012716,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004988623094435649,
      "loss": 5.9471,
      "mean_token_accuracy": 0.16143951714038848,
      "num_tokens": 8572677.0,
      "step": 4595
    },
    {
      "entropy": 6.016867923736572,
      "epoch": 0.406216884493112,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004988591443938813,
      "loss": 5.8775,
      "mean_token_accuracy": 0.1681135416030884,
      "num_tokens": 8581907.0,
      "step": 4600
    },
    {
      "entropy": 6.116613101959229,
      "epoch": 0.4066584245849523,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004988559749589244,
      "loss": 5.9016,
      "mean_token_accuracy": 0.16940733939409255,
      "num_tokens": 8591701.0,
      "step": 4605
    },
    {
      "entropy": 6.058897066116333,
      "epoch": 0.40709996467679266,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004988528011387563,
      "loss": 5.9476,
      "mean_token_accuracy": 0.170088429749012,
      "num_tokens": 8601054.0,
      "step": 4610
    },
    {
      "entropy": 6.01906566619873,
      "epoch": 0.407541504768633,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004988496229334392,
      "loss": 5.8599,
      "mean_token_accuracy": 0.1751582607626915,
      "num_tokens": 8610346.0,
      "step": 4615
    },
    {
      "entropy": 6.039183759689331,
      "epoch": 0.4079830448604733,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004988464403430352,
      "loss": 5.8486,
      "mean_token_accuracy": 0.1733414739370346,
      "num_tokens": 8620823.0,
      "step": 4620
    },
    {
      "entropy": 6.101271772384644,
      "epoch": 0.40842458495231365,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004988432533676067,
      "loss": 6.0104,
      "mean_token_accuracy": 0.15902107506990432,
      "num_tokens": 8630184.0,
      "step": 4625
    },
    {
      "entropy": 6.092225360870361,
      "epoch": 0.408866125044154,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004988400620072163,
      "loss": 5.9966,
      "mean_token_accuracy": 0.15500882863998414,
      "num_tokens": 8640064.0,
      "step": 4630
    },
    {
      "entropy": 6.104148864746094,
      "epoch": 0.40930766513599437,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004988368662619263,
      "loss": 5.8964,
      "mean_token_accuracy": 0.16778270304203033,
      "num_tokens": 8650503.0,
      "step": 4635
    },
    {
      "entropy": 6.148624897003174,
      "epoch": 0.4097492052278347,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004988336661317994,
      "loss": 5.9207,
      "mean_token_accuracy": 0.16135725378990173,
      "num_tokens": 8659125.0,
      "step": 4640
    },
    {
      "entropy": 6.012713432312012,
      "epoch": 0.41019074531967503,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004988304616168984,
      "loss": 5.946,
      "mean_token_accuracy": 0.16297379732131959,
      "num_tokens": 8668193.0,
      "step": 4645
    },
    {
      "entropy": 6.092987585067749,
      "epoch": 0.41063228541151536,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004988272527172858,
      "loss": 5.8977,
      "mean_token_accuracy": 0.16841595321893693,
      "num_tokens": 8677515.0,
      "step": 4650
    },
    {
      "entropy": 6.112038326263428,
      "epoch": 0.4110738255033557,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004988240394330246,
      "loss": 5.9334,
      "mean_token_accuracy": 0.16429426819086074,
      "num_tokens": 8687549.0,
      "step": 4655
    },
    {
      "entropy": 6.041188097000122,
      "epoch": 0.411515365595196,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004988208217641778,
      "loss": 5.9235,
      "mean_token_accuracy": 0.1723731204867363,
      "num_tokens": 8697371.0,
      "step": 4660
    },
    {
      "entropy": 6.059947109222412,
      "epoch": 0.41195690568703636,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004988175997108086,
      "loss": 5.9241,
      "mean_token_accuracy": 0.16697203665971755,
      "num_tokens": 8707193.0,
      "step": 4665
    },
    {
      "entropy": 6.085496950149536,
      "epoch": 0.41239844577887674,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004988143732729797,
      "loss": 5.9383,
      "mean_token_accuracy": 0.16928174197673798,
      "num_tokens": 8716052.0,
      "step": 4670
    },
    {
      "entropy": 6.014765739440918,
      "epoch": 0.4128399858707171,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004988111424507546,
      "loss": 5.8757,
      "mean_token_accuracy": 0.17379332482814788,
      "num_tokens": 8726140.0,
      "step": 4675
    },
    {
      "entropy": 6.012660837173462,
      "epoch": 0.4132815259625574,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004988079072441964,
      "loss": 5.8323,
      "mean_token_accuracy": 0.1784330740571022,
      "num_tokens": 8735299.0,
      "step": 4680
    },
    {
      "entropy": 6.058017349243164,
      "epoch": 0.41372306605439774,
      "grad_norm": 1.0,
      "learning_rate": 0.0004988046676533687,
      "loss": 5.9309,
      "mean_token_accuracy": 0.16514215022325515,
      "num_tokens": 8744686.0,
      "step": 4685
    },
    {
      "entropy": 6.061055088043213,
      "epoch": 0.41416460614623807,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004988014236783347,
      "loss": 5.9154,
      "mean_token_accuracy": 0.17677023261785507,
      "num_tokens": 8754942.0,
      "step": 4690
    },
    {
      "entropy": 6.153789758682251,
      "epoch": 0.4146061462380784,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004987981753191582,
      "loss": 5.9172,
      "mean_token_accuracy": 0.16312515884637832,
      "num_tokens": 8764054.0,
      "step": 4695
    },
    {
      "entropy": 5.930011558532715,
      "epoch": 0.41504768632991873,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004987949225759027,
      "loss": 5.7808,
      "mean_token_accuracy": 0.17439974397420882,
      "num_tokens": 8773050.0,
      "step": 4700
    },
    {
      "entropy": 6.110511398315429,
      "epoch": 0.4154892264217591,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004987916654486321,
      "loss": 5.9053,
      "mean_token_accuracy": 0.16541254967451097,
      "num_tokens": 8782476.0,
      "step": 4705
    },
    {
      "entropy": 6.106545448303223,
      "epoch": 0.41593076651359945,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004987884039374099,
      "loss": 6.0026,
      "mean_token_accuracy": 0.16032444387674333,
      "num_tokens": 8791147.0,
      "step": 4710
    },
    {
      "entropy": 6.089293193817139,
      "epoch": 0.4163723066054398,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004987851380423001,
      "loss": 5.962,
      "mean_token_accuracy": 0.1648245297372341,
      "num_tokens": 8801151.0,
      "step": 4715
    },
    {
      "entropy": 6.077416801452637,
      "epoch": 0.4168138466972801,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004987818677633668,
      "loss": 5.914,
      "mean_token_accuracy": 0.1638292446732521,
      "num_tokens": 8809587.0,
      "step": 4720
    },
    {
      "entropy": 5.994811487197876,
      "epoch": 0.41725538678912044,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000498778593100674,
      "loss": 5.7869,
      "mean_token_accuracy": 0.17192702740430832,
      "num_tokens": 8818168.0,
      "step": 4725
    },
    {
      "entropy": 5.973134136199951,
      "epoch": 0.4176969268809608,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004987753140542857,
      "loss": 5.8466,
      "mean_token_accuracy": 0.1726938620209694,
      "num_tokens": 8827477.0,
      "step": 4730
    },
    {
      "entropy": 6.047493362426758,
      "epoch": 0.4181384669728011,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004987720306242664,
      "loss": 5.8354,
      "mean_token_accuracy": 0.17470157742500306,
      "num_tokens": 8837067.0,
      "step": 4735
    },
    {
      "entropy": 6.049209785461426,
      "epoch": 0.4185800070646415,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004987687428106803,
      "loss": 5.9488,
      "mean_token_accuracy": 0.1697684958577156,
      "num_tokens": 8845790.0,
      "step": 4740
    },
    {
      "entropy": 6.073341274261475,
      "epoch": 0.4190215471564818,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004987654506135917,
      "loss": 5.882,
      "mean_token_accuracy": 0.17220733910799027,
      "num_tokens": 8855242.0,
      "step": 4745
    },
    {
      "entropy": 6.10594334602356,
      "epoch": 0.41946308724832215,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004987621540330652,
      "loss": 5.906,
      "mean_token_accuracy": 0.17124312967061997,
      "num_tokens": 8864459.0,
      "step": 4750
    },
    {
      "entropy": 6.104751348495483,
      "epoch": 0.4199046273401625,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004987588530691653,
      "loss": 5.9673,
      "mean_token_accuracy": 0.16231338381767274,
      "num_tokens": 8875028.0,
      "step": 4755
    },
    {
      "entropy": 5.991171360015869,
      "epoch": 0.4203461674320028,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004987555477219569,
      "loss": 5.8217,
      "mean_token_accuracy": 0.17759732604026796,
      "num_tokens": 8883857.0,
      "step": 4760
    },
    {
      "entropy": 6.0481287956237795,
      "epoch": 0.42078770752384315,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004987522379915045,
      "loss": 5.7978,
      "mean_token_accuracy": 0.17629640698432922,
      "num_tokens": 8893499.0,
      "step": 4765
    },
    {
      "entropy": 6.0255521774292,
      "epoch": 0.4212292476156835,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000498748923877873,
      "loss": 5.8209,
      "mean_token_accuracy": 0.17817992717027664,
      "num_tokens": 8903368.0,
      "step": 4770
    },
    {
      "entropy": 6.027704524993896,
      "epoch": 0.42167078770752386,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004987456053811273,
      "loss": 5.9307,
      "mean_token_accuracy": 0.16877609640359878,
      "num_tokens": 8912701.0,
      "step": 4775
    },
    {
      "entropy": 6.072631692886352,
      "epoch": 0.4221123277993642,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004987422825013325,
      "loss": 5.879,
      "mean_token_accuracy": 0.16829278767108918,
      "num_tokens": 8921962.0,
      "step": 4780
    },
    {
      "entropy": 6.109710693359375,
      "epoch": 0.4225538678912045,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004987389552385536,
      "loss": 5.8662,
      "mean_token_accuracy": 0.17203690707683564,
      "num_tokens": 8931923.0,
      "step": 4785
    },
    {
      "entropy": 6.003576421737671,
      "epoch": 0.42299540798304486,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004987356235928558,
      "loss": 5.9099,
      "mean_token_accuracy": 0.16035820543766022,
      "num_tokens": 8940403.0,
      "step": 4790
    },
    {
      "entropy": 6.0650804996490475,
      "epoch": 0.4234369480748852,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004987322875643044,
      "loss": 5.8359,
      "mean_token_accuracy": 0.1735415205359459,
      "num_tokens": 8949377.0,
      "step": 4795
    },
    {
      "entropy": 6.1672426700592045,
      "epoch": 0.4238784881667255,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004987289471529647,
      "loss": 5.9921,
      "mean_token_accuracy": 0.15903109014034272,
      "num_tokens": 8958719.0,
      "step": 4800
    },
    {
      "entropy": 6.056859874725342,
      "epoch": 0.42432002825856585,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004987256023589022,
      "loss": 5.9481,
      "mean_token_accuracy": 0.15579934269189835,
      "num_tokens": 8968226.0,
      "step": 4805
    },
    {
      "entropy": 6.062025690078736,
      "epoch": 0.42476156835040624,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004987222531821824,
      "loss": 5.8949,
      "mean_token_accuracy": 0.16687395423650742,
      "num_tokens": 8976670.0,
      "step": 4810
    },
    {
      "entropy": 5.9757143497467045,
      "epoch": 0.42520310844224657,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004987188996228709,
      "loss": 5.8281,
      "mean_token_accuracy": 0.17414531707763672,
      "num_tokens": 8986185.0,
      "step": 4815
    },
    {
      "entropy": 6.0142176151275635,
      "epoch": 0.4256446485340869,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004987155416810334,
      "loss": 5.8315,
      "mean_token_accuracy": 0.17239081412553786,
      "num_tokens": 8995124.0,
      "step": 4820
    },
    {
      "entropy": 6.050481128692627,
      "epoch": 0.42608618862592723,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004987121793567356,
      "loss": 5.9421,
      "mean_token_accuracy": 0.1673525243997574,
      "num_tokens": 9004380.0,
      "step": 4825
    },
    {
      "entropy": 6.11012635231018,
      "epoch": 0.42652772871776756,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004987088126500436,
      "loss": 5.8117,
      "mean_token_accuracy": 0.17067271023988723,
      "num_tokens": 9013791.0,
      "step": 4830
    },
    {
      "entropy": 6.019365072250366,
      "epoch": 0.4269692688096079,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000498705441561023,
      "loss": 5.8511,
      "mean_token_accuracy": 0.17690863311290742,
      "num_tokens": 9023076.0,
      "step": 4835
    },
    {
      "entropy": 6.13967719078064,
      "epoch": 0.4274108089014482,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004987020660897401,
      "loss": 5.9362,
      "mean_token_accuracy": 0.16226004511117936,
      "num_tokens": 9032720.0,
      "step": 4840
    },
    {
      "entropy": 6.107657623291016,
      "epoch": 0.4278523489932886,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000498698686236261,
      "loss": 5.908,
      "mean_token_accuracy": 0.16728319376707076,
      "num_tokens": 9042652.0,
      "step": 4845
    },
    {
      "entropy": 6.1251993656158445,
      "epoch": 0.42829388908512894,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004986953020006519,
      "loss": 5.9835,
      "mean_token_accuracy": 0.16414731591939927,
      "num_tokens": 9052172.0,
      "step": 4850
    },
    {
      "entropy": 6.0267579555511475,
      "epoch": 0.4287354291769693,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004986919133829788,
      "loss": 5.8619,
      "mean_token_accuracy": 0.1743416041135788,
      "num_tokens": 9061798.0,
      "step": 4855
    },
    {
      "entropy": 5.954681015014648,
      "epoch": 0.4291769692688096,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004986885203833086,
      "loss": 5.7878,
      "mean_token_accuracy": 0.17132241576910018,
      "num_tokens": 9070429.0,
      "step": 4860
    },
    {
      "entropy": 6.0399816036224365,
      "epoch": 0.42961850936064994,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004986851230017075,
      "loss": 5.8992,
      "mean_token_accuracy": 0.1731168210506439,
      "num_tokens": 9080200.0,
      "step": 4865
    },
    {
      "entropy": 6.017161512374878,
      "epoch": 0.43006004945249027,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004986817212382419,
      "loss": 5.8932,
      "mean_token_accuracy": 0.1726991668343544,
      "num_tokens": 9089555.0,
      "step": 4870
    },
    {
      "entropy": 6.037603855133057,
      "epoch": 0.4305015895443306,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004986783150929786,
      "loss": 5.8551,
      "mean_token_accuracy": 0.16833554804325104,
      "num_tokens": 9099091.0,
      "step": 4875
    },
    {
      "entropy": 6.059744453430175,
      "epoch": 0.430943129636171,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004986749045659845,
      "loss": 5.956,
      "mean_token_accuracy": 0.15879134684801102,
      "num_tokens": 9109008.0,
      "step": 4880
    },
    {
      "entropy": 6.035340404510498,
      "epoch": 0.4313846697280113,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004986714896573261,
      "loss": 5.8079,
      "mean_token_accuracy": 0.16968925893306733,
      "num_tokens": 9117882.0,
      "step": 4885
    },
    {
      "entropy": 6.090554904937744,
      "epoch": 0.43182620981985165,
      "grad_norm": 1.0,
      "learning_rate": 0.0004986680703670704,
      "loss": 5.9384,
      "mean_token_accuracy": 0.1664467602968216,
      "num_tokens": 9126860.0,
      "step": 4890
    },
    {
      "entropy": 6.059638452529907,
      "epoch": 0.432267749911692,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004986646466952845,
      "loss": 5.8479,
      "mean_token_accuracy": 0.1714713305234909,
      "num_tokens": 9135819.0,
      "step": 4895
    },
    {
      "entropy": 6.017725610733033,
      "epoch": 0.4327092900035323,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004986612186420353,
      "loss": 5.8381,
      "mean_token_accuracy": 0.17259113788604735,
      "num_tokens": 9145302.0,
      "step": 4900
    },
    {
      "entropy": 6.003063488006592,
      "epoch": 0.43315083009537264,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004986577862073901,
      "loss": 5.8874,
      "mean_token_accuracy": 0.16548449397087098,
      "num_tokens": 9154667.0,
      "step": 4905
    },
    {
      "entropy": 6.125681400299072,
      "epoch": 0.43359237018721297,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004986543493914159,
      "loss": 5.9333,
      "mean_token_accuracy": 0.15938420295715333,
      "num_tokens": 9164562.0,
      "step": 4910
    },
    {
      "entropy": 6.060342788696289,
      "epoch": 0.43403391027905336,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004986509081941805,
      "loss": 5.958,
      "mean_token_accuracy": 0.16365616768598557,
      "num_tokens": 9174872.0,
      "step": 4915
    },
    {
      "entropy": 5.960330772399902,
      "epoch": 0.4344754503708937,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004986474626157507,
      "loss": 5.6941,
      "mean_token_accuracy": 0.1887632668018341,
      "num_tokens": 9184322.0,
      "step": 4920
    },
    {
      "entropy": 5.985001468658448,
      "epoch": 0.434916990462734,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004986440126561945,
      "loss": 5.8471,
      "mean_token_accuracy": 0.17162098586559296,
      "num_tokens": 9194450.0,
      "step": 4925
    },
    {
      "entropy": 6.089518976211548,
      "epoch": 0.43535853055457435,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004986405583155792,
      "loss": 5.8902,
      "mean_token_accuracy": 0.16652678400278093,
      "num_tokens": 9203658.0,
      "step": 4930
    },
    {
      "entropy": 6.0208498477935795,
      "epoch": 0.4358000706464147,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004986370995939725,
      "loss": 5.8667,
      "mean_token_accuracy": 0.1689656123518944,
      "num_tokens": 9213609.0,
      "step": 4935
    },
    {
      "entropy": 6.101765394210815,
      "epoch": 0.436241610738255,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004986336364914423,
      "loss": 5.8749,
      "mean_token_accuracy": 0.16693566143512725,
      "num_tokens": 9222704.0,
      "step": 4940
    },
    {
      "entropy": 6.069900178909302,
      "epoch": 0.43668315083009535,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004986301690080564,
      "loss": 5.8994,
      "mean_token_accuracy": 0.16185810565948486,
      "num_tokens": 9231599.0,
      "step": 4945
    },
    {
      "entropy": 6.067622566223145,
      "epoch": 0.43712469092193573,
      "grad_norm": 0.875,
      "learning_rate": 0.0004986266971438826,
      "loss": 5.8988,
      "mean_token_accuracy": 0.1730104938149452,
      "num_tokens": 9241886.0,
      "step": 4950
    },
    {
      "entropy": 6.030080842971802,
      "epoch": 0.43756623101377606,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000498623220898989,
      "loss": 5.9147,
      "mean_token_accuracy": 0.1641816407442093,
      "num_tokens": 9251085.0,
      "step": 4955
    },
    {
      "entropy": 6.146408605575561,
      "epoch": 0.4380077711056164,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004986197402734436,
      "loss": 5.881,
      "mean_token_accuracy": 0.16509923040866853,
      "num_tokens": 9259601.0,
      "step": 4960
    },
    {
      "entropy": 6.047398519515991,
      "epoch": 0.4384493111974567,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004986162552673148,
      "loss": 5.9035,
      "mean_token_accuracy": 0.17131353914737701,
      "num_tokens": 9268935.0,
      "step": 4965
    },
    {
      "entropy": 6.043163251876831,
      "epoch": 0.43889085128929706,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004986127658806706,
      "loss": 5.9097,
      "mean_token_accuracy": 0.1679410979151726,
      "num_tokens": 9277647.0,
      "step": 4970
    },
    {
      "entropy": 6.058499574661255,
      "epoch": 0.4393323913811374,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004986092721135796,
      "loss": 5.8651,
      "mean_token_accuracy": 0.1684943124651909,
      "num_tokens": 9286610.0,
      "step": 4975
    },
    {
      "entropy": 6.079809713363647,
      "epoch": 0.4397739314729777,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004986057739661101,
      "loss": 5.9556,
      "mean_token_accuracy": 0.16726439744234084,
      "num_tokens": 9295946.0,
      "step": 4980
    },
    {
      "entropy": 5.996550035476685,
      "epoch": 0.4402154715648181,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004986022714383307,
      "loss": 5.8543,
      "mean_token_accuracy": 0.16506249457597733,
      "num_tokens": 9304903.0,
      "step": 4985
    },
    {
      "entropy": 6.0311790943145756,
      "epoch": 0.44065701165665844,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004985987645303099,
      "loss": 5.7545,
      "mean_token_accuracy": 0.17690206319093704,
      "num_tokens": 9313606.0,
      "step": 4990
    },
    {
      "entropy": 5.978774547576904,
      "epoch": 0.44109855174849877,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004985952532421164,
      "loss": 5.8664,
      "mean_token_accuracy": 0.16727809906005858,
      "num_tokens": 9322815.0,
      "step": 4995
    },
    {
      "entropy": 6.001399135589599,
      "epoch": 0.4415400918403391,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004985917375738193,
      "loss": 5.7948,
      "mean_token_accuracy": 0.16886604577302933,
      "num_tokens": 9332630.0,
      "step": 5000
    },
    {
      "entropy": 5.953770589828491,
      "epoch": 0.44198163193217943,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004985882175254871,
      "loss": 5.8331,
      "mean_token_accuracy": 0.16362490952014924,
      "num_tokens": 9342216.0,
      "step": 5005
    },
    {
      "entropy": 6.037289476394653,
      "epoch": 0.44242317202401976,
      "grad_norm": 0.875,
      "learning_rate": 0.0004985846930971887,
      "loss": 5.8741,
      "mean_token_accuracy": 0.16757041811943055,
      "num_tokens": 9352295.0,
      "step": 5010
    },
    {
      "entropy": 6.042323303222656,
      "epoch": 0.4428647121158601,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004985811642889937,
      "loss": 5.8725,
      "mean_token_accuracy": 0.1704294890165329,
      "num_tokens": 9361462.0,
      "step": 5015
    },
    {
      "entropy": 6.0555445671081545,
      "epoch": 0.4433062522077005,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004985776311009705,
      "loss": 5.8104,
      "mean_token_accuracy": 0.16330993324518203,
      "num_tokens": 9370379.0,
      "step": 5020
    },
    {
      "entropy": 6.038263320922852,
      "epoch": 0.4437477922995408,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004985740935331888,
      "loss": 5.8682,
      "mean_token_accuracy": 0.17026212811470032,
      "num_tokens": 9379255.0,
      "step": 5025
    },
    {
      "entropy": 6.011457347869873,
      "epoch": 0.44418933239138114,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004985705515857177,
      "loss": 5.8265,
      "mean_token_accuracy": 0.16707434952259065,
      "num_tokens": 9389313.0,
      "step": 5030
    },
    {
      "entropy": 5.883806419372559,
      "epoch": 0.4446308724832215,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004985670052586268,
      "loss": 5.6301,
      "mean_token_accuracy": 0.1897162079811096,
      "num_tokens": 9397778.0,
      "step": 5035
    },
    {
      "entropy": 6.027179956436157,
      "epoch": 0.4450724125750618,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004985634545519853,
      "loss": 5.9104,
      "mean_token_accuracy": 0.16823799014091492,
      "num_tokens": 9407831.0,
      "step": 5040
    },
    {
      "entropy": 6.010887002944946,
      "epoch": 0.44551395266690214,
      "grad_norm": 1.125,
      "learning_rate": 0.0004985598994658629,
      "loss": 5.8031,
      "mean_token_accuracy": 0.17146655917167664,
      "num_tokens": 9418458.0,
      "step": 5045
    },
    {
      "entropy": 5.885775423049926,
      "epoch": 0.44595549275874247,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004985563400003291,
      "loss": 5.7187,
      "mean_token_accuracy": 0.18735787272453308,
      "num_tokens": 9426911.0,
      "step": 5050
    },
    {
      "entropy": 5.971336841583252,
      "epoch": 0.44639703285058285,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004985527761554539,
      "loss": 5.9022,
      "mean_token_accuracy": 0.17282261699438095,
      "num_tokens": 9435896.0,
      "step": 5055
    },
    {
      "entropy": 6.102570009231568,
      "epoch": 0.4468385729424232,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000498549207931307,
      "loss": 5.8913,
      "mean_token_accuracy": 0.16545256823301316,
      "num_tokens": 9445272.0,
      "step": 5060
    },
    {
      "entropy": 6.019857549667359,
      "epoch": 0.4472801130342635,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004985456353279581,
      "loss": 5.872,
      "mean_token_accuracy": 0.16560457795858383,
      "num_tokens": 9455377.0,
      "step": 5065
    },
    {
      "entropy": 6.038298034667969,
      "epoch": 0.44772165312610385,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004985420583454774,
      "loss": 5.9011,
      "mean_token_accuracy": 0.16640974432229996,
      "num_tokens": 9464918.0,
      "step": 5070
    },
    {
      "entropy": 6.034631872177124,
      "epoch": 0.4481631932179442,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004985384769839349,
      "loss": 5.8991,
      "mean_token_accuracy": 0.15917696356773375,
      "num_tokens": 9473322.0,
      "step": 5075
    },
    {
      "entropy": 6.0339454174041744,
      "epoch": 0.4486047333097845,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004985348912434008,
      "loss": 5.8089,
      "mean_token_accuracy": 0.17205141931772233,
      "num_tokens": 9482255.0,
      "step": 5080
    },
    {
      "entropy": 6.015864276885987,
      "epoch": 0.44904627340162484,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004985313011239452,
      "loss": 5.8479,
      "mean_token_accuracy": 0.1718486800789833,
      "num_tokens": 9491709.0,
      "step": 5085
    },
    {
      "entropy": 5.916902494430542,
      "epoch": 0.4494878134934652,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004985277066256388,
      "loss": 5.8391,
      "mean_token_accuracy": 0.17222511470317842,
      "num_tokens": 9500594.0,
      "step": 5090
    },
    {
      "entropy": 6.0166843891143795,
      "epoch": 0.44992935358530556,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004985241077485515,
      "loss": 5.8302,
      "mean_token_accuracy": 0.17186546325683594,
      "num_tokens": 9509088.0,
      "step": 5095
    },
    {
      "entropy": 6.054204368591309,
      "epoch": 0.4503708936771459,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004985205044927541,
      "loss": 5.8254,
      "mean_token_accuracy": 0.17474611550569535,
      "num_tokens": 9517776.0,
      "step": 5100
    },
    {
      "entropy": 6.00213885307312,
      "epoch": 0.4508124337689862,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004985168968583173,
      "loss": 5.8802,
      "mean_token_accuracy": 0.16938357502222062,
      "num_tokens": 9527080.0,
      "step": 5105
    },
    {
      "entropy": 5.956983804702759,
      "epoch": 0.45125397386082655,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004985132848453114,
      "loss": 5.7284,
      "mean_token_accuracy": 0.18465249985456467,
      "num_tokens": 9536358.0,
      "step": 5110
    },
    {
      "entropy": 5.965890550613404,
      "epoch": 0.4516955139526669,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004985096684538075,
      "loss": 5.8272,
      "mean_token_accuracy": 0.16931069046258926,
      "num_tokens": 9545528.0,
      "step": 5115
    },
    {
      "entropy": 6.022493124008179,
      "epoch": 0.4521370540445072,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004985060476838763,
      "loss": 5.8626,
      "mean_token_accuracy": 0.1728214368224144,
      "num_tokens": 9554433.0,
      "step": 5120
    },
    {
      "entropy": 5.963240098953247,
      "epoch": 0.4525785941363476,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004985024225355887,
      "loss": 5.7917,
      "mean_token_accuracy": 0.16836365759372712,
      "num_tokens": 9563932.0,
      "step": 5125
    },
    {
      "entropy": 6.0266313552856445,
      "epoch": 0.45302013422818793,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004984987930090158,
      "loss": 5.8077,
      "mean_token_accuracy": 0.17788426131010054,
      "num_tokens": 9572829.0,
      "step": 5130
    },
    {
      "entropy": 5.9038893699646,
      "epoch": 0.45346167432002826,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004984951591042285,
      "loss": 5.8342,
      "mean_token_accuracy": 0.17775831371545792,
      "num_tokens": 9583597.0,
      "step": 5135
    },
    {
      "entropy": 6.081721448898316,
      "epoch": 0.4539032144118686,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004984915208212983,
      "loss": 5.905,
      "mean_token_accuracy": 0.1649731293320656,
      "num_tokens": 9593114.0,
      "step": 5140
    },
    {
      "entropy": 6.008826160430909,
      "epoch": 0.4543447545037089,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004984878781602964,
      "loss": 5.8081,
      "mean_token_accuracy": 0.16601281613111496,
      "num_tokens": 9601454.0,
      "step": 5145
    },
    {
      "entropy": 5.970934009552002,
      "epoch": 0.45478629459554926,
      "grad_norm": 0.78125,
      "learning_rate": 0.0004984842311212939,
      "loss": 5.845,
      "mean_token_accuracy": 0.17320733666419982,
      "num_tokens": 9611994.0,
      "step": 5150
    },
    {
      "entropy": 6.006623125076294,
      "epoch": 0.4552278346873896,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004984805797043625,
      "loss": 5.8176,
      "mean_token_accuracy": 0.17202675640583037,
      "num_tokens": 9621318.0,
      "step": 5155
    },
    {
      "entropy": 5.9953593730926515,
      "epoch": 0.45566937477923,
      "grad_norm": 0.875,
      "learning_rate": 0.0004984769239095736,
      "loss": 5.8221,
      "mean_token_accuracy": 0.17728879749774934,
      "num_tokens": 9630270.0,
      "step": 5160
    },
    {
      "entropy": 6.006083202362061,
      "epoch": 0.4561109148710703,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004984732637369989,
      "loss": 5.8543,
      "mean_token_accuracy": 0.16921540051698686,
      "num_tokens": 9640391.0,
      "step": 5165
    },
    {
      "entropy": 5.991392326354981,
      "epoch": 0.45655245496291064,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004984695991867099,
      "loss": 5.8739,
      "mean_token_accuracy": 0.16853009909391403,
      "num_tokens": 9648827.0,
      "step": 5170
    },
    {
      "entropy": 5.993249130249024,
      "epoch": 0.45699399505475097,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004984659302587788,
      "loss": 5.7851,
      "mean_token_accuracy": 0.16915750354528428,
      "num_tokens": 9657940.0,
      "step": 5175
    },
    {
      "entropy": 5.999341154098511,
      "epoch": 0.4574355351465913,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000498462256953277,
      "loss": 5.913,
      "mean_token_accuracy": 0.16240039765834807,
      "num_tokens": 9668296.0,
      "step": 5180
    },
    {
      "entropy": 6.074361944198609,
      "epoch": 0.45787707523843163,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004984585792702767,
      "loss": 5.8598,
      "mean_token_accuracy": 0.16403512805700302,
      "num_tokens": 9677914.0,
      "step": 5185
    },
    {
      "entropy": 6.065802478790284,
      "epoch": 0.45831861533027196,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004984548972098501,
      "loss": 6.0056,
      "mean_token_accuracy": 0.15658557415008545,
      "num_tokens": 9687339.0,
      "step": 5190
    },
    {
      "entropy": 5.965339040756225,
      "epoch": 0.45876015542211235,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000498451210772069,
      "loss": 5.7932,
      "mean_token_accuracy": 0.17470138892531395,
      "num_tokens": 9696889.0,
      "step": 5195
    },
    {
      "entropy": 6.006475973129272,
      "epoch": 0.4592016955139527,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004984475199570058,
      "loss": 5.7387,
      "mean_token_accuracy": 0.17829496264457703,
      "num_tokens": 9705498.0,
      "step": 5200
    },
    {
      "entropy": 6.030996513366699,
      "epoch": 0.459643235605793,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004984438247647329,
      "loss": 5.8224,
      "mean_token_accuracy": 0.17188162505626678,
      "num_tokens": 9713936.0,
      "step": 5205
    },
    {
      "entropy": 6.003251218795777,
      "epoch": 0.46008477569763334,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004984401251953223,
      "loss": 5.885,
      "mean_token_accuracy": 0.1700436070561409,
      "num_tokens": 9723924.0,
      "step": 5210
    },
    {
      "entropy": 6.122557973861694,
      "epoch": 0.4605263157894737,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004984364212488469,
      "loss": 5.9228,
      "mean_token_accuracy": 0.16665101200342178,
      "num_tokens": 9733427.0,
      "step": 5215
    },
    {
      "entropy": 5.952805805206299,
      "epoch": 0.460967855881314,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004984327129253789,
      "loss": 5.7258,
      "mean_token_accuracy": 0.1850974217057228,
      "num_tokens": 9742054.0,
      "step": 5220
    },
    {
      "entropy": 5.9985393524169925,
      "epoch": 0.46140939597315433,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004984290002249914,
      "loss": 5.9011,
      "mean_token_accuracy": 0.1632828250527382,
      "num_tokens": 9752447.0,
      "step": 5225
    },
    {
      "entropy": 6.011101341247558,
      "epoch": 0.4618509360649947,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004984252831477567,
      "loss": 5.7107,
      "mean_token_accuracy": 0.17958685308694838,
      "num_tokens": 9760878.0,
      "step": 5230
    },
    {
      "entropy": 5.887458419799804,
      "epoch": 0.46229247615683505,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004984215616937477,
      "loss": 5.7921,
      "mean_token_accuracy": 0.17403154075145721,
      "num_tokens": 9770200.0,
      "step": 5235
    },
    {
      "entropy": 6.054266548156738,
      "epoch": 0.4627340162486754,
      "grad_norm": 0.875,
      "learning_rate": 0.0004984178358630374,
      "loss": 5.8944,
      "mean_token_accuracy": 0.17439596951007844,
      "num_tokens": 9780303.0,
      "step": 5240
    },
    {
      "entropy": 5.999644231796265,
      "epoch": 0.4631755563405157,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004984141056556989,
      "loss": 5.7899,
      "mean_token_accuracy": 0.17770855873823166,
      "num_tokens": 9790248.0,
      "step": 5245
    },
    {
      "entropy": 6.080256414413452,
      "epoch": 0.46361709643235605,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004984103710718051,
      "loss": 5.8931,
      "mean_token_accuracy": 0.16303637325763704,
      "num_tokens": 9799345.0,
      "step": 5250
    },
    {
      "entropy": 5.97104778289795,
      "epoch": 0.4640586365241964,
      "grad_norm": 0.953125,
      "learning_rate": 0.000498406632111429,
      "loss": 5.7595,
      "mean_token_accuracy": 0.17320360243320465,
      "num_tokens": 9808561.0,
      "step": 5255
    },
    {
      "entropy": 5.9949586391448975,
      "epoch": 0.4645001766160367,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004984028887746443,
      "loss": 5.8166,
      "mean_token_accuracy": 0.17151765376329423,
      "num_tokens": 9818324.0,
      "step": 5260
    },
    {
      "entropy": 5.9564426898956295,
      "epoch": 0.4649417167078771,
      "grad_norm": 1.0,
      "learning_rate": 0.0004983991410615239,
      "loss": 5.7875,
      "mean_token_accuracy": 0.1792932629585266,
      "num_tokens": 9827935.0,
      "step": 5265
    },
    {
      "entropy": 5.999061918258667,
      "epoch": 0.4653832567997174,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004983953889721414,
      "loss": 5.8047,
      "mean_token_accuracy": 0.17208529710769654,
      "num_tokens": 9837118.0,
      "step": 5270
    },
    {
      "entropy": 6.022691106796264,
      "epoch": 0.46582479689155776,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004983916325065703,
      "loss": 5.8701,
      "mean_token_accuracy": 0.1652819335460663,
      "num_tokens": 9846197.0,
      "step": 5275
    },
    {
      "entropy": 6.097032976150513,
      "epoch": 0.4662663369833981,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004983878716648842,
      "loss": 5.9403,
      "mean_token_accuracy": 0.15975897908210754,
      "num_tokens": 9856120.0,
      "step": 5280
    },
    {
      "entropy": 6.038593721389771,
      "epoch": 0.4667078770752384,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004983841064471567,
      "loss": 5.8877,
      "mean_token_accuracy": 0.16843740940093993,
      "num_tokens": 9865599.0,
      "step": 5285
    },
    {
      "entropy": 6.003615522384644,
      "epoch": 0.46714941716707875,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004983803368534617,
      "loss": 5.7839,
      "mean_token_accuracy": 0.1708475798368454,
      "num_tokens": 9876471.0,
      "step": 5290
    },
    {
      "entropy": 6.113279008865357,
      "epoch": 0.4675909572589191,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004983765628838728,
      "loss": 5.8396,
      "mean_token_accuracy": 0.16985518038272857,
      "num_tokens": 9887680.0,
      "step": 5295
    },
    {
      "entropy": 6.086295223236084,
      "epoch": 0.46803249735075947,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004983727845384641,
      "loss": 5.9189,
      "mean_token_accuracy": 0.1621912255883217,
      "num_tokens": 9897366.0,
      "step": 5300
    },
    {
      "entropy": 5.999127626419067,
      "epoch": 0.4684740374425998,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004983690018173096,
      "loss": 5.8913,
      "mean_token_accuracy": 0.1667093500494957,
      "num_tokens": 9907437.0,
      "step": 5305
    },
    {
      "entropy": 5.950143623352051,
      "epoch": 0.46891557753444013,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004983652147204834,
      "loss": 5.6476,
      "mean_token_accuracy": 0.18681340366601945,
      "num_tokens": 9915766.0,
      "step": 5310
    },
    {
      "entropy": 5.981765604019165,
      "epoch": 0.46935711762628046,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004983614232480598,
      "loss": 5.8873,
      "mean_token_accuracy": 0.16824992448091508,
      "num_tokens": 9925389.0,
      "step": 5315
    },
    {
      "entropy": 6.03397536277771,
      "epoch": 0.4697986577181208,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004983576274001127,
      "loss": 5.9331,
      "mean_token_accuracy": 0.16481596678495408,
      "num_tokens": 9935798.0,
      "step": 5320
    },
    {
      "entropy": 6.047959184646606,
      "epoch": 0.4702401978099611,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000498353827176717,
      "loss": 5.7848,
      "mean_token_accuracy": 0.17924853414297104,
      "num_tokens": 9945579.0,
      "step": 5325
    },
    {
      "entropy": 5.996797943115235,
      "epoch": 0.4706817379018015,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004983500225779466,
      "loss": 5.7843,
      "mean_token_accuracy": 0.175889290869236,
      "num_tokens": 9955248.0,
      "step": 5330
    },
    {
      "entropy": 5.982284498214722,
      "epoch": 0.47112327799364184,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004983462136038764,
      "loss": 5.816,
      "mean_token_accuracy": 0.17963929623365402,
      "num_tokens": 9965078.0,
      "step": 5335
    },
    {
      "entropy": 6.037516069412232,
      "epoch": 0.4715648180854822,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004983424002545809,
      "loss": 5.7763,
      "mean_token_accuracy": 0.18421670794487,
      "num_tokens": 9973932.0,
      "step": 5340
    },
    {
      "entropy": 6.022534418106079,
      "epoch": 0.4720063581773225,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004983385825301348,
      "loss": 5.8297,
      "mean_token_accuracy": 0.17369676381349564,
      "num_tokens": 9983315.0,
      "step": 5345
    },
    {
      "entropy": 5.988554048538208,
      "epoch": 0.47244789826916284,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004983347604306129,
      "loss": 5.7781,
      "mean_token_accuracy": 0.1726129099726677,
      "num_tokens": 9992987.0,
      "step": 5350
    },
    {
      "entropy": 5.954887008666992,
      "epoch": 0.47288943836100317,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004983309339560899,
      "loss": 5.8602,
      "mean_token_accuracy": 0.17085562348365785,
      "num_tokens": 10002268.0,
      "step": 5355
    },
    {
      "entropy": 6.0079731941223145,
      "epoch": 0.4733309784528435,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004983271031066412,
      "loss": 5.8679,
      "mean_token_accuracy": 0.16448251605033876,
      "num_tokens": 10011772.0,
      "step": 5360
    },
    {
      "entropy": 6.009686517715454,
      "epoch": 0.4737725185446839,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004983232678823414,
      "loss": 5.868,
      "mean_token_accuracy": 0.1690191462635994,
      "num_tokens": 10021069.0,
      "step": 5365
    },
    {
      "entropy": 6.010422468185425,
      "epoch": 0.4742140586365242,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004983194282832657,
      "loss": 5.7833,
      "mean_token_accuracy": 0.17588813751935958,
      "num_tokens": 10029706.0,
      "step": 5370
    },
    {
      "entropy": 5.897851991653442,
      "epoch": 0.47465559872836455,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004983155843094895,
      "loss": 5.7517,
      "mean_token_accuracy": 0.17830378711223602,
      "num_tokens": 10039633.0,
      "step": 5375
    },
    {
      "entropy": 5.99191837310791,
      "epoch": 0.4750971388202049,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004983117359610881,
      "loss": 5.8065,
      "mean_token_accuracy": 0.1708494946360588,
      "num_tokens": 10048675.0,
      "step": 5380
    },
    {
      "entropy": 6.01197624206543,
      "epoch": 0.4755386789120452,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004983078832381367,
      "loss": 5.7896,
      "mean_token_accuracy": 0.17758023142814636,
      "num_tokens": 10057447.0,
      "step": 5385
    },
    {
      "entropy": 6.0020026683807375,
      "epoch": 0.47598021900388554,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004983040261407109,
      "loss": 5.862,
      "mean_token_accuracy": 0.16202759742736816,
      "num_tokens": 10067020.0,
      "step": 5390
    },
    {
      "entropy": 6.078158807754517,
      "epoch": 0.47642175909572587,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004983001646688863,
      "loss": 5.8989,
      "mean_token_accuracy": 0.16627535820007325,
      "num_tokens": 10076466.0,
      "step": 5395
    },
    {
      "entropy": 6.0067445755004885,
      "epoch": 0.47686329918756626,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004982962988227383,
      "loss": 5.7734,
      "mean_token_accuracy": 0.17013956755399703,
      "num_tokens": 10085384.0,
      "step": 5400
    },
    {
      "entropy": 5.968350505828857,
      "epoch": 0.4773048392794066,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000498292428602343,
      "loss": 5.897,
      "mean_token_accuracy": 0.1669442757964134,
      "num_tokens": 10094015.0,
      "step": 5405
    },
    {
      "entropy": 5.968395566940307,
      "epoch": 0.4777463793712469,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004982885540077758,
      "loss": 5.7796,
      "mean_token_accuracy": 0.17206281870603563,
      "num_tokens": 10103107.0,
      "step": 5410
    },
    {
      "entropy": 5.966505622863769,
      "epoch": 0.47818791946308725,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004982846750391129,
      "loss": 5.7145,
      "mean_token_accuracy": 0.18297846764326095,
      "num_tokens": 10112408.0,
      "step": 5415
    },
    {
      "entropy": 6.003802013397217,
      "epoch": 0.4786294595549276,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004982807916964303,
      "loss": 5.7725,
      "mean_token_accuracy": 0.17201074063777924,
      "num_tokens": 10121605.0,
      "step": 5420
    },
    {
      "entropy": 5.9246360778808596,
      "epoch": 0.4790709996467679,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000498276903979804,
      "loss": 5.7731,
      "mean_token_accuracy": 0.17475963085889817,
      "num_tokens": 10130484.0,
      "step": 5425
    },
    {
      "entropy": 5.951443338394165,
      "epoch": 0.47951253973860825,
      "grad_norm": 0.90625,
      "learning_rate": 0.00049827301188931,
      "loss": 5.7344,
      "mean_token_accuracy": 0.18052613735198975,
      "num_tokens": 10140002.0,
      "step": 5430
    },
    {
      "entropy": 6.060378789901733,
      "epoch": 0.47995407983044863,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004982691154250247,
      "loss": 5.8956,
      "mean_token_accuracy": 0.17365120649337767,
      "num_tokens": 10150287.0,
      "step": 5435
    },
    {
      "entropy": 5.939712142944336,
      "epoch": 0.48039561992228896,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004982652145870245,
      "loss": 5.7209,
      "mean_token_accuracy": 0.1840263232588768,
      "num_tokens": 10160615.0,
      "step": 5440
    },
    {
      "entropy": 5.987045478820801,
      "epoch": 0.4808371600141293,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004982613093753856,
      "loss": 5.8025,
      "mean_token_accuracy": 0.17736782282590866,
      "num_tokens": 10170271.0,
      "step": 5445
    },
    {
      "entropy": 5.9765979766845705,
      "epoch": 0.4812787001059696,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004982573997901847,
      "loss": 5.7628,
      "mean_token_accuracy": 0.17974358052015305,
      "num_tokens": 10179663.0,
      "step": 5450
    },
    {
      "entropy": 6.029351854324341,
      "epoch": 0.48172024019780996,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004982534858314982,
      "loss": 5.845,
      "mean_token_accuracy": 0.1695121109485626,
      "num_tokens": 10188608.0,
      "step": 5455
    },
    {
      "entropy": 6.038789987564087,
      "epoch": 0.4821617802896503,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004982495674994031,
      "loss": 5.8093,
      "mean_token_accuracy": 0.16905225962400436,
      "num_tokens": 10198462.0,
      "step": 5460
    },
    {
      "entropy": 5.905494213104248,
      "epoch": 0.4826033203814906,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004982456447939758,
      "loss": 5.7539,
      "mean_token_accuracy": 0.18186217844486235,
      "num_tokens": 10208362.0,
      "step": 5465
    },
    {
      "entropy": 6.002645826339721,
      "epoch": 0.483044860473331,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004982417177152933,
      "loss": 5.7952,
      "mean_token_accuracy": 0.17442271411418914,
      "num_tokens": 10217915.0,
      "step": 5470
    },
    {
      "entropy": 6.008253574371338,
      "epoch": 0.48348640056517134,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004982377862634325,
      "loss": 5.8407,
      "mean_token_accuracy": 0.16862571984529495,
      "num_tokens": 10226913.0,
      "step": 5475
    },
    {
      "entropy": 5.935338640213013,
      "epoch": 0.48392794065701167,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004982338504384705,
      "loss": 5.8078,
      "mean_token_accuracy": 0.17525406777858735,
      "num_tokens": 10236516.0,
      "step": 5480
    },
    {
      "entropy": 5.997871446609497,
      "epoch": 0.484369480748852,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004982299102404843,
      "loss": 5.8276,
      "mean_token_accuracy": 0.18046585321426392,
      "num_tokens": 10245492.0,
      "step": 5485
    },
    {
      "entropy": 5.913363599777222,
      "epoch": 0.48481102084069233,
      "grad_norm": 1.015625,
      "learning_rate": 0.000498225965669551,
      "loss": 5.6818,
      "mean_token_accuracy": 0.19123907536268234,
      "num_tokens": 10254094.0,
      "step": 5490
    },
    {
      "entropy": 5.917245578765869,
      "epoch": 0.48525256093253266,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004982220167257482,
      "loss": 5.7687,
      "mean_token_accuracy": 0.1767420396208763,
      "num_tokens": 10263645.0,
      "step": 5495
    },
    {
      "entropy": 5.899789619445801,
      "epoch": 0.485694101024373,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004982180634091529,
      "loss": 5.7196,
      "mean_token_accuracy": 0.18229301422834396,
      "num_tokens": 10273416.0,
      "step": 5500
    },
    {
      "entropy": 5.976984024047852,
      "epoch": 0.4861356411162134,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004982141057198427,
      "loss": 5.8398,
      "mean_token_accuracy": 0.17539370357990264,
      "num_tokens": 10283194.0,
      "step": 5505
    },
    {
      "entropy": 5.983163928985595,
      "epoch": 0.4865771812080537,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004982101436578952,
      "loss": 5.8025,
      "mean_token_accuracy": 0.17517516165971755,
      "num_tokens": 10292188.0,
      "step": 5510
    },
    {
      "entropy": 5.889424133300781,
      "epoch": 0.48701872129989404,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004982061772233878,
      "loss": 5.7229,
      "mean_token_accuracy": 0.18908909112215042,
      "num_tokens": 10301539.0,
      "step": 5515
    },
    {
      "entropy": 5.965572834014893,
      "epoch": 0.4874602613917344,
      "grad_norm": 1.0,
      "learning_rate": 0.0004982022064163984,
      "loss": 5.7355,
      "mean_token_accuracy": 0.18447270691394807,
      "num_tokens": 10310379.0,
      "step": 5520
    },
    {
      "entropy": 6.091221475601197,
      "epoch": 0.4879018014835747,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004981982312370047,
      "loss": 5.9364,
      "mean_token_accuracy": 0.16318108290433883,
      "num_tokens": 10320622.0,
      "step": 5525
    },
    {
      "entropy": 5.95211443901062,
      "epoch": 0.48834334157541504,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004981942516852847,
      "loss": 5.7453,
      "mean_token_accuracy": 0.17574492841959,
      "num_tokens": 10330525.0,
      "step": 5530
    },
    {
      "entropy": 5.932955980300903,
      "epoch": 0.48878488166725537,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004981902677613161,
      "loss": 5.7909,
      "mean_token_accuracy": 0.17560216039419174,
      "num_tokens": 10340288.0,
      "step": 5535
    },
    {
      "entropy": 6.035536384582519,
      "epoch": 0.48922642175909575,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004981862794651771,
      "loss": 5.8825,
      "mean_token_accuracy": 0.1646736666560173,
      "num_tokens": 10349836.0,
      "step": 5540
    },
    {
      "entropy": 6.015440130233765,
      "epoch": 0.4896679618509361,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004981822867969459,
      "loss": 5.8088,
      "mean_token_accuracy": 0.17500588893890381,
      "num_tokens": 10359481.0,
      "step": 5545
    },
    {
      "entropy": 6.0031482696533205,
      "epoch": 0.4901095019427764,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004981782897567006,
      "loss": 5.7763,
      "mean_token_accuracy": 0.17515885084867477,
      "num_tokens": 10369837.0,
      "step": 5550
    },
    {
      "entropy": 5.974392652511597,
      "epoch": 0.49055104203461675,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004981742883445195,
      "loss": 5.7395,
      "mean_token_accuracy": 0.17979752272367477,
      "num_tokens": 10379276.0,
      "step": 5555
    },
    {
      "entropy": 6.046759605407715,
      "epoch": 0.4909925821264571,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000498170282560481,
      "loss": 5.8989,
      "mean_token_accuracy": 0.16546373218297958,
      "num_tokens": 10388878.0,
      "step": 5560
    },
    {
      "entropy": 6.05845193862915,
      "epoch": 0.4914341222182974,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004981662724046637,
      "loss": 5.8647,
      "mean_token_accuracy": 0.16656183153390886,
      "num_tokens": 10399148.0,
      "step": 5565
    },
    {
      "entropy": 5.8881865501403805,
      "epoch": 0.49187566231013774,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000498162257877146,
      "loss": 5.7904,
      "mean_token_accuracy": 0.17556624859571457,
      "num_tokens": 10409002.0,
      "step": 5570
    },
    {
      "entropy": 5.949180030822754,
      "epoch": 0.4923172024019781,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004981582389780065,
      "loss": 5.8411,
      "mean_token_accuracy": 0.17601792961359025,
      "num_tokens": 10418083.0,
      "step": 5575
    },
    {
      "entropy": 6.075670337677002,
      "epoch": 0.49275874249381846,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004981542157073241,
      "loss": 5.8447,
      "mean_token_accuracy": 0.171254663169384,
      "num_tokens": 10428134.0,
      "step": 5580
    },
    {
      "entropy": 5.985844087600708,
      "epoch": 0.4932002825856588,
      "grad_norm": 1.0,
      "learning_rate": 0.0004981501880651775,
      "loss": 5.8326,
      "mean_token_accuracy": 0.1712258756160736,
      "num_tokens": 10437436.0,
      "step": 5585
    },
    {
      "entropy": 5.9775690078735355,
      "epoch": 0.4936418226774991,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004981461560516457,
      "loss": 5.7736,
      "mean_token_accuracy": 0.17167493402957917,
      "num_tokens": 10446176.0,
      "step": 5590
    },
    {
      "entropy": 6.065248298645019,
      "epoch": 0.49408336276933945,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004981421196668075,
      "loss": 5.9154,
      "mean_token_accuracy": 0.16895957440137863,
      "num_tokens": 10455953.0,
      "step": 5595
    },
    {
      "entropy": 6.053670883178711,
      "epoch": 0.4945249028611798,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004981380789107422,
      "loss": 5.7647,
      "mean_token_accuracy": 0.17600456476211548,
      "num_tokens": 10465129.0,
      "step": 5600
    },
    {
      "entropy": 5.887149000167847,
      "epoch": 0.4949664429530201,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004981340337835287,
      "loss": 5.7692,
      "mean_token_accuracy": 0.1786765456199646,
      "num_tokens": 10474376.0,
      "step": 5605
    },
    {
      "entropy": 5.976278829574585,
      "epoch": 0.4954079830448605,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004981299842852464,
      "loss": 5.8378,
      "mean_token_accuracy": 0.17351005226373672,
      "num_tokens": 10484849.0,
      "step": 5610
    },
    {
      "entropy": 5.9031822204589846,
      "epoch": 0.49584952313670083,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004981259304159747,
      "loss": 5.7426,
      "mean_token_accuracy": 0.18532529473304749,
      "num_tokens": 10493950.0,
      "step": 5615
    },
    {
      "entropy": 5.928320455551147,
      "epoch": 0.49629106322854116,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004981218721757929,
      "loss": 5.6874,
      "mean_token_accuracy": 0.1819420799612999,
      "num_tokens": 10502890.0,
      "step": 5620
    },
    {
      "entropy": 5.977579784393311,
      "epoch": 0.4967326033203815,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004981178095647805,
      "loss": 5.6809,
      "mean_token_accuracy": 0.17615050375461577,
      "num_tokens": 10511702.0,
      "step": 5625
    },
    {
      "entropy": 5.920915508270264,
      "epoch": 0.4971741434122218,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004981137425830171,
      "loss": 5.7728,
      "mean_token_accuracy": 0.17457375377416612,
      "num_tokens": 10520212.0,
      "step": 5630
    },
    {
      "entropy": 5.900805234909058,
      "epoch": 0.49761568350406216,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004981096712305825,
      "loss": 5.7985,
      "mean_token_accuracy": 0.17847424298524855,
      "num_tokens": 10529652.0,
      "step": 5635
    },
    {
      "entropy": 6.063334560394287,
      "epoch": 0.4980572235959025,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004981055955075561,
      "loss": 5.856,
      "mean_token_accuracy": 0.17167602330446244,
      "num_tokens": 10537980.0,
      "step": 5640
    },
    {
      "entropy": 5.985257768630982,
      "epoch": 0.4984987636877429,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004981015154140181,
      "loss": 5.716,
      "mean_token_accuracy": 0.18206259310245515,
      "num_tokens": 10546854.0,
      "step": 5645
    },
    {
      "entropy": 5.934383678436279,
      "epoch": 0.4989403037795832,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004980974309500483,
      "loss": 5.7564,
      "mean_token_accuracy": 0.17640744894742966,
      "num_tokens": 10555691.0,
      "step": 5650
    },
    {
      "entropy": 5.953382968902588,
      "epoch": 0.49938184387142354,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004980933421157267,
      "loss": 5.8198,
      "mean_token_accuracy": 0.17145697474479676,
      "num_tokens": 10564750.0,
      "step": 5655
    },
    {
      "entropy": 5.992902565002441,
      "epoch": 0.49982338396326387,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004980892489111334,
      "loss": 5.8708,
      "mean_token_accuracy": 0.17010241001844406,
      "num_tokens": 10573515.0,
      "step": 5660
    },
    {
      "entropy": 5.965075302124023,
      "epoch": 0.5002649240551043,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004980851513363486,
      "loss": 5.7427,
      "mean_token_accuracy": 0.183037468791008,
      "num_tokens": 10582730.0,
      "step": 5665
    },
    {
      "entropy": 5.837370443344116,
      "epoch": 0.5007064641469445,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004980810493914526,
      "loss": 5.6517,
      "mean_token_accuracy": 0.19263290315866471,
      "num_tokens": 10591267.0,
      "step": 5670
    },
    {
      "entropy": 5.930122137069702,
      "epoch": 0.5011480042387849,
      "grad_norm": 1.0,
      "learning_rate": 0.0004980769430765256,
      "loss": 5.7238,
      "mean_token_accuracy": 0.1784021809697151,
      "num_tokens": 10599776.0,
      "step": 5675
    },
    {
      "entropy": 5.913628816604614,
      "epoch": 0.5015895443306252,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004980728323916484,
      "loss": 5.7442,
      "mean_token_accuracy": 0.1708422601222992,
      "num_tokens": 10609670.0,
      "step": 5680
    },
    {
      "entropy": 5.911869525909424,
      "epoch": 0.5020310844224656,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004980687173369009,
      "loss": 5.687,
      "mean_token_accuracy": 0.1840984344482422,
      "num_tokens": 10619032.0,
      "step": 5685
    },
    {
      "entropy": 5.96634349822998,
      "epoch": 0.5024726245143059,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004980645979123644,
      "loss": 5.7824,
      "mean_token_accuracy": 0.17758573591709137,
      "num_tokens": 10628851.0,
      "step": 5690
    },
    {
      "entropy": 5.955354261398315,
      "epoch": 0.5029141646061462,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004980604741181192,
      "loss": 5.7578,
      "mean_token_accuracy": 0.1791350230574608,
      "num_tokens": 10638335.0,
      "step": 5695
    },
    {
      "entropy": 5.9715251445770265,
      "epoch": 0.5033557046979866,
      "grad_norm": 0.875,
      "learning_rate": 0.0004980563459542461,
      "loss": 5.8441,
      "mean_token_accuracy": 0.17314625829458236,
      "num_tokens": 10648450.0,
      "step": 5700
    },
    {
      "entropy": 6.0982170581817625,
      "epoch": 0.5037972447898269,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004980522134208261,
      "loss": 5.8715,
      "mean_token_accuracy": 0.16750158816576005,
      "num_tokens": 10658700.0,
      "step": 5705
    },
    {
      "entropy": 6.037292432785034,
      "epoch": 0.5042387848816673,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004980480765179401,
      "loss": 5.8785,
      "mean_token_accuracy": 0.1714808985590935,
      "num_tokens": 10667952.0,
      "step": 5710
    },
    {
      "entropy": 5.895547342300415,
      "epoch": 0.5046803249735076,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004980439352456692,
      "loss": 5.7019,
      "mean_token_accuracy": 0.17903856039047242,
      "num_tokens": 10677228.0,
      "step": 5715
    },
    {
      "entropy": 5.980213356018067,
      "epoch": 0.505121865065348,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004980397896040944,
      "loss": 5.7934,
      "mean_token_accuracy": 0.17691175937652587,
      "num_tokens": 10686183.0,
      "step": 5720
    },
    {
      "entropy": 6.027280044555664,
      "epoch": 0.5055634051571882,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004980356395932969,
      "loss": 5.8348,
      "mean_token_accuracy": 0.1764046236872673,
      "num_tokens": 10695293.0,
      "step": 5725
    },
    {
      "entropy": 5.946255016326904,
      "epoch": 0.5060049452490286,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004980314852133581,
      "loss": 5.7923,
      "mean_token_accuracy": 0.1797493353486061,
      "num_tokens": 10704853.0,
      "step": 5730
    },
    {
      "entropy": 5.957155179977417,
      "epoch": 0.506446485340869,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004980273264643594,
      "loss": 5.7572,
      "mean_token_accuracy": 0.1728769674897194,
      "num_tokens": 10714307.0,
      "step": 5735
    },
    {
      "entropy": 5.847817277908325,
      "epoch": 0.5068880254327093,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004980231633463822,
      "loss": 5.6558,
      "mean_token_accuracy": 0.17468281239271163,
      "num_tokens": 10723513.0,
      "step": 5740
    },
    {
      "entropy": 5.904374122619629,
      "epoch": 0.5073295655245497,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004980189958595081,
      "loss": 5.8291,
      "mean_token_accuracy": 0.16950504183769227,
      "num_tokens": 10732809.0,
      "step": 5745
    },
    {
      "entropy": 6.001342058181763,
      "epoch": 0.5077711056163899,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004980148240038186,
      "loss": 5.8531,
      "mean_token_accuracy": 0.17709660679101943,
      "num_tokens": 10742960.0,
      "step": 5750
    },
    {
      "entropy": 6.103294467926025,
      "epoch": 0.5082126457082303,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004980106477793957,
      "loss": 5.9156,
      "mean_token_accuracy": 0.1706822007894516,
      "num_tokens": 10752835.0,
      "step": 5755
    },
    {
      "entropy": 5.984300756454468,
      "epoch": 0.5086541858000706,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004980064671863209,
      "loss": 5.7683,
      "mean_token_accuracy": 0.17604003846645355,
      "num_tokens": 10762139.0,
      "step": 5760
    },
    {
      "entropy": 5.982193517684936,
      "epoch": 0.509095725891911,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004980022822246763,
      "loss": 5.8479,
      "mean_token_accuracy": 0.17236626744270325,
      "num_tokens": 10771961.0,
      "step": 5765
    },
    {
      "entropy": 6.084481191635132,
      "epoch": 0.5095372659837514,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004979980928945439,
      "loss": 5.8726,
      "mean_token_accuracy": 0.16543770730495452,
      "num_tokens": 10780163.0,
      "step": 5770
    },
    {
      "entropy": 6.0536665439605715,
      "epoch": 0.5099788060755917,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004979938991960056,
      "loss": 5.8429,
      "mean_token_accuracy": 0.17202573120594025,
      "num_tokens": 10788339.0,
      "step": 5775
    },
    {
      "entropy": 6.004624271392823,
      "epoch": 0.510420346167432,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004979897011291436,
      "loss": 5.8412,
      "mean_token_accuracy": 0.16892923414707184,
      "num_tokens": 10797315.0,
      "step": 5780
    },
    {
      "entropy": 6.031994342803955,
      "epoch": 0.5108618862592723,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004979854986940402,
      "loss": 5.7524,
      "mean_token_accuracy": 0.17269293516874312,
      "num_tokens": 10806864.0,
      "step": 5785
    },
    {
      "entropy": 6.027312898635865,
      "epoch": 0.5113034263511127,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004979812918907777,
      "loss": 5.7983,
      "mean_token_accuracy": 0.17434913963079451,
      "num_tokens": 10816417.0,
      "step": 5790
    },
    {
      "entropy": 5.969706630706787,
      "epoch": 0.511744966442953,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004979770807194385,
      "loss": 5.8605,
      "mean_token_accuracy": 0.17248998880386351,
      "num_tokens": 10825692.0,
      "step": 5795
    },
    {
      "entropy": 5.984615087509155,
      "epoch": 0.5121865065347934,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004979728651801051,
      "loss": 5.8402,
      "mean_token_accuracy": 0.1778946951031685,
      "num_tokens": 10834652.0,
      "step": 5800
    },
    {
      "entropy": 6.018920278549194,
      "epoch": 0.5126280466266337,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004979686452728602,
      "loss": 5.828,
      "mean_token_accuracy": 0.17260900139808655,
      "num_tokens": 10844491.0,
      "step": 5805
    },
    {
      "entropy": 5.937997484207154,
      "epoch": 0.513069586718474,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004979644209977863,
      "loss": 5.7696,
      "mean_token_accuracy": 0.17836492955684663,
      "num_tokens": 10853948.0,
      "step": 5810
    },
    {
      "entropy": 5.9561933994293215,
      "epoch": 0.5135111268103144,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004979601923549661,
      "loss": 5.7795,
      "mean_token_accuracy": 0.17643316686153412,
      "num_tokens": 10861745.0,
      "step": 5815
    },
    {
      "entropy": 6.041404867172242,
      "epoch": 0.5139526669021547,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004979559593444826,
      "loss": 5.8554,
      "mean_token_accuracy": 0.16654422879219055,
      "num_tokens": 10871239.0,
      "step": 5820
    },
    {
      "entropy": 5.95691614151001,
      "epoch": 0.5143942069939951,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004979517219664187,
      "loss": 5.7184,
      "mean_token_accuracy": 0.1781800225377083,
      "num_tokens": 10880883.0,
      "step": 5825
    },
    {
      "entropy": 5.979454565048218,
      "epoch": 0.5148357470858353,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004979474802208572,
      "loss": 5.7715,
      "mean_token_accuracy": 0.17757208198308944,
      "num_tokens": 10890002.0,
      "step": 5830
    },
    {
      "entropy": 5.9080863952636715,
      "epoch": 0.5152772871776757,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004979432341078816,
      "loss": 5.7001,
      "mean_token_accuracy": 0.17272792011499405,
      "num_tokens": 10898299.0,
      "step": 5835
    },
    {
      "entropy": 6.025524568557739,
      "epoch": 0.5157188272695161,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004979389836275746,
      "loss": 5.91,
      "mean_token_accuracy": 0.16205677539110183,
      "num_tokens": 10907918.0,
      "step": 5840
    },
    {
      "entropy": 6.026790571212769,
      "epoch": 0.5161603673613564,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004979347287800198,
      "loss": 5.7872,
      "mean_token_accuracy": 0.1693910390138626,
      "num_tokens": 10917318.0,
      "step": 5845
    },
    {
      "entropy": 5.962571716308593,
      "epoch": 0.5166019074531968,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004979304695653005,
      "loss": 5.7259,
      "mean_token_accuracy": 0.1753496468067169,
      "num_tokens": 10926338.0,
      "step": 5850
    },
    {
      "entropy": 5.909990406036377,
      "epoch": 0.5170434475450371,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004979262059835001,
      "loss": 5.7636,
      "mean_token_accuracy": 0.17829854786396027,
      "num_tokens": 10935244.0,
      "step": 5855
    },
    {
      "entropy": 6.022508668899536,
      "epoch": 0.5174849876368774,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004979219380347021,
      "loss": 5.8021,
      "mean_token_accuracy": 0.17361603677272797,
      "num_tokens": 10943764.0,
      "step": 5860
    },
    {
      "entropy": 6.012854290008545,
      "epoch": 0.5179265277287177,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004979176657189901,
      "loss": 5.7735,
      "mean_token_accuracy": 0.17107923626899718,
      "num_tokens": 10952580.0,
      "step": 5865
    },
    {
      "entropy": 5.9548540115356445,
      "epoch": 0.5183680678205581,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004979133890364477,
      "loss": 5.804,
      "mean_token_accuracy": 0.17859138250350953,
      "num_tokens": 10961745.0,
      "step": 5870
    },
    {
      "entropy": 5.8955892562866214,
      "epoch": 0.5188096079123985,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000497909107987159,
      "loss": 5.7239,
      "mean_token_accuracy": 0.17332278788089753,
      "num_tokens": 10971018.0,
      "step": 5875
    },
    {
      "entropy": 5.957081127166748,
      "epoch": 0.5192511480042388,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004979048225712074,
      "loss": 5.7574,
      "mean_token_accuracy": 0.17847395539283753,
      "num_tokens": 10980070.0,
      "step": 5880
    },
    {
      "entropy": 6.00935115814209,
      "epoch": 0.5196926880960792,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004979005327886772,
      "loss": 5.7406,
      "mean_token_accuracy": 0.17666334360837938,
      "num_tokens": 10989090.0,
      "step": 5885
    },
    {
      "entropy": 5.992536735534668,
      "epoch": 0.5201342281879194,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004978962386396525,
      "loss": 5.7908,
      "mean_token_accuracy": 0.17853841334581375,
      "num_tokens": 10998652.0,
      "step": 5890
    },
    {
      "entropy": 5.848595428466797,
      "epoch": 0.5205757682797598,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004978919401242171,
      "loss": 5.6844,
      "mean_token_accuracy": 0.18879187554121019,
      "num_tokens": 11006323.0,
      "step": 5895
    },
    {
      "entropy": 5.797189474105835,
      "epoch": 0.5210173083716001,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004978876372424553,
      "loss": 5.6297,
      "mean_token_accuracy": 0.1844954565167427,
      "num_tokens": 11015054.0,
      "step": 5900
    },
    {
      "entropy": 5.956575059890747,
      "epoch": 0.5214588484634405,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004978833299944515,
      "loss": 5.8139,
      "mean_token_accuracy": 0.170148167014122,
      "num_tokens": 11025787.0,
      "step": 5905
    },
    {
      "entropy": 6.084104728698731,
      "epoch": 0.5219003885552809,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004978790183802901,
      "loss": 5.9223,
      "mean_token_accuracy": 0.16725478023290635,
      "num_tokens": 11036091.0,
      "step": 5910
    },
    {
      "entropy": 5.951149892807007,
      "epoch": 0.5223419286471211,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004978747024000554,
      "loss": 5.8703,
      "mean_token_accuracy": 0.16429074257612228,
      "num_tokens": 11046442.0,
      "step": 5915
    },
    {
      "entropy": 6.039041662216187,
      "epoch": 0.5227834687389615,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004978703820538321,
      "loss": 5.9039,
      "mean_token_accuracy": 0.16753809303045272,
      "num_tokens": 11056907.0,
      "step": 5920
    },
    {
      "entropy": 6.0201164245605465,
      "epoch": 0.5232250088308018,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004978660573417048,
      "loss": 5.7536,
      "mean_token_accuracy": 0.17897529453039168,
      "num_tokens": 11065298.0,
      "step": 5925
    },
    {
      "entropy": 5.924715948104859,
      "epoch": 0.5236665489226422,
      "grad_norm": 0.96875,
      "learning_rate": 0.000497861728263758,
      "loss": 5.7376,
      "mean_token_accuracy": 0.1859300434589386,
      "num_tokens": 11074509.0,
      "step": 5930
    },
    {
      "entropy": 5.973381519317627,
      "epoch": 0.5241080890144825,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004978573948200769,
      "loss": 5.7772,
      "mean_token_accuracy": 0.1776812806725502,
      "num_tokens": 11082433.0,
      "step": 5935
    },
    {
      "entropy": 5.972002792358398,
      "epoch": 0.5245496291063229,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004978530570107461,
      "loss": 5.7476,
      "mean_token_accuracy": 0.17605585902929305,
      "num_tokens": 11090838.0,
      "step": 5940
    },
    {
      "entropy": 5.965470218658448,
      "epoch": 0.5249911691981632,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004978487148358506,
      "loss": 5.7421,
      "mean_token_accuracy": 0.18264487236738206,
      "num_tokens": 11100048.0,
      "step": 5945
    },
    {
      "entropy": 5.874642705917358,
      "epoch": 0.5254327092900035,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004978443682954756,
      "loss": 5.7466,
      "mean_token_accuracy": 0.17708369791507722,
      "num_tokens": 11109871.0,
      "step": 5950
    },
    {
      "entropy": 5.954572439193726,
      "epoch": 0.5258742493818439,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004978400173897061,
      "loss": 5.7382,
      "mean_token_accuracy": 0.1706852361559868,
      "num_tokens": 11119450.0,
      "step": 5955
    },
    {
      "entropy": 6.011610317230224,
      "epoch": 0.5263157894736842,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004978356621186275,
      "loss": 5.7024,
      "mean_token_accuracy": 0.19025731831789017,
      "num_tokens": 11128321.0,
      "step": 5960
    },
    {
      "entropy": 5.893124532699585,
      "epoch": 0.5267573295655246,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004978313024823249,
      "loss": 5.7508,
      "mean_token_accuracy": 0.1797892838716507,
      "num_tokens": 11138657.0,
      "step": 5965
    },
    {
      "entropy": 5.876752948760986,
      "epoch": 0.5271988696573648,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004978269384808839,
      "loss": 5.7308,
      "mean_token_accuracy": 0.1822565272450447,
      "num_tokens": 11147704.0,
      "step": 5970
    },
    {
      "entropy": 5.982156419754029,
      "epoch": 0.5276404097492052,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004978225701143898,
      "loss": 5.8386,
      "mean_token_accuracy": 0.1702994167804718,
      "num_tokens": 11156979.0,
      "step": 5975
    },
    {
      "entropy": 5.976283884048462,
      "epoch": 0.5280819498410456,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004978181973829284,
      "loss": 5.7233,
      "mean_token_accuracy": 0.17943737357854844,
      "num_tokens": 11165539.0,
      "step": 5980
    },
    {
      "entropy": 6.0231208324432375,
      "epoch": 0.5285234899328859,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004978138202865851,
      "loss": 5.7906,
      "mean_token_accuracy": 0.16854341477155685,
      "num_tokens": 11175911.0,
      "step": 5985
    },
    {
      "entropy": 5.924521636962891,
      "epoch": 0.5289650300247263,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004978094388254459,
      "loss": 5.8217,
      "mean_token_accuracy": 0.16968544870615004,
      "num_tokens": 11185365.0,
      "step": 5990
    },
    {
      "entropy": 5.926220464706421,
      "epoch": 0.5294065701165666,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004978050529995965,
      "loss": 5.7211,
      "mean_token_accuracy": 0.17357961237430572,
      "num_tokens": 11195318.0,
      "step": 5995
    },
    {
      "entropy": 5.973835372924805,
      "epoch": 0.5298481102084069,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004978006628091228,
      "loss": 5.8353,
      "mean_token_accuracy": 0.17482065111398698,
      "num_tokens": 11205246.0,
      "step": 6000
    },
    {
      "epoch": 0.5298481102084069,
      "eval_entropy": 5.7665525598407195,
      "eval_loss": 5.782026290893555,
      "eval_mean_token_accuracy": 0.18332941638198202,
      "eval_num_tokens": 11205246.0,
      "eval_runtime": 20.2512,
      "eval_samples_per_second": 1743.846,
      "eval_steps_per_second": 218.012,
      "step": 6000
    },
    {
      "entropy": 5.990317106246948,
      "epoch": 0.5302896503002472,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004977962682541108,
      "loss": 5.781,
      "mean_token_accuracy": 0.17299736738204957,
      "num_tokens": 11215183.0,
      "step": 6005
    },
    {
      "entropy": 5.952910041809082,
      "epoch": 0.5307311903920876,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004977918693346467,
      "loss": 5.7518,
      "mean_token_accuracy": 0.17333558350801467,
      "num_tokens": 11224127.0,
      "step": 6010
    },
    {
      "entropy": 5.951988935470581,
      "epoch": 0.531172730483928,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004977874660508165,
      "loss": 5.7806,
      "mean_token_accuracy": 0.17980630993843078,
      "num_tokens": 11233014.0,
      "step": 6015
    },
    {
      "entropy": 5.9942632675170895,
      "epoch": 0.5316142705757683,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004977830584027065,
      "loss": 5.7706,
      "mean_token_accuracy": 0.1789461463689804,
      "num_tokens": 11242251.0,
      "step": 6020
    },
    {
      "entropy": 6.00589075088501,
      "epoch": 0.5320558106676087,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004977786463904033,
      "loss": 5.8286,
      "mean_token_accuracy": 0.17321041971445084,
      "num_tokens": 11251947.0,
      "step": 6025
    },
    {
      "entropy": 6.000608110427857,
      "epoch": 0.5324973507594489,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004977742300139929,
      "loss": 5.8648,
      "mean_token_accuracy": 0.16567921191453933,
      "num_tokens": 11261125.0,
      "step": 6030
    },
    {
      "entropy": 6.012487840652466,
      "epoch": 0.5329388908512893,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004977698092735621,
      "loss": 5.739,
      "mean_token_accuracy": 0.17683637589216233,
      "num_tokens": 11270334.0,
      "step": 6035
    },
    {
      "entropy": 5.920803642272949,
      "epoch": 0.5333804309431296,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004977653841691975,
      "loss": 5.7561,
      "mean_token_accuracy": 0.17879996448755264,
      "num_tokens": 11280497.0,
      "step": 6040
    },
    {
      "entropy": 5.9857405662536625,
      "epoch": 0.53382197103497,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004977609547009857,
      "loss": 5.6924,
      "mean_token_accuracy": 0.17791507840156556,
      "num_tokens": 11288636.0,
      "step": 6045
    },
    {
      "entropy": 5.947061967849732,
      "epoch": 0.5342635111268104,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004977565208690134,
      "loss": 5.8118,
      "mean_token_accuracy": 0.17096064984798431,
      "num_tokens": 11297519.0,
      "step": 6050
    },
    {
      "entropy": 5.937090826034546,
      "epoch": 0.5347050512186506,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004977520826733677,
      "loss": 5.7466,
      "mean_token_accuracy": 0.1841330721974373,
      "num_tokens": 11306280.0,
      "step": 6055
    },
    {
      "entropy": 5.941117191314698,
      "epoch": 0.535146591310491,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004977476401141353,
      "loss": 5.7608,
      "mean_token_accuracy": 0.17097747325897217,
      "num_tokens": 11315366.0,
      "step": 6060
    },
    {
      "entropy": 5.977073001861572,
      "epoch": 0.5355881314023313,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004977431931914032,
      "loss": 5.7688,
      "mean_token_accuracy": 0.1683377668261528,
      "num_tokens": 11324664.0,
      "step": 6065
    },
    {
      "entropy": 6.0049549579620365,
      "epoch": 0.5360296714941717,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004977387419052586,
      "loss": 5.752,
      "mean_token_accuracy": 0.1675189822912216,
      "num_tokens": 11334118.0,
      "step": 6070
    },
    {
      "entropy": 6.066574954986573,
      "epoch": 0.536471211586012,
      "grad_norm": 0.83984375,
      "learning_rate": 0.000497734286255789,
      "loss": 5.8661,
      "mean_token_accuracy": 0.16980284303426743,
      "num_tokens": 11342684.0,
      "step": 6075
    },
    {
      "entropy": 5.942904949188232,
      "epoch": 0.5369127516778524,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004977298262430813,
      "loss": 5.787,
      "mean_token_accuracy": 0.18056662529706954,
      "num_tokens": 11351298.0,
      "step": 6080
    },
    {
      "entropy": 5.964593315124512,
      "epoch": 0.5373542917696927,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004977253618672229,
      "loss": 5.7353,
      "mean_token_accuracy": 0.17694929987192154,
      "num_tokens": 11360264.0,
      "step": 6085
    },
    {
      "entropy": 5.946565723419189,
      "epoch": 0.537795831861533,
      "grad_norm": 0.875,
      "learning_rate": 0.0004977208931283013,
      "loss": 5.7171,
      "mean_token_accuracy": 0.179603773355484,
      "num_tokens": 11369815.0,
      "step": 6090
    },
    {
      "entropy": 5.9731690883636475,
      "epoch": 0.5382373719533734,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004977164200264042,
      "loss": 5.7954,
      "mean_token_accuracy": 0.17322832047939302,
      "num_tokens": 11379131.0,
      "step": 6095
    },
    {
      "entropy": 5.961735391616822,
      "epoch": 0.5386789120452137,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004977119425616191,
      "loss": 5.6836,
      "mean_token_accuracy": 0.18143999129533767,
      "num_tokens": 11387692.0,
      "step": 6100
    },
    {
      "entropy": 5.944207382202149,
      "epoch": 0.5391204521370541,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004977074607340336,
      "loss": 5.7364,
      "mean_token_accuracy": 0.17622338831424714,
      "num_tokens": 11396278.0,
      "step": 6105
    },
    {
      "entropy": 5.873842573165893,
      "epoch": 0.5395619922288943,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004977029745437357,
      "loss": 5.6798,
      "mean_token_accuracy": 0.1845750167965889,
      "num_tokens": 11404790.0,
      "step": 6110
    },
    {
      "entropy": 5.9296238899230955,
      "epoch": 0.5400035323207347,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000497698483990813,
      "loss": 5.7827,
      "mean_token_accuracy": 0.17046503573656083,
      "num_tokens": 11414614.0,
      "step": 6115
    },
    {
      "entropy": 5.982711362838745,
      "epoch": 0.5404450724125751,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004976939890753539,
      "loss": 5.7841,
      "mean_token_accuracy": 0.17147547155618667,
      "num_tokens": 11423696.0,
      "step": 6120
    },
    {
      "entropy": 5.949193668365479,
      "epoch": 0.5408866125044154,
      "grad_norm": 1.046875,
      "learning_rate": 0.000497689489797446,
      "loss": 5.7296,
      "mean_token_accuracy": 0.1846212074160576,
      "num_tokens": 11432669.0,
      "step": 6125
    },
    {
      "entropy": 5.935843181610108,
      "epoch": 0.5413281525962558,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004976849861571778,
      "loss": 5.7638,
      "mean_token_accuracy": 0.17586904317140578,
      "num_tokens": 11442132.0,
      "step": 6130
    },
    {
      "entropy": 5.958372163772583,
      "epoch": 0.541769692688096,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004976804781546373,
      "loss": 5.7237,
      "mean_token_accuracy": 0.17114574760198592,
      "num_tokens": 11451821.0,
      "step": 6135
    },
    {
      "entropy": 5.9710346221923825,
      "epoch": 0.5422112327799364,
      "grad_norm": 1.0,
      "learning_rate": 0.0004976759657899129,
      "loss": 5.7422,
      "mean_token_accuracy": 0.1799530491232872,
      "num_tokens": 11460242.0,
      "step": 6140
    },
    {
      "entropy": 5.869692325592041,
      "epoch": 0.5426527728717767,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000497671449063093,
      "loss": 5.7729,
      "mean_token_accuracy": 0.18124603480100632,
      "num_tokens": 11469588.0,
      "step": 6145
    },
    {
      "entropy": 6.013301944732666,
      "epoch": 0.5430943129636171,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000497666927974266,
      "loss": 5.8456,
      "mean_token_accuracy": 0.16814137101173401,
      "num_tokens": 11480401.0,
      "step": 6150
    },
    {
      "entropy": 6.017386198043823,
      "epoch": 0.5435358530554575,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004976624025235205,
      "loss": 5.7602,
      "mean_token_accuracy": 0.1720780089497566,
      "num_tokens": 11489117.0,
      "step": 6155
    },
    {
      "entropy": 5.8355712890625,
      "epoch": 0.5439773931472978,
      "grad_norm": 0.875,
      "learning_rate": 0.0004976578727109453,
      "loss": 5.7019,
      "mean_token_accuracy": 0.18260259926319122,
      "num_tokens": 11497612.0,
      "step": 6160
    },
    {
      "entropy": 5.938451480865479,
      "epoch": 0.5444189332391381,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004976533385366289,
      "loss": 5.7813,
      "mean_token_accuracy": 0.17646318674087524,
      "num_tokens": 11507024.0,
      "step": 6165
    },
    {
      "entropy": 5.959741735458374,
      "epoch": 0.5448604733309784,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004976488000006602,
      "loss": 5.7019,
      "mean_token_accuracy": 0.18477845042943955,
      "num_tokens": 11516873.0,
      "step": 6170
    },
    {
      "entropy": 6.025303840637207,
      "epoch": 0.5453020134228188,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004976442571031282,
      "loss": 5.8072,
      "mean_token_accuracy": 0.18425012081861497,
      "num_tokens": 11526128.0,
      "step": 6175
    },
    {
      "entropy": 5.902001094818115,
      "epoch": 0.5457435535146591,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004976397098441218,
      "loss": 5.8113,
      "mean_token_accuracy": 0.18117952942848206,
      "num_tokens": 11535030.0,
      "step": 6180
    },
    {
      "entropy": 5.852960729598999,
      "epoch": 0.5461850936064995,
      "grad_norm": 1.0,
      "learning_rate": 0.0004976351582237301,
      "loss": 5.6982,
      "mean_token_accuracy": 0.1818488821387291,
      "num_tokens": 11543113.0,
      "step": 6185
    },
    {
      "entropy": 5.96119122505188,
      "epoch": 0.5466266336983399,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004976306022420423,
      "loss": 5.7295,
      "mean_token_accuracy": 0.17677829265594483,
      "num_tokens": 11552744.0,
      "step": 6190
    },
    {
      "entropy": 6.032695865631103,
      "epoch": 0.5470681737901801,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004976260418991476,
      "loss": 5.7739,
      "mean_token_accuracy": 0.1760561227798462,
      "num_tokens": 11561274.0,
      "step": 6195
    },
    {
      "entropy": 6.003128910064698,
      "epoch": 0.5475097138820205,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004976214771951354,
      "loss": 5.8,
      "mean_token_accuracy": 0.17939124554395675,
      "num_tokens": 11570485.0,
      "step": 6200
    },
    {
      "entropy": 5.972816276550293,
      "epoch": 0.5479512539738608,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004976169081300949,
      "loss": 5.7707,
      "mean_token_accuracy": 0.17227309495210646,
      "num_tokens": 11579380.0,
      "step": 6205
    },
    {
      "entropy": 5.934311819076538,
      "epoch": 0.5483927940657012,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004976123347041159,
      "loss": 5.75,
      "mean_token_accuracy": 0.1758207708597183,
      "num_tokens": 11588686.0,
      "step": 6210
    },
    {
      "entropy": 5.94596152305603,
      "epoch": 0.5488343341575415,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004976077569172879,
      "loss": 5.6867,
      "mean_token_accuracy": 0.18259273767471312,
      "num_tokens": 11598547.0,
      "step": 6215
    },
    {
      "entropy": 5.933783531188965,
      "epoch": 0.5492758742493818,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004976031747697006,
      "loss": 5.7744,
      "mean_token_accuracy": 0.1780215844511986,
      "num_tokens": 11608407.0,
      "step": 6220
    },
    {
      "entropy": 5.873191547393799,
      "epoch": 0.5497174143412222,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004975985882614435,
      "loss": 5.7298,
      "mean_token_accuracy": 0.1796346575021744,
      "num_tokens": 11618641.0,
      "step": 6225
    },
    {
      "entropy": 5.968663263320923,
      "epoch": 0.5501589544330625,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004975939973926068,
      "loss": 5.7934,
      "mean_token_accuracy": 0.16603399068117142,
      "num_tokens": 11627082.0,
      "step": 6230
    },
    {
      "entropy": 5.934994173049927,
      "epoch": 0.5506004945249029,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004975894021632803,
      "loss": 5.6769,
      "mean_token_accuracy": 0.1807957574725151,
      "num_tokens": 11636045.0,
      "step": 6235
    },
    {
      "entropy": 5.929260015487671,
      "epoch": 0.5510420346167432,
      "grad_norm": 0.9375,
      "learning_rate": 0.000497584802573554,
      "loss": 5.7589,
      "mean_token_accuracy": 0.1748830959200859,
      "num_tokens": 11646516.0,
      "step": 6240
    },
    {
      "entropy": 5.919200325012207,
      "epoch": 0.5514835747085836,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000497580198623518,
      "loss": 5.6835,
      "mean_token_accuracy": 0.18040877282619477,
      "num_tokens": 11655512.0,
      "step": 6245
    },
    {
      "entropy": 5.988286733627319,
      "epoch": 0.5519251148004238,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004975755903132623,
      "loss": 5.8009,
      "mean_token_accuracy": 0.17675041109323503,
      "num_tokens": 11665641.0,
      "step": 6250
    },
    {
      "entropy": 5.927127695083618,
      "epoch": 0.5523666548922642,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004975709776428776,
      "loss": 5.7674,
      "mean_token_accuracy": 0.17123859375715256,
      "num_tokens": 11674585.0,
      "step": 6255
    },
    {
      "entropy": 5.883406162261963,
      "epoch": 0.5528081949841046,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000497566360612454,
      "loss": 5.6318,
      "mean_token_accuracy": 0.18883141577243806,
      "num_tokens": 11683498.0,
      "step": 6260
    },
    {
      "entropy": 5.945862483978272,
      "epoch": 0.5532497350759449,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000497561739222082,
      "loss": 5.6778,
      "mean_token_accuracy": 0.1816217005252838,
      "num_tokens": 11692663.0,
      "step": 6265
    },
    {
      "entropy": 5.911245727539063,
      "epoch": 0.5536912751677853,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004975571134718521,
      "loss": 5.8389,
      "mean_token_accuracy": 0.17130470722913743,
      "num_tokens": 11701906.0,
      "step": 6270
    },
    {
      "entropy": 5.953613424301148,
      "epoch": 0.5541328152596255,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004975524833618548,
      "loss": 5.7533,
      "mean_token_accuracy": 0.18487675338983536,
      "num_tokens": 11710932.0,
      "step": 6275
    },
    {
      "entropy": 5.97131118774414,
      "epoch": 0.5545743553514659,
      "grad_norm": 0.984375,
      "learning_rate": 0.000497547848892181,
      "loss": 5.7541,
      "mean_token_accuracy": 0.18125775456428528,
      "num_tokens": 11719953.0,
      "step": 6280
    },
    {
      "entropy": 5.979850387573242,
      "epoch": 0.5550158954433062,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004975432100629215,
      "loss": 5.6412,
      "mean_token_accuracy": 0.18299894779920578,
      "num_tokens": 11728706.0,
      "step": 6285
    },
    {
      "entropy": 5.969555902481079,
      "epoch": 0.5554574355351466,
      "grad_norm": 1.078125,
      "learning_rate": 0.000497538566874167,
      "loss": 5.8242,
      "mean_token_accuracy": 0.1693728670477867,
      "num_tokens": 11738537.0,
      "step": 6290
    },
    {
      "entropy": 5.960725116729736,
      "epoch": 0.555898975626987,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004975339193260084,
      "loss": 5.6493,
      "mean_token_accuracy": 0.18169963359832764,
      "num_tokens": 11747503.0,
      "step": 6295
    },
    {
      "entropy": 5.951407527923584,
      "epoch": 0.5563405157188273,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004975292674185371,
      "loss": 5.7519,
      "mean_token_accuracy": 0.17961287796497344,
      "num_tokens": 11756501.0,
      "step": 6300
    },
    {
      "entropy": 5.8904026508331295,
      "epoch": 0.5567820558106676,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004975246111518439,
      "loss": 5.7268,
      "mean_token_accuracy": 0.1724578857421875,
      "num_tokens": 11765483.0,
      "step": 6305
    },
    {
      "entropy": 6.040583896636963,
      "epoch": 0.5572235959025079,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004975199505260202,
      "loss": 5.8402,
      "mean_token_accuracy": 0.1671855017542839,
      "num_tokens": 11774941.0,
      "step": 6310
    },
    {
      "entropy": 6.042414903640747,
      "epoch": 0.5576651359943483,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004975152855411571,
      "loss": 5.8157,
      "mean_token_accuracy": 0.16865744739770888,
      "num_tokens": 11785080.0,
      "step": 6315
    },
    {
      "entropy": 5.949933290481567,
      "epoch": 0.5581066760861886,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004975106161973462,
      "loss": 5.707,
      "mean_token_accuracy": 0.17533884942531586,
      "num_tokens": 11793933.0,
      "step": 6320
    },
    {
      "entropy": 5.896859550476075,
      "epoch": 0.558548216178029,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000497505942494679,
      "loss": 5.7346,
      "mean_token_accuracy": 0.17790014147758484,
      "num_tokens": 11803592.0,
      "step": 6325
    },
    {
      "entropy": 5.902767181396484,
      "epoch": 0.5589897562698694,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004975012644332467,
      "loss": 5.6507,
      "mean_token_accuracy": 0.18465393632650376,
      "num_tokens": 11812397.0,
      "step": 6330
    },
    {
      "entropy": 5.903921031951905,
      "epoch": 0.5594312963617096,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004974965820131413,
      "loss": 5.7665,
      "mean_token_accuracy": 0.173906309902668,
      "num_tokens": 11821734.0,
      "step": 6335
    },
    {
      "entropy": 5.908196878433228,
      "epoch": 0.55987283645355,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004974918952344544,
      "loss": 5.6429,
      "mean_token_accuracy": 0.18138528764247894,
      "num_tokens": 11832107.0,
      "step": 6340
    },
    {
      "entropy": 5.960652589797974,
      "epoch": 0.5603143765453903,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004974872040972777,
      "loss": 5.6944,
      "mean_token_accuracy": 0.17450693845748902,
      "num_tokens": 11841983.0,
      "step": 6345
    },
    {
      "entropy": 5.822365188598633,
      "epoch": 0.5607559166372307,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004974825086017033,
      "loss": 5.6839,
      "mean_token_accuracy": 0.18381367623806,
      "num_tokens": 11851836.0,
      "step": 6350
    },
    {
      "entropy": 5.940042352676391,
      "epoch": 0.561197456729071,
      "grad_norm": 0.890625,
      "learning_rate": 0.000497477808747823,
      "loss": 5.7497,
      "mean_token_accuracy": 0.17437803000211716,
      "num_tokens": 11861109.0,
      "step": 6355
    },
    {
      "entropy": 5.960816669464111,
      "epoch": 0.5616389968209113,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004974731045357292,
      "loss": 5.845,
      "mean_token_accuracy": 0.1676926702260971,
      "num_tokens": 11871015.0,
      "step": 6360
    },
    {
      "entropy": 5.906099796295166,
      "epoch": 0.5620805369127517,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004974683959655136,
      "loss": 5.7129,
      "mean_token_accuracy": 0.18293437659740447,
      "num_tokens": 11881306.0,
      "step": 6365
    },
    {
      "entropy": 5.926035451889038,
      "epoch": 0.562522077004592,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004974636830372686,
      "loss": 5.648,
      "mean_token_accuracy": 0.188501039147377,
      "num_tokens": 11890251.0,
      "step": 6370
    },
    {
      "entropy": 5.925776243209839,
      "epoch": 0.5629636170964324,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004974589657510866,
      "loss": 5.7729,
      "mean_token_accuracy": 0.17742838859558105,
      "num_tokens": 11899662.0,
      "step": 6375
    },
    {
      "entropy": 5.8874077796936035,
      "epoch": 0.5634051571882727,
      "grad_norm": 1.0,
      "learning_rate": 0.0004974542441070601,
      "loss": 5.727,
      "mean_token_accuracy": 0.18206923007965087,
      "num_tokens": 11908699.0,
      "step": 6380
    },
    {
      "entropy": 5.9278564453125,
      "epoch": 0.563846697280113,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004974495181052814,
      "loss": 5.7072,
      "mean_token_accuracy": 0.18245860040187836,
      "num_tokens": 11916637.0,
      "step": 6385
    },
    {
      "entropy": 5.957434034347534,
      "epoch": 0.5642882373719533,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004974447877458433,
      "loss": 5.7751,
      "mean_token_accuracy": 0.17324685156345368,
      "num_tokens": 11926063.0,
      "step": 6390
    },
    {
      "entropy": 5.929507160186768,
      "epoch": 0.5647297774637937,
      "grad_norm": 1.125,
      "learning_rate": 0.0004974400530288381,
      "loss": 5.6611,
      "mean_token_accuracy": 0.18200212568044663,
      "num_tokens": 11934861.0,
      "step": 6395
    },
    {
      "entropy": 5.959511756896973,
      "epoch": 0.5651713175556341,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004974353139543589,
      "loss": 5.7622,
      "mean_token_accuracy": 0.1750079497694969,
      "num_tokens": 11944403.0,
      "step": 6400
    },
    {
      "entropy": 5.965158271789551,
      "epoch": 0.5656128576474744,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004974305705224983,
      "loss": 5.8406,
      "mean_token_accuracy": 0.17911516278982162,
      "num_tokens": 11953306.0,
      "step": 6405
    },
    {
      "entropy": 5.934712839126587,
      "epoch": 0.5660543977393148,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004974258227333493,
      "loss": 5.7194,
      "mean_token_accuracy": 0.17788851857185364,
      "num_tokens": 11962727.0,
      "step": 6410
    },
    {
      "entropy": 5.954853057861328,
      "epoch": 0.566495937831155,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004974210705870052,
      "loss": 5.8619,
      "mean_token_accuracy": 0.17795796543359757,
      "num_tokens": 11972375.0,
      "step": 6415
    },
    {
      "entropy": 5.850558567047119,
      "epoch": 0.5669374779229954,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004974163140835586,
      "loss": 5.6381,
      "mean_token_accuracy": 0.18238368183374404,
      "num_tokens": 11981989.0,
      "step": 6420
    },
    {
      "entropy": 5.973133182525634,
      "epoch": 0.5673790180148357,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004974115532231029,
      "loss": 5.6917,
      "mean_token_accuracy": 0.18118394017219544,
      "num_tokens": 11990865.0,
      "step": 6425
    },
    {
      "entropy": 5.942868566513061,
      "epoch": 0.5678205581066761,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004974067880057315,
      "loss": 5.6916,
      "mean_token_accuracy": 0.1895127236843109,
      "num_tokens": 12000697.0,
      "step": 6430
    },
    {
      "entropy": 5.884861660003662,
      "epoch": 0.5682620981985165,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004974020184315375,
      "loss": 5.7739,
      "mean_token_accuracy": 0.17718288153409958,
      "num_tokens": 12010560.0,
      "step": 6435
    },
    {
      "entropy": 5.981086111068725,
      "epoch": 0.5687036382903567,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004973972445006144,
      "loss": 5.7798,
      "mean_token_accuracy": 0.1795758858323097,
      "num_tokens": 12020760.0,
      "step": 6440
    },
    {
      "entropy": 5.845777130126953,
      "epoch": 0.5691451783821971,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004973924662130559,
      "loss": 5.6409,
      "mean_token_accuracy": 0.18322015553712845,
      "num_tokens": 12030467.0,
      "step": 6445
    },
    {
      "entropy": 5.929102373123169,
      "epoch": 0.5695867184740374,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004973876835689554,
      "loss": 5.7441,
      "mean_token_accuracy": 0.17533728629350662,
      "num_tokens": 12039897.0,
      "step": 6450
    },
    {
      "entropy": 5.986848783493042,
      "epoch": 0.5700282585658778,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004973828965684066,
      "loss": 5.7667,
      "mean_token_accuracy": 0.1707102909684181,
      "num_tokens": 12048997.0,
      "step": 6455
    },
    {
      "entropy": 5.889280796051025,
      "epoch": 0.5704697986577181,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004973781052115035,
      "loss": 5.6626,
      "mean_token_accuracy": 0.18899978697299957,
      "num_tokens": 12057493.0,
      "step": 6460
    },
    {
      "entropy": 5.862435913085937,
      "epoch": 0.5709113387495585,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004973733094983396,
      "loss": 5.7348,
      "mean_token_accuracy": 0.18229613900184632,
      "num_tokens": 12066329.0,
      "step": 6465
    },
    {
      "entropy": 5.959417295455933,
      "epoch": 0.5713528788413988,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004973685094290092,
      "loss": 5.7654,
      "mean_token_accuracy": 0.1740923374891281,
      "num_tokens": 12076022.0,
      "step": 6470
    },
    {
      "entropy": 5.901052379608155,
      "epoch": 0.5717944189332391,
      "grad_norm": 0.85546875,
      "learning_rate": 0.000497363705003606,
      "loss": 5.7036,
      "mean_token_accuracy": 0.17817170470952987,
      "num_tokens": 12085750.0,
      "step": 6475
    },
    {
      "entropy": 5.895412921905518,
      "epoch": 0.5722359590250795,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004973588962222244,
      "loss": 5.753,
      "mean_token_accuracy": 0.17770908027887344,
      "num_tokens": 12094935.0,
      "step": 6480
    },
    {
      "entropy": 5.895012998580933,
      "epoch": 0.5726774991169198,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004973540830849585,
      "loss": 5.6811,
      "mean_token_accuracy": 0.17887165248394013,
      "num_tokens": 12104970.0,
      "step": 6485
    },
    {
      "entropy": 5.959931230545044,
      "epoch": 0.5731190392087602,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004973492655919025,
      "loss": 5.7748,
      "mean_token_accuracy": 0.17862619161605836,
      "num_tokens": 12116195.0,
      "step": 6490
    },
    {
      "entropy": 5.988841438293457,
      "epoch": 0.5735605793006004,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004973444437431508,
      "loss": 5.7261,
      "mean_token_accuracy": 0.18133638501167298,
      "num_tokens": 12125485.0,
      "step": 6495
    },
    {
      "entropy": 5.886534547805786,
      "epoch": 0.5740021193924408,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004973396175387981,
      "loss": 5.7432,
      "mean_token_accuracy": 0.18072778433561326,
      "num_tokens": 12135474.0,
      "step": 6500
    },
    {
      "entropy": 5.9477166652679445,
      "epoch": 0.5744436594842812,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004973347869789385,
      "loss": 5.7713,
      "mean_token_accuracy": 0.17266511768102646,
      "num_tokens": 12144945.0,
      "step": 6505
    },
    {
      "entropy": 5.9193422317504885,
      "epoch": 0.5748851995761215,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004973299520636671,
      "loss": 5.7062,
      "mean_token_accuracy": 0.18106334358453752,
      "num_tokens": 12154893.0,
      "step": 6510
    },
    {
      "entropy": 5.919309759140015,
      "epoch": 0.5753267396679619,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004973251127930782,
      "loss": 5.7894,
      "mean_token_accuracy": 0.17616706043481828,
      "num_tokens": 12164047.0,
      "step": 6515
    },
    {
      "entropy": 5.9936785221099855,
      "epoch": 0.5757682797598022,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004973202691672669,
      "loss": 5.7679,
      "mean_token_accuracy": 0.1752978190779686,
      "num_tokens": 12172965.0,
      "step": 6520
    },
    {
      "entropy": 5.984523725509644,
      "epoch": 0.5762098198516425,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004973154211863279,
      "loss": 5.7988,
      "mean_token_accuracy": 0.17396161705255508,
      "num_tokens": 12183534.0,
      "step": 6525
    },
    {
      "entropy": 5.926711177825927,
      "epoch": 0.5766513599434828,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004973105688503563,
      "loss": 5.7586,
      "mean_token_accuracy": 0.17371885776519774,
      "num_tokens": 12194370.0,
      "step": 6530
    },
    {
      "entropy": 5.94952425956726,
      "epoch": 0.5770929000353232,
      "grad_norm": 0.875,
      "learning_rate": 0.000497305712159447,
      "loss": 5.7548,
      "mean_token_accuracy": 0.17564807534217836,
      "num_tokens": 12203765.0,
      "step": 6535
    },
    {
      "entropy": 5.950043344497681,
      "epoch": 0.5775344401271636,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004973008511136954,
      "loss": 5.6843,
      "mean_token_accuracy": 0.18619055598974227,
      "num_tokens": 12212889.0,
      "step": 6540
    },
    {
      "entropy": 5.9376091957092285,
      "epoch": 0.5779759802190039,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004972959857131963,
      "loss": 5.7568,
      "mean_token_accuracy": 0.17935430109500886,
      "num_tokens": 12222055.0,
      "step": 6545
    },
    {
      "entropy": 5.945558071136475,
      "epoch": 0.5784175203108443,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004972911159580454,
      "loss": 5.7504,
      "mean_token_accuracy": 0.1824483186006546,
      "num_tokens": 12232092.0,
      "step": 6550
    },
    {
      "entropy": 5.967494964599609,
      "epoch": 0.5788590604026845,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000497286241848338,
      "loss": 5.7899,
      "mean_token_accuracy": 0.17304221391677857,
      "num_tokens": 12241411.0,
      "step": 6555
    },
    {
      "entropy": 5.939413595199585,
      "epoch": 0.5793006004945249,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004972813633841696,
      "loss": 5.6636,
      "mean_token_accuracy": 0.17755235135555267,
      "num_tokens": 12249735.0,
      "step": 6560
    },
    {
      "entropy": 5.896793413162231,
      "epoch": 0.5797421405863652,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004972764805656355,
      "loss": 5.6744,
      "mean_token_accuracy": 0.18935427516698838,
      "num_tokens": 12259132.0,
      "step": 6565
    },
    {
      "entropy": 5.91446475982666,
      "epoch": 0.5801836806782056,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004972715933928318,
      "loss": 5.7427,
      "mean_token_accuracy": 0.18480055928230285,
      "num_tokens": 12269401.0,
      "step": 6570
    },
    {
      "entropy": 5.961280679702758,
      "epoch": 0.580625220770046,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004972667018658539,
      "loss": 5.7714,
      "mean_token_accuracy": 0.1721562162041664,
      "num_tokens": 12279163.0,
      "step": 6575
    },
    {
      "entropy": 5.897429990768432,
      "epoch": 0.5810667608618862,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004972618059847976,
      "loss": 5.6763,
      "mean_token_accuracy": 0.18488114923238755,
      "num_tokens": 12287942.0,
      "step": 6580
    },
    {
      "entropy": 5.941003751754761,
      "epoch": 0.5815083009537266,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004972569057497591,
      "loss": 5.675,
      "mean_token_accuracy": 0.17982326596975326,
      "num_tokens": 12296717.0,
      "step": 6585
    },
    {
      "entropy": 5.910008764266967,
      "epoch": 0.5819498410455669,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004972520011608341,
      "loss": 5.6576,
      "mean_token_accuracy": 0.18751820027828217,
      "num_tokens": 12305983.0,
      "step": 6590
    },
    {
      "entropy": 5.849145221710205,
      "epoch": 0.5823913811374073,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004972470922181189,
      "loss": 5.7389,
      "mean_token_accuracy": 0.17131290584802628,
      "num_tokens": 12315338.0,
      "step": 6595
    },
    {
      "entropy": 5.961781024932861,
      "epoch": 0.5828329212292476,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004972421789217094,
      "loss": 5.8171,
      "mean_token_accuracy": 0.1724231258034706,
      "num_tokens": 12324828.0,
      "step": 6600
    },
    {
      "entropy": 6.009534645080566,
      "epoch": 0.583274461321088,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004972372612717022,
      "loss": 5.686,
      "mean_token_accuracy": 0.18648577928543092,
      "num_tokens": 12332844.0,
      "step": 6605
    },
    {
      "entropy": 5.874768495559692,
      "epoch": 0.5837160014129283,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004972323392681934,
      "loss": 5.7203,
      "mean_token_accuracy": 0.18115541487932205,
      "num_tokens": 12342062.0,
      "step": 6610
    },
    {
      "entropy": 5.906319189071655,
      "epoch": 0.5841575415047686,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004972274129112794,
      "loss": 5.7949,
      "mean_token_accuracy": 0.17332905381917954,
      "num_tokens": 12352071.0,
      "step": 6615
    },
    {
      "entropy": 6.078907680511475,
      "epoch": 0.584599081596609,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004972224822010567,
      "loss": 5.8644,
      "mean_token_accuracy": 0.1694015458226204,
      "num_tokens": 12362025.0,
      "step": 6620
    },
    {
      "entropy": 5.988093709945678,
      "epoch": 0.5850406216884493,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004972175471376222,
      "loss": 5.8597,
      "mean_token_accuracy": 0.16808632612228394,
      "num_tokens": 12371547.0,
      "step": 6625
    },
    {
      "entropy": 5.961208009719849,
      "epoch": 0.5854821617802897,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004972126077210722,
      "loss": 5.7425,
      "mean_token_accuracy": 0.17402835786342621,
      "num_tokens": 12381385.0,
      "step": 6630
    },
    {
      "entropy": 5.982282543182373,
      "epoch": 0.5859237018721299,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004972076639515035,
      "loss": 5.8134,
      "mean_token_accuracy": 0.17358503490686417,
      "num_tokens": 12391302.0,
      "step": 6635
    },
    {
      "entropy": 5.828851175308228,
      "epoch": 0.5863652419639703,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004972027158290131,
      "loss": 5.6969,
      "mean_token_accuracy": 0.1809057265520096,
      "num_tokens": 12400388.0,
      "step": 6640
    },
    {
      "entropy": 5.867154169082641,
      "epoch": 0.5868067820558107,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000497197763353698,
      "loss": 5.7531,
      "mean_token_accuracy": 0.17146917283535004,
      "num_tokens": 12409695.0,
      "step": 6645
    },
    {
      "entropy": 5.974938201904297,
      "epoch": 0.587248322147651,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000497192806525655,
      "loss": 5.6163,
      "mean_token_accuracy": 0.19104450494050979,
      "num_tokens": 12418728.0,
      "step": 6650
    },
    {
      "entropy": 5.9528045654296875,
      "epoch": 0.5876898622394914,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004971878453449811,
      "loss": 5.7788,
      "mean_token_accuracy": 0.18384409248828887,
      "num_tokens": 12428072.0,
      "step": 6655
    },
    {
      "entropy": 5.993191766738891,
      "epoch": 0.5881314023313317,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000497182879811774,
      "loss": 5.7131,
      "mean_token_accuracy": 0.18061271756887437,
      "num_tokens": 12436624.0,
      "step": 6660
    },
    {
      "entropy": 5.881068277359009,
      "epoch": 0.588572942423172,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004971779099261304,
      "loss": 5.7265,
      "mean_token_accuracy": 0.17819689810276032,
      "num_tokens": 12446370.0,
      "step": 6665
    },
    {
      "entropy": 5.889710187911987,
      "epoch": 0.5890144825150123,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000497172935688148,
      "loss": 5.785,
      "mean_token_accuracy": 0.17816216349601746,
      "num_tokens": 12456705.0,
      "step": 6670
    },
    {
      "entropy": 5.939290475845337,
      "epoch": 0.5894560226068527,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004971679570979242,
      "loss": 5.7909,
      "mean_token_accuracy": 0.17827335596084595,
      "num_tokens": 12466150.0,
      "step": 6675
    },
    {
      "entropy": 5.957861614227295,
      "epoch": 0.5898975626986931,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004971629741555565,
      "loss": 5.6405,
      "mean_token_accuracy": 0.18321777880191803,
      "num_tokens": 12475051.0,
      "step": 6680
    },
    {
      "entropy": 5.9179764747619625,
      "epoch": 0.5903391027905334,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004971579868611423,
      "loss": 5.7193,
      "mean_token_accuracy": 0.18429083675146102,
      "num_tokens": 12485427.0,
      "step": 6685
    },
    {
      "entropy": 5.848150014877319,
      "epoch": 0.5907806428823738,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004971529952147797,
      "loss": 5.6583,
      "mean_token_accuracy": 0.18403666615486144,
      "num_tokens": 12494489.0,
      "step": 6690
    },
    {
      "entropy": 5.954574680328369,
      "epoch": 0.591222182974214,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000497147999216566,
      "loss": 5.6894,
      "mean_token_accuracy": 0.1826411232352257,
      "num_tokens": 12503656.0,
      "step": 6695
    },
    {
      "entropy": 5.95741491317749,
      "epoch": 0.5916637230660544,
      "grad_norm": 1.25,
      "learning_rate": 0.0004971429988665996,
      "loss": 5.7271,
      "mean_token_accuracy": 0.17112142443656922,
      "num_tokens": 12512940.0,
      "step": 6700
    },
    {
      "entropy": 5.92096962928772,
      "epoch": 0.5921052631578947,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000497137994164978,
      "loss": 5.6653,
      "mean_token_accuracy": 0.18557649552822114,
      "num_tokens": 12523186.0,
      "step": 6705
    },
    {
      "entropy": 5.926599788665771,
      "epoch": 0.5925468032497351,
      "grad_norm": 1.0,
      "learning_rate": 0.0004971329851117996,
      "loss": 5.709,
      "mean_token_accuracy": 0.1801753506064415,
      "num_tokens": 12533018.0,
      "step": 6710
    },
    {
      "entropy": 5.996683883666992,
      "epoch": 0.5929883433415755,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004971279717071622,
      "loss": 5.7826,
      "mean_token_accuracy": 0.17882051169872284,
      "num_tokens": 12542889.0,
      "step": 6715
    },
    {
      "entropy": 5.967419385910034,
      "epoch": 0.5934298834334157,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004971229539511643,
      "loss": 5.7956,
      "mean_token_accuracy": 0.17490288019180297,
      "num_tokens": 12553625.0,
      "step": 6720
    },
    {
      "entropy": 5.867678117752075,
      "epoch": 0.5938714235252561,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000497117931843904,
      "loss": 5.7149,
      "mean_token_accuracy": 0.17882569432258605,
      "num_tokens": 12563012.0,
      "step": 6725
    },
    {
      "entropy": 6.035449504852295,
      "epoch": 0.5943129636170964,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004971129053854797,
      "loss": 5.8473,
      "mean_token_accuracy": 0.1676483556628227,
      "num_tokens": 12572892.0,
      "step": 6730
    },
    {
      "entropy": 5.953235340118408,
      "epoch": 0.5947545037089368,
      "grad_norm": 0.9375,
      "learning_rate": 0.00049710787457599,
      "loss": 5.6586,
      "mean_token_accuracy": 0.1844158798456192,
      "num_tokens": 12581550.0,
      "step": 6735
    },
    {
      "entropy": 5.877220201492309,
      "epoch": 0.5951960438007771,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004971028394155332,
      "loss": 5.7398,
      "mean_token_accuracy": 0.17651234865188598,
      "num_tokens": 12590599.0,
      "step": 6740
    },
    {
      "entropy": 5.981046724319458,
      "epoch": 0.5956375838926175,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004970977999042082,
      "loss": 5.7146,
      "mean_token_accuracy": 0.18010410368442537,
      "num_tokens": 12599497.0,
      "step": 6745
    },
    {
      "entropy": 5.942370986938476,
      "epoch": 0.5960791239844578,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004970927560421137,
      "loss": 5.7958,
      "mean_token_accuracy": 0.17954052835702897,
      "num_tokens": 12608273.0,
      "step": 6750
    },
    {
      "entropy": 5.882658863067627,
      "epoch": 0.5965206640762981,
      "grad_norm": 1.125,
      "learning_rate": 0.0004970877078293483,
      "loss": 5.6383,
      "mean_token_accuracy": 0.18409765362739564,
      "num_tokens": 12617287.0,
      "step": 6755
    },
    {
      "entropy": 5.927299499511719,
      "epoch": 0.5969622041681385,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004970826552660111,
      "loss": 5.7166,
      "mean_token_accuracy": 0.17487230598926545,
      "num_tokens": 12626874.0,
      "step": 6760
    },
    {
      "entropy": 5.94651575088501,
      "epoch": 0.5974037442599788,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004970775983522009,
      "loss": 5.7682,
      "mean_token_accuracy": 0.17655521929264067,
      "num_tokens": 12635218.0,
      "step": 6765
    },
    {
      "entropy": 5.948738765716553,
      "epoch": 0.5978452843518192,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004970725370880168,
      "loss": 5.7085,
      "mean_token_accuracy": 0.17957753241062163,
      "num_tokens": 12644395.0,
      "step": 6770
    },
    {
      "entropy": 5.971686220169067,
      "epoch": 0.5982868244436594,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004970674714735581,
      "loss": 5.7582,
      "mean_token_accuracy": 0.1807936668395996,
      "num_tokens": 12654540.0,
      "step": 6775
    },
    {
      "entropy": 5.892604112625122,
      "epoch": 0.5987283645354998,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004970624015089239,
      "loss": 5.5915,
      "mean_token_accuracy": 0.18906807750463486,
      "num_tokens": 12663546.0,
      "step": 6780
    },
    {
      "entropy": 5.866572380065918,
      "epoch": 0.5991699046273402,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004970573271942135,
      "loss": 5.7395,
      "mean_token_accuracy": 0.18058196157217027,
      "num_tokens": 12672353.0,
      "step": 6785
    },
    {
      "entropy": 5.866556930541992,
      "epoch": 0.5996114447191805,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004970522485295265,
      "loss": 5.6094,
      "mean_token_accuracy": 0.19295398741960526,
      "num_tokens": 12680833.0,
      "step": 6790
    },
    {
      "entropy": 5.851094532012939,
      "epoch": 0.6000529848110209,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004970471655149621,
      "loss": 5.5789,
      "mean_token_accuracy": 0.1865946739912033,
      "num_tokens": 12690033.0,
      "step": 6795
    },
    {
      "entropy": 5.816410923004151,
      "epoch": 0.6004945249028611,
      "grad_norm": 1.015625,
      "learning_rate": 0.00049704207815062,
      "loss": 5.6477,
      "mean_token_accuracy": 0.18757187128067015,
      "num_tokens": 12700127.0,
      "step": 6800
    },
    {
      "entropy": 5.935033321380615,
      "epoch": 0.6009360649947015,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004970369864366,
      "loss": 5.6766,
      "mean_token_accuracy": 0.18124886006116867,
      "num_tokens": 12709803.0,
      "step": 6805
    },
    {
      "entropy": 5.913188362121582,
      "epoch": 0.6013776050865418,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004970318903730017,
      "loss": 5.6686,
      "mean_token_accuracy": 0.18383529633283616,
      "num_tokens": 12718876.0,
      "step": 6810
    },
    {
      "entropy": 5.947503089904785,
      "epoch": 0.6018191451783822,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004970267899599249,
      "loss": 5.7802,
      "mean_token_accuracy": 0.17513394802808763,
      "num_tokens": 12728745.0,
      "step": 6815
    },
    {
      "entropy": 5.921718597412109,
      "epoch": 0.6022606852702226,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004970216851974695,
      "loss": 5.7881,
      "mean_token_accuracy": 0.17280806452035904,
      "num_tokens": 12738764.0,
      "step": 6820
    },
    {
      "entropy": 5.870873689651489,
      "epoch": 0.6027022253620629,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004970165760857356,
      "loss": 5.7253,
      "mean_token_accuracy": 0.17291287034749986,
      "num_tokens": 12747888.0,
      "step": 6825
    },
    {
      "entropy": 6.020090627670288,
      "epoch": 0.6031437654539032,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004970114626248232,
      "loss": 5.7377,
      "mean_token_accuracy": 0.18093507140874862,
      "num_tokens": 12757763.0,
      "step": 6830
    },
    {
      "entropy": 5.94991626739502,
      "epoch": 0.6035853055457435,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004970063448148326,
      "loss": 5.77,
      "mean_token_accuracy": 0.17415328770875932,
      "num_tokens": 12767924.0,
      "step": 6835
    },
    {
      "entropy": 5.948456811904907,
      "epoch": 0.6040268456375839,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004970012226558639,
      "loss": 5.7911,
      "mean_token_accuracy": 0.17666855156421662,
      "num_tokens": 12777140.0,
      "step": 6840
    },
    {
      "entropy": 5.975889825820923,
      "epoch": 0.6044683857294242,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004969960961480176,
      "loss": 5.7618,
      "mean_token_accuracy": 0.1819861873984337,
      "num_tokens": 12786731.0,
      "step": 6845
    },
    {
      "entropy": 5.902105522155762,
      "epoch": 0.6049099258212646,
      "grad_norm": 1.0,
      "learning_rate": 0.0004969909652913939,
      "loss": 5.6598,
      "mean_token_accuracy": 0.18601689636707305,
      "num_tokens": 12796839.0,
      "step": 6850
    },
    {
      "entropy": 5.9462145328521725,
      "epoch": 0.605351465913105,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004969858300860935,
      "loss": 5.7836,
      "mean_token_accuracy": 0.17639988511800767,
      "num_tokens": 12805944.0,
      "step": 6855
    },
    {
      "entropy": 5.95946044921875,
      "epoch": 0.6057930060049452,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004969806905322169,
      "loss": 5.7202,
      "mean_token_accuracy": 0.1749048963189125,
      "num_tokens": 12814692.0,
      "step": 6860
    },
    {
      "entropy": 5.877088022232056,
      "epoch": 0.6062345460967856,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004969755466298649,
      "loss": 5.6688,
      "mean_token_accuracy": 0.17678387016057967,
      "num_tokens": 12824506.0,
      "step": 6865
    },
    {
      "entropy": 5.915852737426758,
      "epoch": 0.6066760861886259,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004969703983791382,
      "loss": 5.7392,
      "mean_token_accuracy": 0.18122698962688447,
      "num_tokens": 12835157.0,
      "step": 6870
    },
    {
      "entropy": 5.91579122543335,
      "epoch": 0.6071176262804663,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004969652457801376,
      "loss": 5.6357,
      "mean_token_accuracy": 0.1784660905599594,
      "num_tokens": 12844281.0,
      "step": 6875
    },
    {
      "entropy": 5.997165012359619,
      "epoch": 0.6075591663723066,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004969600888329641,
      "loss": 5.7117,
      "mean_token_accuracy": 0.18102120906114577,
      "num_tokens": 12854486.0,
      "step": 6880
    },
    {
      "entropy": 5.928967142105103,
      "epoch": 0.608000706464147,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004969549275377186,
      "loss": 5.701,
      "mean_token_accuracy": 0.17599090039730073,
      "num_tokens": 12863574.0,
      "step": 6885
    },
    {
      "entropy": 5.84666748046875,
      "epoch": 0.6084422465559873,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004969497618945024,
      "loss": 5.6549,
      "mean_token_accuracy": 0.18378470838069916,
      "num_tokens": 12872450.0,
      "step": 6890
    },
    {
      "entropy": 5.866508674621582,
      "epoch": 0.6088837866478276,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004969445919034165,
      "loss": 5.6066,
      "mean_token_accuracy": 0.19402503222227097,
      "num_tokens": 12881299.0,
      "step": 6895
    },
    {
      "entropy": 5.8215114116668705,
      "epoch": 0.609325326739668,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004969394175645623,
      "loss": 5.5739,
      "mean_token_accuracy": 0.18549795001745223,
      "num_tokens": 12889943.0,
      "step": 6900
    },
    {
      "entropy": 5.840115118026733,
      "epoch": 0.6097668668315083,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004969342388780411,
      "loss": 5.6779,
      "mean_token_accuracy": 0.18420430421829223,
      "num_tokens": 12899497.0,
      "step": 6905
    },
    {
      "entropy": 5.927986431121826,
      "epoch": 0.6102084069233487,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004969290558439546,
      "loss": 5.6608,
      "mean_token_accuracy": 0.177994304895401,
      "num_tokens": 12908677.0,
      "step": 6910
    },
    {
      "entropy": 5.953958320617676,
      "epoch": 0.6106499470151889,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004969238684624038,
      "loss": 5.7974,
      "mean_token_accuracy": 0.17282726615667343,
      "num_tokens": 12918468.0,
      "step": 6915
    },
    {
      "entropy": 5.8817911624908445,
      "epoch": 0.6110914871070293,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004969186767334908,
      "loss": 5.677,
      "mean_token_accuracy": 0.18767081946134567,
      "num_tokens": 12928527.0,
      "step": 6920
    },
    {
      "entropy": 5.910523414611816,
      "epoch": 0.6115330271988697,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004969134806573172,
      "loss": 5.6662,
      "mean_token_accuracy": 0.1900987759232521,
      "num_tokens": 12937021.0,
      "step": 6925
    },
    {
      "entropy": 5.804785871505738,
      "epoch": 0.61197456729071,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004969082802339846,
      "loss": 5.5893,
      "mean_token_accuracy": 0.1889987915754318,
      "num_tokens": 12945570.0,
      "step": 6930
    },
    {
      "entropy": 5.893771982192993,
      "epoch": 0.6124161073825504,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000496903075463595,
      "loss": 5.7471,
      "mean_token_accuracy": 0.18079857677221298,
      "num_tokens": 12954594.0,
      "step": 6935
    },
    {
      "entropy": 5.93186502456665,
      "epoch": 0.6128576474743906,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004968978663462504,
      "loss": 5.6933,
      "mean_token_accuracy": 0.18866560161113738,
      "num_tokens": 12963328.0,
      "step": 6940
    },
    {
      "entropy": 5.902009201049805,
      "epoch": 0.613299187566231,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004968926528820529,
      "loss": 5.7045,
      "mean_token_accuracy": 0.1821183055639267,
      "num_tokens": 12972806.0,
      "step": 6945
    },
    {
      "entropy": 5.9072239875793455,
      "epoch": 0.6137407276580713,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004968874350711044,
      "loss": 5.7147,
      "mean_token_accuracy": 0.17737514525651932,
      "num_tokens": 12981940.0,
      "step": 6950
    },
    {
      "entropy": 5.902434349060059,
      "epoch": 0.6141822677499117,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004968822129135072,
      "loss": 5.6743,
      "mean_token_accuracy": 0.18073017597198487,
      "num_tokens": 12990501.0,
      "step": 6955
    },
    {
      "entropy": 5.960966110229492,
      "epoch": 0.6146238078417521,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004968769864093637,
      "loss": 5.701,
      "mean_token_accuracy": 0.17968421429395676,
      "num_tokens": 12999734.0,
      "step": 6960
    },
    {
      "entropy": 6.008027362823486,
      "epoch": 0.6150653479335924,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004968717555587762,
      "loss": 5.7938,
      "mean_token_accuracy": 0.17155789136886596,
      "num_tokens": 13009323.0,
      "step": 6965
    },
    {
      "entropy": 5.979251575469971,
      "epoch": 0.6155068880254327,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004968665203618472,
      "loss": 5.7889,
      "mean_token_accuracy": 0.18105098307132722,
      "num_tokens": 13018136.0,
      "step": 6970
    },
    {
      "entropy": 5.955159854888916,
      "epoch": 0.615948428117273,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004968612808186793,
      "loss": 5.7365,
      "mean_token_accuracy": 0.17707560658454896,
      "num_tokens": 13026494.0,
      "step": 6975
    },
    {
      "entropy": 5.939317274093628,
      "epoch": 0.6163899682091134,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004968560369293751,
      "loss": 5.7283,
      "mean_token_accuracy": 0.17934589833021164,
      "num_tokens": 13036548.0,
      "step": 6980
    },
    {
      "entropy": 5.967477321624756,
      "epoch": 0.6168315083009537,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004968507886940372,
      "loss": 5.7605,
      "mean_token_accuracy": 0.18426348268985748,
      "num_tokens": 13045653.0,
      "step": 6985
    },
    {
      "entropy": 6.00388650894165,
      "epoch": 0.6172730483927941,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004968455361127686,
      "loss": 5.8066,
      "mean_token_accuracy": 0.17793480753898622,
      "num_tokens": 13056204.0,
      "step": 6990
    },
    {
      "entropy": 5.94718656539917,
      "epoch": 0.6177145884846345,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004968402791856721,
      "loss": 5.7655,
      "mean_token_accuracy": 0.1818525716662407,
      "num_tokens": 13067215.0,
      "step": 6995
    },
    {
      "entropy": 5.893466234207153,
      "epoch": 0.6181561285764747,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004968350179128507,
      "loss": 5.7287,
      "mean_token_accuracy": 0.17613782733678818,
      "num_tokens": 13076824.0,
      "step": 7000
    },
    {
      "entropy": 5.9018186092376705,
      "epoch": 0.6185976686683151,
      "grad_norm": 0.875,
      "learning_rate": 0.0004968297522944075,
      "loss": 5.6894,
      "mean_token_accuracy": 0.1825672522187233,
      "num_tokens": 13086205.0,
      "step": 7005
    },
    {
      "entropy": 5.870771694183349,
      "epoch": 0.6190392087601554,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004968244823304455,
      "loss": 5.6302,
      "mean_token_accuracy": 0.18527472019195557,
      "num_tokens": 13096381.0,
      "step": 7010
    },
    {
      "entropy": 5.887761640548706,
      "epoch": 0.6194807488519958,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004968192080210683,
      "loss": 5.6947,
      "mean_token_accuracy": 0.18954527229070664,
      "num_tokens": 13106533.0,
      "step": 7015
    },
    {
      "entropy": 5.844252872467041,
      "epoch": 0.619922288943836,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004968139293663787,
      "loss": 5.6596,
      "mean_token_accuracy": 0.18514977842569352,
      "num_tokens": 13116605.0,
      "step": 7020
    },
    {
      "entropy": 5.889920091629028,
      "epoch": 0.6203638290356764,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004968086463664806,
      "loss": 5.6567,
      "mean_token_accuracy": 0.1948334753513336,
      "num_tokens": 13125316.0,
      "step": 7025
    },
    {
      "entropy": 6.018892908096314,
      "epoch": 0.6208053691275168,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000496803359021477,
      "loss": 5.8357,
      "mean_token_accuracy": 0.18200814127922058,
      "num_tokens": 13135595.0,
      "step": 7030
    },
    {
      "entropy": 5.931438970565796,
      "epoch": 0.6212469092193571,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004967980673314719,
      "loss": 5.763,
      "mean_token_accuracy": 0.1763605758547783,
      "num_tokens": 13144395.0,
      "step": 7035
    },
    {
      "entropy": 5.873738098144531,
      "epoch": 0.6216884493111975,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004967927712965688,
      "loss": 5.7099,
      "mean_token_accuracy": 0.17793530225753784,
      "num_tokens": 13153906.0,
      "step": 7040
    },
    {
      "entropy": 5.951040601730346,
      "epoch": 0.6221299894030378,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004967874709168714,
      "loss": 5.6495,
      "mean_token_accuracy": 0.18728768527507783,
      "num_tokens": 13163840.0,
      "step": 7045
    },
    {
      "entropy": 5.889948749542237,
      "epoch": 0.6225715294948782,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004967821661924837,
      "loss": 5.6492,
      "mean_token_accuracy": 0.18374582082033158,
      "num_tokens": 13173224.0,
      "step": 7050
    },
    {
      "entropy": 5.907984256744385,
      "epoch": 0.6230130695867184,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004967768571235094,
      "loss": 5.7068,
      "mean_token_accuracy": 0.1758308783173561,
      "num_tokens": 13182664.0,
      "step": 7055
    },
    {
      "entropy": 5.906645822525024,
      "epoch": 0.6234546096785588,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004967715437100526,
      "loss": 5.5975,
      "mean_token_accuracy": 0.1858988508582115,
      "num_tokens": 13191076.0,
      "step": 7060
    },
    {
      "entropy": 5.988809823989868,
      "epoch": 0.6238961497703992,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004967662259522174,
      "loss": 5.8038,
      "mean_token_accuracy": 0.1728333130478859,
      "num_tokens": 13201075.0,
      "step": 7065
    },
    {
      "entropy": 5.931052255630493,
      "epoch": 0.6243376898622395,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000496760903850108,
      "loss": 5.7399,
      "mean_token_accuracy": 0.18440952450037001,
      "num_tokens": 13210602.0,
      "step": 7070
    },
    {
      "entropy": 5.948591327667236,
      "epoch": 0.6247792299540799,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004967555774038286,
      "loss": 5.7262,
      "mean_token_accuracy": 0.17679778188467027,
      "num_tokens": 13219601.0,
      "step": 7075
    },
    {
      "entropy": 5.946039438247681,
      "epoch": 0.6252207700459201,
      "grad_norm": 0.76953125,
      "learning_rate": 0.0004967502466134836,
      "loss": 5.7772,
      "mean_token_accuracy": 0.1736842855811119,
      "num_tokens": 13230675.0,
      "step": 7080
    },
    {
      "entropy": 5.926424598693847,
      "epoch": 0.6256623101377605,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004967449114791773,
      "loss": 5.7228,
      "mean_token_accuracy": 0.18056838661432267,
      "num_tokens": 13240349.0,
      "step": 7085
    },
    {
      "entropy": 5.9506090641021725,
      "epoch": 0.6261038502296008,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004967395720010143,
      "loss": 5.7857,
      "mean_token_accuracy": 0.16999740153551102,
      "num_tokens": 13249867.0,
      "step": 7090
    },
    {
      "entropy": 5.962334060668946,
      "epoch": 0.6265453903214412,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004967342281790992,
      "loss": 5.7031,
      "mean_token_accuracy": 0.18430460393428802,
      "num_tokens": 13259143.0,
      "step": 7095
    },
    {
      "entropy": 5.884312629699707,
      "epoch": 0.6269869304132816,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004967288800135368,
      "loss": 5.6321,
      "mean_token_accuracy": 0.18402725905179979,
      "num_tokens": 13268673.0,
      "step": 7100
    },
    {
      "entropy": 5.897164916992187,
      "epoch": 0.6274284705051218,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004967235275044315,
      "loss": 5.6694,
      "mean_token_accuracy": 0.1823020711541176,
      "num_tokens": 13278035.0,
      "step": 7105
    },
    {
      "entropy": 5.945112085342407,
      "epoch": 0.6278700105969622,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004967181706518886,
      "loss": 5.7944,
      "mean_token_accuracy": 0.16981353759765624,
      "num_tokens": 13286950.0,
      "step": 7110
    },
    {
      "entropy": 5.843374967575073,
      "epoch": 0.6283115506888025,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004967128094560128,
      "loss": 5.6535,
      "mean_token_accuracy": 0.1964161604642868,
      "num_tokens": 13296425.0,
      "step": 7115
    },
    {
      "entropy": 6.004785966873169,
      "epoch": 0.6287530907806429,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004967074439169092,
      "loss": 5.7793,
      "mean_token_accuracy": 0.17232254892587662,
      "num_tokens": 13306094.0,
      "step": 7120
    },
    {
      "entropy": 5.918289661407471,
      "epoch": 0.6291946308724832,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004967020740346827,
      "loss": 5.7147,
      "mean_token_accuracy": 0.1865583762526512,
      "num_tokens": 13315181.0,
      "step": 7125
    },
    {
      "entropy": 5.852770471572876,
      "epoch": 0.6296361709643236,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004966966998094387,
      "loss": 5.6237,
      "mean_token_accuracy": 0.1867580607533455,
      "num_tokens": 13324108.0,
      "step": 7130
    },
    {
      "entropy": 5.890590524673462,
      "epoch": 0.630077711056164,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004966913212412825,
      "loss": 5.6933,
      "mean_token_accuracy": 0.18645042479038237,
      "num_tokens": 13333101.0,
      "step": 7135
    },
    {
      "entropy": 5.852170133590699,
      "epoch": 0.6305192511480042,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004966859383303194,
      "loss": 5.6105,
      "mean_token_accuracy": 0.19161757826805115,
      "num_tokens": 13341804.0,
      "step": 7140
    },
    {
      "entropy": 5.885224866867065,
      "epoch": 0.6309607912398446,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004966805510766549,
      "loss": 5.6598,
      "mean_token_accuracy": 0.18454794138669967,
      "num_tokens": 13350911.0,
      "step": 7145
    },
    {
      "entropy": 5.943019104003906,
      "epoch": 0.6314023313316849,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004966751594803944,
      "loss": 5.6698,
      "mean_token_accuracy": 0.18281008452177047,
      "num_tokens": 13360657.0,
      "step": 7150
    },
    {
      "entropy": 5.802535104751587,
      "epoch": 0.6318438714235253,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004966697635416437,
      "loss": 5.6219,
      "mean_token_accuracy": 0.18310205191373824,
      "num_tokens": 13370462.0,
      "step": 7155
    },
    {
      "entropy": 5.953017282485962,
      "epoch": 0.6322854115153655,
      "grad_norm": 1.0,
      "learning_rate": 0.0004966643632605082,
      "loss": 5.7309,
      "mean_token_accuracy": 0.17605062574148178,
      "num_tokens": 13379446.0,
      "step": 7160
    },
    {
      "entropy": 5.932848262786865,
      "epoch": 0.6327269516072059,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000496658958637094,
      "loss": 5.7173,
      "mean_token_accuracy": 0.18260978460311889,
      "num_tokens": 13389238.0,
      "step": 7165
    },
    {
      "entropy": 5.886120319366455,
      "epoch": 0.6331684916990463,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000496653549671507,
      "loss": 5.71,
      "mean_token_accuracy": 0.1812327668070793,
      "num_tokens": 13398613.0,
      "step": 7170
    },
    {
      "entropy": 5.922674942016601,
      "epoch": 0.6336100317908866,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004966481363638529,
      "loss": 5.7151,
      "mean_token_accuracy": 0.1866316482424736,
      "num_tokens": 13407600.0,
      "step": 7175
    },
    {
      "entropy": 5.776447343826294,
      "epoch": 0.634051571882727,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004966427187142378,
      "loss": 5.6108,
      "mean_token_accuracy": 0.18910133689641953,
      "num_tokens": 13415974.0,
      "step": 7180
    },
    {
      "entropy": 5.91736102104187,
      "epoch": 0.6344931119745673,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004966372967227681,
      "loss": 5.6084,
      "mean_token_accuracy": 0.1872192531824112,
      "num_tokens": 13425628.0,
      "step": 7185
    },
    {
      "entropy": 5.85201473236084,
      "epoch": 0.6349346520664076,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004966318703895496,
      "loss": 5.6374,
      "mean_token_accuracy": 0.18410265296697617,
      "num_tokens": 13435573.0,
      "step": 7190
    },
    {
      "entropy": 5.8759300231933596,
      "epoch": 0.6353761921582479,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004966264397146888,
      "loss": 5.6048,
      "mean_token_accuracy": 0.1849142238497734,
      "num_tokens": 13443969.0,
      "step": 7195
    },
    {
      "entropy": 5.881554222106933,
      "epoch": 0.6358177322500883,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004966210046982921,
      "loss": 5.6813,
      "mean_token_accuracy": 0.18429166376590728,
      "num_tokens": 13452823.0,
      "step": 7200
    },
    {
      "entropy": 5.915408277511597,
      "epoch": 0.6362592723419287,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004966155653404661,
      "loss": 5.711,
      "mean_token_accuracy": 0.17995356470346452,
      "num_tokens": 13461090.0,
      "step": 7205
    },
    {
      "entropy": 5.9935754299163815,
      "epoch": 0.636700812433769,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004966101216413172,
      "loss": 5.7096,
      "mean_token_accuracy": 0.16896679252386093,
      "num_tokens": 13471425.0,
      "step": 7210
    },
    {
      "entropy": 5.949064111709594,
      "epoch": 0.6371423525256094,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004966046736009519,
      "loss": 5.7538,
      "mean_token_accuracy": 0.180668406188488,
      "num_tokens": 13479755.0,
      "step": 7215
    },
    {
      "entropy": 5.84395956993103,
      "epoch": 0.6375838926174496,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004965992212194772,
      "loss": 5.6444,
      "mean_token_accuracy": 0.1895698606967926,
      "num_tokens": 13488854.0,
      "step": 7220
    },
    {
      "entropy": 5.914668321609497,
      "epoch": 0.63802543270929,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004965937644969998,
      "loss": 5.7083,
      "mean_token_accuracy": 0.17636025100946426,
      "num_tokens": 13497987.0,
      "step": 7225
    },
    {
      "entropy": 5.91913480758667,
      "epoch": 0.6384669728011303,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004965883034336266,
      "loss": 5.7369,
      "mean_token_accuracy": 0.178534634411335,
      "num_tokens": 13506960.0,
      "step": 7230
    },
    {
      "entropy": 5.8813330173492435,
      "epoch": 0.6389085128929707,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004965828380294645,
      "loss": 5.6579,
      "mean_token_accuracy": 0.18809425383806228,
      "num_tokens": 13515907.0,
      "step": 7235
    },
    {
      "entropy": 5.9257501602172855,
      "epoch": 0.6393500529848111,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004965773682846207,
      "loss": 5.7611,
      "mean_token_accuracy": 0.1841880515217781,
      "num_tokens": 13524876.0,
      "step": 7240
    },
    {
      "entropy": 5.961688566207886,
      "epoch": 0.6397915930766513,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004965718941992021,
      "loss": 5.6709,
      "mean_token_accuracy": 0.18797953724861144,
      "num_tokens": 13532891.0,
      "step": 7245
    },
    {
      "entropy": 5.994545555114746,
      "epoch": 0.6402331331684917,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004965664157733162,
      "loss": 5.7375,
      "mean_token_accuracy": 0.18454429805278777,
      "num_tokens": 13542409.0,
      "step": 7250
    },
    {
      "entropy": 5.864827823638916,
      "epoch": 0.640674673260332,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004965609330070703,
      "loss": 5.7061,
      "mean_token_accuracy": 0.18280179649591446,
      "num_tokens": 13551947.0,
      "step": 7255
    },
    {
      "entropy": 5.94833664894104,
      "epoch": 0.6411162133521724,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004965554459005717,
      "loss": 5.7365,
      "mean_token_accuracy": 0.17946964353322983,
      "num_tokens": 13560812.0,
      "step": 7260
    },
    {
      "entropy": 5.968832635879517,
      "epoch": 0.6415577534440127,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004965499544539278,
      "loss": 5.7258,
      "mean_token_accuracy": 0.18233955651521683,
      "num_tokens": 13570197.0,
      "step": 7265
    },
    {
      "entropy": 5.9128265380859375,
      "epoch": 0.641999293535853,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004965444586672463,
      "loss": 5.6503,
      "mean_token_accuracy": 0.18139110803604125,
      "num_tokens": 13579614.0,
      "step": 7270
    },
    {
      "entropy": 5.899955463409424,
      "epoch": 0.6424408336276934,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000496538958540635,
      "loss": 5.6852,
      "mean_token_accuracy": 0.18292355984449388,
      "num_tokens": 13589511.0,
      "step": 7275
    },
    {
      "entropy": 5.9232399463653564,
      "epoch": 0.6428823737195337,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004965334540742015,
      "loss": 5.6729,
      "mean_token_accuracy": 0.18746213912963866,
      "num_tokens": 13597718.0,
      "step": 7280
    },
    {
      "entropy": 5.885659217834473,
      "epoch": 0.6433239138113741,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004965279452680535,
      "loss": 5.7669,
      "mean_token_accuracy": 0.17772504538297654,
      "num_tokens": 13607599.0,
      "step": 7285
    },
    {
      "entropy": 5.995887422561646,
      "epoch": 0.6437654539032144,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004965224321222991,
      "loss": 5.7932,
      "mean_token_accuracy": 0.1795261546969414,
      "num_tokens": 13616513.0,
      "step": 7290
    },
    {
      "entropy": 5.901009464263916,
      "epoch": 0.6442069939950548,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004965169146370462,
      "loss": 5.6643,
      "mean_token_accuracy": 0.18733725994825362,
      "num_tokens": 13625505.0,
      "step": 7295
    },
    {
      "entropy": 5.942426013946533,
      "epoch": 0.644648534086895,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004965113928124031,
      "loss": 5.7231,
      "mean_token_accuracy": 0.17932885885238647,
      "num_tokens": 13635340.0,
      "step": 7300
    },
    {
      "entropy": 5.905205011367798,
      "epoch": 0.6450900741787354,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004965058666484775,
      "loss": 5.6533,
      "mean_token_accuracy": 0.1771724820137024,
      "num_tokens": 13644886.0,
      "step": 7305
    },
    {
      "entropy": 5.897004556655884,
      "epoch": 0.6455316142705758,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004965003361453782,
      "loss": 5.675,
      "mean_token_accuracy": 0.18789544999599456,
      "num_tokens": 13654093.0,
      "step": 7310
    },
    {
      "entropy": 5.986611986160279,
      "epoch": 0.6459731543624161,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004964948013032132,
      "loss": 5.7582,
      "mean_token_accuracy": 0.1729942262172699,
      "num_tokens": 13663095.0,
      "step": 7315
    },
    {
      "entropy": 5.9798657417297365,
      "epoch": 0.6464146944542565,
      "grad_norm": 0.8125,
      "learning_rate": 0.000496489262122091,
      "loss": 5.78,
      "mean_token_accuracy": 0.17928575426340104,
      "num_tokens": 13673624.0,
      "step": 7320
    },
    {
      "entropy": 5.979729175567627,
      "epoch": 0.6468562345460968,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004964837186021201,
      "loss": 5.6401,
      "mean_token_accuracy": 0.1817734122276306,
      "num_tokens": 13683145.0,
      "step": 7325
    },
    {
      "entropy": 5.907504510879517,
      "epoch": 0.6472977746379371,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004964781707434093,
      "loss": 5.6629,
      "mean_token_accuracy": 0.17742054164409637,
      "num_tokens": 13691963.0,
      "step": 7330
    },
    {
      "entropy": 5.77130274772644,
      "epoch": 0.6477393147297774,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004964726185460669,
      "loss": 5.5599,
      "mean_token_accuracy": 0.18250133246183395,
      "num_tokens": 13700926.0,
      "step": 7335
    },
    {
      "entropy": 5.937017917633057,
      "epoch": 0.6481808548216178,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004964670620102019,
      "loss": 5.7943,
      "mean_token_accuracy": 0.17252365052700042,
      "num_tokens": 13709824.0,
      "step": 7340
    },
    {
      "entropy": 5.945567226409912,
      "epoch": 0.6486223949134582,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004964615011359231,
      "loss": 5.6856,
      "mean_token_accuracy": 0.17581720650196075,
      "num_tokens": 13718789.0,
      "step": 7345
    },
    {
      "entropy": 5.929153537750244,
      "epoch": 0.6490639350052985,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004964559359233395,
      "loss": 5.6952,
      "mean_token_accuracy": 0.18001507371664047,
      "num_tokens": 13728180.0,
      "step": 7350
    },
    {
      "entropy": 5.902329540252685,
      "epoch": 0.6495054750971389,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004964503663725601,
      "loss": 5.8018,
      "mean_token_accuracy": 0.17223318219184874,
      "num_tokens": 13737819.0,
      "step": 7355
    },
    {
      "entropy": 5.906650400161743,
      "epoch": 0.6499470151889791,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004964447924836938,
      "loss": 5.7688,
      "mean_token_accuracy": 0.17573816925287247,
      "num_tokens": 13747519.0,
      "step": 7360
    },
    {
      "entropy": 5.989020252227784,
      "epoch": 0.6503885552808195,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004964392142568501,
      "loss": 5.7569,
      "mean_token_accuracy": 0.1694555476307869,
      "num_tokens": 13757821.0,
      "step": 7365
    },
    {
      "entropy": 5.8534832954406735,
      "epoch": 0.6508300953726598,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004964336316921381,
      "loss": 5.5698,
      "mean_token_accuracy": 0.1853997752070427,
      "num_tokens": 13766560.0,
      "step": 7370
    },
    {
      "entropy": 5.892873287200928,
      "epoch": 0.6512716354645002,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004964280447896672,
      "loss": 5.6008,
      "mean_token_accuracy": 0.18718067109584807,
      "num_tokens": 13775560.0,
      "step": 7375
    },
    {
      "entropy": 5.887672233581543,
      "epoch": 0.6517131755563406,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004964224535495468,
      "loss": 5.6739,
      "mean_token_accuracy": 0.18187779188156128,
      "num_tokens": 13784391.0,
      "step": 7380
    },
    {
      "entropy": 5.880650043487549,
      "epoch": 0.6521547156481808,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004964168579718864,
      "loss": 5.6567,
      "mean_token_accuracy": 0.18475085198879243,
      "num_tokens": 13794008.0,
      "step": 7385
    },
    {
      "entropy": 5.972517251968384,
      "epoch": 0.6525962557400212,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004964112580567959,
      "loss": 5.7224,
      "mean_token_accuracy": 0.17432813197374344,
      "num_tokens": 13804321.0,
      "step": 7390
    },
    {
      "entropy": 5.9455334663391115,
      "epoch": 0.6530377958318615,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004964056538043845,
      "loss": 5.7159,
      "mean_token_accuracy": 0.17365975230932235,
      "num_tokens": 13813492.0,
      "step": 7395
    },
    {
      "entropy": 5.919697284698486,
      "epoch": 0.6534793359237019,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004964000452147624,
      "loss": 5.7034,
      "mean_token_accuracy": 0.1820142850279808,
      "num_tokens": 13822471.0,
      "step": 7400
    },
    {
      "entropy": 5.901232194900513,
      "epoch": 0.6539208760155422,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004963944322880394,
      "loss": 5.7557,
      "mean_token_accuracy": 0.17409348785877227,
      "num_tokens": 13832789.0,
      "step": 7405
    },
    {
      "entropy": 5.958118200302124,
      "epoch": 0.6543624161073825,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004963888150243253,
      "loss": 5.709,
      "mean_token_accuracy": 0.1774753138422966,
      "num_tokens": 13842230.0,
      "step": 7410
    },
    {
      "entropy": 5.988575839996338,
      "epoch": 0.6548039561992229,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004963831934237302,
      "loss": 5.6402,
      "mean_token_accuracy": 0.18648962676525116,
      "num_tokens": 13851542.0,
      "step": 7415
    },
    {
      "entropy": 5.891352653503418,
      "epoch": 0.6552454962910632,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004963775674863643,
      "loss": 5.6725,
      "mean_token_accuracy": 0.1804959684610367,
      "num_tokens": 13860753.0,
      "step": 7420
    },
    {
      "entropy": 5.835614538192749,
      "epoch": 0.6556870363829036,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004963719372123378,
      "loss": 5.6639,
      "mean_token_accuracy": 0.18368293344974518,
      "num_tokens": 13870133.0,
      "step": 7425
    },
    {
      "entropy": 5.902837753295898,
      "epoch": 0.6561285764747439,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004963663026017609,
      "loss": 5.7511,
      "mean_token_accuracy": 0.18147976696491241,
      "num_tokens": 13879086.0,
      "step": 7430
    },
    {
      "entropy": 5.9467963695526125,
      "epoch": 0.6565701165665843,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004963606636547441,
      "loss": 5.6768,
      "mean_token_accuracy": 0.18935618698596954,
      "num_tokens": 13888706.0,
      "step": 7435
    },
    {
      "entropy": 5.905650615692139,
      "epoch": 0.6570116566584245,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004963550203713976,
      "loss": 5.7261,
      "mean_token_accuracy": 0.17851834148168563,
      "num_tokens": 13897455.0,
      "step": 7440
    },
    {
      "entropy": 5.907645559310913,
      "epoch": 0.6574531967502649,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004963493727518323,
      "loss": 5.6984,
      "mean_token_accuracy": 0.18513126969337462,
      "num_tokens": 13907697.0,
      "step": 7445
    },
    {
      "entropy": 5.926703357696534,
      "epoch": 0.6578947368421053,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004963437207961585,
      "loss": 5.7227,
      "mean_token_accuracy": 0.1811746746301651,
      "num_tokens": 13917124.0,
      "step": 7450
    },
    {
      "entropy": 5.971260261535645,
      "epoch": 0.6583362769339456,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004963380645044874,
      "loss": 5.6622,
      "mean_token_accuracy": 0.18359404802322388,
      "num_tokens": 13926434.0,
      "step": 7455
    },
    {
      "entropy": 5.838030576705933,
      "epoch": 0.658777817025786,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004963324038769292,
      "loss": 5.6398,
      "mean_token_accuracy": 0.1903223752975464,
      "num_tokens": 13934883.0,
      "step": 7460
    },
    {
      "entropy": 5.900352668762207,
      "epoch": 0.6592193571176262,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004963267389135953,
      "loss": 5.7429,
      "mean_token_accuracy": 0.1787104442715645,
      "num_tokens": 13944514.0,
      "step": 7465
    },
    {
      "entropy": 5.90740818977356,
      "epoch": 0.6596608972094666,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004963210696145964,
      "loss": 5.6602,
      "mean_token_accuracy": 0.179434372484684,
      "num_tokens": 13953434.0,
      "step": 7470
    },
    {
      "entropy": 5.9329437732696535,
      "epoch": 0.6601024373013069,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004963153959800436,
      "loss": 5.6573,
      "mean_token_accuracy": 0.18745767027139665,
      "num_tokens": 13962493.0,
      "step": 7475
    },
    {
      "entropy": 5.919755315780639,
      "epoch": 0.6605439773931473,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000496309718010048,
      "loss": 5.7087,
      "mean_token_accuracy": 0.1846643343567848,
      "num_tokens": 13971193.0,
      "step": 7480
    },
    {
      "entropy": 5.865570878982544,
      "epoch": 0.6609855174849877,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004963040357047209,
      "loss": 5.693,
      "mean_token_accuracy": 0.18486219942569732,
      "num_tokens": 13981829.0,
      "step": 7485
    },
    {
      "entropy": 5.938862991333008,
      "epoch": 0.661427057576828,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004962983490641737,
      "loss": 5.6524,
      "mean_token_accuracy": 0.17965342551469804,
      "num_tokens": 13990262.0,
      "step": 7490
    },
    {
      "entropy": 5.931566143035889,
      "epoch": 0.6618685976686683,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004962926580885177,
      "loss": 5.711,
      "mean_token_accuracy": 0.17974878549575807,
      "num_tokens": 14000100.0,
      "step": 7495
    },
    {
      "entropy": 5.928611373901367,
      "epoch": 0.6623101377605086,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004962869627778642,
      "loss": 5.7297,
      "mean_token_accuracy": 0.18271702378988267,
      "num_tokens": 14009330.0,
      "step": 7500
    },
    {
      "entropy": 5.877379941940307,
      "epoch": 0.662751677852349,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004962812631323252,
      "loss": 5.7004,
      "mean_token_accuracy": 0.17838508486747742,
      "num_tokens": 14018109.0,
      "step": 7505
    },
    {
      "entropy": 5.857556009292603,
      "epoch": 0.6631932179441893,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000496275559152012,
      "loss": 5.5939,
      "mean_token_accuracy": 0.18382253348827363,
      "num_tokens": 14027318.0,
      "step": 7510
    },
    {
      "entropy": 5.90235743522644,
      "epoch": 0.6636347580360297,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004962698508370365,
      "loss": 5.7087,
      "mean_token_accuracy": 0.18098356127738952,
      "num_tokens": 14037061.0,
      "step": 7515
    },
    {
      "entropy": 5.900296354293824,
      "epoch": 0.6640762981278701,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004962641381875103,
      "loss": 5.6673,
      "mean_token_accuracy": 0.17706697136163713,
      "num_tokens": 14046008.0,
      "step": 7520
    },
    {
      "entropy": 5.904057312011719,
      "epoch": 0.6645178382197103,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004962584212035456,
      "loss": 5.6779,
      "mean_token_accuracy": 0.17278148829936982,
      "num_tokens": 14054535.0,
      "step": 7525
    },
    {
      "entropy": 5.935598373413086,
      "epoch": 0.6649593783115507,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004962526998852543,
      "loss": 5.6885,
      "mean_token_accuracy": 0.18802266865968703,
      "num_tokens": 14063680.0,
      "step": 7530
    },
    {
      "entropy": 5.948052835464478,
      "epoch": 0.665400918403391,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004962469742327483,
      "loss": 5.7086,
      "mean_token_accuracy": 0.18357841819524764,
      "num_tokens": 14072206.0,
      "step": 7535
    },
    {
      "entropy": 5.953330135345459,
      "epoch": 0.6658424584952314,
      "grad_norm": 0.90625,
      "learning_rate": 0.00049624124424614,
      "loss": 5.7194,
      "mean_token_accuracy": 0.18233713656663894,
      "num_tokens": 14081943.0,
      "step": 7540
    },
    {
      "entropy": 5.907440185546875,
      "epoch": 0.6662839985870717,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004962355099255415,
      "loss": 5.6778,
      "mean_token_accuracy": 0.18833412528038024,
      "num_tokens": 14091695.0,
      "step": 7545
    },
    {
      "entropy": 5.81530237197876,
      "epoch": 0.666725538678912,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004962297712710653,
      "loss": 5.6031,
      "mean_token_accuracy": 0.1932504266500473,
      "num_tokens": 14100811.0,
      "step": 7550
    },
    {
      "entropy": 5.941954612731934,
      "epoch": 0.6671670787707524,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004962240282828236,
      "loss": 5.7873,
      "mean_token_accuracy": 0.1661756306886673,
      "num_tokens": 14109818.0,
      "step": 7555
    },
    {
      "entropy": 5.935464525222779,
      "epoch": 0.6676086188625927,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000496218280960929,
      "loss": 5.7011,
      "mean_token_accuracy": 0.17778615057468414,
      "num_tokens": 14119811.0,
      "step": 7560
    },
    {
      "entropy": 5.979626893997192,
      "epoch": 0.6680501589544331,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004962125293054941,
      "loss": 5.6657,
      "mean_token_accuracy": 0.18736949414014817,
      "num_tokens": 14128963.0,
      "step": 7565
    },
    {
      "entropy": 5.878252077102661,
      "epoch": 0.6684916990462734,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004962067733166317,
      "loss": 5.7049,
      "mean_token_accuracy": 0.17186809182167054,
      "num_tokens": 14138532.0,
      "step": 7570
    },
    {
      "entropy": 5.826684474945068,
      "epoch": 0.6689332391381138,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004962010129944542,
      "loss": 5.6849,
      "mean_token_accuracy": 0.18159957826137543,
      "num_tokens": 14148107.0,
      "step": 7575
    },
    {
      "entropy": 5.974263906478882,
      "epoch": 0.669374779229954,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004961952483390747,
      "loss": 5.7294,
      "mean_token_accuracy": 0.1763908162713051,
      "num_tokens": 14158823.0,
      "step": 7580
    },
    {
      "entropy": 5.860423851013183,
      "epoch": 0.6698163193217944,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004961894793506062,
      "loss": 5.6734,
      "mean_token_accuracy": 0.1888882637023926,
      "num_tokens": 14168147.0,
      "step": 7585
    },
    {
      "entropy": 5.892768669128418,
      "epoch": 0.6702578594136348,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004961837060291615,
      "loss": 5.7062,
      "mean_token_accuracy": 0.17821233719587326,
      "num_tokens": 14177123.0,
      "step": 7590
    },
    {
      "entropy": 5.907606601715088,
      "epoch": 0.6706993995054751,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004961779283748538,
      "loss": 5.6805,
      "mean_token_accuracy": 0.1794887065887451,
      "num_tokens": 14186614.0,
      "step": 7595
    },
    {
      "entropy": 5.9154106140136715,
      "epoch": 0.6711409395973155,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004961721463877963,
      "loss": 5.6989,
      "mean_token_accuracy": 0.18051617294549943,
      "num_tokens": 14195310.0,
      "step": 7600
    },
    {
      "entropy": 5.919904899597168,
      "epoch": 0.6715824796891557,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004961663600681023,
      "loss": 5.6704,
      "mean_token_accuracy": 0.18061379790306092,
      "num_tokens": 14204321.0,
      "step": 7605
    },
    {
      "entropy": 5.88912878036499,
      "epoch": 0.6720240197809961,
      "grad_norm": 1.015625,
      "learning_rate": 0.000496160569415885,
      "loss": 5.7036,
      "mean_token_accuracy": 0.18224740624427796,
      "num_tokens": 14213615.0,
      "step": 7610
    },
    {
      "entropy": 5.953593444824219,
      "epoch": 0.6724655598728364,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004961547744312579,
      "loss": 5.571,
      "mean_token_accuracy": 0.19122107177972794,
      "num_tokens": 14222715.0,
      "step": 7615
    },
    {
      "entropy": 5.953861474990845,
      "epoch": 0.6729070999646768,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004961489751143346,
      "loss": 5.7101,
      "mean_token_accuracy": 0.18252974301576613,
      "num_tokens": 14232117.0,
      "step": 7620
    },
    {
      "entropy": 5.8538073062896725,
      "epoch": 0.6733486400565172,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004961431714652287,
      "loss": 5.5862,
      "mean_token_accuracy": 0.19271744936704635,
      "num_tokens": 14240606.0,
      "step": 7625
    },
    {
      "entropy": 5.886905479431152,
      "epoch": 0.6737901801483575,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004961373634840538,
      "loss": 5.7103,
      "mean_token_accuracy": 0.1752036228775978,
      "num_tokens": 14249844.0,
      "step": 7630
    },
    {
      "entropy": 5.935414505004883,
      "epoch": 0.6742317202401978,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004961315511709237,
      "loss": 5.7062,
      "mean_token_accuracy": 0.18466105908155442,
      "num_tokens": 14259231.0,
      "step": 7635
    },
    {
      "entropy": 5.972374820709229,
      "epoch": 0.6746732603320381,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004961257345259524,
      "loss": 5.7107,
      "mean_token_accuracy": 0.1831063836812973,
      "num_tokens": 14268508.0,
      "step": 7640
    },
    {
      "entropy": 5.89817533493042,
      "epoch": 0.6751148004238785,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004961199135492535,
      "loss": 5.6311,
      "mean_token_accuracy": 0.18840737044811248,
      "num_tokens": 14278057.0,
      "step": 7645
    },
    {
      "entropy": 5.872119998931884,
      "epoch": 0.6755563405157188,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004961140882409414,
      "loss": 5.6891,
      "mean_token_accuracy": 0.1815480887889862,
      "num_tokens": 14287225.0,
      "step": 7650
    },
    {
      "entropy": 5.89968466758728,
      "epoch": 0.6759978806075592,
      "grad_norm": 0.85546875,
      "learning_rate": 0.00049610825860113,
      "loss": 5.655,
      "mean_token_accuracy": 0.1813758134841919,
      "num_tokens": 14296529.0,
      "step": 7655
    },
    {
      "entropy": 5.920412063598633,
      "epoch": 0.6764394206993996,
      "grad_norm": 1.0,
      "learning_rate": 0.0004961024246299336,
      "loss": 5.6277,
      "mean_token_accuracy": 0.18376876413822174,
      "num_tokens": 14305675.0,
      "step": 7660
    },
    {
      "entropy": 5.9062878608703615,
      "epoch": 0.6768809607912398,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004960965863274665,
      "loss": 5.5924,
      "mean_token_accuracy": 0.18598417639732362,
      "num_tokens": 14313632.0,
      "step": 7665
    },
    {
      "entropy": 5.870595502853393,
      "epoch": 0.6773225008830802,
      "grad_norm": 1.0,
      "learning_rate": 0.000496090743693843,
      "loss": 5.7035,
      "mean_token_accuracy": 0.18137972801923752,
      "num_tokens": 14323353.0,
      "step": 7670
    },
    {
      "entropy": 5.984563302993775,
      "epoch": 0.6777640409749205,
      "grad_norm": 1.0,
      "learning_rate": 0.0004960848967291775,
      "loss": 5.7151,
      "mean_token_accuracy": 0.1740511640906334,
      "num_tokens": 14331953.0,
      "step": 7675
    },
    {
      "entropy": 5.88297929763794,
      "epoch": 0.6782055810667609,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004960790454335846,
      "loss": 5.7209,
      "mean_token_accuracy": 0.17738450318574905,
      "num_tokens": 14342456.0,
      "step": 7680
    },
    {
      "entropy": 5.808986568450928,
      "epoch": 0.6786471211586012,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000496073189807179,
      "loss": 5.5301,
      "mean_token_accuracy": 0.19175227284431456,
      "num_tokens": 14351013.0,
      "step": 7685
    },
    {
      "entropy": 5.941276884078979,
      "epoch": 0.6790886612504415,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004960673298500753,
      "loss": 5.8007,
      "mean_token_accuracy": 0.17323903888463973,
      "num_tokens": 14360887.0,
      "step": 7690
    },
    {
      "entropy": 5.945207118988037,
      "epoch": 0.6795302013422819,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004960614655623883,
      "loss": 5.7007,
      "mean_token_accuracy": 0.1831446796655655,
      "num_tokens": 14370510.0,
      "step": 7695
    },
    {
      "entropy": 5.913691139221191,
      "epoch": 0.6799717414341222,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000496055596944233,
      "loss": 5.6944,
      "mean_token_accuracy": 0.18526873290538787,
      "num_tokens": 14379315.0,
      "step": 7700
    },
    {
      "entropy": 5.881440448760986,
      "epoch": 0.6804132815259626,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004960497239957243,
      "loss": 5.6668,
      "mean_token_accuracy": 0.18388139009475707,
      "num_tokens": 14388537.0,
      "step": 7705
    },
    {
      "entropy": 5.987562608718872,
      "epoch": 0.6808548216178029,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004960438467169772,
      "loss": 5.7216,
      "mean_token_accuracy": 0.17798287868499757,
      "num_tokens": 14398347.0,
      "step": 7710
    },
    {
      "entropy": 5.966675758361816,
      "epoch": 0.6812963617096433,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004960379651081068,
      "loss": 5.6293,
      "mean_token_accuracy": 0.1845902293920517,
      "num_tokens": 14407304.0,
      "step": 7715
    },
    {
      "entropy": 5.903362655639649,
      "epoch": 0.6817379018014835,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004960320791692283,
      "loss": 5.6624,
      "mean_token_accuracy": 0.1791662245988846,
      "num_tokens": 14416641.0,
      "step": 7720
    },
    {
      "entropy": 5.976517868041992,
      "epoch": 0.6821794418933239,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004960261889004572,
      "loss": 5.7595,
      "mean_token_accuracy": 0.17492255866527556,
      "num_tokens": 14426095.0,
      "step": 7725
    },
    {
      "entropy": 5.935105276107788,
      "epoch": 0.6826209819851643,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004960202943019088,
      "loss": 5.6682,
      "mean_token_accuracy": 0.18540456146001816,
      "num_tokens": 14434563.0,
      "step": 7730
    },
    {
      "entropy": 5.794443988800049,
      "epoch": 0.6830625220770046,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004960143953736984,
      "loss": 5.598,
      "mean_token_accuracy": 0.18621402233839035,
      "num_tokens": 14444319.0,
      "step": 7735
    },
    {
      "entropy": 5.9014829158782955,
      "epoch": 0.683504062168845,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004960084921159419,
      "loss": 5.6398,
      "mean_token_accuracy": 0.1842109367251396,
      "num_tokens": 14453381.0,
      "step": 7740
    },
    {
      "entropy": 5.819450187683105,
      "epoch": 0.6839456022606852,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004960025845287546,
      "loss": 5.6404,
      "mean_token_accuracy": 0.18381152004003526,
      "num_tokens": 14462660.0,
      "step": 7745
    },
    {
      "entropy": 5.850420904159546,
      "epoch": 0.6843871423525256,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004959966726122523,
      "loss": 5.5589,
      "mean_token_accuracy": 0.1945916920900345,
      "num_tokens": 14471820.0,
      "step": 7750
    },
    {
      "entropy": 5.949693870544434,
      "epoch": 0.6848286824443659,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000495990756366551,
      "loss": 5.7463,
      "mean_token_accuracy": 0.18212738186120986,
      "num_tokens": 14481663.0,
      "step": 7755
    },
    {
      "entropy": 5.947234582901001,
      "epoch": 0.6852702225362063,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004959848357917664,
      "loss": 5.7543,
      "mean_token_accuracy": 0.17507750540971756,
      "num_tokens": 14491515.0,
      "step": 7760
    },
    {
      "entropy": 5.843775367736816,
      "epoch": 0.6857117626280467,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004959789108880145,
      "loss": 5.5817,
      "mean_token_accuracy": 0.1806255117058754,
      "num_tokens": 14500852.0,
      "step": 7765
    },
    {
      "entropy": 5.86192569732666,
      "epoch": 0.686153302719887,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004959729816554115,
      "loss": 5.6308,
      "mean_token_accuracy": 0.19001229852437973,
      "num_tokens": 14509771.0,
      "step": 7770
    },
    {
      "entropy": 5.83790431022644,
      "epoch": 0.6865948428117273,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004959670480940734,
      "loss": 5.6172,
      "mean_token_accuracy": 0.18619886487722398,
      "num_tokens": 14519377.0,
      "step": 7775
    },
    {
      "entropy": 5.864809226989746,
      "epoch": 0.6870363829035676,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004959611102041165,
      "loss": 5.6722,
      "mean_token_accuracy": 0.18079863488674164,
      "num_tokens": 14528998.0,
      "step": 7780
    },
    {
      "entropy": 5.904181909561157,
      "epoch": 0.687477922995408,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004959551679856571,
      "loss": 5.6997,
      "mean_token_accuracy": 0.1849734291434288,
      "num_tokens": 14538153.0,
      "step": 7785
    },
    {
      "entropy": 5.8900541305542,
      "epoch": 0.6879194630872483,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004959492214388117,
      "loss": 5.6349,
      "mean_token_accuracy": 0.18393201678991317,
      "num_tokens": 14547631.0,
      "step": 7790
    },
    {
      "entropy": 5.935344123840332,
      "epoch": 0.6883610031790887,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004959432705636968,
      "loss": 5.5612,
      "mean_token_accuracy": 0.1882770836353302,
      "num_tokens": 14557685.0,
      "step": 7795
    },
    {
      "entropy": 5.903605842590332,
      "epoch": 0.688802543270929,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004959373153604287,
      "loss": 5.7088,
      "mean_token_accuracy": 0.18453071862459183,
      "num_tokens": 14567360.0,
      "step": 7800
    },
    {
      "entropy": 5.804054594039917,
      "epoch": 0.6892440833627693,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004959313558291243,
      "loss": 5.6375,
      "mean_token_accuracy": 0.18478822112083435,
      "num_tokens": 14576916.0,
      "step": 7805
    },
    {
      "entropy": 5.898049783706665,
      "epoch": 0.6896856234546097,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004959253919699004,
      "loss": 5.555,
      "mean_token_accuracy": 0.1911950170993805,
      "num_tokens": 14586716.0,
      "step": 7810
    },
    {
      "entropy": 5.912984180450439,
      "epoch": 0.69012716354645,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004959194237828735,
      "loss": 5.6252,
      "mean_token_accuracy": 0.1866038516163826,
      "num_tokens": 14595690.0,
      "step": 7815
    },
    {
      "entropy": 5.8310645580291744,
      "epoch": 0.6905687036382904,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004959134512681609,
      "loss": 5.7008,
      "mean_token_accuracy": 0.18006411343812942,
      "num_tokens": 14605481.0,
      "step": 7820
    },
    {
      "entropy": 5.893993234634399,
      "epoch": 0.6910102437301306,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004959074744258794,
      "loss": 5.7398,
      "mean_token_accuracy": 0.17599762678146363,
      "num_tokens": 14615242.0,
      "step": 7825
    },
    {
      "entropy": 5.87747049331665,
      "epoch": 0.691451783821971,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004959014932561461,
      "loss": 5.5771,
      "mean_token_accuracy": 0.19491011202335357,
      "num_tokens": 14624343.0,
      "step": 7830
    },
    {
      "entropy": 5.802929401397705,
      "epoch": 0.6918933239138114,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004958955077590782,
      "loss": 5.5422,
      "mean_token_accuracy": 0.19644290953874588,
      "num_tokens": 14633014.0,
      "step": 7835
    },
    {
      "entropy": 5.872480010986328,
      "epoch": 0.6923348640056517,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004958895179347929,
      "loss": 5.6687,
      "mean_token_accuracy": 0.18705726861953736,
      "num_tokens": 14642796.0,
      "step": 7840
    },
    {
      "entropy": 5.943118047714234,
      "epoch": 0.6927764040974921,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004958835237834075,
      "loss": 5.716,
      "mean_token_accuracy": 0.174401131272316,
      "num_tokens": 14652388.0,
      "step": 7845
    },
    {
      "entropy": 5.9830615520477295,
      "epoch": 0.6932179441893324,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004958775253050396,
      "loss": 5.6614,
      "mean_token_accuracy": 0.18240052610635757,
      "num_tokens": 14661219.0,
      "step": 7850
    },
    {
      "entropy": 5.8752015113830565,
      "epoch": 0.6936594842811727,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004958715224998066,
      "loss": 5.6783,
      "mean_token_accuracy": 0.18555933982133865,
      "num_tokens": 14670878.0,
      "step": 7855
    },
    {
      "entropy": 5.82140998840332,
      "epoch": 0.694101024373013,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004958655153678261,
      "loss": 5.7262,
      "mean_token_accuracy": 0.17967288792133332,
      "num_tokens": 14680893.0,
      "step": 7860
    },
    {
      "entropy": 5.9207089900970455,
      "epoch": 0.6945425644648534,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004958595039092156,
      "loss": 5.6324,
      "mean_token_accuracy": 0.18184648156166078,
      "num_tokens": 14689249.0,
      "step": 7865
    },
    {
      "entropy": 5.896468544006348,
      "epoch": 0.6949841045566938,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004958534881240932,
      "loss": 5.6245,
      "mean_token_accuracy": 0.1842005655169487,
      "num_tokens": 14698724.0,
      "step": 7870
    },
    {
      "entropy": 5.853013896942139,
      "epoch": 0.6954256446485341,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004958474680125766,
      "loss": 5.5998,
      "mean_token_accuracy": 0.181177119910717,
      "num_tokens": 14707756.0,
      "step": 7875
    },
    {
      "entropy": 5.86539044380188,
      "epoch": 0.6958671847403745,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004958414435747837,
      "loss": 5.616,
      "mean_token_accuracy": 0.18928422033786774,
      "num_tokens": 14717416.0,
      "step": 7880
    },
    {
      "entropy": 5.847776317596436,
      "epoch": 0.6963087248322147,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004958354148108324,
      "loss": 5.6802,
      "mean_token_accuracy": 0.182321697473526,
      "num_tokens": 14727321.0,
      "step": 7885
    },
    {
      "entropy": 5.7946008205413815,
      "epoch": 0.6967502649240551,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000495829381720841,
      "loss": 5.5477,
      "mean_token_accuracy": 0.19297181218862533,
      "num_tokens": 14736298.0,
      "step": 7890
    },
    {
      "entropy": 5.8987609386444095,
      "epoch": 0.6971918050158954,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004958233443049275,
      "loss": 5.5849,
      "mean_token_accuracy": 0.18445069193840027,
      "num_tokens": 14745893.0,
      "step": 7895
    },
    {
      "entropy": 5.892021942138672,
      "epoch": 0.6976333451077358,
      "grad_norm": 1.0,
      "learning_rate": 0.0004958173025632103,
      "loss": 5.664,
      "mean_token_accuracy": 0.17899592816829682,
      "num_tokens": 14755277.0,
      "step": 7900
    },
    {
      "entropy": 5.957773065567016,
      "epoch": 0.6980748851995762,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004958112564958079,
      "loss": 5.7665,
      "mean_token_accuracy": 0.17907627373933793,
      "num_tokens": 14764369.0,
      "step": 7905
    },
    {
      "entropy": 5.851858806610108,
      "epoch": 0.6985164252914164,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004958052061028385,
      "loss": 5.6058,
      "mean_token_accuracy": 0.1951722025871277,
      "num_tokens": 14772970.0,
      "step": 7910
    },
    {
      "entropy": 5.824829816818237,
      "epoch": 0.6989579653832568,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004957991513844205,
      "loss": 5.6783,
      "mean_token_accuracy": 0.183780537545681,
      "num_tokens": 14782862.0,
      "step": 7915
    },
    {
      "entropy": 5.947744512557984,
      "epoch": 0.6993995054750971,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004957930923406729,
      "loss": 5.6932,
      "mean_token_accuracy": 0.1842297852039337,
      "num_tokens": 14792042.0,
      "step": 7920
    },
    {
      "entropy": 5.92308783531189,
      "epoch": 0.6998410455669375,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004957870289717142,
      "loss": 5.6239,
      "mean_token_accuracy": 0.18697006553411483,
      "num_tokens": 14801032.0,
      "step": 7925
    },
    {
      "entropy": 5.812475728988647,
      "epoch": 0.7002825856587778,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004957809612776631,
      "loss": 5.6368,
      "mean_token_accuracy": 0.18773610293865203,
      "num_tokens": 14810594.0,
      "step": 7930
    },
    {
      "entropy": 5.906245374679566,
      "epoch": 0.7007241257506182,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004957748892586384,
      "loss": 5.7392,
      "mean_token_accuracy": 0.18038935363292694,
      "num_tokens": 14820570.0,
      "step": 7935
    },
    {
      "entropy": 5.768427610397339,
      "epoch": 0.7011656658424585,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004957688129147594,
      "loss": 5.5486,
      "mean_token_accuracy": 0.1953514665365219,
      "num_tokens": 14830078.0,
      "step": 7940
    },
    {
      "entropy": 5.886784267425537,
      "epoch": 0.7016072059342988,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004957627322461448,
      "loss": 5.6258,
      "mean_token_accuracy": 0.1836659014225006,
      "num_tokens": 14839974.0,
      "step": 7945
    },
    {
      "entropy": 5.916618680953979,
      "epoch": 0.7020487460261392,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004957566472529139,
      "loss": 5.7203,
      "mean_token_accuracy": 0.17270487248897554,
      "num_tokens": 14849551.0,
      "step": 7950
    },
    {
      "entropy": 5.903427076339722,
      "epoch": 0.7024902861179795,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004957505579351858,
      "loss": 5.6697,
      "mean_token_accuracy": 0.17774245142936707,
      "num_tokens": 14859011.0,
      "step": 7955
    },
    {
      "entropy": 5.865008449554443,
      "epoch": 0.7029318262098199,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004957444642930798,
      "loss": 5.6071,
      "mean_token_accuracy": 0.18349462747573853,
      "num_tokens": 14867948.0,
      "step": 7960
    },
    {
      "entropy": 5.898176765441894,
      "epoch": 0.7033733663016601,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004957383663267152,
      "loss": 5.656,
      "mean_token_accuracy": 0.18263181746006013,
      "num_tokens": 14877233.0,
      "step": 7965
    },
    {
      "entropy": 5.915711975097656,
      "epoch": 0.7038149063935005,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004957322640362118,
      "loss": 5.6758,
      "mean_token_accuracy": 0.1814686819911003,
      "num_tokens": 14887439.0,
      "step": 7970
    },
    {
      "entropy": 5.960000324249267,
      "epoch": 0.7042564464853409,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004957261574216887,
      "loss": 5.7501,
      "mean_token_accuracy": 0.17440729439258576,
      "num_tokens": 14897064.0,
      "step": 7975
    },
    {
      "entropy": 5.924387741088867,
      "epoch": 0.7046979865771812,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004957200464832656,
      "loss": 5.7028,
      "mean_token_accuracy": 0.17842092514038085,
      "num_tokens": 14907054.0,
      "step": 7980
    },
    {
      "entropy": 5.930465459823608,
      "epoch": 0.7051395266690216,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004957139312210626,
      "loss": 5.6654,
      "mean_token_accuracy": 0.18382232934236525,
      "num_tokens": 14916969.0,
      "step": 7985
    },
    {
      "entropy": 5.894822502136231,
      "epoch": 0.7055810667608619,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000495707811635199,
      "loss": 5.6861,
      "mean_token_accuracy": 0.18471913486719133,
      "num_tokens": 14926404.0,
      "step": 7990
    },
    {
      "entropy": 5.876964044570923,
      "epoch": 0.7060226068527022,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004957016877257949,
      "loss": 5.6544,
      "mean_token_accuracy": 0.182310651242733,
      "num_tokens": 14935693.0,
      "step": 7995
    },
    {
      "entropy": 5.899308919906616,
      "epoch": 0.7064641469445425,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004956955594929704,
      "loss": 5.596,
      "mean_token_accuracy": 0.184686329215765,
      "num_tokens": 14945147.0,
      "step": 8000
    },
    {
      "entropy": 5.859074115753174,
      "epoch": 0.7069056870363829,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004956894269368454,
      "loss": 5.6438,
      "mean_token_accuracy": 0.18151623457670213,
      "num_tokens": 14954022.0,
      "step": 8005
    },
    {
      "entropy": 5.803120088577271,
      "epoch": 0.7073472271282233,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004956832900575399,
      "loss": 5.6509,
      "mean_token_accuracy": 0.18272694200277328,
      "num_tokens": 14962977.0,
      "step": 8010
    },
    {
      "entropy": 5.9160982131958,
      "epoch": 0.7077887672200636,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004956771488551742,
      "loss": 5.6914,
      "mean_token_accuracy": 0.1767708733677864,
      "num_tokens": 14973420.0,
      "step": 8015
    },
    {
      "entropy": 5.975215291976928,
      "epoch": 0.708230307311904,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004956710033298689,
      "loss": 5.6465,
      "mean_token_accuracy": 0.1861080601811409,
      "num_tokens": 14981762.0,
      "step": 8020
    },
    {
      "entropy": 5.950171566009521,
      "epoch": 0.7086718474037442,
      "grad_norm": 0.9375,
      "learning_rate": 0.000495664853481744,
      "loss": 5.7171,
      "mean_token_accuracy": 0.17843022793531418,
      "num_tokens": 14991449.0,
      "step": 8025
    },
    {
      "entropy": 5.917905616760254,
      "epoch": 0.7091133874955846,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004956586993109201,
      "loss": 5.6514,
      "mean_token_accuracy": 0.18210796564817427,
      "num_tokens": 15000802.0,
      "step": 8030
    },
    {
      "entropy": 5.865891695022583,
      "epoch": 0.7095549275874249,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004956525408175179,
      "loss": 5.6694,
      "mean_token_accuracy": 0.18268949687480926,
      "num_tokens": 15010580.0,
      "step": 8035
    },
    {
      "entropy": 5.844978952407837,
      "epoch": 0.7099964676792653,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004956463780016578,
      "loss": 5.7012,
      "mean_token_accuracy": 0.18287778943777083,
      "num_tokens": 15020059.0,
      "step": 8040
    },
    {
      "entropy": 5.900697183609009,
      "epoch": 0.7104380077711057,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004956402108634607,
      "loss": 5.6449,
      "mean_token_accuracy": 0.1812546208500862,
      "num_tokens": 15030268.0,
      "step": 8045
    },
    {
      "entropy": 5.875079154968262,
      "epoch": 0.7108795478629459,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004956340394030473,
      "loss": 5.6166,
      "mean_token_accuracy": 0.18594399243593215,
      "num_tokens": 15038865.0,
      "step": 8050
    },
    {
      "entropy": 5.94498291015625,
      "epoch": 0.7113210879547863,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004956278636205386,
      "loss": 5.7501,
      "mean_token_accuracy": 0.17342514544725418,
      "num_tokens": 15048427.0,
      "step": 8055
    },
    {
      "entropy": 5.866507053375244,
      "epoch": 0.7117626280466266,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004956216835160556,
      "loss": 5.6176,
      "mean_token_accuracy": 0.1918194204568863,
      "num_tokens": 15057366.0,
      "step": 8060
    },
    {
      "entropy": 5.917425632476807,
      "epoch": 0.712204168138467,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004956154990897191,
      "loss": 5.6597,
      "mean_token_accuracy": 0.18279267549514772,
      "num_tokens": 15067098.0,
      "step": 8065
    },
    {
      "entropy": 5.884406900405883,
      "epoch": 0.7126457082303073,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004956093103416505,
      "loss": 5.6655,
      "mean_token_accuracy": 0.18610380887985228,
      "num_tokens": 15075734.0,
      "step": 8070
    },
    {
      "entropy": 5.898903656005859,
      "epoch": 0.7130872483221476,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004956031172719711,
      "loss": 5.7048,
      "mean_token_accuracy": 0.1823004812002182,
      "num_tokens": 15084625.0,
      "step": 8075
    },
    {
      "entropy": 5.792678642272949,
      "epoch": 0.713528788413988,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004955969198808021,
      "loss": 5.5319,
      "mean_token_accuracy": 0.19952139854431153,
      "num_tokens": 15093256.0,
      "step": 8080
    },
    {
      "entropy": 5.816761541366577,
      "epoch": 0.7139703285058283,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004955907181682648,
      "loss": 5.6,
      "mean_token_accuracy": 0.191490775346756,
      "num_tokens": 15103175.0,
      "step": 8085
    },
    {
      "entropy": 5.834416055679322,
      "epoch": 0.7144118685976687,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004955845121344809,
      "loss": 5.6853,
      "mean_token_accuracy": 0.1825338825583458,
      "num_tokens": 15112789.0,
      "step": 8090
    },
    {
      "entropy": 5.914598226547241,
      "epoch": 0.714853408689509,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004955783017795717,
      "loss": 5.7234,
      "mean_token_accuracy": 0.18739004880189897,
      "num_tokens": 15121869.0,
      "step": 8095
    },
    {
      "entropy": 5.868374967575074,
      "epoch": 0.7152949487813494,
      "grad_norm": 0.875,
      "learning_rate": 0.000495572087103659,
      "loss": 5.6229,
      "mean_token_accuracy": 0.18730900734663009,
      "num_tokens": 15131441.0,
      "step": 8100
    },
    {
      "entropy": 5.806985235214233,
      "epoch": 0.7157364888731896,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004955658681068647,
      "loss": 5.6615,
      "mean_token_accuracy": 0.18200474381446838,
      "num_tokens": 15140641.0,
      "step": 8105
    },
    {
      "entropy": 5.958062267303466,
      "epoch": 0.71617802896503,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004955596447893105,
      "loss": 5.6552,
      "mean_token_accuracy": 0.18732800036668779,
      "num_tokens": 15150073.0,
      "step": 8110
    },
    {
      "entropy": 5.803730249404907,
      "epoch": 0.7166195690568704,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004955534171511181,
      "loss": 5.3936,
      "mean_token_accuracy": 0.20334044098854065,
      "num_tokens": 15159879.0,
      "step": 8115
    },
    {
      "entropy": 5.778502416610718,
      "epoch": 0.7170611091487107,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00049554718519241,
      "loss": 5.6052,
      "mean_token_accuracy": 0.19128098487854003,
      "num_tokens": 15168316.0,
      "step": 8120
    },
    {
      "entropy": 5.765700244903565,
      "epoch": 0.7175026492405511,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004955409489133078,
      "loss": 5.6519,
      "mean_token_accuracy": 0.1813172772526741,
      "num_tokens": 15176700.0,
      "step": 8125
    },
    {
      "entropy": 5.927375173568725,
      "epoch": 0.7179441893323913,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004955347083139338,
      "loss": 5.681,
      "mean_token_accuracy": 0.18737523555755614,
      "num_tokens": 15185997.0,
      "step": 8130
    },
    {
      "entropy": 5.912154483795166,
      "epoch": 0.7183857294242317,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004955284633944104,
      "loss": 5.6616,
      "mean_token_accuracy": 0.18576339930295943,
      "num_tokens": 15195341.0,
      "step": 8135
    },
    {
      "entropy": 5.85348105430603,
      "epoch": 0.718827269516072,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004955222141548597,
      "loss": 5.6797,
      "mean_token_accuracy": 0.18169899731874467,
      "num_tokens": 15205501.0,
      "step": 8140
    },
    {
      "entropy": 5.862333488464356,
      "epoch": 0.7192688096079124,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004955159605954043,
      "loss": 5.6344,
      "mean_token_accuracy": 0.18649196475744248,
      "num_tokens": 15214778.0,
      "step": 8145
    },
    {
      "entropy": 5.957325029373169,
      "epoch": 0.7197103496997528,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004955097027161666,
      "loss": 5.6663,
      "mean_token_accuracy": 0.18574378043413162,
      "num_tokens": 15224201.0,
      "step": 8150
    },
    {
      "entropy": 5.847279453277588,
      "epoch": 0.7201518897915931,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004955034405172693,
      "loss": 5.6008,
      "mean_token_accuracy": 0.18820102363824845,
      "num_tokens": 15233472.0,
      "step": 8155
    },
    {
      "entropy": 5.881206130981445,
      "epoch": 0.7205934298834334,
      "grad_norm": 1.015625,
      "learning_rate": 0.000495497173998835,
      "loss": 5.6868,
      "mean_token_accuracy": 0.17715079635381697,
      "num_tokens": 15242323.0,
      "step": 8160
    },
    {
      "entropy": 5.893172788619995,
      "epoch": 0.7210349699752737,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004954909031609864,
      "loss": 5.6699,
      "mean_token_accuracy": 0.18143803179264067,
      "num_tokens": 15252952.0,
      "step": 8165
    },
    {
      "entropy": 5.937598085403442,
      "epoch": 0.7214765100671141,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004954846280038465,
      "loss": 5.6777,
      "mean_token_accuracy": 0.18283233791589737,
      "num_tokens": 15261815.0,
      "step": 8170
    },
    {
      "entropy": 5.916687059402466,
      "epoch": 0.7219180501589544,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004954783485275381,
      "loss": 5.6998,
      "mean_token_accuracy": 0.177165387570858,
      "num_tokens": 15271026.0,
      "step": 8175
    },
    {
      "entropy": 5.846047449111938,
      "epoch": 0.7223595902507948,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004954720647321843,
      "loss": 5.5364,
      "mean_token_accuracy": 0.19162006825208663,
      "num_tokens": 15279795.0,
      "step": 8180
    },
    {
      "entropy": 5.923773765563965,
      "epoch": 0.7228011303426352,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004954657766179081,
      "loss": 5.6822,
      "mean_token_accuracy": 0.1771241009235382,
      "num_tokens": 15289113.0,
      "step": 8185
    },
    {
      "entropy": 5.846442317962646,
      "epoch": 0.7232426704344754,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004954594841848327,
      "loss": 5.5244,
      "mean_token_accuracy": 0.19908362179994582,
      "num_tokens": 15298271.0,
      "step": 8190
    },
    {
      "entropy": 5.895449018478393,
      "epoch": 0.7236842105263158,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004954531874330814,
      "loss": 5.6342,
      "mean_token_accuracy": 0.18278907835483552,
      "num_tokens": 15307677.0,
      "step": 8195
    },
    {
      "entropy": 5.8396324634552,
      "epoch": 0.7241257506181561,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004954468863627774,
      "loss": 5.6438,
      "mean_token_accuracy": 0.18307971209287643,
      "num_tokens": 15316637.0,
      "step": 8200
    },
    {
      "entropy": 5.833258199691772,
      "epoch": 0.7245672907099965,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004954405809740445,
      "loss": 5.63,
      "mean_token_accuracy": 0.17943803071975709,
      "num_tokens": 15325221.0,
      "step": 8205
    },
    {
      "entropy": 5.881443452835083,
      "epoch": 0.7250088308018368,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004954342712670058,
      "loss": 5.6632,
      "mean_token_accuracy": 0.1839708626270294,
      "num_tokens": 15335877.0,
      "step": 8210
    },
    {
      "entropy": 5.872735929489136,
      "epoch": 0.7254503708936771,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004954279572417852,
      "loss": 5.6677,
      "mean_token_accuracy": 0.18076927959918976,
      "num_tokens": 15345096.0,
      "step": 8215
    },
    {
      "entropy": 5.841041040420532,
      "epoch": 0.7258919109855175,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004954216388985063,
      "loss": 5.5006,
      "mean_token_accuracy": 0.19473296403884888,
      "num_tokens": 15353785.0,
      "step": 8220
    },
    {
      "entropy": 5.870520782470703,
      "epoch": 0.7263334510773578,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004954153162372928,
      "loss": 5.6024,
      "mean_token_accuracy": 0.1930699646472931,
      "num_tokens": 15362761.0,
      "step": 8225
    },
    {
      "entropy": 5.888486576080322,
      "epoch": 0.7267749911691982,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004954089892582686,
      "loss": 5.6081,
      "mean_token_accuracy": 0.18426867425441742,
      "num_tokens": 15371951.0,
      "step": 8230
    },
    {
      "entropy": 5.898847579956055,
      "epoch": 0.7272165312610385,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004954026579615577,
      "loss": 5.6633,
      "mean_token_accuracy": 0.18143970519304276,
      "num_tokens": 15380277.0,
      "step": 8235
    },
    {
      "entropy": 5.8413866519927975,
      "epoch": 0.7276580713528789,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004953963223472841,
      "loss": 5.6742,
      "mean_token_accuracy": 0.18348696678876877,
      "num_tokens": 15390951.0,
      "step": 8240
    },
    {
      "entropy": 5.908440494537354,
      "epoch": 0.7280996114447191,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004953899824155718,
      "loss": 5.6227,
      "mean_token_accuracy": 0.18988921642303466,
      "num_tokens": 15399394.0,
      "step": 8245
    },
    {
      "entropy": 5.865427255630493,
      "epoch": 0.7285411515365595,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004953836381665452,
      "loss": 5.5761,
      "mean_token_accuracy": 0.18975492119789122,
      "num_tokens": 15408644.0,
      "step": 8250
    },
    {
      "entropy": 5.946927690505982,
      "epoch": 0.7289826916283999,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004953772896003284,
      "loss": 5.6464,
      "mean_token_accuracy": 0.18333052694797516,
      "num_tokens": 15417398.0,
      "step": 8255
    },
    {
      "entropy": 5.93438024520874,
      "epoch": 0.7294242317202402,
      "grad_norm": 1.0,
      "learning_rate": 0.0004953709367170458,
      "loss": 5.717,
      "mean_token_accuracy": 0.1741793468594551,
      "num_tokens": 15426532.0,
      "step": 8260
    },
    {
      "entropy": 5.8816978931427,
      "epoch": 0.7298657718120806,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000495364579516822,
      "loss": 5.6952,
      "mean_token_accuracy": 0.17792603224515915,
      "num_tokens": 15436371.0,
      "step": 8265
    },
    {
      "entropy": 5.964777851104737,
      "epoch": 0.7303073119039208,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004953582179997813,
      "loss": 5.7434,
      "mean_token_accuracy": 0.18140242248773575,
      "num_tokens": 15446600.0,
      "step": 8270
    },
    {
      "entropy": 5.91185245513916,
      "epoch": 0.7307488519957612,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004953518521660484,
      "loss": 5.5995,
      "mean_token_accuracy": 0.1830684170126915,
      "num_tokens": 15455547.0,
      "step": 8275
    },
    {
      "entropy": 5.856135416030884,
      "epoch": 0.7311903920876015,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004953454820157481,
      "loss": 5.6275,
      "mean_token_accuracy": 0.18180723786354064,
      "num_tokens": 15464543.0,
      "step": 8280
    },
    {
      "entropy": 5.958267164230347,
      "epoch": 0.7316319321794419,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004953391075490051,
      "loss": 5.7714,
      "mean_token_accuracy": 0.16960600912570953,
      "num_tokens": 15473626.0,
      "step": 8285
    },
    {
      "entropy": 5.923289155960083,
      "epoch": 0.7320734722712823,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004953327287659444,
      "loss": 5.689,
      "mean_token_accuracy": 0.17944410294294358,
      "num_tokens": 15484164.0,
      "step": 8290
    },
    {
      "entropy": 6.006247711181641,
      "epoch": 0.7325150123631226,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004953263456666907,
      "loss": 5.695,
      "mean_token_accuracy": 0.17470264732837676,
      "num_tokens": 15493718.0,
      "step": 8295
    },
    {
      "entropy": 5.919910049438476,
      "epoch": 0.7329565524549629,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004953199582513693,
      "loss": 5.6388,
      "mean_token_accuracy": 0.18529484421014786,
      "num_tokens": 15503339.0,
      "step": 8300
    },
    {
      "entropy": 5.870715141296387,
      "epoch": 0.7333980925468032,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004953135665201052,
      "loss": 5.6187,
      "mean_token_accuracy": 0.18870222121477126,
      "num_tokens": 15513314.0,
      "step": 8305
    },
    {
      "entropy": 5.898948812484742,
      "epoch": 0.7338396326386436,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004953071704730237,
      "loss": 5.6969,
      "mean_token_accuracy": 0.17855761349201202,
      "num_tokens": 15522943.0,
      "step": 8310
    },
    {
      "entropy": 5.9014928340911865,
      "epoch": 0.7342811727304839,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004953007701102499,
      "loss": 5.6605,
      "mean_token_accuracy": 0.18277857601642608,
      "num_tokens": 15531975.0,
      "step": 8315
    },
    {
      "entropy": 5.902299976348877,
      "epoch": 0.7347227128223243,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004952943654319094,
      "loss": 5.5941,
      "mean_token_accuracy": 0.1843016892671585,
      "num_tokens": 15541139.0,
      "step": 8320
    },
    {
      "entropy": 5.9181843280792235,
      "epoch": 0.7351642529141647,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004952879564381276,
      "loss": 5.6706,
      "mean_token_accuracy": 0.1839447870850563,
      "num_tokens": 15550604.0,
      "step": 8325
    },
    {
      "entropy": 5.814180707931518,
      "epoch": 0.7356057930060049,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004952815431290298,
      "loss": 5.5866,
      "mean_token_accuracy": 0.18736102283000947,
      "num_tokens": 15558995.0,
      "step": 8330
    },
    {
      "entropy": 5.975278282165528,
      "epoch": 0.7360473330978453,
      "grad_norm": 1.03125,
      "learning_rate": 0.000495275125504742,
      "loss": 5.6574,
      "mean_token_accuracy": 0.18505442291498184,
      "num_tokens": 15568229.0,
      "step": 8335
    },
    {
      "entropy": 5.837584114074707,
      "epoch": 0.7364888731896856,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004952687035653899,
      "loss": 5.6092,
      "mean_token_accuracy": 0.18216995596885682,
      "num_tokens": 15577042.0,
      "step": 8340
    },
    {
      "entropy": 5.812479686737061,
      "epoch": 0.736930413281526,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004952622773110989,
      "loss": 5.52,
      "mean_token_accuracy": 0.19470138996839523,
      "num_tokens": 15585825.0,
      "step": 8345
    },
    {
      "entropy": 5.863739252090454,
      "epoch": 0.7373719533733663,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004952558467419953,
      "loss": 5.5351,
      "mean_token_accuracy": 0.19085845798254014,
      "num_tokens": 15595486.0,
      "step": 8350
    },
    {
      "entropy": 5.866566705703735,
      "epoch": 0.7378134934652066,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000495249411858205,
      "loss": 5.6003,
      "mean_token_accuracy": 0.17631158828735352,
      "num_tokens": 15605312.0,
      "step": 8355
    },
    {
      "entropy": 5.840285158157348,
      "epoch": 0.738255033557047,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004952429726598538,
      "loss": 5.5771,
      "mean_token_accuracy": 0.19373610317707063,
      "num_tokens": 15615185.0,
      "step": 8360
    },
    {
      "entropy": 5.883241939544678,
      "epoch": 0.7386965736488873,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004952365291470682,
      "loss": 5.6903,
      "mean_token_accuracy": 0.18867729008197784,
      "num_tokens": 15625259.0,
      "step": 8365
    },
    {
      "entropy": 5.9134824752807615,
      "epoch": 0.7391381137407277,
      "grad_norm": 0.875,
      "learning_rate": 0.0004952300813199742,
      "loss": 5.6882,
      "mean_token_accuracy": 0.1814059793949127,
      "num_tokens": 15635040.0,
      "step": 8370
    },
    {
      "entropy": 5.942975568771362,
      "epoch": 0.739579653832568,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004952236291786981,
      "loss": 5.6992,
      "mean_token_accuracy": 0.18363112807273865,
      "num_tokens": 15644996.0,
      "step": 8375
    },
    {
      "entropy": 5.800611877441407,
      "epoch": 0.7400211939244083,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004952171727233665,
      "loss": 5.5721,
      "mean_token_accuracy": 0.1871659651398659,
      "num_tokens": 15654598.0,
      "step": 8380
    },
    {
      "entropy": 5.870110893249512,
      "epoch": 0.7404627340162486,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004952107119541055,
      "loss": 5.5809,
      "mean_token_accuracy": 0.1836706057190895,
      "num_tokens": 15664213.0,
      "step": 8385
    },
    {
      "entropy": 5.889188480377197,
      "epoch": 0.740904274108089,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004952042468710421,
      "loss": 5.6605,
      "mean_token_accuracy": 0.1854677751660347,
      "num_tokens": 15673595.0,
      "step": 8390
    },
    {
      "entropy": 5.942821264266968,
      "epoch": 0.7413458141999294,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004951977774743027,
      "loss": 5.7411,
      "mean_token_accuracy": 0.17860967963933944,
      "num_tokens": 15682753.0,
      "step": 8395
    },
    {
      "entropy": 5.832130432128906,
      "epoch": 0.7417873542917697,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004951913037640139,
      "loss": 5.5636,
      "mean_token_accuracy": 0.1966960161924362,
      "num_tokens": 15691508.0,
      "step": 8400
    },
    {
      "entropy": 5.7541731834411625,
      "epoch": 0.7422288943836101,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004951848257403029,
      "loss": 5.531,
      "mean_token_accuracy": 0.18669057488441468,
      "num_tokens": 15701110.0,
      "step": 8405
    },
    {
      "entropy": 5.792100095748902,
      "epoch": 0.7426704344754503,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004951783434032963,
      "loss": 5.4955,
      "mean_token_accuracy": 0.1957699477672577,
      "num_tokens": 15710565.0,
      "step": 8410
    },
    {
      "entropy": 5.883226823806763,
      "epoch": 0.7431119745672907,
      "grad_norm": 1.125,
      "learning_rate": 0.0004951718567531211,
      "loss": 5.6617,
      "mean_token_accuracy": 0.1794003278017044,
      "num_tokens": 15718364.0,
      "step": 8415
    },
    {
      "entropy": 5.934400129318237,
      "epoch": 0.743553514659131,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004951653657899045,
      "loss": 5.7172,
      "mean_token_accuracy": 0.17798707634210587,
      "num_tokens": 15729127.0,
      "step": 8420
    },
    {
      "entropy": 5.944851684570312,
      "epoch": 0.7439950547509714,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004951588705137737,
      "loss": 5.6884,
      "mean_token_accuracy": 0.17930149734020234,
      "num_tokens": 15738491.0,
      "step": 8425
    },
    {
      "entropy": 5.883695507049561,
      "epoch": 0.7444365948428118,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004951523709248557,
      "loss": 5.5617,
      "mean_token_accuracy": 0.19158599972724916,
      "num_tokens": 15748346.0,
      "step": 8430
    },
    {
      "entropy": 5.836772108078003,
      "epoch": 0.744878134934652,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000495145867023278,
      "loss": 5.5833,
      "mean_token_accuracy": 0.18637182414531708,
      "num_tokens": 15757242.0,
      "step": 8435
    },
    {
      "entropy": 5.957921314239502,
      "epoch": 0.7453196750264924,
      "grad_norm": 0.83203125,
      "learning_rate": 0.000495139358809168,
      "loss": 5.6306,
      "mean_token_accuracy": 0.18897896707057954,
      "num_tokens": 15767236.0,
      "step": 8440
    },
    {
      "entropy": 5.909159517288208,
      "epoch": 0.7457612151183327,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004951328462826532,
      "loss": 5.7068,
      "mean_token_accuracy": 0.17895558327436448,
      "num_tokens": 15776977.0,
      "step": 8445
    },
    {
      "entropy": 5.813918399810791,
      "epoch": 0.7462027552101731,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004951263294438611,
      "loss": 5.5944,
      "mean_token_accuracy": 0.17929491102695466,
      "num_tokens": 15787333.0,
      "step": 8450
    },
    {
      "entropy": 5.803722190856933,
      "epoch": 0.7466442953020134,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004951198082929194,
      "loss": 5.4828,
      "mean_token_accuracy": 0.2009884908795357,
      "num_tokens": 15794773.0,
      "step": 8455
    },
    {
      "entropy": 5.792421913146972,
      "epoch": 0.7470858353938538,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004951132828299558,
      "loss": 5.5766,
      "mean_token_accuracy": 0.19700139313936232,
      "num_tokens": 15803803.0,
      "step": 8460
    },
    {
      "entropy": 5.856327486038208,
      "epoch": 0.7475273754856941,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004951067530550982,
      "loss": 5.6327,
      "mean_token_accuracy": 0.18049795925617218,
      "num_tokens": 15814021.0,
      "step": 8465
    },
    {
      "entropy": 5.89320011138916,
      "epoch": 0.7479689155775344,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004951002189684745,
      "loss": 5.6179,
      "mean_token_accuracy": 0.17834028005599975,
      "num_tokens": 15822974.0,
      "step": 8470
    },
    {
      "entropy": 5.7720318794250485,
      "epoch": 0.7484104556693748,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004950936805702127,
      "loss": 5.4998,
      "mean_token_accuracy": 0.19415683150291443,
      "num_tokens": 15832202.0,
      "step": 8475
    },
    {
      "entropy": 5.834465599060058,
      "epoch": 0.7488519957612151,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004950871378604409,
      "loss": 5.6309,
      "mean_token_accuracy": 0.19115645736455916,
      "num_tokens": 15841223.0,
      "step": 8480
    },
    {
      "entropy": 5.873974132537842,
      "epoch": 0.7492935358530555,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004950805908392872,
      "loss": 5.5861,
      "mean_token_accuracy": 0.1882975786924362,
      "num_tokens": 15850260.0,
      "step": 8485
    },
    {
      "entropy": 5.867472314834595,
      "epoch": 0.7497350759448957,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004950740395068799,
      "loss": 5.6677,
      "mean_token_accuracy": 0.1869749143719673,
      "num_tokens": 15859780.0,
      "step": 8490
    },
    {
      "entropy": 5.826950740814209,
      "epoch": 0.7501766160367361,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004950674838633473,
      "loss": 5.5941,
      "mean_token_accuracy": 0.18690303415060044,
      "num_tokens": 15868331.0,
      "step": 8495
    },
    {
      "entropy": 5.867539358139038,
      "epoch": 0.7506181561285765,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004950609239088178,
      "loss": 5.6468,
      "mean_token_accuracy": 0.17849708199501038,
      "num_tokens": 15877359.0,
      "step": 8500
    },
    {
      "entropy": 5.786456060409546,
      "epoch": 0.7510596962204168,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004950543596434201,
      "loss": 5.555,
      "mean_token_accuracy": 0.19861937165260315,
      "num_tokens": 15886976.0,
      "step": 8505
    },
    {
      "entropy": 5.780098247528076,
      "epoch": 0.7515012363122572,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004950477910672825,
      "loss": 5.6306,
      "mean_token_accuracy": 0.18838491588830947,
      "num_tokens": 15896139.0,
      "step": 8510
    },
    {
      "entropy": 5.891390085220337,
      "epoch": 0.7519427764040975,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000495041218180534,
      "loss": 5.6435,
      "mean_token_accuracy": 0.18974340856075286,
      "num_tokens": 15904672.0,
      "step": 8515
    },
    {
      "entropy": 5.787608051300049,
      "epoch": 0.7523843164959378,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000495034640983303,
      "loss": 5.478,
      "mean_token_accuracy": 0.19494728446006776,
      "num_tokens": 15912716.0,
      "step": 8520
    },
    {
      "entropy": 5.833557415008545,
      "epoch": 0.7528258565877781,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004950280594757186,
      "loss": 5.5967,
      "mean_token_accuracy": 0.18717412501573563,
      "num_tokens": 15921974.0,
      "step": 8525
    },
    {
      "entropy": 5.920584440231323,
      "epoch": 0.7532673966796185,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004950214736579097,
      "loss": 5.6112,
      "mean_token_accuracy": 0.18524771481752395,
      "num_tokens": 15930859.0,
      "step": 8530
    },
    {
      "entropy": 5.784576272964477,
      "epoch": 0.7537089367714589,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004950148835300051,
      "loss": 5.5324,
      "mean_token_accuracy": 0.1943311259150505,
      "num_tokens": 15940052.0,
      "step": 8535
    },
    {
      "entropy": 5.823805046081543,
      "epoch": 0.7541504768632992,
      "grad_norm": 1.0,
      "learning_rate": 0.0004950082890921341,
      "loss": 5.6219,
      "mean_token_accuracy": 0.18529213070869446,
      "num_tokens": 15948422.0,
      "step": 8540
    },
    {
      "entropy": 5.894643259048462,
      "epoch": 0.7545920169551396,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004950016903444261,
      "loss": 5.6906,
      "mean_token_accuracy": 0.17810325771570207,
      "num_tokens": 15957433.0,
      "step": 8545
    },
    {
      "entropy": 5.915259504318238,
      "epoch": 0.7550335570469798,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004949950872870099,
      "loss": 5.6259,
      "mean_token_accuracy": 0.18364077657461167,
      "num_tokens": 15966542.0,
      "step": 8550
    },
    {
      "entropy": 5.895453834533692,
      "epoch": 0.7554750971388202,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000494988479920015,
      "loss": 5.7576,
      "mean_token_accuracy": 0.17443610578775406,
      "num_tokens": 15977067.0,
      "step": 8555
    },
    {
      "entropy": 5.903998708724975,
      "epoch": 0.7559166372306605,
      "grad_norm": 0.875,
      "learning_rate": 0.000494981868243571,
      "loss": 5.6459,
      "mean_token_accuracy": 0.18578411638736725,
      "num_tokens": 15986502.0,
      "step": 8560
    },
    {
      "entropy": 5.942930603027344,
      "epoch": 0.7563581773225009,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004949752522578073,
      "loss": 5.7161,
      "mean_token_accuracy": 0.18248875737190245,
      "num_tokens": 15996788.0,
      "step": 8565
    },
    {
      "entropy": 5.891510248184204,
      "epoch": 0.7567997174143413,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004949686319628535,
      "loss": 5.5889,
      "mean_token_accuracy": 0.18840514570474626,
      "num_tokens": 16007212.0,
      "step": 8570
    },
    {
      "entropy": 5.804558944702149,
      "epoch": 0.7572412575061815,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004949620073588394,
      "loss": 5.6435,
      "mean_token_accuracy": 0.18354785442352295,
      "num_tokens": 16017000.0,
      "step": 8575
    },
    {
      "entropy": 5.941242122650147,
      "epoch": 0.7576827975980219,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004949553784458945,
      "loss": 5.7309,
      "mean_token_accuracy": 0.17222158014774322,
      "num_tokens": 16026647.0,
      "step": 8580
    },
    {
      "entropy": 5.841959381103516,
      "epoch": 0.7581243376898622,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004949487452241489,
      "loss": 5.4496,
      "mean_token_accuracy": 0.2036088615655899,
      "num_tokens": 16034773.0,
      "step": 8585
    },
    {
      "entropy": 5.879746627807617,
      "epoch": 0.7585658777817026,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004949421076937325,
      "loss": 5.6027,
      "mean_token_accuracy": 0.18925876021385193,
      "num_tokens": 16043787.0,
      "step": 8590
    },
    {
      "entropy": 5.76485013961792,
      "epoch": 0.7590074178735429,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004949354658547753,
      "loss": 5.5737,
      "mean_token_accuracy": 0.19907401651144027,
      "num_tokens": 16053130.0,
      "step": 8595
    },
    {
      "entropy": 5.870529079437256,
      "epoch": 0.7594489579653833,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004949288197074074,
      "loss": 5.7198,
      "mean_token_accuracy": 0.17960111498832704,
      "num_tokens": 16062529.0,
      "step": 8600
    },
    {
      "entropy": 5.8299237251281735,
      "epoch": 0.7598904980572236,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004949221692517589,
      "loss": 5.5684,
      "mean_token_accuracy": 0.18861594051122665,
      "num_tokens": 16071932.0,
      "step": 8605
    },
    {
      "entropy": 5.880710124969482,
      "epoch": 0.7603320381490639,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004949155144879602,
      "loss": 5.5314,
      "mean_token_accuracy": 0.19198206812143326,
      "num_tokens": 16081077.0,
      "step": 8610
    },
    {
      "entropy": 5.786740303039551,
      "epoch": 0.7607735782409043,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004949088554161417,
      "loss": 5.4939,
      "mean_token_accuracy": 0.19128501862287522,
      "num_tokens": 16089725.0,
      "step": 8615
    },
    {
      "entropy": 5.847886180877685,
      "epoch": 0.7612151183327446,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004949021920364339,
      "loss": 5.6605,
      "mean_token_accuracy": 0.18638603687286376,
      "num_tokens": 16099362.0,
      "step": 8620
    },
    {
      "entropy": 5.94460711479187,
      "epoch": 0.761656658424585,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004948955243489671,
      "loss": 5.675,
      "mean_token_accuracy": 0.18102553784847258,
      "num_tokens": 16108746.0,
      "step": 8625
    },
    {
      "entropy": 5.966106033325195,
      "epoch": 0.7620981985164252,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004948888523538721,
      "loss": 5.8383,
      "mean_token_accuracy": 0.1732643410563469,
      "num_tokens": 16118902.0,
      "step": 8630
    },
    {
      "entropy": 5.881408309936523,
      "epoch": 0.7625397386082656,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004948821760512795,
      "loss": 5.5767,
      "mean_token_accuracy": 0.1854456201195717,
      "num_tokens": 16127781.0,
      "step": 8635
    },
    {
      "entropy": 5.865966653823852,
      "epoch": 0.762981278700106,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004948754954413201,
      "loss": 5.5434,
      "mean_token_accuracy": 0.20540579557418823,
      "num_tokens": 16136004.0,
      "step": 8640
    },
    {
      "entropy": 5.809378910064697,
      "epoch": 0.7634228187919463,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004948688105241248,
      "loss": 5.5865,
      "mean_token_accuracy": 0.1957641437649727,
      "num_tokens": 16145023.0,
      "step": 8645
    },
    {
      "entropy": 5.872867155075073,
      "epoch": 0.7638643588837867,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004948621212998246,
      "loss": 5.6456,
      "mean_token_accuracy": 0.18539468497037886,
      "num_tokens": 16154242.0,
      "step": 8650
    },
    {
      "entropy": 5.940119457244873,
      "epoch": 0.764305898975627,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004948554277685505,
      "loss": 5.619,
      "mean_token_accuracy": 0.18480096608400345,
      "num_tokens": 16163252.0,
      "step": 8655
    },
    {
      "entropy": 5.909817171096802,
      "epoch": 0.7647474390674673,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004948487299304335,
      "loss": 5.552,
      "mean_token_accuracy": 0.1940522775053978,
      "num_tokens": 16171363.0,
      "step": 8660
    },
    {
      "entropy": 5.8778034210205075,
      "epoch": 0.7651889791593076,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000494842027785605,
      "loss": 5.7388,
      "mean_token_accuracy": 0.18207193166017532,
      "num_tokens": 16181234.0,
      "step": 8665
    },
    {
      "entropy": 5.895607423782349,
      "epoch": 0.765630519251148,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004948353213341961,
      "loss": 5.6517,
      "mean_token_accuracy": 0.1834381639957428,
      "num_tokens": 16189420.0,
      "step": 8670
    },
    {
      "entropy": 5.9458598613739015,
      "epoch": 0.7660720593429884,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004948286105763385,
      "loss": 5.7867,
      "mean_token_accuracy": 0.17419244572520257,
      "num_tokens": 16199202.0,
      "step": 8675
    },
    {
      "entropy": 5.9134142875671385,
      "epoch": 0.7665135994348287,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004948218955121632,
      "loss": 5.6212,
      "mean_token_accuracy": 0.1843989074230194,
      "num_tokens": 16208481.0,
      "step": 8680
    },
    {
      "entropy": 5.904475688934326,
      "epoch": 0.766955139526669,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004948151761418021,
      "loss": 5.7078,
      "mean_token_accuracy": 0.18244910091161728,
      "num_tokens": 16217282.0,
      "step": 8685
    },
    {
      "entropy": 5.880133628845215,
      "epoch": 0.7673966796185093,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004948084524653868,
      "loss": 5.5712,
      "mean_token_accuracy": 0.1891437828540802,
      "num_tokens": 16226567.0,
      "step": 8690
    },
    {
      "entropy": 5.866838026046753,
      "epoch": 0.7678382197103497,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004948017244830489,
      "loss": 5.7673,
      "mean_token_accuracy": 0.17605815678834916,
      "num_tokens": 16236255.0,
      "step": 8695
    },
    {
      "entropy": 5.869735670089722,
      "epoch": 0.76827975980219,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004947949921949201,
      "loss": 5.5621,
      "mean_token_accuracy": 0.19431234449148177,
      "num_tokens": 16244838.0,
      "step": 8700
    },
    {
      "entropy": 5.7782117366790775,
      "epoch": 0.7687212998940304,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004947882556011325,
      "loss": 5.4373,
      "mean_token_accuracy": 0.20313273072242738,
      "num_tokens": 16254101.0,
      "step": 8705
    },
    {
      "entropy": 5.712455606460571,
      "epoch": 0.7691628399858708,
      "grad_norm": 0.859375,
      "learning_rate": 0.000494781514701818,
      "loss": 5.5341,
      "mean_token_accuracy": 0.1988469660282135,
      "num_tokens": 16263496.0,
      "step": 8710
    },
    {
      "entropy": 5.932302951812744,
      "epoch": 0.769604380077711,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004947747694971085,
      "loss": 5.6368,
      "mean_token_accuracy": 0.19015247076749803,
      "num_tokens": 16271884.0,
      "step": 8715
    },
    {
      "entropy": 5.857388734817505,
      "epoch": 0.7700459201695514,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004947680199871363,
      "loss": 5.5507,
      "mean_token_accuracy": 0.19477736204862595,
      "num_tokens": 16281197.0,
      "step": 8720
    },
    {
      "entropy": 5.808503437042236,
      "epoch": 0.7704874602613917,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004947612661720336,
      "loss": 5.646,
      "mean_token_accuracy": 0.18876955658197403,
      "num_tokens": 16289908.0,
      "step": 8725
    },
    {
      "entropy": 5.825726795196533,
      "epoch": 0.7709290003532321,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004947545080519327,
      "loss": 5.599,
      "mean_token_accuracy": 0.18349761068820952,
      "num_tokens": 16299237.0,
      "step": 8730
    },
    {
      "entropy": 5.9406637191772464,
      "epoch": 0.7713705404450724,
      "grad_norm": 1.046875,
      "learning_rate": 0.000494747745626966,
      "loss": 5.6251,
      "mean_token_accuracy": 0.18967668414115907,
      "num_tokens": 16308175.0,
      "step": 8735
    },
    {
      "entropy": 5.865895175933838,
      "epoch": 0.7718120805369127,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004947409788972659,
      "loss": 5.5776,
      "mean_token_accuracy": 0.19288718998432158,
      "num_tokens": 16317121.0,
      "step": 8740
    },
    {
      "entropy": 5.81405348777771,
      "epoch": 0.7722536206287531,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004947342078629649,
      "loss": 5.5723,
      "mean_token_accuracy": 0.18960059136152269,
      "num_tokens": 16325764.0,
      "step": 8745
    },
    {
      "entropy": 5.863961315155029,
      "epoch": 0.7726951607205934,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004947274325241959,
      "loss": 5.5695,
      "mean_token_accuracy": 0.1862097531557083,
      "num_tokens": 16333823.0,
      "step": 8750
    },
    {
      "entropy": 5.800171661376953,
      "epoch": 0.7731367008124338,
      "grad_norm": 1.0,
      "learning_rate": 0.0004947206528810913,
      "loss": 5.5642,
      "mean_token_accuracy": 0.1920272096991539,
      "num_tokens": 16341918.0,
      "step": 8755
    },
    {
      "entropy": 5.817489528656006,
      "epoch": 0.7735782409042741,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004947138689337843,
      "loss": 5.6122,
      "mean_token_accuracy": 0.18937198519706727,
      "num_tokens": 16351904.0,
      "step": 8760
    },
    {
      "entropy": 5.950567388534546,
      "epoch": 0.7740197809961145,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004947070806824074,
      "loss": 5.62,
      "mean_token_accuracy": 0.18496680557727813,
      "num_tokens": 16360921.0,
      "step": 8765
    },
    {
      "entropy": 5.908766412734986,
      "epoch": 0.7744613210879547,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004947002881270937,
      "loss": 5.6423,
      "mean_token_accuracy": 0.1874388188123703,
      "num_tokens": 16370929.0,
      "step": 8770
    },
    {
      "entropy": 5.890651512145996,
      "epoch": 0.7749028611797951,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004946934912679764,
      "loss": 5.6359,
      "mean_token_accuracy": 0.18902267664670944,
      "num_tokens": 16380099.0,
      "step": 8775
    },
    {
      "entropy": 5.898251152038574,
      "epoch": 0.7753444012716355,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004946866901051885,
      "loss": 5.6891,
      "mean_token_accuracy": 0.18321269005537033,
      "num_tokens": 16389109.0,
      "step": 8780
    },
    {
      "entropy": 5.88666090965271,
      "epoch": 0.7757859413634758,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004946798846388634,
      "loss": 5.7083,
      "mean_token_accuracy": 0.17529224157333373,
      "num_tokens": 16398601.0,
      "step": 8785
    },
    {
      "entropy": 5.8759900569915775,
      "epoch": 0.7762274814553162,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004946730748691342,
      "loss": 5.5623,
      "mean_token_accuracy": 0.19695408195257186,
      "num_tokens": 16406950.0,
      "step": 8790
    },
    {
      "entropy": 5.928198385238647,
      "epoch": 0.7766690215471564,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004946662607961344,
      "loss": 5.5653,
      "mean_token_accuracy": 0.1901753857731819,
      "num_tokens": 16414680.0,
      "step": 8795
    },
    {
      "entropy": 5.811094951629639,
      "epoch": 0.7771105616389968,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004946594424199976,
      "loss": 5.63,
      "mean_token_accuracy": 0.18612197637557984,
      "num_tokens": 16424060.0,
      "step": 8800
    },
    {
      "entropy": 5.8213338375091555,
      "epoch": 0.7775521017308371,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004946526197408571,
      "loss": 5.6604,
      "mean_token_accuracy": 0.18676266372203826,
      "num_tokens": 16433691.0,
      "step": 8805
    },
    {
      "entropy": 5.919007301330566,
      "epoch": 0.7779936418226775,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004946457927588467,
      "loss": 5.6535,
      "mean_token_accuracy": 0.18097317218780518,
      "num_tokens": 16443234.0,
      "step": 8810
    },
    {
      "entropy": 5.85936803817749,
      "epoch": 0.7784351819145179,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004946389614741002,
      "loss": 5.6024,
      "mean_token_accuracy": 0.19192105978727342,
      "num_tokens": 16452084.0,
      "step": 8815
    },
    {
      "entropy": 5.853142642974854,
      "epoch": 0.7788767220063582,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004946321258867513,
      "loss": 5.5928,
      "mean_token_accuracy": 0.1917520835995674,
      "num_tokens": 16460135.0,
      "step": 8820
    },
    {
      "entropy": 5.902470922470092,
      "epoch": 0.7793182620981985,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000494625285996934,
      "loss": 5.7215,
      "mean_token_accuracy": 0.18264151215553284,
      "num_tokens": 16469919.0,
      "step": 8825
    },
    {
      "entropy": 5.948286771774292,
      "epoch": 0.7797598021900388,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004946184418047823,
      "loss": 5.6396,
      "mean_token_accuracy": 0.1841666057705879,
      "num_tokens": 16478473.0,
      "step": 8830
    },
    {
      "entropy": 5.9353759765625,
      "epoch": 0.7802013422818792,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004946115933104301,
      "loss": 5.6427,
      "mean_token_accuracy": 0.17996163070201873,
      "num_tokens": 16486878.0,
      "step": 8835
    },
    {
      "entropy": 5.856232929229736,
      "epoch": 0.7806428823737195,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004946047405140119,
      "loss": 5.6813,
      "mean_token_accuracy": 0.1801102951169014,
      "num_tokens": 16496798.0,
      "step": 8840
    },
    {
      "entropy": 5.931548357009888,
      "epoch": 0.7810844224655599,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004945978834156616,
      "loss": 5.5854,
      "mean_token_accuracy": 0.19176071733236313,
      "num_tokens": 16506508.0,
      "step": 8845
    },
    {
      "entropy": 5.789986038208008,
      "epoch": 0.7815259625574003,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004945910220155136,
      "loss": 5.5651,
      "mean_token_accuracy": 0.19404177665710448,
      "num_tokens": 16514979.0,
      "step": 8850
    },
    {
      "entropy": 5.872536373138428,
      "epoch": 0.7819675026492405,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004945841563137025,
      "loss": 5.6417,
      "mean_token_accuracy": 0.1903441235423088,
      "num_tokens": 16524654.0,
      "step": 8855
    },
    {
      "entropy": 5.909209680557251,
      "epoch": 0.7824090427410809,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004945772863103626,
      "loss": 5.7241,
      "mean_token_accuracy": 0.18177213817834853,
      "num_tokens": 16534486.0,
      "step": 8860
    },
    {
      "entropy": 5.937249708175659,
      "epoch": 0.7828505828329212,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004945704120056286,
      "loss": 5.6423,
      "mean_token_accuracy": 0.19361244142055511,
      "num_tokens": 16543718.0,
      "step": 8865
    },
    {
      "entropy": 5.8497367858886715,
      "epoch": 0.7832921229247616,
      "grad_norm": 1.0,
      "learning_rate": 0.000494563533399635,
      "loss": 5.6089,
      "mean_token_accuracy": 0.18692233711481093,
      "num_tokens": 16553159.0,
      "step": 8870
    },
    {
      "entropy": 5.948692417144775,
      "epoch": 0.7837336630166019,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004945566504925167,
      "loss": 5.6642,
      "mean_token_accuracy": 0.1829291120171547,
      "num_tokens": 16562398.0,
      "step": 8875
    },
    {
      "entropy": 5.894802904129028,
      "epoch": 0.7841752031084422,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004945497632844084,
      "loss": 5.6048,
      "mean_token_accuracy": 0.19393257647752762,
      "num_tokens": 16571530.0,
      "step": 8880
    },
    {
      "entropy": 5.789991092681885,
      "epoch": 0.7846167432002826,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004945428717754453,
      "loss": 5.6248,
      "mean_token_accuracy": 0.1875714972615242,
      "num_tokens": 16581074.0,
      "step": 8885
    },
    {
      "entropy": 5.832242250442505,
      "epoch": 0.7850582832921229,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000494535975965762,
      "loss": 5.5046,
      "mean_token_accuracy": 0.1969103291630745,
      "num_tokens": 16589574.0,
      "step": 8890
    },
    {
      "entropy": 5.781494998931885,
      "epoch": 0.7854998233839633,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004945290758554939,
      "loss": 5.5657,
      "mean_token_accuracy": 0.19426276236772538,
      "num_tokens": 16599623.0,
      "step": 8895
    },
    {
      "entropy": 5.818102407455444,
      "epoch": 0.7859413634758036,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000494522171444776,
      "loss": 5.6364,
      "mean_token_accuracy": 0.17526367157697678,
      "num_tokens": 16608559.0,
      "step": 8900
    },
    {
      "entropy": 5.861512613296509,
      "epoch": 0.786382903567644,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004945152627337437,
      "loss": 5.5512,
      "mean_token_accuracy": 0.19917489737272262,
      "num_tokens": 16617148.0,
      "step": 8905
    },
    {
      "entropy": 5.7910514831542965,
      "epoch": 0.7868244436594842,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004945083497225321,
      "loss": 5.5606,
      "mean_token_accuracy": 0.1904294088482857,
      "num_tokens": 16626405.0,
      "step": 8910
    },
    {
      "entropy": 5.840649938583374,
      "epoch": 0.7872659837513246,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004945014324112769,
      "loss": 5.6101,
      "mean_token_accuracy": 0.1881376251578331,
      "num_tokens": 16635116.0,
      "step": 8915
    },
    {
      "entropy": 5.78248348236084,
      "epoch": 0.787707523843165,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004944945108001134,
      "loss": 5.5713,
      "mean_token_accuracy": 0.18788688331842424,
      "num_tokens": 16644381.0,
      "step": 8920
    },
    {
      "entropy": 5.86023359298706,
      "epoch": 0.7881490639350053,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004944875848891773,
      "loss": 5.6207,
      "mean_token_accuracy": 0.19069672822952272,
      "num_tokens": 16654275.0,
      "step": 8925
    },
    {
      "entropy": 5.885168266296387,
      "epoch": 0.7885906040268457,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004944806546786042,
      "loss": 5.7006,
      "mean_token_accuracy": 0.17883567065000533,
      "num_tokens": 16663484.0,
      "step": 8930
    },
    {
      "entropy": 5.799995088577271,
      "epoch": 0.7890321441186859,
      "grad_norm": 0.9375,
      "learning_rate": 0.00049447372016853,
      "loss": 5.64,
      "mean_token_accuracy": 0.18918595016002654,
      "num_tokens": 16672490.0,
      "step": 8935
    },
    {
      "entropy": 5.899666690826416,
      "epoch": 0.7894736842105263,
      "grad_norm": 1.0,
      "learning_rate": 0.0004944667813590904,
      "loss": 5.59,
      "mean_token_accuracy": 0.18535562306642533,
      "num_tokens": 16681242.0,
      "step": 8940
    },
    {
      "entropy": 5.812223339080811,
      "epoch": 0.7899152243023666,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004944598382504214,
      "loss": 5.5646,
      "mean_token_accuracy": 0.1899319499731064,
      "num_tokens": 16690359.0,
      "step": 8945
    },
    {
      "entropy": 5.85008430480957,
      "epoch": 0.790356764394207,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004944528908426589,
      "loss": 5.637,
      "mean_token_accuracy": 0.1899271935224533,
      "num_tokens": 16700774.0,
      "step": 8950
    },
    {
      "entropy": 5.957952642440796,
      "epoch": 0.7907983044860474,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004944459391359391,
      "loss": 5.7111,
      "mean_token_accuracy": 0.1864597260951996,
      "num_tokens": 16710513.0,
      "step": 8955
    },
    {
      "entropy": 5.826300096511841,
      "epoch": 0.7912398445778877,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004944389831303982,
      "loss": 5.6322,
      "mean_token_accuracy": 0.18941447585821153,
      "num_tokens": 16720785.0,
      "step": 8960
    },
    {
      "entropy": 5.864205551147461,
      "epoch": 0.791681384669728,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004944320228261723,
      "loss": 5.6036,
      "mean_token_accuracy": 0.18326730877161027,
      "num_tokens": 16729083.0,
      "step": 8965
    },
    {
      "entropy": 5.831407117843628,
      "epoch": 0.7921229247615683,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000494425058223398,
      "loss": 5.6218,
      "mean_token_accuracy": 0.1872497618198395,
      "num_tokens": 16738313.0,
      "step": 8970
    },
    {
      "entropy": 5.846746397018433,
      "epoch": 0.7925644648534087,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004944180893222117,
      "loss": 5.5538,
      "mean_token_accuracy": 0.19500797092914582,
      "num_tokens": 16747191.0,
      "step": 8975
    },
    {
      "entropy": 5.7963032722473145,
      "epoch": 0.793006004945249,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004944111161227496,
      "loss": 5.5776,
      "mean_token_accuracy": 0.18987296223640443,
      "num_tokens": 16755921.0,
      "step": 8980
    },
    {
      "entropy": 5.910832405090332,
      "epoch": 0.7934475450370894,
      "grad_norm": 1.0,
      "learning_rate": 0.0004944041386251486,
      "loss": 5.7473,
      "mean_token_accuracy": 0.18214513510465621,
      "num_tokens": 16765394.0,
      "step": 8985
    },
    {
      "entropy": 5.920050525665284,
      "epoch": 0.7938890851289298,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004943971568295454,
      "loss": 5.6229,
      "mean_token_accuracy": 0.19057771414518357,
      "num_tokens": 16774824.0,
      "step": 8990
    },
    {
      "entropy": 5.925590467453003,
      "epoch": 0.79433062522077,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004943901707360765,
      "loss": 5.7477,
      "mean_token_accuracy": 0.17744729071855544,
      "num_tokens": 16783945.0,
      "step": 8995
    },
    {
      "entropy": 5.856736326217652,
      "epoch": 0.7947721653126104,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000494383180344879,
      "loss": 5.5277,
      "mean_token_accuracy": 0.19507913142442704,
      "num_tokens": 16792618.0,
      "step": 9000
    },
    {
      "epoch": 0.7947721653126104,
      "eval_entropy": 5.591507935388982,
      "eval_loss": 5.619070529937744,
      "eval_mean_token_accuracy": 0.19495929661475855,
      "eval_num_tokens": 16792618.0,
      "eval_runtime": 20.2421,
      "eval_samples_per_second": 1744.631,
      "eval_steps_per_second": 218.11,
      "step": 9000
    },
    {
      "entropy": 5.825322103500366,
      "epoch": 0.7952137054044507,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004943761856560898,
      "loss": 5.5992,
      "mean_token_accuracy": 0.18942335098981858,
      "num_tokens": 16801210.0,
      "step": 9005
    },
    {
      "entropy": 5.867608642578125,
      "epoch": 0.7956552454962911,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000494369186669846,
      "loss": 5.6397,
      "mean_token_accuracy": 0.18785736858844757,
      "num_tokens": 16810371.0,
      "step": 9010
    },
    {
      "entropy": 5.823724222183228,
      "epoch": 0.7960967855881314,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004943621833862844,
      "loss": 5.545,
      "mean_token_accuracy": 0.19277409613132476,
      "num_tokens": 16818886.0,
      "step": 9015
    },
    {
      "entropy": 5.842554903030395,
      "epoch": 0.7965383256799717,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004943551758055425,
      "loss": 5.6224,
      "mean_token_accuracy": 0.18964377343654631,
      "num_tokens": 16827705.0,
      "step": 9020
    },
    {
      "entropy": 5.8641541481018065,
      "epoch": 0.7969798657718121,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004943481639277574,
      "loss": 5.582,
      "mean_token_accuracy": 0.18107532560825348,
      "num_tokens": 16836313.0,
      "step": 9025
    },
    {
      "entropy": 5.9201160907745365,
      "epoch": 0.7974214058636524,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004943411477530665,
      "loss": 5.6618,
      "mean_token_accuracy": 0.1813226357102394,
      "num_tokens": 16845016.0,
      "step": 9030
    },
    {
      "entropy": 5.87960410118103,
      "epoch": 0.7978629459554928,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004943341272816073,
      "loss": 5.627,
      "mean_token_accuracy": 0.1824415385723114,
      "num_tokens": 16854025.0,
      "step": 9035
    },
    {
      "entropy": 5.828975486755371,
      "epoch": 0.7983044860473331,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004943271025135173,
      "loss": 5.5683,
      "mean_token_accuracy": 0.18993970900774002,
      "num_tokens": 16862859.0,
      "step": 9040
    },
    {
      "entropy": 5.855637502670288,
      "epoch": 0.7987460261391734,
      "grad_norm": 0.875,
      "learning_rate": 0.0004943200734489341,
      "loss": 5.6629,
      "mean_token_accuracy": 0.18977779150009155,
      "num_tokens": 16872226.0,
      "step": 9045
    },
    {
      "entropy": 5.803923225402832,
      "epoch": 0.7991875662310137,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004943130400879953,
      "loss": 5.5409,
      "mean_token_accuracy": 0.1905651032924652,
      "num_tokens": 16881977.0,
      "step": 9050
    },
    {
      "entropy": 5.857625722885132,
      "epoch": 0.7996291063228541,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004943060024308389,
      "loss": 5.6599,
      "mean_token_accuracy": 0.1782623678445816,
      "num_tokens": 16891656.0,
      "step": 9055
    },
    {
      "entropy": 5.903197383880615,
      "epoch": 0.8000706464146945,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004942989604776026,
      "loss": 5.5252,
      "mean_token_accuracy": 0.19363559782505035,
      "num_tokens": 16900944.0,
      "step": 9060
    },
    {
      "entropy": 5.761235761642456,
      "epoch": 0.8005121865065348,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004942919142284244,
      "loss": 5.546,
      "mean_token_accuracy": 0.18910249769687654,
      "num_tokens": 16910154.0,
      "step": 9065
    },
    {
      "entropy": 5.7982673168182375,
      "epoch": 0.8009537265983752,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004942848636834423,
      "loss": 5.6586,
      "mean_token_accuracy": 0.1833169311285019,
      "num_tokens": 16919694.0,
      "step": 9070
    },
    {
      "entropy": 5.882076978683472,
      "epoch": 0.8013952666902154,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004942778088427944,
      "loss": 5.6214,
      "mean_token_accuracy": 0.18945298343896866,
      "num_tokens": 16928530.0,
      "step": 9075
    },
    {
      "entropy": 5.927755212783813,
      "epoch": 0.8018368067820558,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000494270749706619,
      "loss": 5.6551,
      "mean_token_accuracy": 0.18146197944879533,
      "num_tokens": 16938614.0,
      "step": 9080
    },
    {
      "entropy": 5.8563508033752445,
      "epoch": 0.8022783468738961,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004942636862750544,
      "loss": 5.595,
      "mean_token_accuracy": 0.18589434772729874,
      "num_tokens": 16947655.0,
      "step": 9085
    },
    {
      "entropy": 5.8665442943573,
      "epoch": 0.8027198869657365,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004942566185482387,
      "loss": 5.6857,
      "mean_token_accuracy": 0.1815369561314583,
      "num_tokens": 16957101.0,
      "step": 9090
    },
    {
      "entropy": 5.843801975250244,
      "epoch": 0.8031614270575769,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004942495465263106,
      "loss": 5.7167,
      "mean_token_accuracy": 0.1784170463681221,
      "num_tokens": 16966421.0,
      "step": 9095
    },
    {
      "entropy": 5.925464344024658,
      "epoch": 0.8036029671494171,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004942424702094085,
      "loss": 5.6041,
      "mean_token_accuracy": 0.19640391767024995,
      "num_tokens": 16975656.0,
      "step": 9100
    },
    {
      "entropy": 5.887157869338989,
      "epoch": 0.8040445072412575,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004942353895976713,
      "loss": 5.6632,
      "mean_token_accuracy": 0.18462532758712769,
      "num_tokens": 16984409.0,
      "step": 9105
    },
    {
      "entropy": 5.816635465621948,
      "epoch": 0.8044860473330978,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004942283046912374,
      "loss": 5.5567,
      "mean_token_accuracy": 0.1862815201282501,
      "num_tokens": 16992728.0,
      "step": 9110
    },
    {
      "entropy": 5.8094785690307615,
      "epoch": 0.8049275874249382,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004942212154902456,
      "loss": 5.4597,
      "mean_token_accuracy": 0.20306771099567414,
      "num_tokens": 17001616.0,
      "step": 9115
    },
    {
      "entropy": 5.818345212936402,
      "epoch": 0.8053691275167785,
      "grad_norm": 1.0,
      "learning_rate": 0.000494214121994835,
      "loss": 5.641,
      "mean_token_accuracy": 0.18777192533016204,
      "num_tokens": 17010655.0,
      "step": 9120
    },
    {
      "entropy": 5.862553977966309,
      "epoch": 0.8058106676086189,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004942070242051444,
      "loss": 5.5987,
      "mean_token_accuracy": 0.18641175031661988,
      "num_tokens": 17019819.0,
      "step": 9125
    },
    {
      "entropy": 5.934750032424927,
      "epoch": 0.8062522077004592,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004941999221213129,
      "loss": 5.6402,
      "mean_token_accuracy": 0.18712895065546037,
      "num_tokens": 17029218.0,
      "step": 9130
    },
    {
      "entropy": 5.9235235214233395,
      "epoch": 0.8066937477922995,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004941928157434796,
      "loss": 5.5558,
      "mean_token_accuracy": 0.1911984696984291,
      "num_tokens": 17037880.0,
      "step": 9135
    },
    {
      "entropy": 5.789757156372071,
      "epoch": 0.8071352878841399,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004941857050717836,
      "loss": 5.6241,
      "mean_token_accuracy": 0.1851594939827919,
      "num_tokens": 17047934.0,
      "step": 9140
    },
    {
      "entropy": 5.932223892211914,
      "epoch": 0.8075768279759802,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004941785901063642,
      "loss": 5.691,
      "mean_token_accuracy": 0.18222017735242843,
      "num_tokens": 17057258.0,
      "step": 9145
    },
    {
      "entropy": 5.885777521133423,
      "epoch": 0.8080183680678206,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004941714708473611,
      "loss": 5.6141,
      "mean_token_accuracy": 0.18538845628499984,
      "num_tokens": 17066387.0,
      "step": 9150
    },
    {
      "entropy": 5.799456453323364,
      "epoch": 0.8084599081596608,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004941643472949134,
      "loss": 5.5675,
      "mean_token_accuracy": 0.1871365562081337,
      "num_tokens": 17075952.0,
      "step": 9155
    },
    {
      "entropy": 5.7912437915802,
      "epoch": 0.8089014482515012,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004941572194491608,
      "loss": 5.4397,
      "mean_token_accuracy": 0.20129949748516082,
      "num_tokens": 17084209.0,
      "step": 9160
    },
    {
      "entropy": 5.819268655776978,
      "epoch": 0.8093429883433416,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000494150087310243,
      "loss": 5.6811,
      "mean_token_accuracy": 0.18660429269075393,
      "num_tokens": 17093395.0,
      "step": 9165
    },
    {
      "entropy": 5.889335632324219,
      "epoch": 0.8097845284351819,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004941429508782995,
      "loss": 5.5633,
      "mean_token_accuracy": 0.19305034577846528,
      "num_tokens": 17102646.0,
      "step": 9170
    },
    {
      "entropy": 5.912970304489136,
      "epoch": 0.8102260685270223,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004941358101534703,
      "loss": 5.577,
      "mean_token_accuracy": 0.1855246216058731,
      "num_tokens": 17111530.0,
      "step": 9175
    },
    {
      "entropy": 5.789304494857788,
      "epoch": 0.8106676086188626,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004941286651358952,
      "loss": 5.5631,
      "mean_token_accuracy": 0.1946306437253952,
      "num_tokens": 17119659.0,
      "step": 9180
    },
    {
      "entropy": 5.7249321937561035,
      "epoch": 0.8111091487107029,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004941215158257142,
      "loss": 5.5494,
      "mean_token_accuracy": 0.19357234835624695,
      "num_tokens": 17127931.0,
      "step": 9185
    },
    {
      "entropy": 5.942633867263794,
      "epoch": 0.8115506888025432,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004941143622230672,
      "loss": 5.5768,
      "mean_token_accuracy": 0.19068775326013565,
      "num_tokens": 17137374.0,
      "step": 9190
    },
    {
      "entropy": 5.92955060005188,
      "epoch": 0.8119922288943836,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004941072043280946,
      "loss": 5.6504,
      "mean_token_accuracy": 0.18410056233406066,
      "num_tokens": 17147257.0,
      "step": 9195
    },
    {
      "entropy": 5.830786895751953,
      "epoch": 0.812433768986224,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004941000421409365,
      "loss": 5.5152,
      "mean_token_accuracy": 0.19687575548887254,
      "num_tokens": 17156233.0,
      "step": 9200
    },
    {
      "entropy": 5.914954757690429,
      "epoch": 0.8128753090780643,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004940928756617331,
      "loss": 5.713,
      "mean_token_accuracy": 0.1751517355442047,
      "num_tokens": 17166364.0,
      "step": 9205
    },
    {
      "entropy": 5.830341339111328,
      "epoch": 0.8133168491699047,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004940857048906248,
      "loss": 5.6207,
      "mean_token_accuracy": 0.1892386794090271,
      "num_tokens": 17175853.0,
      "step": 9210
    },
    {
      "entropy": 5.878078269958496,
      "epoch": 0.8137583892617449,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004940785298277523,
      "loss": 5.7153,
      "mean_token_accuracy": 0.18135832846164704,
      "num_tokens": 17187849.0,
      "step": 9215
    },
    {
      "entropy": 5.920613622665405,
      "epoch": 0.8141999293535853,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000494071350473256,
      "loss": 5.5822,
      "mean_token_accuracy": 0.18615618199110032,
      "num_tokens": 17196646.0,
      "step": 9220
    },
    {
      "entropy": 5.888037824630738,
      "epoch": 0.8146414694454256,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004940641668272765,
      "loss": 5.7214,
      "mean_token_accuracy": 0.18530243486166,
      "num_tokens": 17205919.0,
      "step": 9225
    },
    {
      "entropy": 5.902007102966309,
      "epoch": 0.815083009537266,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004940569788899544,
      "loss": 5.6713,
      "mean_token_accuracy": 0.18742569386959076,
      "num_tokens": 17215085.0,
      "step": 9230
    },
    {
      "entropy": 5.867893171310425,
      "epoch": 0.8155245496291064,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004940497866614309,
      "loss": 5.6533,
      "mean_token_accuracy": 0.18850154876708985,
      "num_tokens": 17223633.0,
      "step": 9235
    },
    {
      "entropy": 5.853533172607422,
      "epoch": 0.8159660897209466,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004940425901418465,
      "loss": 5.5872,
      "mean_token_accuracy": 0.19215673506259917,
      "num_tokens": 17232915.0,
      "step": 9240
    },
    {
      "entropy": 5.933935165405273,
      "epoch": 0.816407629812787,
      "grad_norm": 1.25,
      "learning_rate": 0.0004940353893313423,
      "loss": 5.676,
      "mean_token_accuracy": 0.18245417177677153,
      "num_tokens": 17242392.0,
      "step": 9245
    },
    {
      "entropy": 5.964029026031494,
      "epoch": 0.8168491699046273,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004940281842300596,
      "loss": 5.7735,
      "mean_token_accuracy": 0.17744607329368592,
      "num_tokens": 17252259.0,
      "step": 9250
    },
    {
      "entropy": 5.921383571624756,
      "epoch": 0.8172907099964677,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004940209748381392,
      "loss": 5.6371,
      "mean_token_accuracy": 0.18096522390842437,
      "num_tokens": 17261584.0,
      "step": 9255
    },
    {
      "entropy": 5.890787363052368,
      "epoch": 0.817732250088308,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004940137611557225,
      "loss": 5.5562,
      "mean_token_accuracy": 0.19099668562412261,
      "num_tokens": 17270901.0,
      "step": 9260
    },
    {
      "entropy": 5.876820230484009,
      "epoch": 0.8181737901801484,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004940065431829508,
      "loss": 5.6184,
      "mean_token_accuracy": 0.18632869124412538,
      "num_tokens": 17280024.0,
      "step": 9265
    },
    {
      "entropy": 5.889107370376587,
      "epoch": 0.8186153302719887,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004939993209199654,
      "loss": 5.6784,
      "mean_token_accuracy": 0.1789259731769562,
      "num_tokens": 17290131.0,
      "step": 9270
    },
    {
      "entropy": 5.96183500289917,
      "epoch": 0.819056870363829,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004939920943669079,
      "loss": 5.6668,
      "mean_token_accuracy": 0.18145134747028352,
      "num_tokens": 17299453.0,
      "step": 9275
    },
    {
      "entropy": 5.8522838115692135,
      "epoch": 0.8194984104556694,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004939848635239198,
      "loss": 5.6284,
      "mean_token_accuracy": 0.18608871698379517,
      "num_tokens": 17309344.0,
      "step": 9280
    },
    {
      "entropy": 5.8321044921875,
      "epoch": 0.8199399505475097,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004939776283911429,
      "loss": 5.5445,
      "mean_token_accuracy": 0.1960724487900734,
      "num_tokens": 17318636.0,
      "step": 9285
    },
    {
      "entropy": 5.826576089859008,
      "epoch": 0.8203814906393501,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004939703889687186,
      "loss": 5.6117,
      "mean_token_accuracy": 0.18429741710424424,
      "num_tokens": 17328146.0,
      "step": 9290
    },
    {
      "entropy": 5.860490322113037,
      "epoch": 0.8208230307311903,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004939631452567891,
      "loss": 5.6699,
      "mean_token_accuracy": 0.18432924002408982,
      "num_tokens": 17338542.0,
      "step": 9295
    },
    {
      "entropy": 5.937232255935669,
      "epoch": 0.8212645708230307,
      "grad_norm": 0.9375,
      "learning_rate": 0.000493955897255496,
      "loss": 5.6259,
      "mean_token_accuracy": 0.1889485612511635,
      "num_tokens": 17347577.0,
      "step": 9300
    },
    {
      "entropy": 5.814497280120849,
      "epoch": 0.8217061109148711,
      "grad_norm": 1.0,
      "learning_rate": 0.0004939486449649814,
      "loss": 5.606,
      "mean_token_accuracy": 0.1901299074292183,
      "num_tokens": 17356554.0,
      "step": 9305
    },
    {
      "entropy": 5.827168941497803,
      "epoch": 0.8221476510067114,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004939413883853873,
      "loss": 5.6382,
      "mean_token_accuracy": 0.18767885267734527,
      "num_tokens": 17366432.0,
      "step": 9310
    },
    {
      "entropy": 5.914646291732788,
      "epoch": 0.8225891910985518,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000493934127516856,
      "loss": 5.619,
      "mean_token_accuracy": 0.19058295339345932,
      "num_tokens": 17376556.0,
      "step": 9315
    },
    {
      "entropy": 5.845771455764771,
      "epoch": 0.823030731190392,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004939268623595297,
      "loss": 5.659,
      "mean_token_accuracy": 0.18527737408876419,
      "num_tokens": 17385382.0,
      "step": 9320
    },
    {
      "entropy": 5.861846494674682,
      "epoch": 0.8234722712822324,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004939195929135506,
      "loss": 5.516,
      "mean_token_accuracy": 0.1856905475258827,
      "num_tokens": 17393229.0,
      "step": 9325
    },
    {
      "entropy": 5.8488586902618405,
      "epoch": 0.8239138113740727,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004939123191790612,
      "loss": 5.5046,
      "mean_token_accuracy": 0.18935805708169937,
      "num_tokens": 17402439.0,
      "step": 9330
    },
    {
      "entropy": 5.842728233337402,
      "epoch": 0.8243553514659131,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004939050411562041,
      "loss": 5.6861,
      "mean_token_accuracy": 0.17682990729808806,
      "num_tokens": 17412349.0,
      "step": 9335
    },
    {
      "entropy": 5.7745805263519285,
      "epoch": 0.8247968915577535,
      "grad_norm": 1.0,
      "learning_rate": 0.0004938977588451216,
      "loss": 5.4547,
      "mean_token_accuracy": 0.20448464900255203,
      "num_tokens": 17420842.0,
      "step": 9340
    },
    {
      "entropy": 5.804004430770874,
      "epoch": 0.8252384316495938,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004938904722459565,
      "loss": 5.5068,
      "mean_token_accuracy": 0.20089084506034852,
      "num_tokens": 17429877.0,
      "step": 9345
    },
    {
      "entropy": 5.835537624359131,
      "epoch": 0.8256799717414341,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004938831813588515,
      "loss": 5.6559,
      "mean_token_accuracy": 0.17827803045511245,
      "num_tokens": 17439508.0,
      "step": 9350
    },
    {
      "entropy": 5.783529901504517,
      "epoch": 0.8261215118332744,
      "grad_norm": 1.0,
      "learning_rate": 0.0004938758861839495,
      "loss": 5.5674,
      "mean_token_accuracy": 0.19206740260124205,
      "num_tokens": 17449004.0,
      "step": 9355
    },
    {
      "entropy": 5.813489484786987,
      "epoch": 0.8265630519251148,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004938685867213934,
      "loss": 5.5891,
      "mean_token_accuracy": 0.18597745895385742,
      "num_tokens": 17458462.0,
      "step": 9360
    },
    {
      "entropy": 5.885543584823608,
      "epoch": 0.8270045920169551,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004938612829713262,
      "loss": 5.6062,
      "mean_token_accuracy": 0.1844778299331665,
      "num_tokens": 17466808.0,
      "step": 9365
    },
    {
      "entropy": 5.891280174255371,
      "epoch": 0.8274461321087955,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004938539749338909,
      "loss": 5.6734,
      "mean_token_accuracy": 0.18365688621997833,
      "num_tokens": 17476441.0,
      "step": 9370
    },
    {
      "entropy": 5.8662230491638185,
      "epoch": 0.8278876722006359,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004938466626092308,
      "loss": 5.5768,
      "mean_token_accuracy": 0.1952987253665924,
      "num_tokens": 17485393.0,
      "step": 9375
    },
    {
      "entropy": 5.807134485244751,
      "epoch": 0.8283292122924761,
      "grad_norm": 1.015625,
      "learning_rate": 0.000493839345997489,
      "loss": 5.6094,
      "mean_token_accuracy": 0.19124255627393721,
      "num_tokens": 17494904.0,
      "step": 9380
    },
    {
      "entropy": 5.896303987503051,
      "epoch": 0.8287707523843165,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004938320250988087,
      "loss": 5.6033,
      "mean_token_accuracy": 0.18640150576829911,
      "num_tokens": 17505137.0,
      "step": 9385
    },
    {
      "entropy": 5.881055641174316,
      "epoch": 0.8292122924761568,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004938246999133337,
      "loss": 5.6649,
      "mean_token_accuracy": 0.18294226974248887,
      "num_tokens": 17514039.0,
      "step": 9390
    },
    {
      "entropy": 5.825823831558227,
      "epoch": 0.8296538325679972,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004938173704412071,
      "loss": 5.5456,
      "mean_token_accuracy": 0.18992763757705688,
      "num_tokens": 17522423.0,
      "step": 9395
    },
    {
      "entropy": 5.826774454116821,
      "epoch": 0.8300953726598375,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004938100366825728,
      "loss": 5.5187,
      "mean_token_accuracy": 0.19853198379278184,
      "num_tokens": 17531958.0,
      "step": 9400
    },
    {
      "entropy": 5.802187108993531,
      "epoch": 0.8305369127516778,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004938026986375742,
      "loss": 5.5884,
      "mean_token_accuracy": 0.19164701849222182,
      "num_tokens": 17540905.0,
      "step": 9405
    },
    {
      "entropy": 5.846265411376953,
      "epoch": 0.8309784528435182,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004937953563063553,
      "loss": 5.6312,
      "mean_token_accuracy": 0.18226529955863952,
      "num_tokens": 17551007.0,
      "step": 9410
    },
    {
      "entropy": 5.79426941871643,
      "epoch": 0.8314199929353585,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004937880096890598,
      "loss": 5.4573,
      "mean_token_accuracy": 0.20695766806602478,
      "num_tokens": 17559403.0,
      "step": 9415
    },
    {
      "entropy": 5.8758955001831055,
      "epoch": 0.8318615330271989,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004937806587858315,
      "loss": 5.6001,
      "mean_token_accuracy": 0.18914103060960769,
      "num_tokens": 17569191.0,
      "step": 9420
    },
    {
      "entropy": 5.900097465515136,
      "epoch": 0.8323030731190392,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004937733035968147,
      "loss": 5.7011,
      "mean_token_accuracy": 0.17884953022003175,
      "num_tokens": 17578538.0,
      "step": 9425
    },
    {
      "entropy": 5.870207452774048,
      "epoch": 0.8327446132108796,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004937659441221531,
      "loss": 5.5631,
      "mean_token_accuracy": 0.1875049278140068,
      "num_tokens": 17587344.0,
      "step": 9430
    },
    {
      "entropy": 5.817720985412597,
      "epoch": 0.8331861533027198,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004937585803619912,
      "loss": 5.5161,
      "mean_token_accuracy": 0.1912759780883789,
      "num_tokens": 17596033.0,
      "step": 9435
    },
    {
      "entropy": 5.836237287521362,
      "epoch": 0.8336276933945602,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004937512123164731,
      "loss": 5.4889,
      "mean_token_accuracy": 0.20142687410116195,
      "num_tokens": 17605860.0,
      "step": 9440
    },
    {
      "entropy": 5.844070720672607,
      "epoch": 0.8340692334864006,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004937438399857431,
      "loss": 5.6416,
      "mean_token_accuracy": 0.18426611870527268,
      "num_tokens": 17615819.0,
      "step": 9445
    },
    {
      "entropy": 5.843596935272217,
      "epoch": 0.8345107735782409,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004937364633699459,
      "loss": 5.6088,
      "mean_token_accuracy": 0.19166117459535598,
      "num_tokens": 17625478.0,
      "step": 9450
    },
    {
      "entropy": 5.911627244949341,
      "epoch": 0.8349523136700813,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004937290824692255,
      "loss": 5.5726,
      "mean_token_accuracy": 0.19171651750802993,
      "num_tokens": 17635329.0,
      "step": 9455
    },
    {
      "entropy": 5.814493846893311,
      "epoch": 0.8353938537619215,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004937216972837269,
      "loss": 5.5565,
      "mean_token_accuracy": 0.1882588431239128,
      "num_tokens": 17644129.0,
      "step": 9460
    },
    {
      "entropy": 5.733139610290527,
      "epoch": 0.8358353938537619,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004937143078135946,
      "loss": 5.4978,
      "mean_token_accuracy": 0.19633229672908784,
      "num_tokens": 17652836.0,
      "step": 9465
    },
    {
      "entropy": 5.855417060852051,
      "epoch": 0.8362769339456022,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004937069140589735,
      "loss": 5.5625,
      "mean_token_accuracy": 0.18842038065195083,
      "num_tokens": 17661222.0,
      "step": 9470
    },
    {
      "entropy": 5.822893047332764,
      "epoch": 0.8367184740374426,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004936995160200083,
      "loss": 5.4958,
      "mean_token_accuracy": 0.20001437067985534,
      "num_tokens": 17669637.0,
      "step": 9475
    },
    {
      "entropy": 5.933269119262695,
      "epoch": 0.837160014129283,
      "grad_norm": 1.0,
      "learning_rate": 0.0004936921136968439,
      "loss": 5.7061,
      "mean_token_accuracy": 0.1770281821489334,
      "num_tokens": 17678589.0,
      "step": 9480
    },
    {
      "entropy": 5.864948320388794,
      "epoch": 0.8376015542211233,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004936847070896254,
      "loss": 5.6684,
      "mean_token_accuracy": 0.1817604660987854,
      "num_tokens": 17689129.0,
      "step": 9485
    },
    {
      "entropy": 5.9451494216918945,
      "epoch": 0.8380430943129636,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004936772961984979,
      "loss": 5.6541,
      "mean_token_accuracy": 0.18763858824968338,
      "num_tokens": 17698864.0,
      "step": 9490
    },
    {
      "entropy": 5.964946937561035,
      "epoch": 0.8384846344048039,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004936698810236065,
      "loss": 5.7361,
      "mean_token_accuracy": 0.17936290800571442,
      "num_tokens": 17707900.0,
      "step": 9495
    },
    {
      "entropy": 5.836407423019409,
      "epoch": 0.8389261744966443,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004936624615650964,
      "loss": 5.5412,
      "mean_token_accuracy": 0.19406685680150987,
      "num_tokens": 17716983.0,
      "step": 9500
    },
    {
      "entropy": 5.838777542114258,
      "epoch": 0.8393677145884846,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004936550378231133,
      "loss": 5.5538,
      "mean_token_accuracy": 0.18904087096452712,
      "num_tokens": 17726503.0,
      "step": 9505
    },
    {
      "entropy": 5.873986625671387,
      "epoch": 0.839809254680325,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004936476097978023,
      "loss": 5.6409,
      "mean_token_accuracy": 0.18771174252033235,
      "num_tokens": 17735708.0,
      "step": 9510
    },
    {
      "entropy": 5.859618520736694,
      "epoch": 0.8402507947721654,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004936401774893088,
      "loss": 5.5591,
      "mean_token_accuracy": 0.18589181452989578,
      "num_tokens": 17745169.0,
      "step": 9515
    },
    {
      "entropy": 5.937458419799805,
      "epoch": 0.8406923348640056,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004936327408977786,
      "loss": 5.7008,
      "mean_token_accuracy": 0.17849522083997726,
      "num_tokens": 17755630.0,
      "step": 9520
    },
    {
      "entropy": 5.868137788772583,
      "epoch": 0.841133874955846,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004936253000233574,
      "loss": 5.5731,
      "mean_token_accuracy": 0.18807740211486818,
      "num_tokens": 17764469.0,
      "step": 9525
    },
    {
      "entropy": 5.833565092086792,
      "epoch": 0.8415754150476863,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004936178548661909,
      "loss": 5.6655,
      "mean_token_accuracy": 0.18485989272594452,
      "num_tokens": 17774977.0,
      "step": 9530
    },
    {
      "entropy": 5.835967254638672,
      "epoch": 0.8420169551395267,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000493610405426425,
      "loss": 5.582,
      "mean_token_accuracy": 0.1799257680773735,
      "num_tokens": 17784250.0,
      "step": 9535
    },
    {
      "entropy": 5.891266679763794,
      "epoch": 0.842458495231367,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004936029517042056,
      "loss": 5.6253,
      "mean_token_accuracy": 0.19154638797044754,
      "num_tokens": 17793858.0,
      "step": 9540
    },
    {
      "entropy": 5.947404050827027,
      "epoch": 0.8429000353232073,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004935954936996787,
      "loss": 5.7057,
      "mean_token_accuracy": 0.1791315421462059,
      "num_tokens": 17804330.0,
      "step": 9545
    },
    {
      "entropy": 5.87277250289917,
      "epoch": 0.8433415754150477,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004935880314129903,
      "loss": 5.6006,
      "mean_token_accuracy": 0.1922000303864479,
      "num_tokens": 17813750.0,
      "step": 9550
    },
    {
      "entropy": 5.731779670715332,
      "epoch": 0.843783115506888,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004935805648442868,
      "loss": 5.5664,
      "mean_token_accuracy": 0.19010436683893203,
      "num_tokens": 17823537.0,
      "step": 9555
    },
    {
      "entropy": 5.836687755584717,
      "epoch": 0.8442246555987284,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004935730939937143,
      "loss": 5.5825,
      "mean_token_accuracy": 0.19426935017108918,
      "num_tokens": 17833298.0,
      "step": 9560
    },
    {
      "entropy": 5.900979709625244,
      "epoch": 0.8446661956905687,
      "grad_norm": 1.0,
      "learning_rate": 0.0004935656188614192,
      "loss": 5.4836,
      "mean_token_accuracy": 0.19837422668933868,
      "num_tokens": 17842339.0,
      "step": 9565
    },
    {
      "entropy": 5.874470138549805,
      "epoch": 0.845107735782409,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004935581394475479,
      "loss": 5.7009,
      "mean_token_accuracy": 0.18750401884317397,
      "num_tokens": 17851827.0,
      "step": 9570
    },
    {
      "entropy": 5.7602321147918705,
      "epoch": 0.8455492758742493,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004935506557522469,
      "loss": 5.5391,
      "mean_token_accuracy": 0.19153136610984803,
      "num_tokens": 17860788.0,
      "step": 9575
    },
    {
      "entropy": 5.866920423507691,
      "epoch": 0.8459908159660897,
      "grad_norm": 0.921875,
      "learning_rate": 0.000493543167775663,
      "loss": 5.5448,
      "mean_token_accuracy": 0.19585472494363784,
      "num_tokens": 17869849.0,
      "step": 9580
    },
    {
      "entropy": 5.86996488571167,
      "epoch": 0.8464323560579301,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004935356755179426,
      "loss": 5.5749,
      "mean_token_accuracy": 0.18391755521297454,
      "num_tokens": 17879431.0,
      "step": 9585
    },
    {
      "entropy": 5.9047904968261715,
      "epoch": 0.8468738961497704,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004935281789792326,
      "loss": 5.6799,
      "mean_token_accuracy": 0.1816534087061882,
      "num_tokens": 17888599.0,
      "step": 9590
    },
    {
      "entropy": 5.820351839065552,
      "epoch": 0.8473154362416108,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004935206781596799,
      "loss": 5.5834,
      "mean_token_accuracy": 0.1976203590631485,
      "num_tokens": 17898662.0,
      "step": 9595
    },
    {
      "entropy": 5.9325127601623535,
      "epoch": 0.847756976333451,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004935131730594315,
      "loss": 5.622,
      "mean_token_accuracy": 0.1850518763065338,
      "num_tokens": 17907810.0,
      "step": 9600
    },
    {
      "entropy": 5.853305768966675,
      "epoch": 0.8481985164252914,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004935056636786342,
      "loss": 5.5828,
      "mean_token_accuracy": 0.18844246566295625,
      "num_tokens": 17917611.0,
      "step": 9605
    },
    {
      "entropy": 5.834033870697022,
      "epoch": 0.8486400565171317,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004934981500174353,
      "loss": 5.6101,
      "mean_token_accuracy": 0.18328014016151428,
      "num_tokens": 17926076.0,
      "step": 9610
    },
    {
      "entropy": 5.88135986328125,
      "epoch": 0.8490815966089721,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004934906320759818,
      "loss": 5.628,
      "mean_token_accuracy": 0.19137745052576066,
      "num_tokens": 17935521.0,
      "step": 9615
    },
    {
      "entropy": 5.8876971244812015,
      "epoch": 0.8495231367008125,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004934831098544211,
      "loss": 5.5865,
      "mean_token_accuracy": 0.19253529310226442,
      "num_tokens": 17944849.0,
      "step": 9620
    },
    {
      "entropy": 5.886591815948487,
      "epoch": 0.8499646767926528,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004934755833529007,
      "loss": 5.5721,
      "mean_token_accuracy": 0.18591910004615783,
      "num_tokens": 17953988.0,
      "step": 9625
    },
    {
      "entropy": 5.807942008972168,
      "epoch": 0.8504062168844931,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004934680525715677,
      "loss": 5.5739,
      "mean_token_accuracy": 0.1885495513677597,
      "num_tokens": 17963319.0,
      "step": 9630
    },
    {
      "entropy": 5.822053098678589,
      "epoch": 0.8508477569763334,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004934605175105698,
      "loss": 5.5384,
      "mean_token_accuracy": 0.20306545943021775,
      "num_tokens": 17972054.0,
      "step": 9635
    },
    {
      "entropy": 5.9353546619415285,
      "epoch": 0.8512892970681738,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004934529781700546,
      "loss": 5.687,
      "mean_token_accuracy": 0.18286622017621995,
      "num_tokens": 17981630.0,
      "step": 9640
    },
    {
      "entropy": 5.850656509399414,
      "epoch": 0.8517308371600141,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004934454345501698,
      "loss": 5.6222,
      "mean_token_accuracy": 0.18302069157361983,
      "num_tokens": 17990987.0,
      "step": 9645
    },
    {
      "entropy": 5.742798328399658,
      "epoch": 0.8521723772518545,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004934378866510633,
      "loss": 5.5217,
      "mean_token_accuracy": 0.19494544416666032,
      "num_tokens": 17999284.0,
      "step": 9650
    },
    {
      "entropy": 5.88429822921753,
      "epoch": 0.8526139173436948,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004934303344728828,
      "loss": 5.5357,
      "mean_token_accuracy": 0.18915798515081406,
      "num_tokens": 18008670.0,
      "step": 9655
    },
    {
      "entropy": 5.821765184402466,
      "epoch": 0.8530554574355351,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004934227780157761,
      "loss": 5.5523,
      "mean_token_accuracy": 0.18990737497806548,
      "num_tokens": 18018374.0,
      "step": 9660
    },
    {
      "entropy": 5.900697135925293,
      "epoch": 0.8534969975273755,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004934152172798916,
      "loss": 5.6485,
      "mean_token_accuracy": 0.187437105178833,
      "num_tokens": 18026943.0,
      "step": 9665
    },
    {
      "entropy": 5.905522871017456,
      "epoch": 0.8539385376192158,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004934076522653772,
      "loss": 5.5986,
      "mean_token_accuracy": 0.18579338490962982,
      "num_tokens": 18036270.0,
      "step": 9670
    },
    {
      "entropy": 5.760067796707153,
      "epoch": 0.8543800777110562,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000493400082972381,
      "loss": 5.5694,
      "mean_token_accuracy": 0.18511977046728134,
      "num_tokens": 18045611.0,
      "step": 9675
    },
    {
      "entropy": 5.7979716777801515,
      "epoch": 0.8548216178028964,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004933925094010516,
      "loss": 5.6022,
      "mean_token_accuracy": 0.18361486494541168,
      "num_tokens": 18055152.0,
      "step": 9680
    },
    {
      "entropy": 5.896060514450073,
      "epoch": 0.8552631578947368,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004933849315515369,
      "loss": 5.5878,
      "mean_token_accuracy": 0.18785662204027176,
      "num_tokens": 18064376.0,
      "step": 9685
    },
    {
      "entropy": 5.780475234985351,
      "epoch": 0.8557046979865772,
      "grad_norm": 0.875,
      "learning_rate": 0.0004933773494239857,
      "loss": 5.5042,
      "mean_token_accuracy": 0.20165061950683594,
      "num_tokens": 18073569.0,
      "step": 9690
    },
    {
      "entropy": 5.783981561660767,
      "epoch": 0.8561462380784175,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004933697630185464,
      "loss": 5.5754,
      "mean_token_accuracy": 0.19270028918981552,
      "num_tokens": 18083264.0,
      "step": 9695
    },
    {
      "entropy": 5.887913942337036,
      "epoch": 0.8565877781702579,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004933621723353677,
      "loss": 5.5802,
      "mean_token_accuracy": 0.18706991970539094,
      "num_tokens": 18093151.0,
      "step": 9700
    },
    {
      "entropy": 5.7990175724029545,
      "epoch": 0.8570293182620982,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000493354577374598,
      "loss": 5.4653,
      "mean_token_accuracy": 0.20603102296590806,
      "num_tokens": 18102087.0,
      "step": 9705
    },
    {
      "entropy": 5.854032421112061,
      "epoch": 0.8574708583539385,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004933469781363865,
      "loss": 5.6354,
      "mean_token_accuracy": 0.18712162971496582,
      "num_tokens": 18111854.0,
      "step": 9710
    },
    {
      "entropy": 5.869991159439087,
      "epoch": 0.8579123984457788,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004933393746208819,
      "loss": 5.6492,
      "mean_token_accuracy": 0.17795441001653672,
      "num_tokens": 18120520.0,
      "step": 9715
    },
    {
      "entropy": 5.8892655849456785,
      "epoch": 0.8583539385376192,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004933317668282331,
      "loss": 5.7079,
      "mean_token_accuracy": 0.17710036486387254,
      "num_tokens": 18130557.0,
      "step": 9720
    },
    {
      "entropy": 5.832837152481079,
      "epoch": 0.8587954786294596,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004933241547585891,
      "loss": 5.4832,
      "mean_token_accuracy": 0.20000069737434387,
      "num_tokens": 18139199.0,
      "step": 9725
    },
    {
      "entropy": 5.786106538772583,
      "epoch": 0.8592370187212999,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004933165384120991,
      "loss": 5.5505,
      "mean_token_accuracy": 0.19100535064935684,
      "num_tokens": 18150039.0,
      "step": 9730
    },
    {
      "entropy": 5.812739372253418,
      "epoch": 0.8596785588131403,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004933089177889124,
      "loss": 5.5447,
      "mean_token_accuracy": 0.18598742187023162,
      "num_tokens": 18159451.0,
      "step": 9735
    },
    {
      "entropy": 5.868243551254272,
      "epoch": 0.8601200989049805,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004933012928891781,
      "loss": 5.6304,
      "mean_token_accuracy": 0.1877829000353813,
      "num_tokens": 18169896.0,
      "step": 9740
    },
    {
      "entropy": 5.8395103931427,
      "epoch": 0.8605616389968209,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004932936637130456,
      "loss": 5.5366,
      "mean_token_accuracy": 0.199994458258152,
      "num_tokens": 18177764.0,
      "step": 9745
    },
    {
      "entropy": 5.842320156097412,
      "epoch": 0.8610031790886612,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004932860302606644,
      "loss": 5.5981,
      "mean_token_accuracy": 0.19198889136314393,
      "num_tokens": 18186940.0,
      "step": 9750
    },
    {
      "entropy": 5.880097675323486,
      "epoch": 0.8614447191805016,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004932783925321838,
      "loss": 5.4928,
      "mean_token_accuracy": 0.19632400572299957,
      "num_tokens": 18195374.0,
      "step": 9755
    },
    {
      "entropy": 5.78803334236145,
      "epoch": 0.861886259272342,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000493270750527754,
      "loss": 5.5754,
      "mean_token_accuracy": 0.19211004823446273,
      "num_tokens": 18204779.0,
      "step": 9760
    },
    {
      "entropy": 5.795253658294678,
      "epoch": 0.8623277993641822,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004932631042475241,
      "loss": 5.5211,
      "mean_token_accuracy": 0.1943109005689621,
      "num_tokens": 18214099.0,
      "step": 9765
    },
    {
      "entropy": 5.85060133934021,
      "epoch": 0.8627693394560226,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004932554536916441,
      "loss": 5.5957,
      "mean_token_accuracy": 0.1854076012969017,
      "num_tokens": 18223190.0,
      "step": 9770
    },
    {
      "entropy": 5.875854444503784,
      "epoch": 0.8632108795478629,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000493247798860264,
      "loss": 5.5688,
      "mean_token_accuracy": 0.19499116837978364,
      "num_tokens": 18232380.0,
      "step": 9775
    },
    {
      "entropy": 5.912772750854492,
      "epoch": 0.8636524196397033,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004932401397535337,
      "loss": 5.6423,
      "mean_token_accuracy": 0.1843586429953575,
      "num_tokens": 18242621.0,
      "step": 9780
    },
    {
      "entropy": 5.894945049285889,
      "epoch": 0.8640939597315436,
      "grad_norm": 0.875,
      "learning_rate": 0.000493232476371603,
      "loss": 5.6958,
      "mean_token_accuracy": 0.18013483583927153,
      "num_tokens": 18253183.0,
      "step": 9785
    },
    {
      "entropy": 5.782314443588257,
      "epoch": 0.864535499823384,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004932248087146223,
      "loss": 5.5122,
      "mean_token_accuracy": 0.19055208265781404,
      "num_tokens": 18261967.0,
      "step": 9790
    },
    {
      "entropy": 5.764515495300293,
      "epoch": 0.8649770399152243,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004932171367827417,
      "loss": 5.5869,
      "mean_token_accuracy": 0.19173365384340285,
      "num_tokens": 18270902.0,
      "step": 9795
    },
    {
      "entropy": 5.841852235794067,
      "epoch": 0.8654185800070646,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004932094605761115,
      "loss": 5.6048,
      "mean_token_accuracy": 0.19432415813207626,
      "num_tokens": 18279387.0,
      "step": 9800
    },
    {
      "entropy": 5.799977827072143,
      "epoch": 0.865860120098905,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004932017800948821,
      "loss": 5.3995,
      "mean_token_accuracy": 0.20187714397907258,
      "num_tokens": 18288363.0,
      "step": 9805
    },
    {
      "entropy": 5.957182359695435,
      "epoch": 0.8663016601907453,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004931940953392038,
      "loss": 5.7171,
      "mean_token_accuracy": 0.17543497085571289,
      "num_tokens": 18297398.0,
      "step": 9810
    },
    {
      "entropy": 5.918104887008667,
      "epoch": 0.8667432002825857,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004931864063092274,
      "loss": 5.569,
      "mean_token_accuracy": 0.19007169157266618,
      "num_tokens": 18306230.0,
      "step": 9815
    },
    {
      "entropy": 5.856356287002564,
      "epoch": 0.8671847403744259,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004931787130051034,
      "loss": 5.6938,
      "mean_token_accuracy": 0.18917493373155594,
      "num_tokens": 18316395.0,
      "step": 9820
    },
    {
      "entropy": 5.824563980102539,
      "epoch": 0.8676262804662663,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004931710154269824,
      "loss": 5.5723,
      "mean_token_accuracy": 0.19153814613819123,
      "num_tokens": 18326020.0,
      "step": 9825
    },
    {
      "entropy": 5.7984528064727785,
      "epoch": 0.8680678205581067,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004931633135750154,
      "loss": 5.5643,
      "mean_token_accuracy": 0.19000247418880462,
      "num_tokens": 18335320.0,
      "step": 9830
    },
    {
      "entropy": 5.848925161361694,
      "epoch": 0.868509360649947,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004931556074493531,
      "loss": 5.6002,
      "mean_token_accuracy": 0.1890251263976097,
      "num_tokens": 18345137.0,
      "step": 9835
    },
    {
      "entropy": 5.887432813644409,
      "epoch": 0.8689509007417874,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004931478970501465,
      "loss": 5.6235,
      "mean_token_accuracy": 0.18818622529506684,
      "num_tokens": 18354289.0,
      "step": 9840
    },
    {
      "entropy": 5.948161697387695,
      "epoch": 0.8693924408336277,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004931401823775466,
      "loss": 5.7096,
      "mean_token_accuracy": 0.1750674292445183,
      "num_tokens": 18363701.0,
      "step": 9845
    },
    {
      "entropy": 5.916717147827148,
      "epoch": 0.869833980925468,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004931324634317047,
      "loss": 5.5695,
      "mean_token_accuracy": 0.18598162978887559,
      "num_tokens": 18373336.0,
      "step": 9850
    },
    {
      "entropy": 5.889935493469238,
      "epoch": 0.8702755210173083,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004931247402127719,
      "loss": 5.5954,
      "mean_token_accuracy": 0.1866972878575325,
      "num_tokens": 18383233.0,
      "step": 9855
    },
    {
      "entropy": 5.881876516342163,
      "epoch": 0.8707170611091487,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004931170127208995,
      "loss": 5.6465,
      "mean_token_accuracy": 0.18168477565050126,
      "num_tokens": 18393214.0,
      "step": 9860
    },
    {
      "entropy": 5.842750072479248,
      "epoch": 0.8711586012009891,
      "grad_norm": 1.0,
      "learning_rate": 0.0004931092809562388,
      "loss": 5.5281,
      "mean_token_accuracy": 0.188180635869503,
      "num_tokens": 18401773.0,
      "step": 9865
    },
    {
      "entropy": 5.790875864028931,
      "epoch": 0.8716001412928294,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004931015449189414,
      "loss": 5.592,
      "mean_token_accuracy": 0.18443243950605392,
      "num_tokens": 18411748.0,
      "step": 9870
    },
    {
      "entropy": 5.86829023361206,
      "epoch": 0.8720416813846698,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004930938046091587,
      "loss": 5.6401,
      "mean_token_accuracy": 0.185184346139431,
      "num_tokens": 18421399.0,
      "step": 9875
    },
    {
      "entropy": 5.869137191772461,
      "epoch": 0.87248322147651,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004930860600270425,
      "loss": 5.626,
      "mean_token_accuracy": 0.1813107594847679,
      "num_tokens": 18430052.0,
      "step": 9880
    },
    {
      "entropy": 5.8621752738952635,
      "epoch": 0.8729247615683504,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004930783111727443,
      "loss": 5.4581,
      "mean_token_accuracy": 0.20144166499376298,
      "num_tokens": 18438268.0,
      "step": 9885
    },
    {
      "entropy": 5.791285467147827,
      "epoch": 0.8733663016601907,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000493070558046416,
      "loss": 5.5989,
      "mean_token_accuracy": 0.185325887799263,
      "num_tokens": 18446964.0,
      "step": 9890
    },
    {
      "entropy": 5.905385112762451,
      "epoch": 0.8738078417520311,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004930628006482097,
      "loss": 5.5936,
      "mean_token_accuracy": 0.17827080488204955,
      "num_tokens": 18455701.0,
      "step": 9895
    },
    {
      "entropy": 5.879530239105224,
      "epoch": 0.8742493818438715,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004930550389782769,
      "loss": 5.5543,
      "mean_token_accuracy": 0.19134498983621598,
      "num_tokens": 18465635.0,
      "step": 9900
    },
    {
      "entropy": 5.808335256576538,
      "epoch": 0.8746909219357117,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004930472730367701,
      "loss": 5.4903,
      "mean_token_accuracy": 0.1899092212319374,
      "num_tokens": 18474833.0,
      "step": 9905
    },
    {
      "entropy": 5.82249641418457,
      "epoch": 0.8751324620275521,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004930395028238412,
      "loss": 5.5671,
      "mean_token_accuracy": 0.19031829833984376,
      "num_tokens": 18482987.0,
      "step": 9910
    },
    {
      "entropy": 5.788164138793945,
      "epoch": 0.8755740021193924,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004930317283396423,
      "loss": 5.4798,
      "mean_token_accuracy": 0.1948615252971649,
      "num_tokens": 18491274.0,
      "step": 9915
    },
    {
      "entropy": 5.836555004119873,
      "epoch": 0.8760155422112328,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000493023949584326,
      "loss": 5.6372,
      "mean_token_accuracy": 0.1819130912423134,
      "num_tokens": 18500811.0,
      "step": 9920
    },
    {
      "entropy": 5.72904372215271,
      "epoch": 0.8764570823030731,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004930161665580445,
      "loss": 5.4097,
      "mean_token_accuracy": 0.20912078022956848,
      "num_tokens": 18509475.0,
      "step": 9925
    },
    {
      "entropy": 5.906229877471924,
      "epoch": 0.8768986223949135,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004930083792609502,
      "loss": 5.5801,
      "mean_token_accuracy": 0.1860847517848015,
      "num_tokens": 18519410.0,
      "step": 9930
    },
    {
      "entropy": 5.686052227020264,
      "epoch": 0.8773401624867538,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004930005876931958,
      "loss": 5.4132,
      "mean_token_accuracy": 0.20658794939517974,
      "num_tokens": 18529194.0,
      "step": 9935
    },
    {
      "entropy": 5.85654730796814,
      "epoch": 0.8777817025785941,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004929927918549339,
      "loss": 5.6348,
      "mean_token_accuracy": 0.18094472736120223,
      "num_tokens": 18539166.0,
      "step": 9940
    },
    {
      "entropy": 5.871091413497925,
      "epoch": 0.8782232426704345,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004929849917463171,
      "loss": 5.5506,
      "mean_token_accuracy": 0.19216870218515397,
      "num_tokens": 18548166.0,
      "step": 9945
    },
    {
      "entropy": 5.802586507797241,
      "epoch": 0.8786647827622748,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004929771873674984,
      "loss": 5.5709,
      "mean_token_accuracy": 0.18824805915355683,
      "num_tokens": 18557041.0,
      "step": 9950
    },
    {
      "entropy": 5.86892409324646,
      "epoch": 0.8791063228541152,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004929693787186305,
      "loss": 5.5759,
      "mean_token_accuracy": 0.1842259109020233,
      "num_tokens": 18566461.0,
      "step": 9955
    },
    {
      "entropy": 5.824876546859741,
      "epoch": 0.8795478629459554,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004929615657998664,
      "loss": 5.5174,
      "mean_token_accuracy": 0.19374087154865266,
      "num_tokens": 18576334.0,
      "step": 9960
    },
    {
      "entropy": 5.834353828430176,
      "epoch": 0.8799894030377958,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004929537486113593,
      "loss": 5.5218,
      "mean_token_accuracy": 0.19154039919376373,
      "num_tokens": 18584882.0,
      "step": 9965
    },
    {
      "entropy": 5.880415534973144,
      "epoch": 0.8804309431296362,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004929459271532621,
      "loss": 5.5657,
      "mean_token_accuracy": 0.19408414959907533,
      "num_tokens": 18593669.0,
      "step": 9970
    },
    {
      "entropy": 5.8575982570648195,
      "epoch": 0.8808724832214765,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004929381014257282,
      "loss": 5.6155,
      "mean_token_accuracy": 0.19106159508228301,
      "num_tokens": 18603357.0,
      "step": 9975
    },
    {
      "entropy": 5.810539817810058,
      "epoch": 0.8813140233133169,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004929302714289107,
      "loss": 5.5186,
      "mean_token_accuracy": 0.19460396021604537,
      "num_tokens": 18611856.0,
      "step": 9980
    },
    {
      "entropy": 5.78901777267456,
      "epoch": 0.8817555634051572,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004929224371629634,
      "loss": 5.5747,
      "mean_token_accuracy": 0.1909811556339264,
      "num_tokens": 18620500.0,
      "step": 9985
    },
    {
      "entropy": 5.798228359222412,
      "epoch": 0.8821971034969975,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004929145986280394,
      "loss": 5.5013,
      "mean_token_accuracy": 0.19727822244167328,
      "num_tokens": 18628836.0,
      "step": 9990
    },
    {
      "entropy": 5.819610834121704,
      "epoch": 0.8826386435888378,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004929067558242922,
      "loss": 5.4532,
      "mean_token_accuracy": 0.20147583782672882,
      "num_tokens": 18636831.0,
      "step": 9995
    },
    {
      "entropy": 5.819434881210327,
      "epoch": 0.8830801836806782,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004928989087518758,
      "loss": 5.5911,
      "mean_token_accuracy": 0.19338216483592988,
      "num_tokens": 18646327.0,
      "step": 10000
    },
    {
      "entropy": 5.841499662399292,
      "epoch": 0.8835217237725186,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004928910574109435,
      "loss": 5.6318,
      "mean_token_accuracy": 0.18724457770586014,
      "num_tokens": 18656125.0,
      "step": 10005
    },
    {
      "entropy": 5.8710731029510494,
      "epoch": 0.8839632638643589,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004928832018016495,
      "loss": 5.6313,
      "mean_token_accuracy": 0.18780807703733443,
      "num_tokens": 18665853.0,
      "step": 10010
    },
    {
      "entropy": 5.9379126071929935,
      "epoch": 0.8844048039561992,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004928753419241472,
      "loss": 5.6719,
      "mean_token_accuracy": 0.18405981361865997,
      "num_tokens": 18676674.0,
      "step": 10015
    },
    {
      "entropy": 5.857097625732422,
      "epoch": 0.8848463440480395,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004928674777785912,
      "loss": 5.4977,
      "mean_token_accuracy": 0.19990508407354354,
      "num_tokens": 18685535.0,
      "step": 10020
    },
    {
      "entropy": 5.807556390762329,
      "epoch": 0.8852878841398799,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004928596093651351,
      "loss": 5.5442,
      "mean_token_accuracy": 0.18528246730566025,
      "num_tokens": 18694186.0,
      "step": 10025
    },
    {
      "entropy": 5.787950134277343,
      "epoch": 0.8857294242317202,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000492851736683933,
      "loss": 5.5218,
      "mean_token_accuracy": 0.19080675393342972,
      "num_tokens": 18703061.0,
      "step": 10030
    },
    {
      "entropy": 5.785179281234742,
      "epoch": 0.8861709643235606,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004928438597351395,
      "loss": 5.4461,
      "mean_token_accuracy": 0.19986527413129807,
      "num_tokens": 18712466.0,
      "step": 10035
    },
    {
      "entropy": 5.904344892501831,
      "epoch": 0.886612504415401,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004928359785189086,
      "loss": 5.5937,
      "mean_token_accuracy": 0.1826832503080368,
      "num_tokens": 18721522.0,
      "step": 10040
    },
    {
      "entropy": 5.783854866027832,
      "epoch": 0.8870540445072412,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004928280930353948,
      "loss": 5.5113,
      "mean_token_accuracy": 0.18971350640058518,
      "num_tokens": 18730488.0,
      "step": 10045
    },
    {
      "entropy": 5.807558202743531,
      "epoch": 0.8874955845990816,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004928202032847525,
      "loss": 5.4634,
      "mean_token_accuracy": 0.19593009650707244,
      "num_tokens": 18740318.0,
      "step": 10050
    },
    {
      "entropy": 5.844618654251098,
      "epoch": 0.8879371246909219,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004928123092671362,
      "loss": 5.5804,
      "mean_token_accuracy": 0.18753331601619722,
      "num_tokens": 18748731.0,
      "step": 10055
    },
    {
      "entropy": 5.775624227523804,
      "epoch": 0.8883786647827623,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004928044109827008,
      "loss": 5.5029,
      "mean_token_accuracy": 0.19979605823755264,
      "num_tokens": 18757847.0,
      "step": 10060
    },
    {
      "entropy": 5.871121978759765,
      "epoch": 0.8888202048746026,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004927965084316007,
      "loss": 5.5828,
      "mean_token_accuracy": 0.18818929344415664,
      "num_tokens": 18767429.0,
      "step": 10065
    },
    {
      "entropy": 5.83223180770874,
      "epoch": 0.889261744966443,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000492788601613991,
      "loss": 5.5099,
      "mean_token_accuracy": 0.19484381079673768,
      "num_tokens": 18776652.0,
      "step": 10070
    },
    {
      "entropy": 5.853069925308228,
      "epoch": 0.8897032850582833,
      "grad_norm": 1.0,
      "learning_rate": 0.0004927806905300265,
      "loss": 5.5848,
      "mean_token_accuracy": 0.18585562855005264,
      "num_tokens": 18785687.0,
      "step": 10075
    },
    {
      "entropy": 5.8013709545135494,
      "epoch": 0.8901448251501236,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000492772775179862,
      "loss": 5.5981,
      "mean_token_accuracy": 0.19164907187223434,
      "num_tokens": 18794924.0,
      "step": 10080
    },
    {
      "entropy": 5.740886402130127,
      "epoch": 0.890586365241964,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004927648555636528,
      "loss": 5.4636,
      "mean_token_accuracy": 0.20413612872362136,
      "num_tokens": 18803359.0,
      "step": 10085
    },
    {
      "entropy": 5.848564958572387,
      "epoch": 0.8910279053338043,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004927569316815539,
      "loss": 5.6846,
      "mean_token_accuracy": 0.18487467914819716,
      "num_tokens": 18813496.0,
      "step": 10090
    },
    {
      "entropy": 5.871120262145996,
      "epoch": 0.8914694454256447,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004927490035337205,
      "loss": 5.5264,
      "mean_token_accuracy": 0.1974919855594635,
      "num_tokens": 18823274.0,
      "step": 10095
    },
    {
      "entropy": 5.836826467514038,
      "epoch": 0.8919109855174849,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000492741071120308,
      "loss": 5.5709,
      "mean_token_accuracy": 0.1899980917572975,
      "num_tokens": 18832436.0,
      "step": 10100
    },
    {
      "entropy": 5.861043167114258,
      "epoch": 0.8923525256093253,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004927331344414717,
      "loss": 5.6186,
      "mean_token_accuracy": 0.19028650522232055,
      "num_tokens": 18841923.0,
      "step": 10105
    },
    {
      "entropy": 5.8604639053344725,
      "epoch": 0.8927940657011657,
      "grad_norm": 1.0,
      "learning_rate": 0.0004927251934973672,
      "loss": 5.6378,
      "mean_token_accuracy": 0.19036390632390976,
      "num_tokens": 18850957.0,
      "step": 10110
    },
    {
      "entropy": 5.9429018020629885,
      "epoch": 0.893235605793006,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004927172482881499,
      "loss": 5.6574,
      "mean_token_accuracy": 0.18523878306150438,
      "num_tokens": 18859622.0,
      "step": 10115
    },
    {
      "entropy": 5.913838958740234,
      "epoch": 0.8936771458848464,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004927092988139755,
      "loss": 5.5954,
      "mean_token_accuracy": 0.18543685376644134,
      "num_tokens": 18868555.0,
      "step": 10120
    },
    {
      "entropy": 5.873710680007934,
      "epoch": 0.8941186859766866,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004927013450749999,
      "loss": 5.6223,
      "mean_token_accuracy": 0.18576771765947342,
      "num_tokens": 18877445.0,
      "step": 10125
    },
    {
      "entropy": 5.796116781234741,
      "epoch": 0.894560226068527,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004926933870713787,
      "loss": 5.4899,
      "mean_token_accuracy": 0.19211601465940475,
      "num_tokens": 18886386.0,
      "step": 10130
    },
    {
      "entropy": 5.885198497772217,
      "epoch": 0.8950017661603673,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004926854248032678,
      "loss": 5.5841,
      "mean_token_accuracy": 0.18588477224111558,
      "num_tokens": 18895016.0,
      "step": 10135
    },
    {
      "entropy": 5.77524151802063,
      "epoch": 0.8954433062522077,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004926774582708234,
      "loss": 5.4651,
      "mean_token_accuracy": 0.19851129949092866,
      "num_tokens": 18904126.0,
      "step": 10140
    },
    {
      "entropy": 5.775547122955322,
      "epoch": 0.8958848463440481,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004926694874742012,
      "loss": 5.5271,
      "mean_token_accuracy": 0.1910506159067154,
      "num_tokens": 18913064.0,
      "step": 10145
    },
    {
      "entropy": 5.818687582015992,
      "epoch": 0.8963263864358884,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004926615124135577,
      "loss": 5.6335,
      "mean_token_accuracy": 0.18564965426921845,
      "num_tokens": 18921189.0,
      "step": 10150
    },
    {
      "entropy": 5.848613119125366,
      "epoch": 0.8967679265277287,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004926535330890488,
      "loss": 5.5471,
      "mean_token_accuracy": 0.19497582465410232,
      "num_tokens": 18930818.0,
      "step": 10155
    },
    {
      "entropy": 5.931802415847779,
      "epoch": 0.897209466619569,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004926455495008311,
      "loss": 5.5703,
      "mean_token_accuracy": 0.18524253219366074,
      "num_tokens": 18940376.0,
      "step": 10160
    },
    {
      "entropy": 5.858433055877685,
      "epoch": 0.8976510067114094,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004926375616490608,
      "loss": 5.5997,
      "mean_token_accuracy": 0.19515008479356766,
      "num_tokens": 18949418.0,
      "step": 10165
    },
    {
      "entropy": 5.832569408416748,
      "epoch": 0.8980925468032497,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004926295695338945,
      "loss": 5.5395,
      "mean_token_accuracy": 0.19587756246328353,
      "num_tokens": 18959315.0,
      "step": 10170
    },
    {
      "entropy": 5.773690319061279,
      "epoch": 0.8985340868950901,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004926215731554887,
      "loss": 5.4795,
      "mean_token_accuracy": 0.19844437688589095,
      "num_tokens": 18968336.0,
      "step": 10175
    },
    {
      "entropy": 5.825376367568969,
      "epoch": 0.8989756269869305,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000492613572514,
      "loss": 5.5255,
      "mean_token_accuracy": 0.19697312414646148,
      "num_tokens": 18977123.0,
      "step": 10180
    },
    {
      "entropy": 5.857279682159424,
      "epoch": 0.8994171670787707,
      "grad_norm": 1.0,
      "learning_rate": 0.0004926055676095851,
      "loss": 5.6546,
      "mean_token_accuracy": 0.1828354775905609,
      "num_tokens": 18986150.0,
      "step": 10185
    },
    {
      "entropy": 5.920010948181153,
      "epoch": 0.8998587071706111,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004925975584424012,
      "loss": 5.6333,
      "mean_token_accuracy": 0.18982790857553483,
      "num_tokens": 18996010.0,
      "step": 10190
    },
    {
      "entropy": 5.860252571105957,
      "epoch": 0.9003002472624514,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004925895450126046,
      "loss": 5.6349,
      "mean_token_accuracy": 0.1805057168006897,
      "num_tokens": 19005100.0,
      "step": 10195
    },
    {
      "entropy": 5.828482389450073,
      "epoch": 0.9007417873542918,
      "grad_norm": 1.0,
      "learning_rate": 0.0004925815273203526,
      "loss": 5.5176,
      "mean_token_accuracy": 0.1949925735592842,
      "num_tokens": 19013844.0,
      "step": 10200
    },
    {
      "entropy": 5.925581741333008,
      "epoch": 0.901183327446132,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004925735053658022,
      "loss": 5.6748,
      "mean_token_accuracy": 0.18633326888084412,
      "num_tokens": 19024218.0,
      "step": 10205
    },
    {
      "entropy": 5.870132875442505,
      "epoch": 0.9016248675379724,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004925654791491106,
      "loss": 5.5969,
      "mean_token_accuracy": 0.18883185982704162,
      "num_tokens": 19034725.0,
      "step": 10210
    },
    {
      "entropy": 5.887659358978271,
      "epoch": 0.9020664076298128,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004925574486704351,
      "loss": 5.6185,
      "mean_token_accuracy": 0.18139611631631852,
      "num_tokens": 19044597.0,
      "step": 10215
    },
    {
      "entropy": 5.978512763977051,
      "epoch": 0.9025079477216531,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004925494139299329,
      "loss": 5.7472,
      "mean_token_accuracy": 0.17511816322803497,
      "num_tokens": 19055124.0,
      "step": 10220
    },
    {
      "entropy": 5.832484722137451,
      "epoch": 0.9029494878134935,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004925413749277613,
      "loss": 5.4823,
      "mean_token_accuracy": 0.1957518771290779,
      "num_tokens": 19064516.0,
      "step": 10225
    },
    {
      "entropy": 5.787093019485473,
      "epoch": 0.9033910279053338,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004925333316640779,
      "loss": 5.5421,
      "mean_token_accuracy": 0.19380549490451812,
      "num_tokens": 19072926.0,
      "step": 10230
    },
    {
      "entropy": 5.859374189376831,
      "epoch": 0.9038325679971742,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004925252841390403,
      "loss": 5.5995,
      "mean_token_accuracy": 0.18605128526687623,
      "num_tokens": 19082313.0,
      "step": 10235
    },
    {
      "entropy": 5.799882936477661,
      "epoch": 0.9042741080890144,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004925172323528061,
      "loss": 5.565,
      "mean_token_accuracy": 0.19741909354925155,
      "num_tokens": 19091891.0,
      "step": 10240
    },
    {
      "entropy": 5.858547878265381,
      "epoch": 0.9047156481808548,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000492509176305533,
      "loss": 5.5876,
      "mean_token_accuracy": 0.1897084191441536,
      "num_tokens": 19100311.0,
      "step": 10245
    },
    {
      "entropy": 5.900732564926147,
      "epoch": 0.9051571882726952,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004925011159973788,
      "loss": 5.5921,
      "mean_token_accuracy": 0.18747150152921677,
      "num_tokens": 19110463.0,
      "step": 10250
    },
    {
      "entropy": 5.870510864257812,
      "epoch": 0.9055987283645355,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004924930514285015,
      "loss": 5.5646,
      "mean_token_accuracy": 0.19496434628963472,
      "num_tokens": 19120723.0,
      "step": 10255
    },
    {
      "entropy": 5.783599853515625,
      "epoch": 0.9060402684563759,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000492484982599059,
      "loss": 5.5296,
      "mean_token_accuracy": 0.19487620741128922,
      "num_tokens": 19130259.0,
      "step": 10260
    },
    {
      "entropy": 5.849772882461548,
      "epoch": 0.9064818085482161,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004924769095092094,
      "loss": 5.6082,
      "mean_token_accuracy": 0.18738152384757994,
      "num_tokens": 19140256.0,
      "step": 10265
    },
    {
      "entropy": 5.847735500335693,
      "epoch": 0.9069233486400565,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004924688321591109,
      "loss": 5.6108,
      "mean_token_accuracy": 0.19002716541290282,
      "num_tokens": 19149165.0,
      "step": 10270
    },
    {
      "entropy": 5.806421995162964,
      "epoch": 0.9073648887318968,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004924607505489215,
      "loss": 5.5302,
      "mean_token_accuracy": 0.1930050030350685,
      "num_tokens": 19159254.0,
      "step": 10275
    },
    {
      "entropy": 5.856638717651367,
      "epoch": 0.9078064288237372,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004924526646787998,
      "loss": 5.5815,
      "mean_token_accuracy": 0.18753056079149247,
      "num_tokens": 19168166.0,
      "step": 10280
    },
    {
      "entropy": 5.875579071044922,
      "epoch": 0.9082479689155776,
      "grad_norm": 0.984375,
      "learning_rate": 0.000492444574548904,
      "loss": 5.5756,
      "mean_token_accuracy": 0.18910295367240906,
      "num_tokens": 19176887.0,
      "step": 10285
    },
    {
      "entropy": 5.816029071807861,
      "epoch": 0.9086895090074179,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004924364801593927,
      "loss": 5.5055,
      "mean_token_accuracy": 0.19390737116336823,
      "num_tokens": 19185510.0,
      "step": 10290
    },
    {
      "entropy": 5.730973148345948,
      "epoch": 0.9091310490992582,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004924283815104243,
      "loss": 5.539,
      "mean_token_accuracy": 0.2019898921251297,
      "num_tokens": 19195459.0,
      "step": 10295
    },
    {
      "entropy": 5.832513618469238,
      "epoch": 0.9095725891910985,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004924202786021578,
      "loss": 5.5633,
      "mean_token_accuracy": 0.19126464575529098,
      "num_tokens": 19204590.0,
      "step": 10300
    },
    {
      "entropy": 5.8629861831665036,
      "epoch": 0.9100141292829389,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004924121714347515,
      "loss": 5.509,
      "mean_token_accuracy": 0.1883266180753708,
      "num_tokens": 19213302.0,
      "step": 10305
    },
    {
      "entropy": 5.921752691268921,
      "epoch": 0.9104556693747792,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004924040600083644,
      "loss": 5.6284,
      "mean_token_accuracy": 0.18644503206014634,
      "num_tokens": 19223172.0,
      "step": 10310
    },
    {
      "entropy": 5.79363374710083,
      "epoch": 0.9108972094666196,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004923959443231554,
      "loss": 5.512,
      "mean_token_accuracy": 0.19097822457551955,
      "num_tokens": 19232234.0,
      "step": 10315
    },
    {
      "entropy": 5.706864833831787,
      "epoch": 0.91133874955846,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004923878243792836,
      "loss": 5.5101,
      "mean_token_accuracy": 0.19105002135038376,
      "num_tokens": 19241070.0,
      "step": 10320
    },
    {
      "entropy": 5.867473459243774,
      "epoch": 0.9117802896503002,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004923797001769079,
      "loss": 5.6053,
      "mean_token_accuracy": 0.1921307310461998,
      "num_tokens": 19251094.0,
      "step": 10325
    },
    {
      "entropy": 5.951294708251953,
      "epoch": 0.9122218297421406,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004923715717161875,
      "loss": 5.712,
      "mean_token_accuracy": 0.17849336862564086,
      "num_tokens": 19260588.0,
      "step": 10330
    },
    {
      "entropy": 5.920758962631226,
      "epoch": 0.9126633698339809,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004923634389972817,
      "loss": 5.7312,
      "mean_token_accuracy": 0.18118082880973815,
      "num_tokens": 19270687.0,
      "step": 10335
    },
    {
      "entropy": 5.824967288970948,
      "epoch": 0.9131049099258213,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004923553020203497,
      "loss": 5.5033,
      "mean_token_accuracy": 0.19493811577558517,
      "num_tokens": 19278739.0,
      "step": 10340
    },
    {
      "entropy": 5.960403347015381,
      "epoch": 0.9135464500176615,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004923471607855511,
      "loss": 5.674,
      "mean_token_accuracy": 0.17853326499462127,
      "num_tokens": 19289184.0,
      "step": 10345
    },
    {
      "entropy": 5.817495632171631,
      "epoch": 0.9139879901095019,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000492339015293045,
      "loss": 5.5119,
      "mean_token_accuracy": 0.20601145029067994,
      "num_tokens": 19298390.0,
      "step": 10350
    },
    {
      "entropy": 5.710802125930786,
      "epoch": 0.9144295302013423,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004923308655429913,
      "loss": 5.4672,
      "mean_token_accuracy": 0.2000226393342018,
      "num_tokens": 19307227.0,
      "step": 10355
    },
    {
      "entropy": 5.861149406433105,
      "epoch": 0.9148710702931826,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004923227115355495,
      "loss": 5.4825,
      "mean_token_accuracy": 0.18927655071020127,
      "num_tokens": 19317701.0,
      "step": 10360
    },
    {
      "entropy": 5.870515537261963,
      "epoch": 0.915312610385023,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004923145532708794,
      "loss": 5.609,
      "mean_token_accuracy": 0.18576886802911757,
      "num_tokens": 19326983.0,
      "step": 10365
    },
    {
      "entropy": 5.838311290740966,
      "epoch": 0.9157541504768633,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004923063907491408,
      "loss": 5.6094,
      "mean_token_accuracy": 0.1861104741692543,
      "num_tokens": 19337157.0,
      "step": 10370
    },
    {
      "entropy": 5.907456350326538,
      "epoch": 0.9161956905687036,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004922982239704936,
      "loss": 5.706,
      "mean_token_accuracy": 0.1902213379740715,
      "num_tokens": 19346845.0,
      "step": 10375
    },
    {
      "entropy": 5.897551155090332,
      "epoch": 0.9166372306605439,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004922900529350978,
      "loss": 5.5666,
      "mean_token_accuracy": 0.19248107224702835,
      "num_tokens": 19356905.0,
      "step": 10380
    },
    {
      "entropy": 5.88671841621399,
      "epoch": 0.9170787707523843,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004922818776431134,
      "loss": 5.5663,
      "mean_token_accuracy": 0.18569775819778442,
      "num_tokens": 19366595.0,
      "step": 10385
    },
    {
      "entropy": 5.902255821228027,
      "epoch": 0.9175203108442247,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004922736980947005,
      "loss": 5.6216,
      "mean_token_accuracy": 0.185403411090374,
      "num_tokens": 19377213.0,
      "step": 10390
    },
    {
      "entropy": 5.812246799468994,
      "epoch": 0.917961850936065,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004922655142900196,
      "loss": 5.4878,
      "mean_token_accuracy": 0.19387887716293334,
      "num_tokens": 19385510.0,
      "step": 10395
    },
    {
      "entropy": 5.769075489044189,
      "epoch": 0.9184033910279054,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004922573262292308,
      "loss": 5.533,
      "mean_token_accuracy": 0.18867291808128356,
      "num_tokens": 19394505.0,
      "step": 10400
    },
    {
      "entropy": 5.840726375579834,
      "epoch": 0.9188449311197456,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004922491339124945,
      "loss": 5.4133,
      "mean_token_accuracy": 0.2043693780899048,
      "num_tokens": 19403140.0,
      "step": 10405
    },
    {
      "entropy": 5.870825862884521,
      "epoch": 0.919286471211586,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004922409373399713,
      "loss": 5.6429,
      "mean_token_accuracy": 0.18787774443626404,
      "num_tokens": 19412709.0,
      "step": 10410
    },
    {
      "entropy": 5.760507488250733,
      "epoch": 0.9197280113034263,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004922327365118216,
      "loss": 5.4797,
      "mean_token_accuracy": 0.19664435982704162,
      "num_tokens": 19422442.0,
      "step": 10415
    },
    {
      "entropy": 5.799489831924438,
      "epoch": 0.9201695513952667,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004922245314282062,
      "loss": 5.5869,
      "mean_token_accuracy": 0.1881786435842514,
      "num_tokens": 19432153.0,
      "step": 10420
    },
    {
      "entropy": 5.823181200027466,
      "epoch": 0.9206110914871071,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004922163220892858,
      "loss": 5.4102,
      "mean_token_accuracy": 0.20729821771383286,
      "num_tokens": 19440811.0,
      "step": 10425
    },
    {
      "entropy": 5.832983493804932,
      "epoch": 0.9210526315789473,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000492208108495221,
      "loss": 5.5976,
      "mean_token_accuracy": 0.18627535849809645,
      "num_tokens": 19451330.0,
      "step": 10430
    },
    {
      "entropy": 5.754718780517578,
      "epoch": 0.9214941716707877,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000492199890646173,
      "loss": 5.4342,
      "mean_token_accuracy": 0.2057944655418396,
      "num_tokens": 19459689.0,
      "step": 10435
    },
    {
      "entropy": 5.9080911636352536,
      "epoch": 0.921935711762628,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004921916685423028,
      "loss": 5.5012,
      "mean_token_accuracy": 0.19368696510791777,
      "num_tokens": 19468110.0,
      "step": 10440
    },
    {
      "entropy": 5.8278931140899655,
      "epoch": 0.9223772518544684,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004921834421837711,
      "loss": 5.5555,
      "mean_token_accuracy": 0.18965283781290054,
      "num_tokens": 19477190.0,
      "step": 10445
    },
    {
      "entropy": 5.878892230987549,
      "epoch": 0.9228187919463087,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004921752115707394,
      "loss": 5.6379,
      "mean_token_accuracy": 0.1885254517197609,
      "num_tokens": 19485757.0,
      "step": 10450
    },
    {
      "entropy": 5.852978515625,
      "epoch": 0.9232603320381491,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004921669767033688,
      "loss": 5.5819,
      "mean_token_accuracy": 0.18951420485973358,
      "num_tokens": 19494846.0,
      "step": 10455
    },
    {
      "entropy": 5.862094497680664,
      "epoch": 0.9237018721299894,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004921587375818207,
      "loss": 5.58,
      "mean_token_accuracy": 0.1880530074238777,
      "num_tokens": 19504307.0,
      "step": 10460
    },
    {
      "entropy": 5.860325145721435,
      "epoch": 0.9241434122218297,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004921504942062564,
      "loss": 5.5183,
      "mean_token_accuracy": 0.18791094273328782,
      "num_tokens": 19512741.0,
      "step": 10465
    },
    {
      "entropy": 5.765936851501465,
      "epoch": 0.9245849523136701,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004921422465768374,
      "loss": 5.43,
      "mean_token_accuracy": 0.1966376557946205,
      "num_tokens": 19521823.0,
      "step": 10470
    },
    {
      "entropy": 5.794504547119141,
      "epoch": 0.9250264924055104,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004921339946937253,
      "loss": 5.6214,
      "mean_token_accuracy": 0.1903781697154045,
      "num_tokens": 19531048.0,
      "step": 10475
    },
    {
      "entropy": 5.863368797302246,
      "epoch": 0.9254680324973508,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004921257385570818,
      "loss": 5.5655,
      "mean_token_accuracy": 0.19053512662649155,
      "num_tokens": 19539543.0,
      "step": 10480
    },
    {
      "entropy": 5.94217700958252,
      "epoch": 0.925909572589191,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004921174781670685,
      "loss": 5.6751,
      "mean_token_accuracy": 0.18787076473236083,
      "num_tokens": 19549242.0,
      "step": 10485
    },
    {
      "entropy": 5.89439754486084,
      "epoch": 0.9263511126810314,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004921092135238473,
      "loss": 5.5648,
      "mean_token_accuracy": 0.18893613815307617,
      "num_tokens": 19558259.0,
      "step": 10490
    },
    {
      "entropy": 5.848619270324707,
      "epoch": 0.9267926527728718,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00049210094462758,
      "loss": 5.5645,
      "mean_token_accuracy": 0.18988049328327178,
      "num_tokens": 19567327.0,
      "step": 10495
    },
    {
      "entropy": 5.851853847503662,
      "epoch": 0.9272341928647121,
      "grad_norm": 1.0,
      "learning_rate": 0.0004920926714784288,
      "loss": 5.5915,
      "mean_token_accuracy": 0.18649715185165405,
      "num_tokens": 19576696.0,
      "step": 10500
    },
    {
      "entropy": 5.78373966217041,
      "epoch": 0.9276757329565525,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004920843940765555,
      "loss": 5.4707,
      "mean_token_accuracy": 0.20074461102485658,
      "num_tokens": 19585708.0,
      "step": 10505
    },
    {
      "entropy": 5.900629997253418,
      "epoch": 0.9281172730483928,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004920761124221224,
      "loss": 5.6463,
      "mean_token_accuracy": 0.19012684673070906,
      "num_tokens": 19595199.0,
      "step": 10510
    },
    {
      "entropy": 5.890163230895996,
      "epoch": 0.9285588131402331,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004920678265152919,
      "loss": 5.5683,
      "mean_token_accuracy": 0.1849339336156845,
      "num_tokens": 19604781.0,
      "step": 10515
    },
    {
      "entropy": 5.79541654586792,
      "epoch": 0.9290003532320734,
      "grad_norm": 0.984375,
      "learning_rate": 0.000492059536356226,
      "loss": 5.5189,
      "mean_token_accuracy": 0.19828343540430068,
      "num_tokens": 19614203.0,
      "step": 10520
    },
    {
      "entropy": 5.859884405136109,
      "epoch": 0.9294418933239138,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004920512419450872,
      "loss": 5.5467,
      "mean_token_accuracy": 0.1924946814775467,
      "num_tokens": 19623664.0,
      "step": 10525
    },
    {
      "entropy": 5.855048704147339,
      "epoch": 0.9298834334157542,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000492042943282038,
      "loss": 5.5473,
      "mean_token_accuracy": 0.19281800091266632,
      "num_tokens": 19632960.0,
      "step": 10530
    },
    {
      "entropy": 5.839187288284302,
      "epoch": 0.9303249735075945,
      "grad_norm": 0.890625,
      "learning_rate": 0.000492034640367241,
      "loss": 5.6609,
      "mean_token_accuracy": 0.18112767189741136,
      "num_tokens": 19643087.0,
      "step": 10535
    },
    {
      "entropy": 5.922775077819824,
      "epoch": 0.9307665135994349,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000492026333200859,
      "loss": 5.6156,
      "mean_token_accuracy": 0.18796712905168533,
      "num_tokens": 19653015.0,
      "step": 10540
    },
    {
      "entropy": 5.935885286331176,
      "epoch": 0.9312080536912751,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004920180217830543,
      "loss": 5.6198,
      "mean_token_accuracy": 0.18706526309251786,
      "num_tokens": 19661569.0,
      "step": 10545
    },
    {
      "entropy": 5.706429433822632,
      "epoch": 0.9316495937831155,
      "grad_norm": 1.0,
      "learning_rate": 0.00049200970611399,
      "loss": 5.4372,
      "mean_token_accuracy": 0.20040739327669144,
      "num_tokens": 19669559.0,
      "step": 10550
    },
    {
      "entropy": 5.738175296783448,
      "epoch": 0.9320911338749558,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004920013861938291,
      "loss": 5.4618,
      "mean_token_accuracy": 0.20482187420129777,
      "num_tokens": 19678766.0,
      "step": 10555
    },
    {
      "entropy": 5.83841872215271,
      "epoch": 0.9325326739667962,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004919930620227345,
      "loss": 5.6172,
      "mean_token_accuracy": 0.18378598093986512,
      "num_tokens": 19688422.0,
      "step": 10560
    },
    {
      "entropy": 5.947232913970947,
      "epoch": 0.9329742140586366,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004919847336008692,
      "loss": 5.6382,
      "mean_token_accuracy": 0.1858677551150322,
      "num_tokens": 19697206.0,
      "step": 10565
    },
    {
      "entropy": 5.870072269439698,
      "epoch": 0.9334157541504768,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004919764009283963,
      "loss": 5.6012,
      "mean_token_accuracy": 0.1842290610074997,
      "num_tokens": 19706065.0,
      "step": 10570
    },
    {
      "entropy": 5.8446617603302,
      "epoch": 0.9338572942423172,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004919680640054792,
      "loss": 5.5312,
      "mean_token_accuracy": 0.19687247425317764,
      "num_tokens": 19715402.0,
      "step": 10575
    },
    {
      "entropy": 5.799044561386109,
      "epoch": 0.9342988343341575,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000491959722832281,
      "loss": 5.4632,
      "mean_token_accuracy": 0.1981448546051979,
      "num_tokens": 19724328.0,
      "step": 10580
    },
    {
      "entropy": 5.792357778549194,
      "epoch": 0.9347403744259979,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004919513774089653,
      "loss": 5.581,
      "mean_token_accuracy": 0.19021185338497162,
      "num_tokens": 19734087.0,
      "step": 10585
    },
    {
      "entropy": 5.932400131225586,
      "epoch": 0.9351819145178382,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004919430277356955,
      "loss": 5.5321,
      "mean_token_accuracy": 0.19613374769687653,
      "num_tokens": 19743010.0,
      "step": 10590
    },
    {
      "entropy": 5.843912744522095,
      "epoch": 0.9356234546096786,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004919346738126351,
      "loss": 5.4913,
      "mean_token_accuracy": 0.19796763360500336,
      "num_tokens": 19752212.0,
      "step": 10595
    },
    {
      "entropy": 5.823946142196656,
      "epoch": 0.9360649947015189,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004919263156399478,
      "loss": 5.588,
      "mean_token_accuracy": 0.19262956976890563,
      "num_tokens": 19762496.0,
      "step": 10600
    },
    {
      "entropy": 5.8475781917572025,
      "epoch": 0.9365065347933592,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004919179532177976,
      "loss": 5.5768,
      "mean_token_accuracy": 0.1893441617488861,
      "num_tokens": 19772665.0,
      "step": 10605
    },
    {
      "entropy": 5.775518417358398,
      "epoch": 0.9369480748851996,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004919095865463479,
      "loss": 5.419,
      "mean_token_accuracy": 0.20027465969324112,
      "num_tokens": 19781639.0,
      "step": 10610
    },
    {
      "entropy": 5.844583129882812,
      "epoch": 0.9373896149770399,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004919012156257627,
      "loss": 5.5176,
      "mean_token_accuracy": 0.19134962409734727,
      "num_tokens": 19791475.0,
      "step": 10615
    },
    {
      "entropy": 5.819388580322266,
      "epoch": 0.9378311550688803,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004918928404562061,
      "loss": 5.5021,
      "mean_token_accuracy": 0.19283542335033416,
      "num_tokens": 19800967.0,
      "step": 10620
    },
    {
      "entropy": 5.782877779006958,
      "epoch": 0.9382726951607206,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004918844610378421,
      "loss": 5.6631,
      "mean_token_accuracy": 0.18687088489532472,
      "num_tokens": 19810303.0,
      "step": 10625
    },
    {
      "entropy": 5.933231925964355,
      "epoch": 0.9387142352525609,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004918760773708348,
      "loss": 5.5558,
      "mean_token_accuracy": 0.18964565843343734,
      "num_tokens": 19819049.0,
      "step": 10630
    },
    {
      "entropy": 5.9463951587677,
      "epoch": 0.9391557753444013,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004918676894553485,
      "loss": 5.5939,
      "mean_token_accuracy": 0.19474973082542418,
      "num_tokens": 19828253.0,
      "step": 10635
    },
    {
      "entropy": 5.834896755218506,
      "epoch": 0.9395973154362416,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004918592972915476,
      "loss": 5.5447,
      "mean_token_accuracy": 0.1950840249657631,
      "num_tokens": 19837673.0,
      "step": 10640
    },
    {
      "entropy": 5.758587121963501,
      "epoch": 0.940038855528082,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004918509008795963,
      "loss": 5.5817,
      "mean_token_accuracy": 0.18384605646133423,
      "num_tokens": 19846289.0,
      "step": 10645
    },
    {
      "entropy": 5.782199668884277,
      "epoch": 0.9404803956199222,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004918425002196594,
      "loss": 5.5483,
      "mean_token_accuracy": 0.19021508246660232,
      "num_tokens": 19855384.0,
      "step": 10650
    },
    {
      "entropy": 5.815890741348267,
      "epoch": 0.9409219357117626,
      "grad_norm": 1.625,
      "learning_rate": 0.000491834095311901,
      "loss": 5.4442,
      "mean_token_accuracy": 0.20328225791454316,
      "num_tokens": 19864688.0,
      "step": 10655
    },
    {
      "entropy": 5.7918627738952635,
      "epoch": 0.941363475803603,
      "grad_norm": 1.09375,
      "learning_rate": 0.000491825686156486,
      "loss": 5.4835,
      "mean_token_accuracy": 0.1945149391889572,
      "num_tokens": 19873692.0,
      "step": 10660
    },
    {
      "entropy": 5.855662870407104,
      "epoch": 0.9418050158954433,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004918172727535792,
      "loss": 5.5903,
      "mean_token_accuracy": 0.17907644957304,
      "num_tokens": 19882063.0,
      "step": 10665
    },
    {
      "entropy": 5.889585542678833,
      "epoch": 0.9422465559872837,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004918088551033454,
      "loss": 5.5299,
      "mean_token_accuracy": 0.19539738744497298,
      "num_tokens": 19891727.0,
      "step": 10670
    },
    {
      "entropy": 5.788696193695069,
      "epoch": 0.942688096079124,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004918004332059493,
      "loss": 5.5827,
      "mean_token_accuracy": 0.1965092420578003,
      "num_tokens": 19901851.0,
      "step": 10675
    },
    {
      "entropy": 5.855720949172974,
      "epoch": 0.9431296361709643,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004917920070615561,
      "loss": 5.5253,
      "mean_token_accuracy": 0.1925203323364258,
      "num_tokens": 19911092.0,
      "step": 10680
    },
    {
      "entropy": 5.8097868919372555,
      "epoch": 0.9435711762628046,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004917835766703307,
      "loss": 5.493,
      "mean_token_accuracy": 0.2018080472946167,
      "num_tokens": 19920842.0,
      "step": 10685
    },
    {
      "entropy": 5.9051590919494625,
      "epoch": 0.944012716354645,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004917751420324384,
      "loss": 5.5925,
      "mean_token_accuracy": 0.18529444336891174,
      "num_tokens": 19930833.0,
      "step": 10690
    },
    {
      "entropy": 5.908380222320557,
      "epoch": 0.9444542564464854,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004917667031480444,
      "loss": 5.6317,
      "mean_token_accuracy": 0.19101576805114745,
      "num_tokens": 19940514.0,
      "step": 10695
    },
    {
      "entropy": 5.79524040222168,
      "epoch": 0.9448957965383257,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004917582600173138,
      "loss": 5.4725,
      "mean_token_accuracy": 0.19828383177518843,
      "num_tokens": 19948949.0,
      "step": 10700
    },
    {
      "entropy": 5.715598058700562,
      "epoch": 0.9453373366301661,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004917498126404123,
      "loss": 5.4482,
      "mean_token_accuracy": 0.19933301210403442,
      "num_tokens": 19958410.0,
      "step": 10705
    },
    {
      "entropy": 5.811657333374024,
      "epoch": 0.9457788767220063,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004917413610175052,
      "loss": 5.4453,
      "mean_token_accuracy": 0.2015089213848114,
      "num_tokens": 19968160.0,
      "step": 10710
    },
    {
      "entropy": 5.925013494491577,
      "epoch": 0.9462204168138467,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004917329051487582,
      "loss": 5.6116,
      "mean_token_accuracy": 0.1815849557518959,
      "num_tokens": 19976811.0,
      "step": 10715
    },
    {
      "entropy": 5.927849149703979,
      "epoch": 0.946661956905687,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004917244450343368,
      "loss": 5.6883,
      "mean_token_accuracy": 0.17899896651506425,
      "num_tokens": 19986326.0,
      "step": 10720
    },
    {
      "entropy": 5.78308892250061,
      "epoch": 0.9471034969975274,
      "grad_norm": 0.875,
      "learning_rate": 0.0004917159806744067,
      "loss": 5.4072,
      "mean_token_accuracy": 0.19397269636392594,
      "num_tokens": 19996255.0,
      "step": 10725
    },
    {
      "entropy": 5.804378938674927,
      "epoch": 0.9475450370893678,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004917075120691338,
      "loss": 5.6274,
      "mean_token_accuracy": 0.18804312944412233,
      "num_tokens": 20004805.0,
      "step": 10730
    },
    {
      "entropy": 5.79609866142273,
      "epoch": 0.947986577181208,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004916990392186841,
      "loss": 5.5205,
      "mean_token_accuracy": 0.19648486524820327,
      "num_tokens": 20014189.0,
      "step": 10735
    },
    {
      "entropy": 5.874725818634033,
      "epoch": 0.9484281172730484,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004916905621232233,
      "loss": 5.5106,
      "mean_token_accuracy": 0.18644982278347016,
      "num_tokens": 20023261.0,
      "step": 10740
    },
    {
      "entropy": 5.811865997314453,
      "epoch": 0.9488696573648887,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004916820807829177,
      "loss": 5.4675,
      "mean_token_accuracy": 0.20290163904428482,
      "num_tokens": 20032433.0,
      "step": 10745
    },
    {
      "entropy": 5.813940620422363,
      "epoch": 0.9493111974567291,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004916735951979335,
      "loss": 5.5916,
      "mean_token_accuracy": 0.18647033274173735,
      "num_tokens": 20041455.0,
      "step": 10750
    },
    {
      "entropy": 5.860266447067261,
      "epoch": 0.9497527375485694,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004916651053684366,
      "loss": 5.6197,
      "mean_token_accuracy": 0.19127440452575684,
      "num_tokens": 20050088.0,
      "step": 10755
    },
    {
      "entropy": 5.873831081390381,
      "epoch": 0.9501942776404098,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004916566112945936,
      "loss": 5.5936,
      "mean_token_accuracy": 0.1924014210700989,
      "num_tokens": 20059114.0,
      "step": 10760
    },
    {
      "entropy": 5.8611561298370365,
      "epoch": 0.9506358177322501,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004916481129765708,
      "loss": 5.5789,
      "mean_token_accuracy": 0.19046310633420943,
      "num_tokens": 20068692.0,
      "step": 10765
    },
    {
      "entropy": 5.838329076766968,
      "epoch": 0.9510773578240904,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004916396104145347,
      "loss": 5.6301,
      "mean_token_accuracy": 0.18845141381025315,
      "num_tokens": 20078380.0,
      "step": 10770
    },
    {
      "entropy": 5.733320808410644,
      "epoch": 0.9515188979159308,
      "grad_norm": 1.0,
      "learning_rate": 0.0004916311036086518,
      "loss": 5.449,
      "mean_token_accuracy": 0.20709470957517623,
      "num_tokens": 20087849.0,
      "step": 10775
    },
    {
      "entropy": 5.778975915908814,
      "epoch": 0.9519604380077711,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004916225925590889,
      "loss": 5.3979,
      "mean_token_accuracy": 0.2057466834783554,
      "num_tokens": 20097253.0,
      "step": 10780
    },
    {
      "entropy": 5.8431253910064695,
      "epoch": 0.9524019780996115,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004916140772660124,
      "loss": 5.5891,
      "mean_token_accuracy": 0.1924793854355812,
      "num_tokens": 20105931.0,
      "step": 10785
    },
    {
      "entropy": 5.866446781158447,
      "epoch": 0.9528435181914517,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004916055577295895,
      "loss": 5.474,
      "mean_token_accuracy": 0.20115301311016082,
      "num_tokens": 20115546.0,
      "step": 10790
    },
    {
      "entropy": 5.872866916656494,
      "epoch": 0.9532850582832921,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004915970339499868,
      "loss": 5.5021,
      "mean_token_accuracy": 0.1942845806479454,
      "num_tokens": 20125009.0,
      "step": 10795
    },
    {
      "entropy": 5.719097089767456,
      "epoch": 0.9537265983751325,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004915885059273714,
      "loss": 5.4817,
      "mean_token_accuracy": 0.19818040132522582,
      "num_tokens": 20134007.0,
      "step": 10800
    },
    {
      "entropy": 5.820677709579468,
      "epoch": 0.9541681384669728,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004915799736619104,
      "loss": 5.4953,
      "mean_token_accuracy": 0.19351319819688798,
      "num_tokens": 20142617.0,
      "step": 10805
    },
    {
      "entropy": 5.808750438690185,
      "epoch": 0.9546096785588132,
      "grad_norm": 1.0,
      "learning_rate": 0.000491571437153771,
      "loss": 5.5193,
      "mean_token_accuracy": 0.19794072806835175,
      "num_tokens": 20151732.0,
      "step": 10810
    },
    {
      "entropy": 5.856479501724243,
      "epoch": 0.9550512186506535,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004915628964031202,
      "loss": 5.6163,
      "mean_token_accuracy": 0.1860255166888237,
      "num_tokens": 20161733.0,
      "step": 10815
    },
    {
      "entropy": 5.874025154113769,
      "epoch": 0.9554927587424938,
      "grad_norm": 1.0,
      "learning_rate": 0.0004915543514101253,
      "loss": 5.5447,
      "mean_token_accuracy": 0.18483669310808182,
      "num_tokens": 20171022.0,
      "step": 10820
    },
    {
      "entropy": 5.85857367515564,
      "epoch": 0.9559342988343341,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004915458021749541,
      "loss": 5.5463,
      "mean_token_accuracy": 0.18598238825798036,
      "num_tokens": 20180809.0,
      "step": 10825
    },
    {
      "entropy": 5.783239555358887,
      "epoch": 0.9563758389261745,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004915372486977736,
      "loss": 5.486,
      "mean_token_accuracy": 0.19979268461465835,
      "num_tokens": 20189614.0,
      "step": 10830
    },
    {
      "entropy": 5.841081809997559,
      "epoch": 0.9568173790180149,
      "grad_norm": 0.875,
      "learning_rate": 0.0004915286909787516,
      "loss": 5.6297,
      "mean_token_accuracy": 0.18066627979278566,
      "num_tokens": 20199136.0,
      "step": 10835
    },
    {
      "entropy": 5.846323442459107,
      "epoch": 0.9572589191098552,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004915201290180556,
      "loss": 5.529,
      "mean_token_accuracy": 0.19071774035692216,
      "num_tokens": 20208768.0,
      "step": 10840
    },
    {
      "entropy": 5.7795311450958256,
      "epoch": 0.9577004592016956,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004915115628158535,
      "loss": 5.512,
      "mean_token_accuracy": 0.19033466279506683,
      "num_tokens": 20218109.0,
      "step": 10845
    },
    {
      "entropy": 5.84051480293274,
      "epoch": 0.9581419992935358,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000491502992372313,
      "loss": 5.5144,
      "mean_token_accuracy": 0.19057008475065232,
      "num_tokens": 20226768.0,
      "step": 10850
    },
    {
      "entropy": 5.81554651260376,
      "epoch": 0.9585835393853762,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000491494417687602,
      "loss": 5.471,
      "mean_token_accuracy": 0.19765634089708328,
      "num_tokens": 20236519.0,
      "step": 10855
    },
    {
      "entropy": 5.762387752532959,
      "epoch": 0.9590250794772165,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004914858387618886,
      "loss": 5.4908,
      "mean_token_accuracy": 0.19534918218851088,
      "num_tokens": 20245611.0,
      "step": 10860
    },
    {
      "entropy": 5.750170135498047,
      "epoch": 0.9594666195690569,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004914772555953406,
      "loss": 5.406,
      "mean_token_accuracy": 0.20583200305700303,
      "num_tokens": 20255089.0,
      "step": 10865
    },
    {
      "entropy": 5.817626190185547,
      "epoch": 0.9599081596608973,
      "grad_norm": 1.0,
      "learning_rate": 0.0004914686681881264,
      "loss": 5.5574,
      "mean_token_accuracy": 0.1936030000448227,
      "num_tokens": 20264794.0,
      "step": 10870
    },
    {
      "entropy": 5.796624183654785,
      "epoch": 0.9603496997527375,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000491460076540414,
      "loss": 5.416,
      "mean_token_accuracy": 0.20647903829813002,
      "num_tokens": 20274088.0,
      "step": 10875
    },
    {
      "entropy": 5.8205211639404295,
      "epoch": 0.9607912398445779,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004914514806523718,
      "loss": 5.6139,
      "mean_token_accuracy": 0.18427006900310516,
      "num_tokens": 20283837.0,
      "step": 10880
    },
    {
      "entropy": 5.822281312942505,
      "epoch": 0.9612327799364182,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004914428805241683,
      "loss": 5.5071,
      "mean_token_accuracy": 0.1921115681529045,
      "num_tokens": 20293287.0,
      "step": 10885
    },
    {
      "entropy": 5.7551898002624515,
      "epoch": 0.9616743200282586,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004914342761559718,
      "loss": 5.4649,
      "mean_token_accuracy": 0.20005710870027543,
      "num_tokens": 20302262.0,
      "step": 10890
    },
    {
      "entropy": 5.847769927978516,
      "epoch": 0.9621158601200989,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004914256675479509,
      "loss": 5.561,
      "mean_token_accuracy": 0.18721564412117003,
      "num_tokens": 20312475.0,
      "step": 10895
    },
    {
      "entropy": 5.871733570098877,
      "epoch": 0.9625574002119393,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004914170547002742,
      "loss": 5.5751,
      "mean_token_accuracy": 0.18982745260000228,
      "num_tokens": 20322243.0,
      "step": 10900
    },
    {
      "entropy": 5.778389406204224,
      "epoch": 0.9629989403037796,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004914084376131106,
      "loss": 5.4688,
      "mean_token_accuracy": 0.19254059046506883,
      "num_tokens": 20331299.0,
      "step": 10905
    },
    {
      "entropy": 5.7243060111999515,
      "epoch": 0.9634404803956199,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004913998162866288,
      "loss": 5.4259,
      "mean_token_accuracy": 0.20277778804302216,
      "num_tokens": 20340800.0,
      "step": 10910
    },
    {
      "entropy": 5.820883941650391,
      "epoch": 0.9638820204874603,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004913911907209976,
      "loss": 5.5237,
      "mean_token_accuracy": 0.19660518169403077,
      "num_tokens": 20350292.0,
      "step": 10915
    },
    {
      "entropy": 5.881887340545655,
      "epoch": 0.9643235605793006,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004913825609163861,
      "loss": 5.6065,
      "mean_token_accuracy": 0.1878852590918541,
      "num_tokens": 20360109.0,
      "step": 10920
    },
    {
      "entropy": 5.729526853561401,
      "epoch": 0.964765100671141,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004913739268729633,
      "loss": 5.4341,
      "mean_token_accuracy": 0.20365360230207444,
      "num_tokens": 20368917.0,
      "step": 10925
    },
    {
      "entropy": 5.871054363250733,
      "epoch": 0.9652066407629812,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004913652885908983,
      "loss": 5.6194,
      "mean_token_accuracy": 0.18444482535123824,
      "num_tokens": 20378580.0,
      "step": 10930
    },
    {
      "entropy": 5.847923469543457,
      "epoch": 0.9656481808548216,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004913566460703603,
      "loss": 5.4444,
      "mean_token_accuracy": 0.19596558958292007,
      "num_tokens": 20387497.0,
      "step": 10935
    },
    {
      "entropy": 5.832335758209228,
      "epoch": 0.966089720946662,
      "grad_norm": 1.0,
      "learning_rate": 0.0004913479993115187,
      "loss": 5.5996,
      "mean_token_accuracy": 0.18531107306480407,
      "num_tokens": 20396586.0,
      "step": 10940
    },
    {
      "entropy": 5.873624277114868,
      "epoch": 0.9665312610385023,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004913393483145428,
      "loss": 5.687,
      "mean_token_accuracy": 0.1777090162038803,
      "num_tokens": 20406183.0,
      "step": 10945
    },
    {
      "entropy": 5.987033367156982,
      "epoch": 0.9669728011303427,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004913306930796022,
      "loss": 5.6099,
      "mean_token_accuracy": 0.1890963390469551,
      "num_tokens": 20415716.0,
      "step": 10950
    },
    {
      "entropy": 5.862414169311523,
      "epoch": 0.967414341222183,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004913220336068662,
      "loss": 5.5392,
      "mean_token_accuracy": 0.1910262554883957,
      "num_tokens": 20424715.0,
      "step": 10955
    },
    {
      "entropy": 5.779024362564087,
      "epoch": 0.9678558813140233,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004913133698965048,
      "loss": 5.5316,
      "mean_token_accuracy": 0.19780020266771317,
      "num_tokens": 20434686.0,
      "step": 10960
    },
    {
      "entropy": 5.89177713394165,
      "epoch": 0.9682974214058636,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004913047019486874,
      "loss": 5.5132,
      "mean_token_accuracy": 0.19592755883932114,
      "num_tokens": 20443619.0,
      "step": 10965
    },
    {
      "entropy": 5.808563947677612,
      "epoch": 0.968738961497704,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004912960297635839,
      "loss": 5.4652,
      "mean_token_accuracy": 0.20202988535165786,
      "num_tokens": 20453540.0,
      "step": 10970
    },
    {
      "entropy": 5.836717319488526,
      "epoch": 0.9691805015895444,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004912873533413641,
      "loss": 5.5787,
      "mean_token_accuracy": 0.1865977704524994,
      "num_tokens": 20463395.0,
      "step": 10975
    },
    {
      "entropy": 5.80081467628479,
      "epoch": 0.9696220416813847,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004912786726821982,
      "loss": 5.5233,
      "mean_token_accuracy": 0.19126374423503875,
      "num_tokens": 20472302.0,
      "step": 10980
    },
    {
      "entropy": 5.828465366363526,
      "epoch": 0.970063581773225,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000491269987786256,
      "loss": 5.5348,
      "mean_token_accuracy": 0.18925092965364457,
      "num_tokens": 20482077.0,
      "step": 10985
    },
    {
      "entropy": 5.882504272460937,
      "epoch": 0.9705051218650653,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004912612986537078,
      "loss": 5.5443,
      "mean_token_accuracy": 0.19295522719621658,
      "num_tokens": 20492177.0,
      "step": 10990
    },
    {
      "entropy": 5.842564868927002,
      "epoch": 0.9709466619569057,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004912526052847238,
      "loss": 5.5354,
      "mean_token_accuracy": 0.19674329757690429,
      "num_tokens": 20502063.0,
      "step": 10995
    },
    {
      "entropy": 5.8237231254577635,
      "epoch": 0.971388202048746,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004912439076794742,
      "loss": 5.501,
      "mean_token_accuracy": 0.1882395938038826,
      "num_tokens": 20510631.0,
      "step": 11000
    },
    {
      "entropy": 5.805966806411743,
      "epoch": 0.9718297421405864,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004912352058381293,
      "loss": 5.5267,
      "mean_token_accuracy": 0.19073240756988524,
      "num_tokens": 20519978.0,
      "step": 11005
    },
    {
      "entropy": 5.838520431518555,
      "epoch": 0.9722712822324268,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004912264997608599,
      "loss": 5.5247,
      "mean_token_accuracy": 0.19575220346450806,
      "num_tokens": 20529346.0,
      "step": 11010
    },
    {
      "entropy": 5.845974016189575,
      "epoch": 0.972712822324267,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004912177894478364,
      "loss": 5.4883,
      "mean_token_accuracy": 0.18457019031047822,
      "num_tokens": 20537619.0,
      "step": 11015
    },
    {
      "entropy": 5.80692048072815,
      "epoch": 0.9731543624161074,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004912090748992291,
      "loss": 5.5325,
      "mean_token_accuracy": 0.19319151490926742,
      "num_tokens": 20547011.0,
      "step": 11020
    },
    {
      "entropy": 5.868833398818969,
      "epoch": 0.9735959025079477,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004912003561152092,
      "loss": 5.4505,
      "mean_token_accuracy": 0.19301554709672927,
      "num_tokens": 20556425.0,
      "step": 11025
    },
    {
      "entropy": 5.82353720664978,
      "epoch": 0.9740374425997881,
      "grad_norm": 1.0,
      "learning_rate": 0.0004911916330959472,
      "loss": 5.4817,
      "mean_token_accuracy": 0.19372252076864244,
      "num_tokens": 20565330.0,
      "step": 11030
    },
    {
      "entropy": 5.782631540298462,
      "epoch": 0.9744789826916284,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004911829058416141,
      "loss": 5.5746,
      "mean_token_accuracy": 0.20310136675834656,
      "num_tokens": 20575320.0,
      "step": 11035
    },
    {
      "entropy": 5.832729387283325,
      "epoch": 0.9749205227834687,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004911741743523808,
      "loss": 5.4792,
      "mean_token_accuracy": 0.20184270590543746,
      "num_tokens": 20585105.0,
      "step": 11040
    },
    {
      "entropy": 5.805664777755737,
      "epoch": 0.9753620628753091,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004911654386284184,
      "loss": 5.5221,
      "mean_token_accuracy": 0.19096221029758453,
      "num_tokens": 20593889.0,
      "step": 11045
    },
    {
      "entropy": 5.816440153121948,
      "epoch": 0.9758036029671494,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000491156698669898,
      "loss": 5.5806,
      "mean_token_accuracy": 0.19517292082309723,
      "num_tokens": 20602322.0,
      "step": 11050
    },
    {
      "entropy": 5.882116222381592,
      "epoch": 0.9762451430589898,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004911479544769908,
      "loss": 5.646,
      "mean_token_accuracy": 0.1793898180127144,
      "num_tokens": 20612077.0,
      "step": 11055
    },
    {
      "entropy": 5.936902141571045,
      "epoch": 0.9766866831508301,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000491139206049868,
      "loss": 5.6115,
      "mean_token_accuracy": 0.19139361679553984,
      "num_tokens": 20620683.0,
      "step": 11060
    },
    {
      "entropy": 5.883061647415161,
      "epoch": 0.9771282232426705,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004911304533887011,
      "loss": 5.6407,
      "mean_token_accuracy": 0.18760707080364228,
      "num_tokens": 20631124.0,
      "step": 11065
    },
    {
      "entropy": 5.846213531494141,
      "epoch": 0.9775697633345107,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004911216964936616,
      "loss": 5.5012,
      "mean_token_accuracy": 0.19065711200237273,
      "num_tokens": 20640515.0,
      "step": 11070
    },
    {
      "entropy": 5.830524349212647,
      "epoch": 0.9780113034263511,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004911129353649211,
      "loss": 5.4786,
      "mean_token_accuracy": 0.19359829723834993,
      "num_tokens": 20649819.0,
      "step": 11075
    },
    {
      "entropy": 5.820531988143921,
      "epoch": 0.9784528435181915,
      "grad_norm": 1.03125,
      "learning_rate": 0.000491104170002651,
      "loss": 5.5104,
      "mean_token_accuracy": 0.18779035657644272,
      "num_tokens": 20657831.0,
      "step": 11080
    },
    {
      "entropy": 5.872558355331421,
      "epoch": 0.9788943836100318,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000491095400407023,
      "loss": 5.6304,
      "mean_token_accuracy": 0.189630950987339,
      "num_tokens": 20666807.0,
      "step": 11085
    },
    {
      "entropy": 5.868529367446899,
      "epoch": 0.9793359237018722,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004910866265782091,
      "loss": 5.4759,
      "mean_token_accuracy": 0.19473941773176193,
      "num_tokens": 20674859.0,
      "step": 11090
    },
    {
      "entropy": 5.881205654144287,
      "epoch": 0.9797774637937124,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004910778485163811,
      "loss": 5.5587,
      "mean_token_accuracy": 0.19576413780450821,
      "num_tokens": 20684060.0,
      "step": 11095
    },
    {
      "entropy": 5.822815370559693,
      "epoch": 0.9802190038855528,
      "grad_norm": 1.015625,
      "learning_rate": 0.000491069066221711,
      "loss": 5.4876,
      "mean_token_accuracy": 0.1994558945298195,
      "num_tokens": 20693446.0,
      "step": 11100
    },
    {
      "entropy": 5.830358982086182,
      "epoch": 0.9806605439773931,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004910602796943707,
      "loss": 5.6416,
      "mean_token_accuracy": 0.18342330753803254,
      "num_tokens": 20702635.0,
      "step": 11105
    },
    {
      "entropy": 5.9078997611999515,
      "epoch": 0.9811020840692335,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004910514889345323,
      "loss": 5.5593,
      "mean_token_accuracy": 0.19043246358633042,
      "num_tokens": 20712256.0,
      "step": 11110
    },
    {
      "entropy": 5.8011908531188965,
      "epoch": 0.9815436241610739,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004910426939423682,
      "loss": 5.5167,
      "mean_token_accuracy": 0.19414657205343247,
      "num_tokens": 20720175.0,
      "step": 11115
    },
    {
      "entropy": 5.811932897567749,
      "epoch": 0.9819851642529142,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004910338947180506,
      "loss": 5.5335,
      "mean_token_accuracy": 0.1913396969437599,
      "num_tokens": 20729849.0,
      "step": 11120
    },
    {
      "entropy": 5.838641405105591,
      "epoch": 0.9824267043447545,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004910250912617519,
      "loss": 5.4708,
      "mean_token_accuracy": 0.19795780032873153,
      "num_tokens": 20739129.0,
      "step": 11125
    },
    {
      "entropy": 5.8044229507446286,
      "epoch": 0.9828682444365948,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004910162835736445,
      "loss": 5.4401,
      "mean_token_accuracy": 0.20202791690826416,
      "num_tokens": 20748632.0,
      "step": 11130
    },
    {
      "entropy": 5.840196371078491,
      "epoch": 0.9833097845284352,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000491007471653901,
      "loss": 5.4914,
      "mean_token_accuracy": 0.1948569595813751,
      "num_tokens": 20758535.0,
      "step": 11135
    },
    {
      "entropy": 5.8420037746429445,
      "epoch": 0.9837513246202755,
      "grad_norm": 0.875,
      "learning_rate": 0.0004909986555026939,
      "loss": 5.5285,
      "mean_token_accuracy": 0.19545526206493377,
      "num_tokens": 20768982.0,
      "step": 11140
    },
    {
      "entropy": 5.8689741611480715,
      "epoch": 0.9841928647121159,
      "grad_norm": 0.984375,
      "learning_rate": 0.000490989835120196,
      "loss": 5.5847,
      "mean_token_accuracy": 0.19294222444295883,
      "num_tokens": 20778936.0,
      "step": 11145
    },
    {
      "entropy": 5.844592666625976,
      "epoch": 0.9846344048039563,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004909810105065801,
      "loss": 5.5818,
      "mean_token_accuracy": 0.18521394282579423,
      "num_tokens": 20788618.0,
      "step": 11150
    },
    {
      "entropy": 5.882683420181275,
      "epoch": 0.9850759448957965,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004909721816620191,
      "loss": 5.5572,
      "mean_token_accuracy": 0.19761741161346436,
      "num_tokens": 20797721.0,
      "step": 11155
    },
    {
      "entropy": 5.8583558082580565,
      "epoch": 0.9855174849876369,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004909633485866857,
      "loss": 5.4983,
      "mean_token_accuracy": 0.19559048116207123,
      "num_tokens": 20807305.0,
      "step": 11160
    },
    {
      "entropy": 5.772975111007691,
      "epoch": 0.9859590250794772,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004909545112807533,
      "loss": 5.4632,
      "mean_token_accuracy": 0.2001844346523285,
      "num_tokens": 20816984.0,
      "step": 11165
    },
    {
      "entropy": 5.901240539550781,
      "epoch": 0.9864005651713176,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004909456697443948,
      "loss": 5.6369,
      "mean_token_accuracy": 0.18369455635547638,
      "num_tokens": 20825738.0,
      "step": 11170
    },
    {
      "entropy": 5.8502623558044435,
      "epoch": 0.9868421052631579,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004909368239777835,
      "loss": 5.5822,
      "mean_token_accuracy": 0.1874794155359268,
      "num_tokens": 20834135.0,
      "step": 11175
    },
    {
      "entropy": 5.9332075119018555,
      "epoch": 0.9872836453549982,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004909279739810926,
      "loss": 5.6398,
      "mean_token_accuracy": 0.18591250330209733,
      "num_tokens": 20843689.0,
      "step": 11180
    },
    {
      "entropy": 5.897461938858032,
      "epoch": 0.9877251854468386,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004909191197544954,
      "loss": 5.5174,
      "mean_token_accuracy": 0.19147263169288636,
      "num_tokens": 20851724.0,
      "step": 11185
    },
    {
      "entropy": 5.9224241256713865,
      "epoch": 0.9881667255386789,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004909102612981655,
      "loss": 5.6748,
      "mean_token_accuracy": 0.18432777673006057,
      "num_tokens": 20862325.0,
      "step": 11190
    },
    {
      "entropy": 5.858700799942016,
      "epoch": 0.9886082656305193,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004909013986122763,
      "loss": 5.5076,
      "mean_token_accuracy": 0.19612646251916885,
      "num_tokens": 20871046.0,
      "step": 11195
    },
    {
      "entropy": 5.917420291900635,
      "epoch": 0.9890498057223596,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004908925316970016,
      "loss": 5.5363,
      "mean_token_accuracy": 0.18652904480695726,
      "num_tokens": 20881605.0,
      "step": 11200
    },
    {
      "entropy": 5.817625427246094,
      "epoch": 0.9894913458142,
      "grad_norm": 0.921875,
      "learning_rate": 0.000490883660552515,
      "loss": 5.4602,
      "mean_token_accuracy": 0.1957890585064888,
      "num_tokens": 20891170.0,
      "step": 11205
    },
    {
      "entropy": 5.833710765838623,
      "epoch": 0.9899328859060402,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004908747851789902,
      "loss": 5.4702,
      "mean_token_accuracy": 0.1861294686794281,
      "num_tokens": 20900307.0,
      "step": 11210
    },
    {
      "entropy": 5.817179775238037,
      "epoch": 0.9903744259978806,
      "grad_norm": 1.0,
      "learning_rate": 0.0004908659055766011,
      "loss": 5.6567,
      "mean_token_accuracy": 0.18575167357921601,
      "num_tokens": 20910623.0,
      "step": 11215
    },
    {
      "entropy": 5.938866853713989,
      "epoch": 0.990815966089721,
      "grad_norm": 1.0,
      "learning_rate": 0.0004908570217455218,
      "loss": 5.5463,
      "mean_token_accuracy": 0.19014012664556504,
      "num_tokens": 20919744.0,
      "step": 11220
    },
    {
      "entropy": 5.842549800872803,
      "epoch": 0.9912575061815613,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004908481336859262,
      "loss": 5.6128,
      "mean_token_accuracy": 0.18795954138040544,
      "num_tokens": 20928701.0,
      "step": 11225
    },
    {
      "entropy": 5.8330910205841064,
      "epoch": 0.9916990462734017,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004908392413979884,
      "loss": 5.5843,
      "mean_token_accuracy": 0.180023954808712,
      "num_tokens": 20937786.0,
      "step": 11230
    },
    {
      "entropy": 5.864147567749024,
      "epoch": 0.9921405863652419,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004908303448818825,
      "loss": 5.4946,
      "mean_token_accuracy": 0.19863616079092025,
      "num_tokens": 20947214.0,
      "step": 11235
    },
    {
      "entropy": 5.832936477661133,
      "epoch": 0.9925821264570823,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000490821444137783,
      "loss": 5.6014,
      "mean_token_accuracy": 0.1971711367368698,
      "num_tokens": 20956041.0,
      "step": 11240
    },
    {
      "entropy": 5.9129539966583256,
      "epoch": 0.9930236665489226,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004908125391658641,
      "loss": 5.569,
      "mean_token_accuracy": 0.19148291647434235,
      "num_tokens": 20966162.0,
      "step": 11245
    },
    {
      "entropy": 5.822826194763183,
      "epoch": 0.993465206640763,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004908036299663003,
      "loss": 5.5614,
      "mean_token_accuracy": 0.1926749274134636,
      "num_tokens": 20976025.0,
      "step": 11250
    },
    {
      "entropy": 5.8316937446594235,
      "epoch": 0.9939067467326034,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004907947165392662,
      "loss": 5.5893,
      "mean_token_accuracy": 0.18729829490184785,
      "num_tokens": 20985606.0,
      "step": 11255
    },
    {
      "entropy": 5.878600978851319,
      "epoch": 0.9943482868244437,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004907857988849363,
      "loss": 5.5732,
      "mean_token_accuracy": 0.1873452693223953,
      "num_tokens": 20994308.0,
      "step": 11260
    },
    {
      "entropy": 5.794317722320557,
      "epoch": 0.994789826916284,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004907768770034853,
      "loss": 5.5024,
      "mean_token_accuracy": 0.19150446206331254,
      "num_tokens": 21003783.0,
      "step": 11265
    },
    {
      "entropy": 5.8252458572387695,
      "epoch": 0.9952313670081243,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000490767950895088,
      "loss": 5.4881,
      "mean_token_accuracy": 0.19247478395700454,
      "num_tokens": 21012870.0,
      "step": 11270
    },
    {
      "entropy": 5.8911336898803714,
      "epoch": 0.9956729070999647,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004907590205599192,
      "loss": 5.5473,
      "mean_token_accuracy": 0.19090282768011094,
      "num_tokens": 21021987.0,
      "step": 11275
    },
    {
      "entropy": 5.820564222335816,
      "epoch": 0.996114447191805,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000490750085998154,
      "loss": 5.5425,
      "mean_token_accuracy": 0.18923987150192262,
      "num_tokens": 21030977.0,
      "step": 11280
    },
    {
      "entropy": 5.850178575515747,
      "epoch": 0.9965559872836454,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004907411472099672,
      "loss": 5.4942,
      "mean_token_accuracy": 0.19618253856897355,
      "num_tokens": 21039259.0,
      "step": 11285
    },
    {
      "entropy": 5.877530908584594,
      "epoch": 0.9969975273754857,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004907322041955341,
      "loss": 5.5794,
      "mean_token_accuracy": 0.18979037553071976,
      "num_tokens": 21049321.0,
      "step": 11290
    },
    {
      "entropy": 5.921963882446289,
      "epoch": 0.997439067467326,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004907232569550298,
      "loss": 5.651,
      "mean_token_accuracy": 0.18208762407302856,
      "num_tokens": 21059469.0,
      "step": 11295
    },
    {
      "entropy": 5.900750780105591,
      "epoch": 0.9978806075591664,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004907143054886297,
      "loss": 5.5703,
      "mean_token_accuracy": 0.1895820155739784,
      "num_tokens": 21068694.0,
      "step": 11300
    },
    {
      "entropy": 5.857248401641845,
      "epoch": 0.9983221476510067,
      "grad_norm": 1.0,
      "learning_rate": 0.000490705349796509,
      "loss": 5.5068,
      "mean_token_accuracy": 0.19134427458047867,
      "num_tokens": 21077232.0,
      "step": 11305
    },
    {
      "entropy": 5.867048740386963,
      "epoch": 0.9987636877428471,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004906963898788431,
      "loss": 5.5268,
      "mean_token_accuracy": 0.1941692277789116,
      "num_tokens": 21086517.0,
      "step": 11310
    },
    {
      "entropy": 5.850531101226807,
      "epoch": 0.9992052278346873,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004906874257358075,
      "loss": 5.5828,
      "mean_token_accuracy": 0.19759476333856582,
      "num_tokens": 21096969.0,
      "step": 11315
    },
    {
      "entropy": 5.858910846710205,
      "epoch": 0.9996467679265277,
      "grad_norm": 1.03125,
      "learning_rate": 0.000490678457367578,
      "loss": 5.5086,
      "mean_token_accuracy": 0.18713059276342392,
      "num_tokens": 21106554.0,
      "step": 11320
    },
    {
      "entropy": 5.785353326797486,
      "epoch": 1.000088308018368,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004906694847743302,
      "loss": 5.5205,
      "mean_token_accuracy": 0.18986949622631072,
      "num_tokens": 21114829.0,
      "step": 11325
    },
    {
      "entropy": 5.84618034362793,
      "epoch": 1.0005298481102085,
      "grad_norm": 1.0,
      "learning_rate": 0.0004906605079562399,
      "loss": 5.458,
      "mean_token_accuracy": 0.19853014200925828,
      "num_tokens": 21123568.0,
      "step": 11330
    },
    {
      "entropy": 5.854660415649414,
      "epoch": 1.0009713882020488,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004906515269134827,
      "loss": 5.5118,
      "mean_token_accuracy": 0.19205471575260163,
      "num_tokens": 21132811.0,
      "step": 11335
    },
    {
      "entropy": 5.89120774269104,
      "epoch": 1.001412928293889,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004906425416462349,
      "loss": 5.5509,
      "mean_token_accuracy": 0.18671243637800217,
      "num_tokens": 21141549.0,
      "step": 11340
    },
    {
      "entropy": 5.838399982452392,
      "epoch": 1.0018544683857293,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004906335521546723,
      "loss": 5.4523,
      "mean_token_accuracy": 0.19941389858722686,
      "num_tokens": 21150603.0,
      "step": 11345
    },
    {
      "entropy": 5.707959413528442,
      "epoch": 1.0022960084775698,
      "grad_norm": 1.0,
      "learning_rate": 0.0004906245584389712,
      "loss": 5.2676,
      "mean_token_accuracy": 0.21225778609514237,
      "num_tokens": 21159842.0,
      "step": 11350
    },
    {
      "entropy": 5.822812271118164,
      "epoch": 1.00273754856941,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004906155604993075,
      "loss": 5.4184,
      "mean_token_accuracy": 0.2034035414457321,
      "num_tokens": 21169139.0,
      "step": 11355
    },
    {
      "entropy": 5.784867143630981,
      "epoch": 1.0031790886612504,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004906065583358576,
      "loss": 5.4127,
      "mean_token_accuracy": 0.19405371248722075,
      "num_tokens": 21178059.0,
      "step": 11360
    },
    {
      "entropy": 5.803128242492676,
      "epoch": 1.0036206287530909,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004905975519487978,
      "loss": 5.4032,
      "mean_token_accuracy": 0.20225947499275207,
      "num_tokens": 21187966.0,
      "step": 11365
    },
    {
      "entropy": 5.765253305435181,
      "epoch": 1.0040621688449312,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004905885413383046,
      "loss": 5.4632,
      "mean_token_accuracy": 0.19135601818561554,
      "num_tokens": 21197460.0,
      "step": 11370
    },
    {
      "entropy": 5.851017570495605,
      "epoch": 1.0045037089367714,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004905795265045546,
      "loss": 5.4939,
      "mean_token_accuracy": 0.1959708407521248,
      "num_tokens": 21206977.0,
      "step": 11375
    },
    {
      "entropy": 5.869827938079834,
      "epoch": 1.0049452490286117,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004905705074477242,
      "loss": 5.4267,
      "mean_token_accuracy": 0.19144175946712494,
      "num_tokens": 21216046.0,
      "step": 11380
    },
    {
      "entropy": 5.856464099884033,
      "epoch": 1.0053867891204522,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004905614841679901,
      "loss": 5.4792,
      "mean_token_accuracy": 0.19193035364151,
      "num_tokens": 21225612.0,
      "step": 11385
    },
    {
      "entropy": 5.785376119613647,
      "epoch": 1.0058283292122925,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004905524566655292,
      "loss": 5.3684,
      "mean_token_accuracy": 0.19887097030878068,
      "num_tokens": 21233641.0,
      "step": 11390
    },
    {
      "entropy": 5.765918016433716,
      "epoch": 1.0062698693041328,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004905434249405183,
      "loss": 5.4281,
      "mean_token_accuracy": 0.19341274499893188,
      "num_tokens": 21243444.0,
      "step": 11395
    },
    {
      "entropy": 5.7887389183044435,
      "epoch": 1.0067114093959733,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004905343889931342,
      "loss": 5.3977,
      "mean_token_accuracy": 0.20267544835805892,
      "num_tokens": 21253473.0,
      "step": 11400
    },
    {
      "entropy": 5.797057723999023,
      "epoch": 1.0071529494878135,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000490525348823554,
      "loss": 5.3677,
      "mean_token_accuracy": 0.21220226287841798,
      "num_tokens": 21262049.0,
      "step": 11405
    },
    {
      "entropy": 5.783495759963989,
      "epoch": 1.0075944895796538,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004905163044319549,
      "loss": 5.4093,
      "mean_token_accuracy": 0.19663029164075851,
      "num_tokens": 21270909.0,
      "step": 11410
    },
    {
      "entropy": 5.816624593734741,
      "epoch": 1.008036029671494,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004905072558185139,
      "loss": 5.5011,
      "mean_token_accuracy": 0.1927421882748604,
      "num_tokens": 21280503.0,
      "step": 11415
    },
    {
      "entropy": 5.871208047866821,
      "epoch": 1.0084775697633346,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004904982029834083,
      "loss": 5.4831,
      "mean_token_accuracy": 0.19088730365037918,
      "num_tokens": 21289957.0,
      "step": 11420
    },
    {
      "entropy": 5.831331872940064,
      "epoch": 1.0089191098551749,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004904891459268154,
      "loss": 5.4584,
      "mean_token_accuracy": 0.19292692244052886,
      "num_tokens": 21298799.0,
      "step": 11425
    },
    {
      "entropy": 5.818262195587158,
      "epoch": 1.0093606499470151,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004904800846489128,
      "loss": 5.4662,
      "mean_token_accuracy": 0.19724620282649993,
      "num_tokens": 21308251.0,
      "step": 11430
    },
    {
      "entropy": 5.905819416046143,
      "epoch": 1.0098021900388556,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004904710191498779,
      "loss": 5.5004,
      "mean_token_accuracy": 0.1947304055094719,
      "num_tokens": 21317705.0,
      "step": 11435
    },
    {
      "entropy": 5.811056852340698,
      "epoch": 1.010243730130696,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004904619494298882,
      "loss": 5.4791,
      "mean_token_accuracy": 0.1861782342195511,
      "num_tokens": 21326867.0,
      "step": 11440
    },
    {
      "entropy": 5.7974344253540036,
      "epoch": 1.0106852702225362,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004904528754891215,
      "loss": 5.4851,
      "mean_token_accuracy": 0.2002352073788643,
      "num_tokens": 21336054.0,
      "step": 11445
    },
    {
      "entropy": 5.823584175109863,
      "epoch": 1.0111268103143765,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004904437973277555,
      "loss": 5.4017,
      "mean_token_accuracy": 0.1982896387577057,
      "num_tokens": 21344457.0,
      "step": 11450
    },
    {
      "entropy": 5.870646190643311,
      "epoch": 1.011568350406217,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004904347149459681,
      "loss": 5.4993,
      "mean_token_accuracy": 0.19316635876893998,
      "num_tokens": 21353639.0,
      "step": 11455
    },
    {
      "entropy": 5.709807586669922,
      "epoch": 1.0120098904980572,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004904256283439371,
      "loss": 5.2969,
      "mean_token_accuracy": 0.20653853714466094,
      "num_tokens": 21362005.0,
      "step": 11460
    },
    {
      "entropy": 5.858596324920654,
      "epoch": 1.0124514305898975,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004904165375218405,
      "loss": 5.4774,
      "mean_token_accuracy": 0.19295329749584197,
      "num_tokens": 21371917.0,
      "step": 11465
    },
    {
      "entropy": 5.855436897277832,
      "epoch": 1.012892970681738,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004904074424798565,
      "loss": 5.4778,
      "mean_token_accuracy": 0.19918979704380035,
      "num_tokens": 21381500.0,
      "step": 11470
    },
    {
      "entropy": 5.782000780105591,
      "epoch": 1.0133345107735783,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004903983432181632,
      "loss": 5.3096,
      "mean_token_accuracy": 0.2015272855758667,
      "num_tokens": 21391280.0,
      "step": 11475
    },
    {
      "entropy": 5.815475749969482,
      "epoch": 1.0137760508654186,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004903892397369388,
      "loss": 5.4567,
      "mean_token_accuracy": 0.19736944139003754,
      "num_tokens": 21400371.0,
      "step": 11480
    },
    {
      "entropy": 5.67401762008667,
      "epoch": 1.0142175909572588,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004903801320363618,
      "loss": 5.4771,
      "mean_token_accuracy": 0.1985686182975769,
      "num_tokens": 21410493.0,
      "step": 11485
    },
    {
      "entropy": 5.90243935585022,
      "epoch": 1.0146591310490993,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004903710201166105,
      "loss": 5.5491,
      "mean_token_accuracy": 0.1836334154009819,
      "num_tokens": 21420516.0,
      "step": 11490
    },
    {
      "entropy": 5.916108655929565,
      "epoch": 1.0151006711409396,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004903619039778634,
      "loss": 5.4953,
      "mean_token_accuracy": 0.18604948073625566,
      "num_tokens": 21429614.0,
      "step": 11495
    },
    {
      "entropy": 5.828366661071778,
      "epoch": 1.0155422112327799,
      "grad_norm": 1.046875,
      "learning_rate": 0.000490352783620299,
      "loss": 5.435,
      "mean_token_accuracy": 0.19844983667135238,
      "num_tokens": 21438595.0,
      "step": 11500
    },
    {
      "entropy": 5.836382675170898,
      "epoch": 1.0159837513246204,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004903436590440961,
      "loss": 5.4777,
      "mean_token_accuracy": 0.1989193946123123,
      "num_tokens": 21447708.0,
      "step": 11505
    },
    {
      "entropy": 5.802752923965454,
      "epoch": 1.0164252914164607,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004903345302494334,
      "loss": 5.4294,
      "mean_token_accuracy": 0.1908747971057892,
      "num_tokens": 21457210.0,
      "step": 11510
    },
    {
      "entropy": 5.840627908706665,
      "epoch": 1.016866831508301,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004903253972364897,
      "loss": 5.4614,
      "mean_token_accuracy": 0.19405657798051834,
      "num_tokens": 21466247.0,
      "step": 11515
    },
    {
      "entropy": 5.7858726501464846,
      "epoch": 1.0173083716001412,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004903162600054439,
      "loss": 5.4048,
      "mean_token_accuracy": 0.20191211253404617,
      "num_tokens": 21475634.0,
      "step": 11520
    },
    {
      "entropy": 5.86365647315979,
      "epoch": 1.0177499116919817,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004903071185564751,
      "loss": 5.522,
      "mean_token_accuracy": 0.19560528099536895,
      "num_tokens": 21485523.0,
      "step": 11525
    },
    {
      "entropy": 5.8727446556091305,
      "epoch": 1.018191451783822,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004902979728897623,
      "loss": 5.472,
      "mean_token_accuracy": 0.19813337475061416,
      "num_tokens": 21494909.0,
      "step": 11530
    },
    {
      "entropy": 5.864111089706421,
      "epoch": 1.0186329918756623,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004902888230054846,
      "loss": 5.5114,
      "mean_token_accuracy": 0.1915932461619377,
      "num_tokens": 21504413.0,
      "step": 11535
    },
    {
      "entropy": 5.755148887634277,
      "epoch": 1.0190745319675028,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004902796689038213,
      "loss": 5.3802,
      "mean_token_accuracy": 0.20285746902227403,
      "num_tokens": 21514068.0,
      "step": 11540
    },
    {
      "entropy": 5.793791389465332,
      "epoch": 1.019516072059343,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004902705105849517,
      "loss": 5.3969,
      "mean_token_accuracy": 0.20098552852869034,
      "num_tokens": 21523536.0,
      "step": 11545
    },
    {
      "entropy": 5.786781597137451,
      "epoch": 1.0199576121511833,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004902613480490553,
      "loss": 5.4692,
      "mean_token_accuracy": 0.19177822917699813,
      "num_tokens": 21532554.0,
      "step": 11550
    },
    {
      "entropy": 5.76055417060852,
      "epoch": 1.0203991522430236,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004902521812963116,
      "loss": 5.3321,
      "mean_token_accuracy": 0.20827980637550353,
      "num_tokens": 21542508.0,
      "step": 11555
    },
    {
      "entropy": 5.842724275588989,
      "epoch": 1.020840692334864,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004902430103268998,
      "loss": 5.4541,
      "mean_token_accuracy": 0.18653589189052583,
      "num_tokens": 21551954.0,
      "step": 11560
    },
    {
      "entropy": 5.823522233963013,
      "epoch": 1.0212822324267044,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004902338351410002,
      "loss": 5.3807,
      "mean_token_accuracy": 0.20463838130235673,
      "num_tokens": 21561676.0,
      "step": 11565
    },
    {
      "entropy": 5.840857124328613,
      "epoch": 1.0217237725185446,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000490224655738792,
      "loss": 5.5141,
      "mean_token_accuracy": 0.18839401602745057,
      "num_tokens": 21571916.0,
      "step": 11570
    },
    {
      "entropy": 5.841112852096558,
      "epoch": 1.0221653126103851,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004902154721204552,
      "loss": 5.4459,
      "mean_token_accuracy": 0.20426111817359924,
      "num_tokens": 21581892.0,
      "step": 11575
    },
    {
      "entropy": 5.79775037765503,
      "epoch": 1.0226068527022254,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004902062842861696,
      "loss": 5.4042,
      "mean_token_accuracy": 0.194574473798275,
      "num_tokens": 21591390.0,
      "step": 11580
    },
    {
      "entropy": 5.82957501411438,
      "epoch": 1.0230483927940657,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004901970922361154,
      "loss": 5.473,
      "mean_token_accuracy": 0.19892579168081284,
      "num_tokens": 21600766.0,
      "step": 11585
    },
    {
      "entropy": 5.746204280853272,
      "epoch": 1.023489932885906,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004901878959704726,
      "loss": 5.4019,
      "mean_token_accuracy": 0.20576784610748292,
      "num_tokens": 21609264.0,
      "step": 11590
    },
    {
      "entropy": 5.889167833328247,
      "epoch": 1.0239314729777464,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004901786954894212,
      "loss": 5.4956,
      "mean_token_accuracy": 0.1914335384964943,
      "num_tokens": 21618409.0,
      "step": 11595
    },
    {
      "entropy": 5.795125818252563,
      "epoch": 1.0243730130695867,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004901694907931416,
      "loss": 5.3851,
      "mean_token_accuracy": 0.19335469454526902,
      "num_tokens": 21628158.0,
      "step": 11600
    },
    {
      "entropy": 5.816567993164062,
      "epoch": 1.024814553161427,
      "grad_norm": 0.953125,
      "learning_rate": 0.000490160281881814,
      "loss": 5.4942,
      "mean_token_accuracy": 0.19107363224029542,
      "num_tokens": 21637859.0,
      "step": 11605
    },
    {
      "entropy": 5.850227165222168,
      "epoch": 1.0252560932532675,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004901510687556188,
      "loss": 5.5057,
      "mean_token_accuracy": 0.18931938260793685,
      "num_tokens": 21647389.0,
      "step": 11610
    },
    {
      "entropy": 5.873205041885376,
      "epoch": 1.0256976333451078,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004901418514147367,
      "loss": 5.4553,
      "mean_token_accuracy": 0.19083526283502578,
      "num_tokens": 21655857.0,
      "step": 11615
    },
    {
      "entropy": 5.807099008560181,
      "epoch": 1.026139173436948,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000490132629859348,
      "loss": 5.4184,
      "mean_token_accuracy": 0.20058273077011107,
      "num_tokens": 21666150.0,
      "step": 11620
    },
    {
      "entropy": 5.834448575973511,
      "epoch": 1.0265807135287883,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004901234040896334,
      "loss": 5.4394,
      "mean_token_accuracy": 0.19685986191034316,
      "num_tokens": 21675662.0,
      "step": 11625
    },
    {
      "entropy": 5.859313535690307,
      "epoch": 1.0270222536206288,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004901141741057737,
      "loss": 5.4975,
      "mean_token_accuracy": 0.187775319814682,
      "num_tokens": 21685721.0,
      "step": 11630
    },
    {
      "entropy": 5.846322774887085,
      "epoch": 1.027463793712469,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004901049399079495,
      "loss": 5.4729,
      "mean_token_accuracy": 0.1979638710618019,
      "num_tokens": 21694507.0,
      "step": 11635
    },
    {
      "entropy": 5.857589817047119,
      "epoch": 1.0279053338043094,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004900957014963421,
      "loss": 5.4899,
      "mean_token_accuracy": 0.20053462982177733,
      "num_tokens": 21704092.0,
      "step": 11640
    },
    {
      "entropy": 5.760619688034057,
      "epoch": 1.0283468738961499,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004900864588711321,
      "loss": 5.3866,
      "mean_token_accuracy": 0.20713716447353364,
      "num_tokens": 21713313.0,
      "step": 11645
    },
    {
      "entropy": 5.814091444015503,
      "epoch": 1.0287884139879901,
      "grad_norm": 1.0,
      "learning_rate": 0.0004900772120325008,
      "loss": 5.4371,
      "mean_token_accuracy": 0.2021331310272217,
      "num_tokens": 21721622.0,
      "step": 11650
    },
    {
      "entropy": 5.838172197341919,
      "epoch": 1.0292299540798304,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004900679609806291,
      "loss": 5.413,
      "mean_token_accuracy": 0.2008321091532707,
      "num_tokens": 21730914.0,
      "step": 11655
    },
    {
      "entropy": 5.813170003890991,
      "epoch": 1.0296714941716707,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004900587057156985,
      "loss": 5.3781,
      "mean_token_accuracy": 0.20447985976934432,
      "num_tokens": 21740124.0,
      "step": 11660
    },
    {
      "entropy": 5.815191602706909,
      "epoch": 1.0301130342635112,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004900494462378902,
      "loss": 5.4655,
      "mean_token_accuracy": 0.19556693136692047,
      "num_tokens": 21750013.0,
      "step": 11665
    },
    {
      "entropy": 5.862353706359864,
      "epoch": 1.0305545743553515,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004900401825473854,
      "loss": 5.4884,
      "mean_token_accuracy": 0.19777150452136993,
      "num_tokens": 21758842.0,
      "step": 11670
    },
    {
      "entropy": 5.808915328979492,
      "epoch": 1.0309961144471917,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004900309146443658,
      "loss": 5.3693,
      "mean_token_accuracy": 0.20384867042303084,
      "num_tokens": 21767002.0,
      "step": 11675
    },
    {
      "entropy": 5.862798595428467,
      "epoch": 1.0314376545390322,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004900216425290128,
      "loss": 5.5014,
      "mean_token_accuracy": 0.19282409995794297,
      "num_tokens": 21775372.0,
      "step": 11680
    },
    {
      "entropy": 5.848205852508545,
      "epoch": 1.0318791946308725,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004900123662015082,
      "loss": 5.4869,
      "mean_token_accuracy": 0.1917622983455658,
      "num_tokens": 21784541.0,
      "step": 11685
    },
    {
      "entropy": 5.81615629196167,
      "epoch": 1.0323207347227128,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004900030856620336,
      "loss": 5.388,
      "mean_token_accuracy": 0.20575797706842422,
      "num_tokens": 21794525.0,
      "step": 11690
    },
    {
      "entropy": 5.841868925094604,
      "epoch": 1.032762274814553,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004899938009107708,
      "loss": 5.4658,
      "mean_token_accuracy": 0.20023476779460908,
      "num_tokens": 21804469.0,
      "step": 11695
    },
    {
      "entropy": 5.763099622726441,
      "epoch": 1.0332038149063936,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004899845119479019,
      "loss": 5.4179,
      "mean_token_accuracy": 0.20264574736356736,
      "num_tokens": 21813986.0,
      "step": 11700
    },
    {
      "entropy": 5.846244621276855,
      "epoch": 1.0336453549982338,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004899752187736085,
      "loss": 5.418,
      "mean_token_accuracy": 0.20018214583396912,
      "num_tokens": 21824083.0,
      "step": 11705
    },
    {
      "entropy": 5.750829124450684,
      "epoch": 1.0340868950900741,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004899659213880728,
      "loss": 5.3799,
      "mean_token_accuracy": 0.2069662794470787,
      "num_tokens": 21833526.0,
      "step": 11710
    },
    {
      "entropy": 5.792848777770996,
      "epoch": 1.0345284351819146,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004899566197914771,
      "loss": 5.3703,
      "mean_token_accuracy": 0.2027594715356827,
      "num_tokens": 21842537.0,
      "step": 11715
    },
    {
      "entropy": 5.841008043289184,
      "epoch": 1.034969975273755,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004899473139840033,
      "loss": 5.446,
      "mean_token_accuracy": 0.194618758559227,
      "num_tokens": 21851604.0,
      "step": 11720
    },
    {
      "entropy": 5.825651741027832,
      "epoch": 1.0354115153655952,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004899380039658339,
      "loss": 5.4928,
      "mean_token_accuracy": 0.19981683492660524,
      "num_tokens": 21861480.0,
      "step": 11725
    },
    {
      "entropy": 5.839599514007569,
      "epoch": 1.0358530554574354,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004899286897371513,
      "loss": 5.4713,
      "mean_token_accuracy": 0.1941080316901207,
      "num_tokens": 21870901.0,
      "step": 11730
    },
    {
      "entropy": 5.751971244812012,
      "epoch": 1.036294595549276,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000489919371298138,
      "loss": 5.3487,
      "mean_token_accuracy": 0.20988212525844574,
      "num_tokens": 21880491.0,
      "step": 11735
    },
    {
      "entropy": 5.80057053565979,
      "epoch": 1.0367361356411162,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004899100486489763,
      "loss": 5.4267,
      "mean_token_accuracy": 0.20379792600870134,
      "num_tokens": 21889833.0,
      "step": 11740
    },
    {
      "entropy": 5.779327392578125,
      "epoch": 1.0371776757329565,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004899007217898491,
      "loss": 5.4792,
      "mean_token_accuracy": 0.19909710586071014,
      "num_tokens": 21899222.0,
      "step": 11745
    },
    {
      "entropy": 5.846719264984131,
      "epoch": 1.037619215824797,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004898913907209389,
      "loss": 5.4592,
      "mean_token_accuracy": 0.196945820748806,
      "num_tokens": 21908848.0,
      "step": 11750
    },
    {
      "entropy": 5.823145055770874,
      "epoch": 1.0380607559166373,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004898820554424285,
      "loss": 5.4181,
      "mean_token_accuracy": 0.19045039564371108,
      "num_tokens": 21917867.0,
      "step": 11755
    },
    {
      "entropy": 5.874126863479614,
      "epoch": 1.0385022960084775,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000489872715954501,
      "loss": 5.4908,
      "mean_token_accuracy": 0.1879080832004547,
      "num_tokens": 21927476.0,
      "step": 11760
    },
    {
      "entropy": 5.924701070785522,
      "epoch": 1.0389438361003178,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004898633722573391,
      "loss": 5.5758,
      "mean_token_accuracy": 0.18321400731801987,
      "num_tokens": 21937068.0,
      "step": 11765
    },
    {
      "entropy": 5.808666515350342,
      "epoch": 1.0393853761921583,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000489854024351126,
      "loss": 5.4365,
      "mean_token_accuracy": 0.1953706905245781,
      "num_tokens": 21946144.0,
      "step": 11770
    },
    {
      "entropy": 5.7899901390075685,
      "epoch": 1.0398269162839986,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004898446722360446,
      "loss": 5.3694,
      "mean_token_accuracy": 0.2072113871574402,
      "num_tokens": 21955346.0,
      "step": 11775
    },
    {
      "entropy": 5.744083833694458,
      "epoch": 1.0402684563758389,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004898353159122784,
      "loss": 5.4414,
      "mean_token_accuracy": 0.20349046885967254,
      "num_tokens": 21965034.0,
      "step": 11780
    },
    {
      "entropy": 5.783421277999878,
      "epoch": 1.0407099964676794,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004898259553800106,
      "loss": 5.4154,
      "mean_token_accuracy": 0.19425492584705353,
      "num_tokens": 21975167.0,
      "step": 11785
    },
    {
      "entropy": 5.900732803344726,
      "epoch": 1.0411515365595196,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004898165906394244,
      "loss": 5.421,
      "mean_token_accuracy": 0.19738931059837342,
      "num_tokens": 21984372.0,
      "step": 11790
    },
    {
      "entropy": 5.803021955490112,
      "epoch": 1.04159307665136,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004898072216907033,
      "loss": 5.4264,
      "mean_token_accuracy": 0.1924193799495697,
      "num_tokens": 21993513.0,
      "step": 11795
    },
    {
      "entropy": 5.823993492126465,
      "epoch": 1.0420346167432002,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004897978485340311,
      "loss": 5.4697,
      "mean_token_accuracy": 0.19322787672281266,
      "num_tokens": 22002593.0,
      "step": 11800
    },
    {
      "entropy": 5.913989877700805,
      "epoch": 1.0424761568350407,
      "grad_norm": 0.953125,
      "learning_rate": 0.000489788471169591,
      "loss": 5.512,
      "mean_token_accuracy": 0.19518644213676453,
      "num_tokens": 22012296.0,
      "step": 11805
    },
    {
      "entropy": 5.894937324523926,
      "epoch": 1.042917696926881,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004897790895975671,
      "loss": 5.5292,
      "mean_token_accuracy": 0.19130600541830062,
      "num_tokens": 22021267.0,
      "step": 11810
    },
    {
      "entropy": 5.827191638946533,
      "epoch": 1.0433592370187212,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004897697038181428,
      "loss": 5.4924,
      "mean_token_accuracy": 0.19480232298374175,
      "num_tokens": 22030878.0,
      "step": 11815
    },
    {
      "entropy": 5.788354349136353,
      "epoch": 1.0438007771105617,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004897603138315022,
      "loss": 5.4657,
      "mean_token_accuracy": 0.19767487049102783,
      "num_tokens": 22040245.0,
      "step": 11820
    },
    {
      "entropy": 5.861823177337646,
      "epoch": 1.044242317202402,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004897509196378293,
      "loss": 5.4704,
      "mean_token_accuracy": 0.19682453721761703,
      "num_tokens": 22050169.0,
      "step": 11825
    },
    {
      "entropy": 5.900480031967163,
      "epoch": 1.0446838572942423,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000489741521237308,
      "loss": 5.4994,
      "mean_token_accuracy": 0.19175503700971602,
      "num_tokens": 22059168.0,
      "step": 11830
    },
    {
      "entropy": 5.776277828216553,
      "epoch": 1.0451253973860826,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004897321186301223,
      "loss": 5.4242,
      "mean_token_accuracy": 0.19125735759735107,
      "num_tokens": 22068326.0,
      "step": 11835
    },
    {
      "entropy": 5.794267559051514,
      "epoch": 1.045566937477923,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004897227118164566,
      "loss": 5.452,
      "mean_token_accuracy": 0.19609965533018112,
      "num_tokens": 22078026.0,
      "step": 11840
    },
    {
      "entropy": 5.873622989654541,
      "epoch": 1.0460084775697633,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000489713300796495,
      "loss": 5.4966,
      "mean_token_accuracy": 0.19001780599355697,
      "num_tokens": 22088230.0,
      "step": 11845
    },
    {
      "entropy": 5.760268783569336,
      "epoch": 1.0464500176616036,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004897038855704221,
      "loss": 5.3219,
      "mean_token_accuracy": 0.20597454011440278,
      "num_tokens": 22096667.0,
      "step": 11850
    },
    {
      "entropy": 5.770038747787476,
      "epoch": 1.0468915577534441,
      "grad_norm": 1.0,
      "learning_rate": 0.0004896944661384222,
      "loss": 5.4385,
      "mean_token_accuracy": 0.19639641493558885,
      "num_tokens": 22106871.0,
      "step": 11855
    },
    {
      "entropy": 5.861229133605957,
      "epoch": 1.0473330978452844,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004896850425006797,
      "loss": 5.3875,
      "mean_token_accuracy": 0.19390991032123567,
      "num_tokens": 22114954.0,
      "step": 11860
    },
    {
      "entropy": 5.826381063461303,
      "epoch": 1.0477746379371247,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004896756146573794,
      "loss": 5.4207,
      "mean_token_accuracy": 0.19825548231601714,
      "num_tokens": 22125293.0,
      "step": 11865
    },
    {
      "entropy": 5.866196918487549,
      "epoch": 1.048216178028965,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004896661826087059,
      "loss": 5.4628,
      "mean_token_accuracy": 0.1978798970580101,
      "num_tokens": 22133731.0,
      "step": 11870
    },
    {
      "entropy": 5.7895458221435545,
      "epoch": 1.0486577181208054,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004896567463548439,
      "loss": 5.4534,
      "mean_token_accuracy": 0.18647223711013794,
      "num_tokens": 22143007.0,
      "step": 11875
    },
    {
      "entropy": 5.852534818649292,
      "epoch": 1.0490992582126457,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004896473058959783,
      "loss": 5.4701,
      "mean_token_accuracy": 0.1962908148765564,
      "num_tokens": 22152044.0,
      "step": 11880
    },
    {
      "entropy": 5.888934659957886,
      "epoch": 1.049540798304486,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004896378612322942,
      "loss": 5.5378,
      "mean_token_accuracy": 0.19280146360397338,
      "num_tokens": 22161473.0,
      "step": 11885
    },
    {
      "entropy": 5.837153434753418,
      "epoch": 1.0499823383963265,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004896284123639763,
      "loss": 5.4335,
      "mean_token_accuracy": 0.19596049040555955,
      "num_tokens": 22171704.0,
      "step": 11890
    },
    {
      "entropy": 5.830124616622925,
      "epoch": 1.0504238784881668,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048961895929121,
      "loss": 5.4361,
      "mean_token_accuracy": 0.19259973168373107,
      "num_tokens": 22180593.0,
      "step": 11895
    },
    {
      "entropy": 5.813496732711792,
      "epoch": 1.050865418580007,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004896095020141802,
      "loss": 5.4278,
      "mean_token_accuracy": 0.2036044180393219,
      "num_tokens": 22189552.0,
      "step": 11900
    },
    {
      "entropy": 5.837124395370483,
      "epoch": 1.0513069586718473,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004896000405330724,
      "loss": 5.4551,
      "mean_token_accuracy": 0.19098978042602538,
      "num_tokens": 22199314.0,
      "step": 11905
    },
    {
      "entropy": 5.787033367156982,
      "epoch": 1.0517484987636878,
      "grad_norm": 1.0,
      "learning_rate": 0.000489590574848072,
      "loss": 5.3893,
      "mean_token_accuracy": 0.21002008467912675,
      "num_tokens": 22208128.0,
      "step": 11910
    },
    {
      "entropy": 5.725189113616944,
      "epoch": 1.052190038855528,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000489581104959364,
      "loss": 5.3562,
      "mean_token_accuracy": 0.20892212986946107,
      "num_tokens": 22217700.0,
      "step": 11915
    },
    {
      "entropy": 5.841599559783935,
      "epoch": 1.0526315789473684,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004895716308671343,
      "loss": 5.4628,
      "mean_token_accuracy": 0.18936695754528046,
      "num_tokens": 22225941.0,
      "step": 11920
    },
    {
      "entropy": 5.795419073104858,
      "epoch": 1.0530731190392089,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004895621525715685,
      "loss": 5.4098,
      "mean_token_accuracy": 0.1998206689953804,
      "num_tokens": 22234935.0,
      "step": 11925
    },
    {
      "entropy": 5.824654388427734,
      "epoch": 1.0535146591310491,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004895526700728521,
      "loss": 5.4884,
      "mean_token_accuracy": 0.19739733785390853,
      "num_tokens": 22244087.0,
      "step": 11930
    },
    {
      "entropy": 5.804003524780273,
      "epoch": 1.0539561992228894,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004895431833711708,
      "loss": 5.4452,
      "mean_token_accuracy": 0.19215094894170762,
      "num_tokens": 22254277.0,
      "step": 11935
    },
    {
      "entropy": 5.788774204254151,
      "epoch": 1.0543977393147297,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004895336924667107,
      "loss": 5.4065,
      "mean_token_accuracy": 0.2060361012816429,
      "num_tokens": 22263150.0,
      "step": 11940
    },
    {
      "entropy": 5.789840364456177,
      "epoch": 1.0548392794065702,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004895241973596576,
      "loss": 5.437,
      "mean_token_accuracy": 0.198403038084507,
      "num_tokens": 22271973.0,
      "step": 11945
    },
    {
      "entropy": 5.766195297241211,
      "epoch": 1.0552808194984105,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004895146980501973,
      "loss": 5.4104,
      "mean_token_accuracy": 0.20491848587989808,
      "num_tokens": 22280637.0,
      "step": 11950
    },
    {
      "entropy": 5.802606105804443,
      "epoch": 1.0557223595902507,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004895051945385163,
      "loss": 5.426,
      "mean_token_accuracy": 0.19119401425123214,
      "num_tokens": 22290601.0,
      "step": 11955
    },
    {
      "entropy": 5.851685810089111,
      "epoch": 1.0561638996820912,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004894956868248004,
      "loss": 5.454,
      "mean_token_accuracy": 0.19932712465524674,
      "num_tokens": 22300212.0,
      "step": 11960
    },
    {
      "entropy": 5.875369596481323,
      "epoch": 1.0566054397739315,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004894861749092359,
      "loss": 5.4377,
      "mean_token_accuracy": 0.19410779625177382,
      "num_tokens": 22309171.0,
      "step": 11965
    },
    {
      "entropy": 5.8332763671875,
      "epoch": 1.0570469798657718,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004894766587920094,
      "loss": 5.4529,
      "mean_token_accuracy": 0.20589271783828736,
      "num_tokens": 22318457.0,
      "step": 11970
    },
    {
      "entropy": 5.868374919891357,
      "epoch": 1.057488519957612,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000489467138473307,
      "loss": 5.4526,
      "mean_token_accuracy": 0.19353863447904587,
      "num_tokens": 22327077.0,
      "step": 11975
    },
    {
      "entropy": 5.820459318161011,
      "epoch": 1.0579300600494526,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004894576139533154,
      "loss": 5.443,
      "mean_token_accuracy": 0.19651559889316558,
      "num_tokens": 22336259.0,
      "step": 11980
    },
    {
      "entropy": 5.801056051254273,
      "epoch": 1.0583716001412928,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000489448085232221,
      "loss": 5.4227,
      "mean_token_accuracy": 0.20796411633491516,
      "num_tokens": 22344978.0,
      "step": 11985
    },
    {
      "entropy": 5.804754590988159,
      "epoch": 1.058813140233133,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004894385523102107,
      "loss": 5.4315,
      "mean_token_accuracy": 0.20169849544763566,
      "num_tokens": 22353950.0,
      "step": 11990
    },
    {
      "entropy": 5.834704208374023,
      "epoch": 1.0592546803249736,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004894290151874711,
      "loss": 5.3831,
      "mean_token_accuracy": 0.19994827657938002,
      "num_tokens": 22363100.0,
      "step": 11995
    },
    {
      "entropy": 5.760499382019043,
      "epoch": 1.0596962204168139,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000489419473864189,
      "loss": 5.3758,
      "mean_token_accuracy": 0.20482891499996186,
      "num_tokens": 22371979.0,
      "step": 12000
    },
    {
      "epoch": 1.0596962204168139,
      "eval_entropy": 5.5627229945980945,
      "eval_loss": 5.526999473571777,
      "eval_mean_token_accuracy": 0.20152225287742956,
      "eval_num_tokens": 22371979.0,
      "eval_runtime": 20.0722,
      "eval_samples_per_second": 1759.395,
      "eval_steps_per_second": 219.956,
      "step": 12000
    },
    {
      "entropy": 5.76117901802063,
      "epoch": 1.0601377605086542,
      "grad_norm": 1.0,
      "learning_rate": 0.0004894099283405515,
      "loss": 5.398,
      "mean_token_accuracy": 0.1994665563106537,
      "num_tokens": 22380987.0,
      "step": 12005
    },
    {
      "entropy": 5.820873498916626,
      "epoch": 1.0605793006004944,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004894003786167452,
      "loss": 5.4924,
      "mean_token_accuracy": 0.19171537905931474,
      "num_tokens": 22389367.0,
      "step": 12010
    },
    {
      "entropy": 5.860400533676147,
      "epoch": 1.061020840692335,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004893908246929576,
      "loss": 5.529,
      "mean_token_accuracy": 0.19122651368379592,
      "num_tokens": 22399158.0,
      "step": 12015
    },
    {
      "entropy": 5.865876626968384,
      "epoch": 1.0614623807841752,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004893812665693755,
      "loss": 5.5236,
      "mean_token_accuracy": 0.19261441379785538,
      "num_tokens": 22408245.0,
      "step": 12020
    },
    {
      "entropy": 5.787412929534912,
      "epoch": 1.0619039208760155,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004893717042461865,
      "loss": 5.4741,
      "mean_token_accuracy": 0.1920081853866577,
      "num_tokens": 22417014.0,
      "step": 12025
    },
    {
      "entropy": 5.860195589065552,
      "epoch": 1.062345460967856,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004893621377235777,
      "loss": 5.4167,
      "mean_token_accuracy": 0.1942832052707672,
      "num_tokens": 22426504.0,
      "step": 12030
    },
    {
      "entropy": 5.927928256988525,
      "epoch": 1.0627870010596963,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004893525670017364,
      "loss": 5.5063,
      "mean_token_accuracy": 0.18941380232572555,
      "num_tokens": 22435859.0,
      "step": 12035
    },
    {
      "entropy": 5.7999958992004395,
      "epoch": 1.0632285411515365,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004893429920808503,
      "loss": 5.4235,
      "mean_token_accuracy": 0.2001870334148407,
      "num_tokens": 22445112.0,
      "step": 12040
    },
    {
      "entropy": 5.772587394714355,
      "epoch": 1.0636700812433768,
      "grad_norm": 0.953125,
      "learning_rate": 0.000489333412961107,
      "loss": 5.4947,
      "mean_token_accuracy": 0.19539366364479066,
      "num_tokens": 22454602.0,
      "step": 12045
    },
    {
      "entropy": 5.907477617263794,
      "epoch": 1.0641116213352173,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004893238296426939,
      "loss": 5.5175,
      "mean_token_accuracy": 0.19273462295532226,
      "num_tokens": 22463918.0,
      "step": 12050
    },
    {
      "entropy": 5.788310766220093,
      "epoch": 1.0645531614270576,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004893142421257989,
      "loss": 5.3967,
      "mean_token_accuracy": 0.1981455311179161,
      "num_tokens": 22473059.0,
      "step": 12055
    },
    {
      "entropy": 5.795585441589355,
      "epoch": 1.0649947015188979,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004893046504106099,
      "loss": 5.4705,
      "mean_token_accuracy": 0.192001049220562,
      "num_tokens": 22482615.0,
      "step": 12060
    },
    {
      "entropy": 5.797793292999268,
      "epoch": 1.0654362416107384,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004892950544973145,
      "loss": 5.3531,
      "mean_token_accuracy": 0.2064661979675293,
      "num_tokens": 22492141.0,
      "step": 12065
    },
    {
      "entropy": 5.835184335708618,
      "epoch": 1.0658777817025786,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004892854543861009,
      "loss": 5.3335,
      "mean_token_accuracy": 0.20551654249429702,
      "num_tokens": 22502084.0,
      "step": 12070
    },
    {
      "entropy": 5.767578411102295,
      "epoch": 1.066319321794419,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004892758500771572,
      "loss": 5.3795,
      "mean_token_accuracy": 0.19579511880874634,
      "num_tokens": 22510565.0,
      "step": 12075
    },
    {
      "entropy": 5.784418725967408,
      "epoch": 1.0667608618862592,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004892662415706714,
      "loss": 5.4789,
      "mean_token_accuracy": 0.19152628928422927,
      "num_tokens": 22520089.0,
      "step": 12080
    },
    {
      "entropy": 5.802230167388916,
      "epoch": 1.0672024019780997,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004892566288668319,
      "loss": 5.408,
      "mean_token_accuracy": 0.1952907308936119,
      "num_tokens": 22529008.0,
      "step": 12085
    },
    {
      "entropy": 5.837222623825073,
      "epoch": 1.06764394206994,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004892470119658267,
      "loss": 5.4007,
      "mean_token_accuracy": 0.20102533251047133,
      "num_tokens": 22538040.0,
      "step": 12090
    },
    {
      "entropy": 5.809632682800293,
      "epoch": 1.0680854821617802,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004892373908678445,
      "loss": 5.453,
      "mean_token_accuracy": 0.19296614080667496,
      "num_tokens": 22548786.0,
      "step": 12095
    },
    {
      "entropy": 5.81814136505127,
      "epoch": 1.0685270222536207,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004892277655730736,
      "loss": 5.4443,
      "mean_token_accuracy": 0.19625641107559205,
      "num_tokens": 22559841.0,
      "step": 12100
    },
    {
      "entropy": 5.818791198730469,
      "epoch": 1.068968562345461,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004892181360817027,
      "loss": 5.3932,
      "mean_token_accuracy": 0.20388062298297882,
      "num_tokens": 22569338.0,
      "step": 12105
    },
    {
      "entropy": 5.7399450778961185,
      "epoch": 1.0694101024373013,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004892085023939204,
      "loss": 5.4412,
      "mean_token_accuracy": 0.1898152142763138,
      "num_tokens": 22578184.0,
      "step": 12110
    },
    {
      "entropy": 5.8836915493011475,
      "epoch": 1.0698516425291416,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004891988645099151,
      "loss": 5.5404,
      "mean_token_accuracy": 0.18727049976587296,
      "num_tokens": 22587007.0,
      "step": 12115
    },
    {
      "entropy": 5.89788327217102,
      "epoch": 1.070293182620982,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004891892224298761,
      "loss": 5.4614,
      "mean_token_accuracy": 0.19142129123210908,
      "num_tokens": 22596898.0,
      "step": 12120
    },
    {
      "entropy": 5.7848447322845455,
      "epoch": 1.0707347227128223,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000489179576153992,
      "loss": 5.4116,
      "mean_token_accuracy": 0.20285348743200302,
      "num_tokens": 22605627.0,
      "step": 12125
    },
    {
      "entropy": 5.75311050415039,
      "epoch": 1.0711762628046626,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004891699256824518,
      "loss": 5.3397,
      "mean_token_accuracy": 0.2017929509282112,
      "num_tokens": 22615269.0,
      "step": 12130
    },
    {
      "entropy": 5.791806364059449,
      "epoch": 1.071617802896503,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004891602710154446,
      "loss": 5.3771,
      "mean_token_accuracy": 0.2027752161026001,
      "num_tokens": 22625048.0,
      "step": 12135
    },
    {
      "entropy": 5.756436395645141,
      "epoch": 1.0720593429883434,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004891506121531593,
      "loss": 5.3711,
      "mean_token_accuracy": 0.20226978957653047,
      "num_tokens": 22634146.0,
      "step": 12140
    },
    {
      "entropy": 5.775600814819336,
      "epoch": 1.0725008830801837,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004891409490957855,
      "loss": 5.4698,
      "mean_token_accuracy": 0.19833989888429643,
      "num_tokens": 22642915.0,
      "step": 12145
    },
    {
      "entropy": 5.828043651580811,
      "epoch": 1.072942423172024,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004891312818435122,
      "loss": 5.4121,
      "mean_token_accuracy": 0.1949666276574135,
      "num_tokens": 22651546.0,
      "step": 12150
    },
    {
      "entropy": 5.841364526748658,
      "epoch": 1.0733839632638644,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000489121610396529,
      "loss": 5.4101,
      "mean_token_accuracy": 0.2037117898464203,
      "num_tokens": 22660712.0,
      "step": 12155
    },
    {
      "entropy": 5.8018913745880125,
      "epoch": 1.0738255033557047,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004891119347550251,
      "loss": 5.4237,
      "mean_token_accuracy": 0.20187718719244002,
      "num_tokens": 22669428.0,
      "step": 12160
    },
    {
      "entropy": 5.8284533500671385,
      "epoch": 1.074267043447545,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004891022549191902,
      "loss": 5.4883,
      "mean_token_accuracy": 0.1896545797586441,
      "num_tokens": 22679126.0,
      "step": 12165
    },
    {
      "entropy": 5.762653779983521,
      "epoch": 1.0747085835393855,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004890925708892138,
      "loss": 5.3786,
      "mean_token_accuracy": 0.2046400487422943,
      "num_tokens": 22688640.0,
      "step": 12170
    },
    {
      "entropy": 5.825605392456055,
      "epoch": 1.0751501236312258,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000489082882665286,
      "loss": 5.371,
      "mean_token_accuracy": 0.2033366784453392,
      "num_tokens": 22698158.0,
      "step": 12175
    },
    {
      "entropy": 5.729478168487549,
      "epoch": 1.075591663723066,
      "grad_norm": 0.984375,
      "learning_rate": 0.000489073190247596,
      "loss": 5.3131,
      "mean_token_accuracy": 0.2104171559214592,
      "num_tokens": 22707649.0,
      "step": 12180
    },
    {
      "entropy": 5.794720220565796,
      "epoch": 1.0760332038149063,
      "grad_norm": 1.0,
      "learning_rate": 0.000489063493636334,
      "loss": 5.4907,
      "mean_token_accuracy": 0.19447026252746583,
      "num_tokens": 22717123.0,
      "step": 12185
    },
    {
      "entropy": 5.749238109588623,
      "epoch": 1.0764747439067468,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004890537928316901,
      "loss": 5.4214,
      "mean_token_accuracy": 0.19723988473415374,
      "num_tokens": 22725921.0,
      "step": 12190
    },
    {
      "entropy": 5.85083327293396,
      "epoch": 1.076916283998587,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004890440878338538,
      "loss": 5.5186,
      "mean_token_accuracy": 0.19249518662691117,
      "num_tokens": 22735798.0,
      "step": 12195
    },
    {
      "entropy": 5.897446632385254,
      "epoch": 1.0773578240904274,
      "grad_norm": 1.0,
      "learning_rate": 0.0004890343786430158,
      "loss": 5.5028,
      "mean_token_accuracy": 0.19175300002098083,
      "num_tokens": 22745583.0,
      "step": 12200
    },
    {
      "entropy": 5.774647665023804,
      "epoch": 1.0777993641822678,
      "grad_norm": 1.03125,
      "learning_rate": 0.000489024665259366,
      "loss": 5.3356,
      "mean_token_accuracy": 0.2094974547624588,
      "num_tokens": 22754693.0,
      "step": 12205
    },
    {
      "entropy": 5.683815670013428,
      "epoch": 1.0782409042741081,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004890149476830947,
      "loss": 5.3428,
      "mean_token_accuracy": 0.206331467628479,
      "num_tokens": 22763416.0,
      "step": 12210
    },
    {
      "entropy": 5.8208612442016605,
      "epoch": 1.0786824443659484,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004890052259143924,
      "loss": 5.424,
      "mean_token_accuracy": 0.20380556285381318,
      "num_tokens": 22772411.0,
      "step": 12215
    },
    {
      "entropy": 5.7585056781768795,
      "epoch": 1.0791239844577887,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004889954999534493,
      "loss": 5.3263,
      "mean_token_accuracy": 0.2080024152994156,
      "num_tokens": 22780148.0,
      "step": 12220
    },
    {
      "entropy": 5.727291059494019,
      "epoch": 1.0795655245496292,
      "grad_norm": 1.0,
      "learning_rate": 0.0004889857698004561,
      "loss": 5.4002,
      "mean_token_accuracy": 0.19845105707645416,
      "num_tokens": 22789453.0,
      "step": 12225
    },
    {
      "entropy": 5.865949726104736,
      "epoch": 1.0800070646414694,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004889760354556033,
      "loss": 5.4346,
      "mean_token_accuracy": 0.20274429768323898,
      "num_tokens": 22798800.0,
      "step": 12230
    },
    {
      "entropy": 5.830954694747925,
      "epoch": 1.0804486047333097,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004889662969190817,
      "loss": 5.5088,
      "mean_token_accuracy": 0.19512779414653778,
      "num_tokens": 22807766.0,
      "step": 12235
    },
    {
      "entropy": 5.779366540908813,
      "epoch": 1.0808901448251502,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004889565541910822,
      "loss": 5.3753,
      "mean_token_accuracy": 0.1985454335808754,
      "num_tokens": 22817316.0,
      "step": 12240
    },
    {
      "entropy": 5.798041868209839,
      "epoch": 1.0813316849169905,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004889468072717953,
      "loss": 5.4642,
      "mean_token_accuracy": 0.2018450379371643,
      "num_tokens": 22827082.0,
      "step": 12245
    },
    {
      "entropy": 5.798942756652832,
      "epoch": 1.0817732250088308,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004889370561614122,
      "loss": 5.4404,
      "mean_token_accuracy": 0.1997752696275711,
      "num_tokens": 22836114.0,
      "step": 12250
    },
    {
      "entropy": 5.87620096206665,
      "epoch": 1.082214765100671,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004889273008601238,
      "loss": 5.5317,
      "mean_token_accuracy": 0.19290950894355774,
      "num_tokens": 22845653.0,
      "step": 12255
    },
    {
      "entropy": 5.8952672481536865,
      "epoch": 1.0826563051925115,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004889175413681213,
      "loss": 5.4682,
      "mean_token_accuracy": 0.19856464862823486,
      "num_tokens": 22855472.0,
      "step": 12260
    },
    {
      "entropy": 5.781057119369507,
      "epoch": 1.0830978452843518,
      "grad_norm": 1.0,
      "learning_rate": 0.0004889077776855957,
      "loss": 5.3743,
      "mean_token_accuracy": 0.2052726164460182,
      "num_tokens": 22864223.0,
      "step": 12265
    },
    {
      "entropy": 5.784585189819336,
      "epoch": 1.083539385376192,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004888980098127385,
      "loss": 5.4608,
      "mean_token_accuracy": 0.19553003907203675,
      "num_tokens": 22872868.0,
      "step": 12270
    },
    {
      "entropy": 5.870864582061768,
      "epoch": 1.0839809254680326,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004888882377497409,
      "loss": 5.4855,
      "mean_token_accuracy": 0.1898757666349411,
      "num_tokens": 22881867.0,
      "step": 12275
    },
    {
      "entropy": 5.829045009613037,
      "epoch": 1.0844224655598729,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004888784614967943,
      "loss": 5.4357,
      "mean_token_accuracy": 0.19417086243629456,
      "num_tokens": 22890904.0,
      "step": 12280
    },
    {
      "entropy": 5.743711328506469,
      "epoch": 1.0848640056517131,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004888686810540905,
      "loss": 5.3738,
      "mean_token_accuracy": 0.20171961784362794,
      "num_tokens": 22900128.0,
      "step": 12285
    },
    {
      "entropy": 5.797252750396728,
      "epoch": 1.0853055457435534,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004888588964218208,
      "loss": 5.4636,
      "mean_token_accuracy": 0.19641826003789903,
      "num_tokens": 22910101.0,
      "step": 12290
    },
    {
      "entropy": 5.815429210662842,
      "epoch": 1.085747085835394,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004888491076001768,
      "loss": 5.4332,
      "mean_token_accuracy": 0.19740083068609238,
      "num_tokens": 22919372.0,
      "step": 12295
    },
    {
      "entropy": 5.834107255935669,
      "epoch": 1.0861886259272342,
      "grad_norm": 1.0,
      "learning_rate": 0.0004888393145893505,
      "loss": 5.4362,
      "mean_token_accuracy": 0.19701868295669556,
      "num_tokens": 22929053.0,
      "step": 12300
    },
    {
      "entropy": 5.711127853393554,
      "epoch": 1.0866301660190745,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004888295173895335,
      "loss": 5.3759,
      "mean_token_accuracy": 0.19772911816835403,
      "num_tokens": 22938173.0,
      "step": 12305
    },
    {
      "entropy": 5.830015850067139,
      "epoch": 1.087071706110915,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004888197160009181,
      "loss": 5.5007,
      "mean_token_accuracy": 0.19715835601091386,
      "num_tokens": 22947076.0,
      "step": 12310
    },
    {
      "entropy": 5.806154012680054,
      "epoch": 1.0875132462027552,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000488809910423696,
      "loss": 5.422,
      "mean_token_accuracy": 0.20476145893335343,
      "num_tokens": 22956267.0,
      "step": 12315
    },
    {
      "entropy": 5.898472595214844,
      "epoch": 1.0879547862945955,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004888001006580592,
      "loss": 5.5187,
      "mean_token_accuracy": 0.18518722355365752,
      "num_tokens": 22966751.0,
      "step": 12320
    },
    {
      "entropy": 5.761488008499145,
      "epoch": 1.0883963263864358,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004887902867042002,
      "loss": 5.4251,
      "mean_token_accuracy": 0.20033389180898667,
      "num_tokens": 22975862.0,
      "step": 12325
    },
    {
      "entropy": 5.790075588226318,
      "epoch": 1.0888378664782763,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004887804685623111,
      "loss": 5.3893,
      "mean_token_accuracy": 0.19841696918010712,
      "num_tokens": 22985085.0,
      "step": 12330
    },
    {
      "entropy": 5.807966661453247,
      "epoch": 1.0892794065701166,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004887706462325841,
      "loss": 5.5217,
      "mean_token_accuracy": 0.19649834632873536,
      "num_tokens": 22993718.0,
      "step": 12335
    },
    {
      "entropy": 5.7946618556976315,
      "epoch": 1.0897209466619568,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004887608197152117,
      "loss": 5.3899,
      "mean_token_accuracy": 0.20841073095798493,
      "num_tokens": 23002611.0,
      "step": 12340
    },
    {
      "entropy": 5.859386825561524,
      "epoch": 1.0901624867537973,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004887509890103864,
      "loss": 5.5398,
      "mean_token_accuracy": 0.1920286536216736,
      "num_tokens": 23012333.0,
      "step": 12345
    },
    {
      "entropy": 5.856261110305786,
      "epoch": 1.0906040268456376,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004887411541183009,
      "loss": 5.4803,
      "mean_token_accuracy": 0.20197425931692123,
      "num_tokens": 23020998.0,
      "step": 12350
    },
    {
      "entropy": 5.842471599578857,
      "epoch": 1.091045566937478,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004887313150391477,
      "loss": 5.4516,
      "mean_token_accuracy": 0.19660209268331527,
      "num_tokens": 23029520.0,
      "step": 12355
    },
    {
      "entropy": 5.792888975143432,
      "epoch": 1.0914871070293182,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004887214717731196,
      "loss": 5.5197,
      "mean_token_accuracy": 0.1904631733894348,
      "num_tokens": 23039224.0,
      "step": 12360
    },
    {
      "entropy": 5.844955587387085,
      "epoch": 1.0919286471211587,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004887116243204094,
      "loss": 5.4067,
      "mean_token_accuracy": 0.1981590136885643,
      "num_tokens": 23047851.0,
      "step": 12365
    },
    {
      "entropy": 5.739581966400147,
      "epoch": 1.092370187212999,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048870177268121,
      "loss": 5.4264,
      "mean_token_accuracy": 0.19422738552093505,
      "num_tokens": 23057242.0,
      "step": 12370
    },
    {
      "entropy": 5.841652679443359,
      "epoch": 1.0928117273048392,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004886919168557145,
      "loss": 5.5422,
      "mean_token_accuracy": 0.19310810416936874,
      "num_tokens": 23066873.0,
      "step": 12375
    },
    {
      "entropy": 5.895767641067505,
      "epoch": 1.0932532673966797,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004886820568441158,
      "loss": 5.5146,
      "mean_token_accuracy": 0.1886664405465126,
      "num_tokens": 23076448.0,
      "step": 12380
    },
    {
      "entropy": 5.869763708114624,
      "epoch": 1.09369480748852,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004886721926466072,
      "loss": 5.4382,
      "mean_token_accuracy": 0.1931287795305252,
      "num_tokens": 23084759.0,
      "step": 12385
    },
    {
      "entropy": 5.808056640625,
      "epoch": 1.0941363475803603,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004886623242633817,
      "loss": 5.3883,
      "mean_token_accuracy": 0.200452321767807,
      "num_tokens": 23093236.0,
      "step": 12390
    },
    {
      "entropy": 5.792661809921265,
      "epoch": 1.0945778876722005,
      "grad_norm": 1.0,
      "learning_rate": 0.000488652451694633,
      "loss": 5.4178,
      "mean_token_accuracy": 0.20028265118598937,
      "num_tokens": 23101621.0,
      "step": 12395
    },
    {
      "entropy": 5.79798321723938,
      "epoch": 1.095019427764041,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004886425749405542,
      "loss": 5.4712,
      "mean_token_accuracy": 0.19113578498363495,
      "num_tokens": 23110882.0,
      "step": 12400
    },
    {
      "entropy": 5.850175237655639,
      "epoch": 1.0954609678558813,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004886326940013388,
      "loss": 5.4552,
      "mean_token_accuracy": 0.19971661865711213,
      "num_tokens": 23120669.0,
      "step": 12405
    },
    {
      "entropy": 5.820252561569214,
      "epoch": 1.0959025079477216,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004886228088771805,
      "loss": 5.3615,
      "mean_token_accuracy": 0.19930000305175782,
      "num_tokens": 23129524.0,
      "step": 12410
    },
    {
      "entropy": 5.752618837356567,
      "epoch": 1.096344048039562,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004886129195682729,
      "loss": 5.4056,
      "mean_token_accuracy": 0.20056315809488295,
      "num_tokens": 23139144.0,
      "step": 12415
    },
    {
      "entropy": 5.736331605911255,
      "epoch": 1.0967855881314024,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004886030260748096,
      "loss": 5.3601,
      "mean_token_accuracy": 0.199568709731102,
      "num_tokens": 23148600.0,
      "step": 12420
    },
    {
      "entropy": 5.717127895355224,
      "epoch": 1.0972271282232426,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004885931283969845,
      "loss": 5.3429,
      "mean_token_accuracy": 0.20439921617507933,
      "num_tokens": 23156831.0,
      "step": 12425
    },
    {
      "entropy": 5.88601131439209,
      "epoch": 1.097668668315083,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004885832265349915,
      "loss": 5.5168,
      "mean_token_accuracy": 0.18720456659793855,
      "num_tokens": 23166453.0,
      "step": 12430
    },
    {
      "entropy": 5.818928289413452,
      "epoch": 1.0981102084069234,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004885733204890246,
      "loss": 5.3952,
      "mean_token_accuracy": 0.19314671605825423,
      "num_tokens": 23175832.0,
      "step": 12435
    },
    {
      "entropy": 5.794879484176636,
      "epoch": 1.0985517484987637,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004885634102592778,
      "loss": 5.3893,
      "mean_token_accuracy": 0.19797425717115402,
      "num_tokens": 23185349.0,
      "step": 12440
    },
    {
      "entropy": 5.81860728263855,
      "epoch": 1.098993288590604,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004885534958459454,
      "loss": 5.5292,
      "mean_token_accuracy": 0.2007902055978775,
      "num_tokens": 23195391.0,
      "step": 12445
    },
    {
      "entropy": 5.7743274688720705,
      "epoch": 1.0994348286824445,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004885435772492213,
      "loss": 5.5042,
      "mean_token_accuracy": 0.1974121704697609,
      "num_tokens": 23204689.0,
      "step": 12450
    },
    {
      "entropy": 5.77637882232666,
      "epoch": 1.0998763687742847,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004885336544693,
      "loss": 5.4345,
      "mean_token_accuracy": 0.19952351003885269,
      "num_tokens": 23213182.0,
      "step": 12455
    },
    {
      "entropy": 5.736325311660766,
      "epoch": 1.100317908866125,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004885237275063758,
      "loss": 5.4312,
      "mean_token_accuracy": 0.19424163848161696,
      "num_tokens": 23222058.0,
      "step": 12460
    },
    {
      "entropy": 5.875259733200073,
      "epoch": 1.1007594489579653,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004885137963606433,
      "loss": 5.4669,
      "mean_token_accuracy": 0.2015385299921036,
      "num_tokens": 23231747.0,
      "step": 12465
    },
    {
      "entropy": 5.866737127304077,
      "epoch": 1.1012009890498058,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004885038610322969,
      "loss": 5.4063,
      "mean_token_accuracy": 0.19817569553852082,
      "num_tokens": 23240402.0,
      "step": 12470
    },
    {
      "entropy": 5.813427448272705,
      "epoch": 1.101642529141646,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004884939215215314,
      "loss": 5.4635,
      "mean_token_accuracy": 0.20315919071435928,
      "num_tokens": 23249344.0,
      "step": 12475
    },
    {
      "entropy": 5.760272884368897,
      "epoch": 1.1020840692334863,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004884839778285413,
      "loss": 5.4649,
      "mean_token_accuracy": 0.19652874320745467,
      "num_tokens": 23258157.0,
      "step": 12480
    },
    {
      "entropy": 5.8123131275177,
      "epoch": 1.1025256093253268,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004884740299535216,
      "loss": 5.4448,
      "mean_token_accuracy": 0.18557992577552795,
      "num_tokens": 23267798.0,
      "step": 12485
    },
    {
      "entropy": 5.865637493133545,
      "epoch": 1.1029671494171671,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000488464077896667,
      "loss": 5.485,
      "mean_token_accuracy": 0.19723096042871474,
      "num_tokens": 23276429.0,
      "step": 12490
    },
    {
      "entropy": 5.79170560836792,
      "epoch": 1.1034086895090074,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004884541216581724,
      "loss": 5.3605,
      "mean_token_accuracy": 0.20017392039299012,
      "num_tokens": 23285878.0,
      "step": 12495
    },
    {
      "entropy": 5.742550230026245,
      "epoch": 1.1038502296008477,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000488444161238233,
      "loss": 5.4062,
      "mean_token_accuracy": 0.20699722319841385,
      "num_tokens": 23295683.0,
      "step": 12500
    },
    {
      "entropy": 5.8427427291870115,
      "epoch": 1.1042917696926882,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004884341966370439,
      "loss": 5.4578,
      "mean_token_accuracy": 0.19892906546592712,
      "num_tokens": 23305539.0,
      "step": 12505
    },
    {
      "entropy": 5.772446918487549,
      "epoch": 1.1047333097845284,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004884242278548003,
      "loss": 5.4352,
      "mean_token_accuracy": 0.19846121519804,
      "num_tokens": 23314889.0,
      "step": 12510
    },
    {
      "entropy": 5.848331785202026,
      "epoch": 1.1051748498763687,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004884142548916975,
      "loss": 5.4896,
      "mean_token_accuracy": 0.1966036155819893,
      "num_tokens": 23324425.0,
      "step": 12515
    },
    {
      "entropy": 5.777439594268799,
      "epoch": 1.1056163899682092,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004884042777479308,
      "loss": 5.3814,
      "mean_token_accuracy": 0.19838157445192336,
      "num_tokens": 23332994.0,
      "step": 12520
    },
    {
      "entropy": 5.768653202056885,
      "epoch": 1.1060579300600495,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004883942964236957,
      "loss": 5.3175,
      "mean_token_accuracy": 0.20157167315483093,
      "num_tokens": 23342108.0,
      "step": 12525
    },
    {
      "entropy": 5.787688446044922,
      "epoch": 1.1064994701518898,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004883843109191877,
      "loss": 5.3548,
      "mean_token_accuracy": 0.2093011975288391,
      "num_tokens": 23351439.0,
      "step": 12530
    },
    {
      "entropy": 5.833836698532105,
      "epoch": 1.10694101024373,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004883743212346021,
      "loss": 5.4974,
      "mean_token_accuracy": 0.19144345372915267,
      "num_tokens": 23360591.0,
      "step": 12535
    },
    {
      "entropy": 5.824067211151123,
      "epoch": 1.1073825503355705,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004883643273701353,
      "loss": 5.4278,
      "mean_token_accuracy": 0.1985139235854149,
      "num_tokens": 23369854.0,
      "step": 12540
    },
    {
      "entropy": 5.834714078903199,
      "epoch": 1.1078240904274108,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004883543293259826,
      "loss": 5.5058,
      "mean_token_accuracy": 0.19206878244876863,
      "num_tokens": 23378071.0,
      "step": 12545
    },
    {
      "entropy": 5.8821605205535885,
      "epoch": 1.108265630519251,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004883443271023397,
      "loss": 5.4892,
      "mean_token_accuracy": 0.1873974084854126,
      "num_tokens": 23387083.0,
      "step": 12550
    },
    {
      "entropy": 5.770686149597168,
      "epoch": 1.1087071706110916,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000488334320699403,
      "loss": 5.3106,
      "mean_token_accuracy": 0.2036597326397896,
      "num_tokens": 23395008.0,
      "step": 12555
    },
    {
      "entropy": 5.762484693527222,
      "epoch": 1.1091487107029319,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004883243101173681,
      "loss": 5.4334,
      "mean_token_accuracy": 0.19761996120214462,
      "num_tokens": 23404193.0,
      "step": 12560
    },
    {
      "entropy": 5.816957092285156,
      "epoch": 1.1095902507947721,
      "grad_norm": 1.0,
      "learning_rate": 0.0004883142953564315,
      "loss": 5.4231,
      "mean_token_accuracy": 0.19985931515693664,
      "num_tokens": 23413353.0,
      "step": 12565
    },
    {
      "entropy": 5.853426456451416,
      "epoch": 1.1100317908866124,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000488304276416789,
      "loss": 5.4938,
      "mean_token_accuracy": 0.19763486087322235,
      "num_tokens": 23422797.0,
      "step": 12570
    },
    {
      "entropy": 5.828439140319825,
      "epoch": 1.110473330978453,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004882942532986372,
      "loss": 5.5575,
      "mean_token_accuracy": 0.19543415904045106,
      "num_tokens": 23432705.0,
      "step": 12575
    },
    {
      "entropy": 5.825581216812134,
      "epoch": 1.1109148710702932,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004882842260021721,
      "loss": 5.4109,
      "mean_token_accuracy": 0.19751026034355162,
      "num_tokens": 23441613.0,
      "step": 12580
    },
    {
      "entropy": 5.7938634872436525,
      "epoch": 1.1113564111621335,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048827419452759045,
      "loss": 5.4806,
      "mean_token_accuracy": 0.19247215688228608,
      "num_tokens": 23450975.0,
      "step": 12585
    },
    {
      "entropy": 5.847980928421021,
      "epoch": 1.111797951253974,
      "grad_norm": 1.0,
      "learning_rate": 0.00048826415887508857,
      "loss": 5.4778,
      "mean_token_accuracy": 0.19287350475788118,
      "num_tokens": 23460646.0,
      "step": 12590
    },
    {
      "entropy": 5.9207230567932125,
      "epoch": 1.1122394913458142,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048825411904486303,
      "loss": 5.5497,
      "mean_token_accuracy": 0.18543048352003097,
      "num_tokens": 23470454.0,
      "step": 12595
    },
    {
      "entropy": 5.771830892562866,
      "epoch": 1.1126810314376545,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004882440750371108,
      "loss": 5.4194,
      "mean_token_accuracy": 0.20379505902528763,
      "num_tokens": 23480013.0,
      "step": 12600
    },
    {
      "entropy": 5.859670925140381,
      "epoch": 1.1131225715294948,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048823402685202824,
      "loss": 5.4247,
      "mean_token_accuracy": 0.19428976625204086,
      "num_tokens": 23489485.0,
      "step": 12605
    },
    {
      "entropy": 5.772221231460572,
      "epoch": 1.1135641116213353,
      "grad_norm": 1.0,
      "learning_rate": 0.0004882239744898124,
      "loss": 5.4156,
      "mean_token_accuracy": 0.2106514871120453,
      "num_tokens": 23498147.0,
      "step": 12610
    },
    {
      "entropy": 5.678761100769043,
      "epoch": 1.1140056517131756,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004882139179506602,
      "loss": 5.2947,
      "mean_token_accuracy": 0.20896689891815184,
      "num_tokens": 23506614.0,
      "step": 12615
    },
    {
      "entropy": 5.903944253921509,
      "epoch": 1.1144471918050158,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004882038572347685,
      "loss": 5.5749,
      "mean_token_accuracy": 0.19001982361078262,
      "num_tokens": 23514927.0,
      "step": 12620
    },
    {
      "entropy": 5.88766074180603,
      "epoch": 1.1148887318968563,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048819379234233464,
      "loss": 5.5857,
      "mean_token_accuracy": 0.1913965031504631,
      "num_tokens": 23525298.0,
      "step": 12625
    },
    {
      "entropy": 5.938671970367432,
      "epoch": 1.1153302719886966,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004881837232735556,
      "loss": 5.486,
      "mean_token_accuracy": 0.19408829957246782,
      "num_tokens": 23533995.0,
      "step": 12630
    },
    {
      "entropy": 5.736942863464355,
      "epoch": 1.1157718120805369,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004881736500286286,
      "loss": 5.4361,
      "mean_token_accuracy": 0.2004469931125641,
      "num_tokens": 23543900.0,
      "step": 12635
    },
    {
      "entropy": 5.816015386581421,
      "epoch": 1.1162133521723772,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048816357260775115,
      "loss": 5.5347,
      "mean_token_accuracy": 0.19486338496208191,
      "num_tokens": 23553790.0,
      "step": 12640
    },
    {
      "entropy": 5.809583520889282,
      "epoch": 1.1166548922642177,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004881534910111204,
      "loss": 5.4425,
      "mean_token_accuracy": 0.20176970213651657,
      "num_tokens": 23563473.0,
      "step": 12645
    },
    {
      "entropy": 5.83355016708374,
      "epoch": 1.117096432356058,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048814340523893406,
      "loss": 5.3513,
      "mean_token_accuracy": 0.21166206598281861,
      "num_tokens": 23572495.0,
      "step": 12650
    },
    {
      "entropy": 5.870338773727417,
      "epoch": 1.1175379724478982,
      "grad_norm": 0.875,
      "learning_rate": 0.00048813331529138965,
      "loss": 5.525,
      "mean_token_accuracy": 0.18290655612945556,
      "num_tokens": 23582945.0,
      "step": 12655
    },
    {
      "entropy": 5.860627746582031,
      "epoch": 1.1179795125397387,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004881232211686848,
      "loss": 5.5324,
      "mean_token_accuracy": 0.19012165814638138,
      "num_tokens": 23592370.0,
      "step": 12660
    },
    {
      "entropy": 5.8514732837677,
      "epoch": 1.118421052631579,
      "grad_norm": 1.0,
      "learning_rate": 0.0004881131228710172,
      "loss": 5.4836,
      "mean_token_accuracy": 0.18865697979927062,
      "num_tokens": 23601589.0,
      "step": 12665
    },
    {
      "entropy": 5.815238523483276,
      "epoch": 1.1188625927234193,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004881030203985848,
      "loss": 5.3981,
      "mean_token_accuracy": 0.19706527441740035,
      "num_tokens": 23611456.0,
      "step": 12670
    },
    {
      "entropy": 5.828798818588257,
      "epoch": 1.1193041328152595,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048809291375158544,
      "loss": 5.4252,
      "mean_token_accuracy": 0.19195756763219834,
      "num_tokens": 23620362.0,
      "step": 12675
    },
    {
      "entropy": 5.8671183586120605,
      "epoch": 1.1197456729071,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048808280293021703,
      "loss": 5.4716,
      "mean_token_accuracy": 0.19548938572406768,
      "num_tokens": 23629647.0,
      "step": 12680
    },
    {
      "entropy": 5.781561756134034,
      "epoch": 1.1201872129989403,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048807268793467774,
      "loss": 5.3719,
      "mean_token_accuracy": 0.19756822139024735,
      "num_tokens": 23639201.0,
      "step": 12685
    },
    {
      "entropy": 5.771619558334351,
      "epoch": 1.1206287530907806,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048806256876516565,
      "loss": 5.3465,
      "mean_token_accuracy": 0.20281566083431243,
      "num_tokens": 23648166.0,
      "step": 12690
    },
    {
      "entropy": 5.842432594299316,
      "epoch": 1.121070293182621,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048805244542187906,
      "loss": 5.5464,
      "mean_token_accuracy": 0.19111703783273698,
      "num_tokens": 23658478.0,
      "step": 12695
    },
    {
      "entropy": 5.863788509368897,
      "epoch": 1.1215118332744614,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004880423179050161,
      "loss": 5.3773,
      "mean_token_accuracy": 0.20170405209064485,
      "num_tokens": 23667934.0,
      "step": 12700
    },
    {
      "entropy": 5.767390775680542,
      "epoch": 1.1219533733663016,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004880321862147755,
      "loss": 5.4125,
      "mean_token_accuracy": 0.1967923253774643,
      "num_tokens": 23676164.0,
      "step": 12705
    },
    {
      "entropy": 5.8280364036560055,
      "epoch": 1.122394913458142,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048802205035135537,
      "loss": 5.5141,
      "mean_token_accuracy": 0.1940636858344078,
      "num_tokens": 23686223.0,
      "step": 12710
    },
    {
      "entropy": 5.829761028289795,
      "epoch": 1.1228364535499824,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048801191031495444,
      "loss": 5.5021,
      "mean_token_accuracy": 0.19006700962781906,
      "num_tokens": 23695043.0,
      "step": 12715
    },
    {
      "entropy": 5.852439260482788,
      "epoch": 1.1232779936418227,
      "grad_norm": 1.0,
      "learning_rate": 0.0004880017661057714,
      "loss": 5.5161,
      "mean_token_accuracy": 0.18853760063648223,
      "num_tokens": 23704248.0,
      "step": 12720
    },
    {
      "entropy": 5.858014678955078,
      "epoch": 1.123719533733663,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004879916177240049,
      "loss": 5.3975,
      "mean_token_accuracy": 0.19637110084295273,
      "num_tokens": 23713724.0,
      "step": 12725
    },
    {
      "entropy": 5.9444338321685795,
      "epoch": 1.1241610738255035,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004879814651698537,
      "loss": 5.63,
      "mean_token_accuracy": 0.17654719054698945,
      "num_tokens": 23723039.0,
      "step": 12730
    },
    {
      "entropy": 5.799656867980957,
      "epoch": 1.1246026139173437,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048797130844351673,
      "loss": 5.3791,
      "mean_token_accuracy": 0.19843216091394425,
      "num_tokens": 23732024.0,
      "step": 12735
    },
    {
      "entropy": 5.818072175979614,
      "epoch": 1.125044154009184,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048796114754519294,
      "loss": 5.5055,
      "mean_token_accuracy": 0.1969237744808197,
      "num_tokens": 23740739.0,
      "step": 12740
    },
    {
      "entropy": 5.825812911987304,
      "epoch": 1.1254856941010243,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048795098247508145,
      "loss": 5.4227,
      "mean_token_accuracy": 0.1959856018424034,
      "num_tokens": 23750405.0,
      "step": 12745
    },
    {
      "entropy": 5.781939506530762,
      "epoch": 1.1259272341928648,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004879408132333813,
      "loss": 5.4088,
      "mean_token_accuracy": 0.19907349944114686,
      "num_tokens": 23759318.0,
      "step": 12750
    },
    {
      "entropy": 5.84839825630188,
      "epoch": 1.126368774284705,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048793063982029166,
      "loss": 5.5572,
      "mean_token_accuracy": 0.18858904838562013,
      "num_tokens": 23769202.0,
      "step": 12755
    },
    {
      "entropy": 5.808833837509155,
      "epoch": 1.1268103143765453,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000487920462236012,
      "loss": 5.3966,
      "mean_token_accuracy": 0.20316065698862076,
      "num_tokens": 23778478.0,
      "step": 12760
    },
    {
      "entropy": 5.830659866333008,
      "epoch": 1.1272518544683856,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004879102804807415,
      "loss": 5.5015,
      "mean_token_accuracy": 0.19673565477132798,
      "num_tokens": 23788044.0,
      "step": 12765
    },
    {
      "entropy": 5.824640274047852,
      "epoch": 1.127693394560226,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048790009455467967,
      "loss": 5.4895,
      "mean_token_accuracy": 0.18861141949892044,
      "num_tokens": 23796351.0,
      "step": 12770
    },
    {
      "entropy": 5.808917093276977,
      "epoch": 1.1281349346520664,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004878899044580261,
      "loss": 5.4903,
      "mean_token_accuracy": 0.19645975679159164,
      "num_tokens": 23805553.0,
      "step": 12775
    },
    {
      "entropy": 5.785381555557251,
      "epoch": 1.1285764747439067,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048787971019098044,
      "loss": 5.3402,
      "mean_token_accuracy": 0.2021778866648674,
      "num_tokens": 23813742.0,
      "step": 12780
    },
    {
      "entropy": 5.764051151275635,
      "epoch": 1.1290180148357472,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004878695117537423,
      "loss": 5.4249,
      "mean_token_accuracy": 0.1989069864153862,
      "num_tokens": 23822135.0,
      "step": 12785
    },
    {
      "entropy": 5.809045600891113,
      "epoch": 1.1294595549275874,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004878593091465114,
      "loss": 5.4892,
      "mean_token_accuracy": 0.19658098220825196,
      "num_tokens": 23832021.0,
      "step": 12790
    },
    {
      "entropy": 5.8431620597839355,
      "epoch": 1.1299010950194277,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004878491023694877,
      "loss": 5.437,
      "mean_token_accuracy": 0.1971711277961731,
      "num_tokens": 23842432.0,
      "step": 12795
    },
    {
      "entropy": 5.868091726303101,
      "epoch": 1.1303426351112682,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004878388914228713,
      "loss": 5.5085,
      "mean_token_accuracy": 0.1874347910284996,
      "num_tokens": 23852172.0,
      "step": 12800
    },
    {
      "entropy": 5.765841627120972,
      "epoch": 1.1307841752031085,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004878286763068619,
      "loss": 5.3804,
      "mean_token_accuracy": 0.20340241342782975,
      "num_tokens": 23861178.0,
      "step": 12805
    },
    {
      "entropy": 5.834781455993652,
      "epoch": 1.1312257152949488,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004878184570216598,
      "loss": 5.4792,
      "mean_token_accuracy": 0.19326757192611693,
      "num_tokens": 23869998.0,
      "step": 12810
    },
    {
      "entropy": 5.782759332656861,
      "epoch": 1.131667255386789,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048780823356746523,
      "loss": 5.4056,
      "mean_token_accuracy": 0.1978372737765312,
      "num_tokens": 23880404.0,
      "step": 12815
    },
    {
      "entropy": 5.795777750015259,
      "epoch": 1.1321087954786295,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004877980059444784,
      "loss": 5.3639,
      "mean_token_accuracy": 0.20314182341098785,
      "num_tokens": 23889643.0,
      "step": 12820
    },
    {
      "entropy": 5.888383579254151,
      "epoch": 1.1325503355704698,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004877877741528996,
      "loss": 5.437,
      "mean_token_accuracy": 0.19794621467590331,
      "num_tokens": 23899264.0,
      "step": 12825
    },
    {
      "entropy": 5.832461547851563,
      "epoch": 1.13299187566231,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004877775381929294,
      "loss": 5.5,
      "mean_token_accuracy": 0.19860878586769104,
      "num_tokens": 23908685.0,
      "step": 12830
    },
    {
      "entropy": 5.782173013687133,
      "epoch": 1.1334334157541504,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004877672980647682,
      "loss": 5.4654,
      "mean_token_accuracy": 0.1986728385090828,
      "num_tokens": 23917838.0,
      "step": 12835
    },
    {
      "entropy": 5.813170957565307,
      "epoch": 1.1338749558459909,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048775705376861663,
      "loss": 5.4574,
      "mean_token_accuracy": 0.1970720797777176,
      "num_tokens": 23927498.0,
      "step": 12840
    },
    {
      "entropy": 5.8912781238555905,
      "epoch": 1.1343164959378311,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004877468053046753,
      "loss": 5.479,
      "mean_token_accuracy": 0.20567873418331145,
      "num_tokens": 23936205.0,
      "step": 12845
    },
    {
      "entropy": 5.78642258644104,
      "epoch": 1.1347580360296714,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048773655267314516,
      "loss": 5.3761,
      "mean_token_accuracy": 0.2015999123454094,
      "num_tokens": 23945897.0,
      "step": 12850
    },
    {
      "entropy": 5.814234733581543,
      "epoch": 1.135199576121512,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004877262958742268,
      "loss": 5.4265,
      "mean_token_accuracy": 0.1943811669945717,
      "num_tokens": 23954873.0,
      "step": 12855
    },
    {
      "entropy": 5.825142288208008,
      "epoch": 1.1356411162133522,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048771603490812144,
      "loss": 5.5165,
      "mean_token_accuracy": 0.19840817600488664,
      "num_tokens": 23964631.0,
      "step": 12860
    },
    {
      "entropy": 5.790509223937988,
      "epoch": 1.1360826563051925,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004877057697750299,
      "loss": 5.4067,
      "mean_token_accuracy": 0.19818204790353774,
      "num_tokens": 23973935.0,
      "step": 12865
    },
    {
      "entropy": 5.870797109603882,
      "epoch": 1.136524196397033,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004876955004751533,
      "loss": 5.5083,
      "mean_token_accuracy": 0.19151266813278198,
      "num_tokens": 23984340.0,
      "step": 12870
    },
    {
      "entropy": 5.942030000686645,
      "epoch": 1.1369657364888732,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048768522700869284,
      "loss": 5.5811,
      "mean_token_accuracy": 0.18916677087545394,
      "num_tokens": 23994387.0,
      "step": 12875
    },
    {
      "entropy": 5.837050342559815,
      "epoch": 1.1374072765807135,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048767494937584964,
      "loss": 5.3994,
      "mean_token_accuracy": 0.20068531930446626,
      "num_tokens": 24003391.0,
      "step": 12880
    },
    {
      "entropy": 5.788455772399902,
      "epoch": 1.1378488166725538,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048766466757682527,
      "loss": 5.3702,
      "mean_token_accuracy": 0.2022908478975296,
      "num_tokens": 24013501.0,
      "step": 12885
    },
    {
      "entropy": 5.749296617507935,
      "epoch": 1.1382903567643943,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004876543816118209,
      "loss": 5.3519,
      "mean_token_accuracy": 0.20435864776372908,
      "num_tokens": 24022675.0,
      "step": 12890
    },
    {
      "entropy": 5.877848482131958,
      "epoch": 1.1387318968562345,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048764409148103815,
      "loss": 5.5236,
      "mean_token_accuracy": 0.18736222982406617,
      "num_tokens": 24032542.0,
      "step": 12895
    },
    {
      "entropy": 5.7748109817504885,
      "epoch": 1.1391734369480748,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004876337971846787,
      "loss": 5.4126,
      "mean_token_accuracy": 0.19638270884752274,
      "num_tokens": 24041342.0,
      "step": 12900
    },
    {
      "entropy": 5.769016361236572,
      "epoch": 1.139614977039915,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000487623498722944,
      "loss": 5.4094,
      "mean_token_accuracy": 0.2004293456673622,
      "num_tokens": 24051429.0,
      "step": 12905
    },
    {
      "entropy": 5.872424840927124,
      "epoch": 1.1400565171317556,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004876131960960358,
      "loss": 5.434,
      "mean_token_accuracy": 0.19928459078073502,
      "num_tokens": 24061074.0,
      "step": 12910
    },
    {
      "entropy": 5.837236070632935,
      "epoch": 1.1404980572235959,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048760288930415615,
      "loss": 5.45,
      "mean_token_accuracy": 0.1986504539847374,
      "num_tokens": 24069622.0,
      "step": 12915
    },
    {
      "entropy": 5.816545200347901,
      "epoch": 1.1409395973154361,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048759257834750673,
      "loss": 5.4509,
      "mean_token_accuracy": 0.19788366854190825,
      "num_tokens": 24078393.0,
      "step": 12920
    },
    {
      "entropy": 5.840530014038086,
      "epoch": 1.1413811374072766,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004875822632262897,
      "loss": 5.4457,
      "mean_token_accuracy": 0.19962743520736695,
      "num_tokens": 24089157.0,
      "step": 12925
    },
    {
      "entropy": 5.794300603866577,
      "epoch": 1.141822677499117,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048757194394070697,
      "loss": 5.3888,
      "mean_token_accuracy": 0.2025819480419159,
      "num_tokens": 24098022.0,
      "step": 12930
    },
    {
      "entropy": 5.736756801605225,
      "epoch": 1.1422642175909572,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048756162049096077,
      "loss": 5.3845,
      "mean_token_accuracy": 0.19877981543540954,
      "num_tokens": 24107629.0,
      "step": 12935
    },
    {
      "entropy": 5.840106248855591,
      "epoch": 1.1427057576827977,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048755129287725334,
      "loss": 5.4387,
      "mean_token_accuracy": 0.19928207397460937,
      "num_tokens": 24116903.0,
      "step": 12940
    },
    {
      "entropy": 5.804965496063232,
      "epoch": 1.143147297774638,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048754096109978684,
      "loss": 5.393,
      "mean_token_accuracy": 0.19839546978473663,
      "num_tokens": 24126433.0,
      "step": 12945
    },
    {
      "entropy": 5.881588792800903,
      "epoch": 1.1435888378664782,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004875306251587639,
      "loss": 5.5899,
      "mean_token_accuracy": 0.18981537818908692,
      "num_tokens": 24136603.0,
      "step": 12950
    },
    {
      "entropy": 5.862384891510009,
      "epoch": 1.1440303779583185,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004875202850543869,
      "loss": 5.478,
      "mean_token_accuracy": 0.19205156415700914,
      "num_tokens": 24145656.0,
      "step": 12955
    },
    {
      "entropy": 5.814039945602417,
      "epoch": 1.144471918050159,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048750994078685837,
      "loss": 5.5157,
      "mean_token_accuracy": 0.19038341641426088,
      "num_tokens": 24155044.0,
      "step": 12960
    },
    {
      "entropy": 5.813987731933594,
      "epoch": 1.1449134581419993,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004874995923563809,
      "loss": 5.428,
      "mean_token_accuracy": 0.19938230216503144,
      "num_tokens": 24163987.0,
      "step": 12965
    },
    {
      "entropy": 5.818346261978149,
      "epoch": 1.1453549982338396,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004874892397631573,
      "loss": 5.3913,
      "mean_token_accuracy": 0.19698755741119384,
      "num_tokens": 24173542.0,
      "step": 12970
    },
    {
      "entropy": 5.792830610275269,
      "epoch": 1.1457965383256798,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004874788830073904,
      "loss": 5.3938,
      "mean_token_accuracy": 0.19177045077085494,
      "num_tokens": 24182785.0,
      "step": 12975
    },
    {
      "entropy": 5.849016237258911,
      "epoch": 1.1462380784175203,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000487468522089283,
      "loss": 5.4595,
      "mean_token_accuracy": 0.19423977583646773,
      "num_tokens": 24193477.0,
      "step": 12980
    },
    {
      "entropy": 5.897625732421875,
      "epoch": 1.1466796185093606,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048745815700903806,
      "loss": 5.4289,
      "mean_token_accuracy": 0.19301562160253524,
      "num_tokens": 24202289.0,
      "step": 12985
    },
    {
      "entropy": 5.756063318252563,
      "epoch": 1.147121158601201,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048744778776685865,
      "loss": 5.3365,
      "mean_token_accuracy": 0.2100023940205574,
      "num_tokens": 24212199.0,
      "step": 12990
    },
    {
      "entropy": 5.833570289611816,
      "epoch": 1.1475626986930414,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048743741436294794,
      "loss": 5.4837,
      "mean_token_accuracy": 0.1986352488398552,
      "num_tokens": 24220984.0,
      "step": 12995
    },
    {
      "entropy": 5.813465881347656,
      "epoch": 1.1480042387848817,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004874270367975091,
      "loss": 5.4207,
      "mean_token_accuracy": 0.19989872574806214,
      "num_tokens": 24229599.0,
      "step": 13000
    },
    {
      "entropy": 5.848421335220337,
      "epoch": 1.148445778876722,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004874166550707454,
      "loss": 5.4741,
      "mean_token_accuracy": 0.19258294403553008,
      "num_tokens": 24238721.0,
      "step": 13005
    },
    {
      "entropy": 5.874206924438477,
      "epoch": 1.1488873189685624,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00048740626918286026,
      "loss": 5.4957,
      "mean_token_accuracy": 0.18595235794782639,
      "num_tokens": 24248431.0,
      "step": 13010
    },
    {
      "entropy": 5.847845363616943,
      "epoch": 1.1493288590604027,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048739587913405706,
      "loss": 5.4279,
      "mean_token_accuracy": 0.2011121317744255,
      "num_tokens": 24257770.0,
      "step": 13015
    },
    {
      "entropy": 5.872161769866944,
      "epoch": 1.149770399152243,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048738548492453944,
      "loss": 5.4185,
      "mean_token_accuracy": 0.20037979483604432,
      "num_tokens": 24266402.0,
      "step": 13020
    },
    {
      "entropy": 5.807389497756958,
      "epoch": 1.1502119392440833,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004873750865545109,
      "loss": 5.3918,
      "mean_token_accuracy": 0.200962033867836,
      "num_tokens": 24275164.0,
      "step": 13025
    },
    {
      "entropy": 5.817873430252075,
      "epoch": 1.1506534793359238,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048736468402417525,
      "loss": 5.4917,
      "mean_token_accuracy": 0.1918845921754837,
      "num_tokens": 24285650.0,
      "step": 13030
    },
    {
      "entropy": 5.835181093215942,
      "epoch": 1.151095019427764,
      "grad_norm": 1.0,
      "learning_rate": 0.00048735427733373617,
      "loss": 5.4529,
      "mean_token_accuracy": 0.1928752064704895,
      "num_tokens": 24294745.0,
      "step": 13035
    },
    {
      "entropy": 5.873867845535278,
      "epoch": 1.1515365595196043,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048734386648339767,
      "loss": 5.4198,
      "mean_token_accuracy": 0.20396704971790314,
      "num_tokens": 24303837.0,
      "step": 13040
    },
    {
      "entropy": 5.818719005584716,
      "epoch": 1.1519780996114446,
      "grad_norm": 1.0,
      "learning_rate": 0.0004873334514733634,
      "loss": 5.5055,
      "mean_token_accuracy": 0.19916922450065613,
      "num_tokens": 24313417.0,
      "step": 13045
    },
    {
      "entropy": 5.8238574981689455,
      "epoch": 1.152419639703285,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004873230323038378,
      "loss": 5.4209,
      "mean_token_accuracy": 0.20040738433599473,
      "num_tokens": 24322439.0,
      "step": 13050
    },
    {
      "entropy": 5.865116310119629,
      "epoch": 1.1528611797951254,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004873126089750246,
      "loss": 5.497,
      "mean_token_accuracy": 0.19367455840110778,
      "num_tokens": 24332509.0,
      "step": 13055
    },
    {
      "entropy": 5.831319856643677,
      "epoch": 1.1533027198869656,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004873021814871282,
      "loss": 5.5122,
      "mean_token_accuracy": 0.18654235303401948,
      "num_tokens": 24342252.0,
      "step": 13060
    },
    {
      "entropy": 5.872332239151001,
      "epoch": 1.1537442599788061,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004872917498403528,
      "loss": 5.4796,
      "mean_token_accuracy": 0.20358186960220337,
      "num_tokens": 24352459.0,
      "step": 13065
    },
    {
      "entropy": 5.791362237930298,
      "epoch": 1.1541858000706464,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004872813140349028,
      "loss": 5.3531,
      "mean_token_accuracy": 0.20301172137260437,
      "num_tokens": 24361788.0,
      "step": 13070
    },
    {
      "entropy": 5.795202779769897,
      "epoch": 1.1546273401624867,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004872708740709825,
      "loss": 5.4713,
      "mean_token_accuracy": 0.19351743459701537,
      "num_tokens": 24372117.0,
      "step": 13075
    },
    {
      "entropy": 5.746895122528076,
      "epoch": 1.1550688802543272,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048726042994879656,
      "loss": 5.4098,
      "mean_token_accuracy": 0.199046029150486,
      "num_tokens": 24380833.0,
      "step": 13080
    },
    {
      "entropy": 5.847612810134888,
      "epoch": 1.1555104203461675,
      "grad_norm": 1.0,
      "learning_rate": 0.00048724998166854944,
      "loss": 5.4902,
      "mean_token_accuracy": 0.19934733361005783,
      "num_tokens": 24390258.0,
      "step": 13085
    },
    {
      "entropy": 5.8716787815094,
      "epoch": 1.1559519604380077,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000487239529230446,
      "loss": 5.5235,
      "mean_token_accuracy": 0.19019153714179993,
      "num_tokens": 24399359.0,
      "step": 13090
    },
    {
      "entropy": 5.849550867080689,
      "epoch": 1.156393500529848,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048722907263469084,
      "loss": 5.4572,
      "mean_token_accuracy": 0.20486650615930557,
      "num_tokens": 24408012.0,
      "step": 13095
    },
    {
      "entropy": 5.8003052234649655,
      "epoch": 1.1568350406216885,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004872186118814889,
      "loss": 5.4607,
      "mean_token_accuracy": 0.19436182975769042,
      "num_tokens": 24416804.0,
      "step": 13100
    },
    {
      "entropy": 5.683313798904419,
      "epoch": 1.1572765807135288,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048720814697104497,
      "loss": 5.2753,
      "mean_token_accuracy": 0.2063358947634697,
      "num_tokens": 24425401.0,
      "step": 13105
    },
    {
      "entropy": 5.722922849655151,
      "epoch": 1.157718120805369,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048719767790356415,
      "loss": 5.3285,
      "mean_token_accuracy": 0.2067764699459076,
      "num_tokens": 24434473.0,
      "step": 13110
    },
    {
      "entropy": 5.92434253692627,
      "epoch": 1.1581596608972093,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048718720467925156,
      "loss": 5.5765,
      "mean_token_accuracy": 0.17944076806306838,
      "num_tokens": 24443154.0,
      "step": 13115
    },
    {
      "entropy": 5.810152959823609,
      "epoch": 1.1586012009890498,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004871767272983122,
      "loss": 5.399,
      "mean_token_accuracy": 0.20420801043510436,
      "num_tokens": 24451308.0,
      "step": 13120
    },
    {
      "entropy": 5.8501222133636475,
      "epoch": 1.1590427410808901,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004871662457609515,
      "loss": 5.4117,
      "mean_token_accuracy": 0.20110041052103042,
      "num_tokens": 24459997.0,
      "step": 13125
    },
    {
      "entropy": 5.81738691329956,
      "epoch": 1.1594842811727304,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004871557600673747,
      "loss": 5.4513,
      "mean_token_accuracy": 0.19508708715438844,
      "num_tokens": 24469448.0,
      "step": 13130
    },
    {
      "entropy": 5.780637216567993,
      "epoch": 1.1599258212645709,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048714527021778714,
      "loss": 5.4361,
      "mean_token_accuracy": 0.2001718685030937,
      "num_tokens": 24478304.0,
      "step": 13135
    },
    {
      "entropy": 5.796525287628174,
      "epoch": 1.1603673613564112,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004871347762123945,
      "loss": 5.3226,
      "mean_token_accuracy": 0.2066611498594284,
      "num_tokens": 24487300.0,
      "step": 13140
    },
    {
      "entropy": 5.790245914459229,
      "epoch": 1.1608089014482514,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048712427805140224,
      "loss": 5.4006,
      "mean_token_accuracy": 0.1998492196202278,
      "num_tokens": 24495893.0,
      "step": 13145
    },
    {
      "entropy": 5.792077970504761,
      "epoch": 1.161250441540092,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004871137757350159,
      "loss": 5.3971,
      "mean_token_accuracy": 0.19801437556743623,
      "num_tokens": 24504603.0,
      "step": 13150
    },
    {
      "entropy": 5.737908554077149,
      "epoch": 1.1616919816319322,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048710326926344153,
      "loss": 5.3665,
      "mean_token_accuracy": 0.2058656930923462,
      "num_tokens": 24513148.0,
      "step": 13155
    },
    {
      "entropy": 5.87685341835022,
      "epoch": 1.1621335217237725,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048709275863688455,
      "loss": 5.532,
      "mean_token_accuracy": 0.18852106034755706,
      "num_tokens": 24524006.0,
      "step": 13160
    },
    {
      "entropy": 5.790403509140015,
      "epoch": 1.1625750618156128,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048708224385555116,
      "loss": 5.3681,
      "mean_token_accuracy": 0.20684442520141602,
      "num_tokens": 24532468.0,
      "step": 13165
    },
    {
      "entropy": 5.831372356414795,
      "epoch": 1.1630166019074533,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004870717249196472,
      "loss": 5.4618,
      "mean_token_accuracy": 0.1951098173856735,
      "num_tokens": 24540887.0,
      "step": 13170
    },
    {
      "entropy": 5.873029279708862,
      "epoch": 1.1634581419992935,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004870612018293788,
      "loss": 5.4052,
      "mean_token_accuracy": 0.19676822125911714,
      "num_tokens": 24550565.0,
      "step": 13175
    },
    {
      "entropy": 5.828689861297607,
      "epoch": 1.1638996820911338,
      "grad_norm": 0.9375,
      "learning_rate": 0.000487050674584952,
      "loss": 5.3904,
      "mean_token_accuracy": 0.20156201124191284,
      "num_tokens": 24560283.0,
      "step": 13180
    },
    {
      "entropy": 5.754884147644043,
      "epoch": 1.164341222182974,
      "grad_norm": 1.109375,
      "learning_rate": 0.000487040143186573,
      "loss": 5.3399,
      "mean_token_accuracy": 0.20346760600805283,
      "num_tokens": 24569527.0,
      "step": 13185
    },
    {
      "entropy": 5.808266115188599,
      "epoch": 1.1647827622748146,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048702960763444833,
      "loss": 5.4317,
      "mean_token_accuracy": 0.19581593424081803,
      "num_tokens": 24578702.0,
      "step": 13190
    },
    {
      "entropy": 5.844560813903809,
      "epoch": 1.1652243023666549,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004870190679287842,
      "loss": 5.4059,
      "mean_token_accuracy": 0.20308234989643098,
      "num_tokens": 24588252.0,
      "step": 13195
    },
    {
      "entropy": 5.775644016265869,
      "epoch": 1.1656658424584951,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004870085240697871,
      "loss": 5.3954,
      "mean_token_accuracy": 0.19660432189702987,
      "num_tokens": 24596989.0,
      "step": 13200
    },
    {
      "entropy": 5.801691913604737,
      "epoch": 1.1661073825503356,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048699797605766365,
      "loss": 5.512,
      "mean_token_accuracy": 0.19366351664066314,
      "num_tokens": 24607051.0,
      "step": 13205
    },
    {
      "entropy": 5.912935543060303,
      "epoch": 1.166548922642176,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048698742389262036,
      "loss": 5.5146,
      "mean_token_accuracy": 0.19013242274522782,
      "num_tokens": 24616561.0,
      "step": 13210
    },
    {
      "entropy": 5.820238876342773,
      "epoch": 1.1669904627340162,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048697686757486397,
      "loss": 5.4595,
      "mean_token_accuracy": 0.1996126264333725,
      "num_tokens": 24625942.0,
      "step": 13215
    },
    {
      "entropy": 5.82216625213623,
      "epoch": 1.1674320028258567,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004869663071046013,
      "loss": 5.4288,
      "mean_token_accuracy": 0.19304953664541244,
      "num_tokens": 24635646.0,
      "step": 13220
    },
    {
      "entropy": 5.787486600875854,
      "epoch": 1.167873542917697,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048695574248203935,
      "loss": 5.4145,
      "mean_token_accuracy": 0.20401907563209534,
      "num_tokens": 24645147.0,
      "step": 13225
    },
    {
      "entropy": 5.8801429748535154,
      "epoch": 1.1683150830095372,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004869451737073848,
      "loss": 5.4355,
      "mean_token_accuracy": 0.19647326022386552,
      "num_tokens": 24654796.0,
      "step": 13230
    },
    {
      "entropy": 5.8170005798339846,
      "epoch": 1.1687566231013775,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004869346007808449,
      "loss": 5.4041,
      "mean_token_accuracy": 0.1959619030356407,
      "num_tokens": 24663765.0,
      "step": 13235
    },
    {
      "entropy": 5.8346545696258545,
      "epoch": 1.169198163193218,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048692402370262674,
      "loss": 5.4977,
      "mean_token_accuracy": 0.18846199810504913,
      "num_tokens": 24673758.0,
      "step": 13240
    },
    {
      "entropy": 5.835272121429443,
      "epoch": 1.1696397032850583,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048691344247293747,
      "loss": 5.3756,
      "mean_token_accuracy": 0.19807292222976686,
      "num_tokens": 24682751.0,
      "step": 13245
    },
    {
      "entropy": 5.814353513717651,
      "epoch": 1.1700812433768986,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048690285709198435,
      "loss": 5.466,
      "mean_token_accuracy": 0.19401862025260924,
      "num_tokens": 24691375.0,
      "step": 13250
    },
    {
      "entropy": 5.8145843029022215,
      "epoch": 1.1705227834687388,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004868922675599747,
      "loss": 5.4288,
      "mean_token_accuracy": 0.20230161547660827,
      "num_tokens": 24699852.0,
      "step": 13255
    },
    {
      "entropy": 5.785641479492187,
      "epoch": 1.1709643235605793,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004868816738771162,
      "loss": 5.3133,
      "mean_token_accuracy": 0.2019198477268219,
      "num_tokens": 24708857.0,
      "step": 13260
    },
    {
      "entropy": 5.882981061935425,
      "epoch": 1.1714058636524196,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004868710760436161,
      "loss": 5.5256,
      "mean_token_accuracy": 0.18815020024776458,
      "num_tokens": 24718262.0,
      "step": 13265
    },
    {
      "entropy": 5.868311262130737,
      "epoch": 1.1718474037442599,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048686047405968217,
      "loss": 5.4504,
      "mean_token_accuracy": 0.20123834460973739,
      "num_tokens": 24727458.0,
      "step": 13270
    },
    {
      "entropy": 5.921502304077149,
      "epoch": 1.1722889438361004,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048684986792552206,
      "loss": 5.5357,
      "mean_token_accuracy": 0.1842239499092102,
      "num_tokens": 24737297.0,
      "step": 13275
    },
    {
      "entropy": 5.888692378997803,
      "epoch": 1.1727304839279407,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048683925764134346,
      "loss": 5.4004,
      "mean_token_accuracy": 0.20313832759857178,
      "num_tokens": 24746939.0,
      "step": 13280
    },
    {
      "entropy": 5.844958543777466,
      "epoch": 1.173172024019781,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004868286432073543,
      "loss": 5.5455,
      "mean_token_accuracy": 0.19567487835884095,
      "num_tokens": 24756907.0,
      "step": 13285
    },
    {
      "entropy": 5.803437566757202,
      "epoch": 1.1736135641116214,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048681802462376253,
      "loss": 5.3591,
      "mean_token_accuracy": 0.20153522938489915,
      "num_tokens": 24765997.0,
      "step": 13290
    },
    {
      "entropy": 5.88543848991394,
      "epoch": 1.1740551042034617,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048680740189077605,
      "loss": 5.4523,
      "mean_token_accuracy": 0.1859741896390915,
      "num_tokens": 24775090.0,
      "step": 13295
    },
    {
      "entropy": 5.79713716506958,
      "epoch": 1.174496644295302,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048679677500860314,
      "loss": 5.4017,
      "mean_token_accuracy": 0.1957128718495369,
      "num_tokens": 24784709.0,
      "step": 13300
    },
    {
      "entropy": 5.894897937774658,
      "epoch": 1.1749381843871423,
      "grad_norm": 1.0,
      "learning_rate": 0.0004867861439774519,
      "loss": 5.4576,
      "mean_token_accuracy": 0.2024698406457901,
      "num_tokens": 24794473.0,
      "step": 13305
    },
    {
      "entropy": 5.848655700683594,
      "epoch": 1.1753797244789828,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048677550879753037,
      "loss": 5.5414,
      "mean_token_accuracy": 0.1884506344795227,
      "num_tokens": 24804718.0,
      "step": 13310
    },
    {
      "entropy": 5.8445494174957275,
      "epoch": 1.175821264570823,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048676486946904715,
      "loss": 5.4439,
      "mean_token_accuracy": 0.19045979976654054,
      "num_tokens": 24813632.0,
      "step": 13315
    },
    {
      "entropy": 5.774673843383789,
      "epoch": 1.1762628046626633,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004867542259922105,
      "loss": 5.353,
      "mean_token_accuracy": 0.20659865587949752,
      "num_tokens": 24823395.0,
      "step": 13320
    },
    {
      "entropy": 5.863358068466186,
      "epoch": 1.1767043447545038,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004867435783672291,
      "loss": 5.4796,
      "mean_token_accuracy": 0.1941860318183899,
      "num_tokens": 24832869.0,
      "step": 13325
    },
    {
      "entropy": 5.850523662567139,
      "epoch": 1.177145884846344,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004867329265943114,
      "loss": 5.4138,
      "mean_token_accuracy": 0.19960284531116484,
      "num_tokens": 24841706.0,
      "step": 13330
    },
    {
      "entropy": 5.869169807434082,
      "epoch": 1.1775874249381844,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004867222706736661,
      "loss": 5.4558,
      "mean_token_accuracy": 0.1981481581926346,
      "num_tokens": 24851149.0,
      "step": 13335
    },
    {
      "entropy": 5.851470041275024,
      "epoch": 1.1780289650300246,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004867116106055019,
      "loss": 5.4275,
      "mean_token_accuracy": 0.19862915128469466,
      "num_tokens": 24860002.0,
      "step": 13340
    },
    {
      "entropy": 5.807071018218994,
      "epoch": 1.1784705051218651,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048670094639002766,
      "loss": 5.4879,
      "mean_token_accuracy": 0.19347885251045227,
      "num_tokens": 24869457.0,
      "step": 13345
    },
    {
      "entropy": 5.80014934539795,
      "epoch": 1.1789120452137054,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004866902780274523,
      "loss": 5.4097,
      "mean_token_accuracy": 0.1995387315750122,
      "num_tokens": 24879527.0,
      "step": 13350
    },
    {
      "entropy": 5.80926661491394,
      "epoch": 1.1793535853055457,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004866796055179847,
      "loss": 5.43,
      "mean_token_accuracy": 0.2028554856777191,
      "num_tokens": 24889401.0,
      "step": 13355
    },
    {
      "entropy": 5.846627616882325,
      "epoch": 1.1797951253973862,
      "grad_norm": 1.0,
      "learning_rate": 0.0004866689288618342,
      "loss": 5.4328,
      "mean_token_accuracy": 0.2031128704547882,
      "num_tokens": 24897911.0,
      "step": 13360
    },
    {
      "entropy": 5.832061147689819,
      "epoch": 1.1802366654892265,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004866582480592096,
      "loss": 5.4038,
      "mean_token_accuracy": 0.20462207198143006,
      "num_tokens": 24906969.0,
      "step": 13365
    },
    {
      "entropy": 5.841529798507691,
      "epoch": 1.1806782055810667,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004866475631103203,
      "loss": 5.4327,
      "mean_token_accuracy": 0.20104020237922668,
      "num_tokens": 24915713.0,
      "step": 13370
    },
    {
      "entropy": 5.822256994247437,
      "epoch": 1.181119745672907,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048663687401537573,
      "loss": 5.4291,
      "mean_token_accuracy": 0.19380810260772705,
      "num_tokens": 24925484.0,
      "step": 13375
    },
    {
      "entropy": 5.839131879806518,
      "epoch": 1.1815612857647475,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048662618077458507,
      "loss": 5.4626,
      "mean_token_accuracy": 0.19699358493089675,
      "num_tokens": 24934751.0,
      "step": 13380
    },
    {
      "entropy": 5.82229175567627,
      "epoch": 1.1820028258565878,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004866154833881579,
      "loss": 5.5052,
      "mean_token_accuracy": 0.19295311272144317,
      "num_tokens": 24945059.0,
      "step": 13385
    },
    {
      "entropy": 5.831136798858642,
      "epoch": 1.182444365948428,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004866047818563038,
      "loss": 5.4574,
      "mean_token_accuracy": 0.19463883936405182,
      "num_tokens": 24954678.0,
      "step": 13390
    },
    {
      "entropy": 5.803921413421631,
      "epoch": 1.1828859060402686,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004865940761792323,
      "loss": 5.4535,
      "mean_token_accuracy": 0.19376982003450394,
      "num_tokens": 24964093.0,
      "step": 13395
    },
    {
      "entropy": 5.877642822265625,
      "epoch": 1.1833274461321088,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004865833663571533,
      "loss": 5.3945,
      "mean_token_accuracy": 0.205782550573349,
      "num_tokens": 24973903.0,
      "step": 13400
    },
    {
      "entropy": 5.778797292709351,
      "epoch": 1.183768986223949,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048657265239027635,
      "loss": 5.4001,
      "mean_token_accuracy": 0.20267367660999297,
      "num_tokens": 24982458.0,
      "step": 13405
    },
    {
      "entropy": 5.778594207763672,
      "epoch": 1.1842105263157894,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004865619342788116,
      "loss": 5.4687,
      "mean_token_accuracy": 0.1938798740506172,
      "num_tokens": 24992233.0,
      "step": 13410
    },
    {
      "entropy": 5.855686664581299,
      "epoch": 1.1846520664076299,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048655121202296873,
      "loss": 5.3952,
      "mean_token_accuracy": 0.19196417480707167,
      "num_tokens": 25001855.0,
      "step": 13415
    },
    {
      "entropy": 5.785927772521973,
      "epoch": 1.1850936064994702,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048654048562295807,
      "loss": 5.2949,
      "mean_token_accuracy": 0.2096828281879425,
      "num_tokens": 25010950.0,
      "step": 13420
    },
    {
      "entropy": 5.814057970046997,
      "epoch": 1.1855351465913104,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004865297550789895,
      "loss": 5.4715,
      "mean_token_accuracy": 0.18954818695783615,
      "num_tokens": 25020816.0,
      "step": 13425
    },
    {
      "entropy": 5.9006274223327635,
      "epoch": 1.185976686683151,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004865190203912734,
      "loss": 5.4941,
      "mean_token_accuracy": 0.19649789482355118,
      "num_tokens": 25030236.0,
      "step": 13430
    },
    {
      "entropy": 5.812299203872681,
      "epoch": 1.1864182267749912,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048650828156001995,
      "loss": 5.3908,
      "mean_token_accuracy": 0.19811263233423232,
      "num_tokens": 25040236.0,
      "step": 13435
    },
    {
      "entropy": 5.826809358596802,
      "epoch": 1.1868597668668315,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004864975385854396,
      "loss": 5.5173,
      "mean_token_accuracy": 0.19030961990356446,
      "num_tokens": 25049835.0,
      "step": 13440
    },
    {
      "entropy": 5.840577840805054,
      "epoch": 1.1873013069586718,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048648679146774275,
      "loss": 5.4732,
      "mean_token_accuracy": 0.19010272473096848,
      "num_tokens": 25058947.0,
      "step": 13445
    },
    {
      "entropy": 5.888551378250122,
      "epoch": 1.1877428470505123,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004864760402071399,
      "loss": 5.4357,
      "mean_token_accuracy": 0.1927935943007469,
      "num_tokens": 25069128.0,
      "step": 13450
    },
    {
      "entropy": 5.83781099319458,
      "epoch": 1.1881843871423525,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048646528480384175,
      "loss": 5.4931,
      "mean_token_accuracy": 0.18901222944259644,
      "num_tokens": 25079717.0,
      "step": 13455
    },
    {
      "entropy": 5.816900825500488,
      "epoch": 1.1886259272341928,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004864545252580588,
      "loss": 5.4085,
      "mean_token_accuracy": 0.2013822078704834,
      "num_tokens": 25088656.0,
      "step": 13460
    },
    {
      "entropy": 5.830460453033448,
      "epoch": 1.1890674673260333,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048644376157000213,
      "loss": 5.4242,
      "mean_token_accuracy": 0.19458205997943878,
      "num_tokens": 25098066.0,
      "step": 13465
    },
    {
      "entropy": 5.811070537567138,
      "epoch": 1.1895090074178736,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004864329937398823,
      "loss": 5.4383,
      "mean_token_accuracy": 0.19935468584299088,
      "num_tokens": 25107530.0,
      "step": 13470
    },
    {
      "entropy": 5.857369756698608,
      "epoch": 1.1899505475097139,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048642222176791043,
      "loss": 5.38,
      "mean_token_accuracy": 0.20258511751890182,
      "num_tokens": 25116226.0,
      "step": 13475
    },
    {
      "entropy": 5.8830688953399655,
      "epoch": 1.1903920876015541,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004864114456542975,
      "loss": 5.5573,
      "mean_token_accuracy": 0.18713556379079818,
      "num_tokens": 25125619.0,
      "step": 13480
    },
    {
      "entropy": 5.823305940628051,
      "epoch": 1.1908336276933946,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004864006653992546,
      "loss": 5.4095,
      "mean_token_accuracy": 0.20192137360572815,
      "num_tokens": 25134782.0,
      "step": 13485
    },
    {
      "entropy": 5.832695245742798,
      "epoch": 1.191275167785235,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048638988100299283,
      "loss": 5.4645,
      "mean_token_accuracy": 0.20090868026018144,
      "num_tokens": 25144709.0,
      "step": 13490
    },
    {
      "entropy": 5.8342750549316404,
      "epoch": 1.1917167078770752,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004863790924657236,
      "loss": 5.3343,
      "mean_token_accuracy": 0.1985476642847061,
      "num_tokens": 25153771.0,
      "step": 13495
    },
    {
      "entropy": 5.809533023834229,
      "epoch": 1.1921582479689157,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000486368299787658,
      "loss": 5.4937,
      "mean_token_accuracy": 0.19303596913814544,
      "num_tokens": 25163217.0,
      "step": 13500
    },
    {
      "entropy": 5.884714937210083,
      "epoch": 1.192599788060756,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048635750296900784,
      "loss": 5.4969,
      "mean_token_accuracy": 0.19041768461465836,
      "num_tokens": 25172762.0,
      "step": 13505
    },
    {
      "entropy": 5.850723314285278,
      "epoch": 1.1930413281525962,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004863467020099842,
      "loss": 5.4249,
      "mean_token_accuracy": 0.19885963052511216,
      "num_tokens": 25182349.0,
      "step": 13510
    },
    {
      "entropy": 5.875653553009033,
      "epoch": 1.1934828682444365,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048633589691079895,
      "loss": 5.4968,
      "mean_token_accuracy": 0.19478487372398376,
      "num_tokens": 25191971.0,
      "step": 13515
    },
    {
      "entropy": 5.709768342971802,
      "epoch": 1.193924408336277,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004863250876716637,
      "loss": 5.2704,
      "mean_token_accuracy": 0.20902594178915024,
      "num_tokens": 25200476.0,
      "step": 13520
    },
    {
      "entropy": 5.825209093093872,
      "epoch": 1.1943659484281173,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004863142742927902,
      "loss": 5.3754,
      "mean_token_accuracy": 0.19799334704875945,
      "num_tokens": 25209748.0,
      "step": 13525
    },
    {
      "entropy": 5.873579788208008,
      "epoch": 1.1948074885199576,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004863034567743901,
      "loss": 5.4729,
      "mean_token_accuracy": 0.19570437520742417,
      "num_tokens": 25220140.0,
      "step": 13530
    },
    {
      "entropy": 5.757551956176758,
      "epoch": 1.195249028611798,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004862926351166756,
      "loss": 5.3545,
      "mean_token_accuracy": 0.20685475468635559,
      "num_tokens": 25228424.0,
      "step": 13535
    },
    {
      "entropy": 5.827530431747436,
      "epoch": 1.1956905687036383,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004862818093198585,
      "loss": 5.4502,
      "mean_token_accuracy": 0.19304365217685698,
      "num_tokens": 25237448.0,
      "step": 13540
    },
    {
      "entropy": 5.762707901000977,
      "epoch": 1.1961321087954786,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004862709793841509,
      "loss": 5.3744,
      "mean_token_accuracy": 0.20597589612007142,
      "num_tokens": 25247124.0,
      "step": 13545
    },
    {
      "entropy": 5.749617958068848,
      "epoch": 1.1965736488873189,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048626014530976495,
      "loss": 5.4288,
      "mean_token_accuracy": 0.19768060892820358,
      "num_tokens": 25257363.0,
      "step": 13550
    },
    {
      "entropy": 5.8215718269348145,
      "epoch": 1.1970151889791594,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048624930709691296,
      "loss": 5.4882,
      "mean_token_accuracy": 0.19138520658016206,
      "num_tokens": 25266837.0,
      "step": 13555
    },
    {
      "entropy": 5.813835334777832,
      "epoch": 1.1974567290709996,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048623846474580716,
      "loss": 5.522,
      "mean_token_accuracy": 0.18580807745456696,
      "num_tokens": 25275767.0,
      "step": 13560
    },
    {
      "entropy": 5.833759355545044,
      "epoch": 1.19789826916284,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048622761825665994,
      "loss": 5.4664,
      "mean_token_accuracy": 0.19401149749755858,
      "num_tokens": 25285212.0,
      "step": 13565
    },
    {
      "entropy": 5.826594257354737,
      "epoch": 1.1983398092546804,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004862167676296839,
      "loss": 5.3651,
      "mean_token_accuracy": 0.1992584228515625,
      "num_tokens": 25294698.0,
      "step": 13570
    },
    {
      "entropy": 5.805987977981568,
      "epoch": 1.1987813493465207,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004862059128650915,
      "loss": 5.4243,
      "mean_token_accuracy": 0.20547815412282944,
      "num_tokens": 25304343.0,
      "step": 13575
    },
    {
      "entropy": 5.828811454772949,
      "epoch": 1.199222889438361,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048619505396309525,
      "loss": 5.3728,
      "mean_token_accuracy": 0.19790565669536592,
      "num_tokens": 25312542.0,
      "step": 13580
    },
    {
      "entropy": 5.837662649154663,
      "epoch": 1.1996644295302012,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004861841909239081,
      "loss": 5.468,
      "mean_token_accuracy": 0.1954321324825287,
      "num_tokens": 25322665.0,
      "step": 13585
    },
    {
      "entropy": 5.810902881622314,
      "epoch": 1.2001059696220417,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048617332374774276,
      "loss": 5.4302,
      "mean_token_accuracy": 0.19678546935319902,
      "num_tokens": 25332234.0,
      "step": 13590
    },
    {
      "entropy": 5.852847099304199,
      "epoch": 1.200547509713882,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048616245243481216,
      "loss": 5.4885,
      "mean_token_accuracy": 0.19284352660179138,
      "num_tokens": 25341462.0,
      "step": 13595
    },
    {
      "entropy": 5.844420433044434,
      "epoch": 1.2009890498057223,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048615157698532913,
      "loss": 5.4582,
      "mean_token_accuracy": 0.19950661957263946,
      "num_tokens": 25351771.0,
      "step": 13600
    },
    {
      "entropy": 5.7953503131866455,
      "epoch": 1.2014305898975628,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048614069739950683,
      "loss": 5.4097,
      "mean_token_accuracy": 0.1965176656842232,
      "num_tokens": 25360108.0,
      "step": 13605
    },
    {
      "entropy": 5.77955436706543,
      "epoch": 1.201872129989403,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004861298136775583,
      "loss": 5.4063,
      "mean_token_accuracy": 0.20105797052383423,
      "num_tokens": 25369635.0,
      "step": 13610
    },
    {
      "entropy": 5.78814663887024,
      "epoch": 1.2023136700812433,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004861189258196969,
      "loss": 5.3736,
      "mean_token_accuracy": 0.20672035217285156,
      "num_tokens": 25379623.0,
      "step": 13615
    },
    {
      "entropy": 5.868280029296875,
      "epoch": 1.2027552101730836,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004861080338261357,
      "loss": 5.4723,
      "mean_token_accuracy": 0.1960707888007164,
      "num_tokens": 25388591.0,
      "step": 13620
    },
    {
      "entropy": 5.8137904644012455,
      "epoch": 1.2031967502649241,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004860971376970882,
      "loss": 5.3815,
      "mean_token_accuracy": 0.20670562088489533,
      "num_tokens": 25397865.0,
      "step": 13625
    },
    {
      "entropy": 5.714167308807373,
      "epoch": 1.2036382903567644,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048608623743276794,
      "loss": 5.3254,
      "mean_token_accuracy": 0.204473876953125,
      "num_tokens": 25407143.0,
      "step": 13630
    },
    {
      "entropy": 5.8470540046691895,
      "epoch": 1.2040798304486047,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048607533303338823,
      "loss": 5.4608,
      "mean_token_accuracy": 0.19450926780700684,
      "num_tokens": 25415721.0,
      "step": 13635
    },
    {
      "entropy": 5.823157882690429,
      "epoch": 1.2045213705404452,
      "grad_norm": 1.0,
      "learning_rate": 0.00048606442449916284,
      "loss": 5.4912,
      "mean_token_accuracy": 0.20159126520156861,
      "num_tokens": 25425319.0,
      "step": 13640
    },
    {
      "entropy": 5.821358633041382,
      "epoch": 1.2049629106322854,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048605351183030537,
      "loss": 5.4381,
      "mean_token_accuracy": 0.20114302784204482,
      "num_tokens": 25435477.0,
      "step": 13645
    },
    {
      "entropy": 5.8079132556915285,
      "epoch": 1.2054044507241257,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004860425950270296,
      "loss": 5.4326,
      "mean_token_accuracy": 0.19563105404376985,
      "num_tokens": 25445633.0,
      "step": 13650
    },
    {
      "entropy": 5.82715630531311,
      "epoch": 1.205845990815966,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048603167408954953,
      "loss": 5.3841,
      "mean_token_accuracy": 0.20574440360069274,
      "num_tokens": 25455228.0,
      "step": 13655
    },
    {
      "entropy": 5.825229167938232,
      "epoch": 1.2062875309078065,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048602074901807894,
      "loss": 5.3951,
      "mean_token_accuracy": 0.19768795669078826,
      "num_tokens": 25463992.0,
      "step": 13660
    },
    {
      "entropy": 5.783743762969971,
      "epoch": 1.2067290709996468,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048600981981283185,
      "loss": 5.3778,
      "mean_token_accuracy": 0.20332661867141724,
      "num_tokens": 25472985.0,
      "step": 13665
    },
    {
      "entropy": 5.781201410293579,
      "epoch": 1.207170611091487,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004859988864740225,
      "loss": 5.3969,
      "mean_token_accuracy": 0.1971193641424179,
      "num_tokens": 25482165.0,
      "step": 13670
    },
    {
      "entropy": 5.787679672241211,
      "epoch": 1.2076121511833275,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004859879490018649,
      "loss": 5.4879,
      "mean_token_accuracy": 0.20230044424533844,
      "num_tokens": 25490643.0,
      "step": 13675
    },
    {
      "entropy": 5.849478387832642,
      "epoch": 1.2080536912751678,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004859770073965733,
      "loss": 5.3876,
      "mean_token_accuracy": 0.19996831715106964,
      "num_tokens": 25499954.0,
      "step": 13680
    },
    {
      "entropy": 5.858248329162597,
      "epoch": 1.208495231367008,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048596606165836223,
      "loss": 5.4277,
      "mean_token_accuracy": 0.19958026111125945,
      "num_tokens": 25509512.0,
      "step": 13685
    },
    {
      "entropy": 5.812077331542969,
      "epoch": 1.2089367714588484,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048595511178744594,
      "loss": 5.3457,
      "mean_token_accuracy": 0.198810675740242,
      "num_tokens": 25518909.0,
      "step": 13690
    },
    {
      "entropy": 5.803182697296142,
      "epoch": 1.2093783115506889,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000485944157784039,
      "loss": 5.4479,
      "mean_token_accuracy": 0.1966306149959564,
      "num_tokens": 25528221.0,
      "step": 13695
    },
    {
      "entropy": 5.84412260055542,
      "epoch": 1.2098198516425291,
      "grad_norm": 1.03125,
      "learning_rate": 0.000485933199648356,
      "loss": 5.4186,
      "mean_token_accuracy": 0.19448938667774202,
      "num_tokens": 25537428.0,
      "step": 13700
    },
    {
      "entropy": 5.828204774856568,
      "epoch": 1.2102613917343694,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048592223738061155,
      "loss": 5.4051,
      "mean_token_accuracy": 0.19483884423971176,
      "num_tokens": 25547608.0,
      "step": 13705
    },
    {
      "entropy": 5.83042197227478,
      "epoch": 1.21070293182621,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048591127098102045,
      "loss": 5.4949,
      "mean_token_accuracy": 0.1958155959844589,
      "num_tokens": 25557482.0,
      "step": 13710
    },
    {
      "entropy": 5.7836699962615965,
      "epoch": 1.2111444719180502,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004859003004497974,
      "loss": 5.3566,
      "mean_token_accuracy": 0.2017431765794754,
      "num_tokens": 25566967.0,
      "step": 13715
    },
    {
      "entropy": 5.862409543991089,
      "epoch": 1.2115860120098905,
      "grad_norm": 1.0,
      "learning_rate": 0.00048588932578715754,
      "loss": 5.3814,
      "mean_token_accuracy": 0.20105107128620148,
      "num_tokens": 25576114.0,
      "step": 13720
    },
    {
      "entropy": 5.899003696441651,
      "epoch": 1.2120275521017307,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048587834699331567,
      "loss": 5.6072,
      "mean_token_accuracy": 0.18123642206192017,
      "num_tokens": 25585953.0,
      "step": 13725
    },
    {
      "entropy": 5.890837287902832,
      "epoch": 1.2124690921935712,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000485867364068487,
      "loss": 5.5367,
      "mean_token_accuracy": 0.1876230001449585,
      "num_tokens": 25595445.0,
      "step": 13730
    },
    {
      "entropy": 5.896729278564453,
      "epoch": 1.2129106322854115,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048585637701288646,
      "loss": 5.4138,
      "mean_token_accuracy": 0.19927503615617753,
      "num_tokens": 25605883.0,
      "step": 13735
    },
    {
      "entropy": 5.8271716117858885,
      "epoch": 1.2133521723772518,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004858453858267295,
      "loss": 5.459,
      "mean_token_accuracy": 0.19328728169202805,
      "num_tokens": 25615653.0,
      "step": 13740
    },
    {
      "entropy": 5.766813802719116,
      "epoch": 1.2137937124690923,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004858343905102314,
      "loss": 5.3864,
      "mean_token_accuracy": 0.20176917165517808,
      "num_tokens": 25625043.0,
      "step": 13745
    },
    {
      "entropy": 5.84102349281311,
      "epoch": 1.2142352525609326,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048582339106360736,
      "loss": 5.4138,
      "mean_token_accuracy": 0.20222438126802444,
      "num_tokens": 25633642.0,
      "step": 13750
    },
    {
      "entropy": 5.866560459136963,
      "epoch": 1.2146767926527728,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048581238748707304,
      "loss": 5.4686,
      "mean_token_accuracy": 0.1934058651328087,
      "num_tokens": 25642154.0,
      "step": 13755
    },
    {
      "entropy": 5.808893013000488,
      "epoch": 1.2151183327446131,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000485801379780844,
      "loss": 5.3959,
      "mean_token_accuracy": 0.19736346006393432,
      "num_tokens": 25650250.0,
      "step": 13760
    },
    {
      "entropy": 5.797326946258545,
      "epoch": 1.2155598728364536,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048579036794513584,
      "loss": 5.4163,
      "mean_token_accuracy": 0.20288915634155275,
      "num_tokens": 25659789.0,
      "step": 13765
    },
    {
      "entropy": 5.815957260131836,
      "epoch": 1.216001412928294,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048577935198016413,
      "loss": 5.4441,
      "mean_token_accuracy": 0.19575237184762956,
      "num_tokens": 25668658.0,
      "step": 13770
    },
    {
      "entropy": 5.835956859588623,
      "epoch": 1.2164429530201342,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004857683318861449,
      "loss": 5.45,
      "mean_token_accuracy": 0.19296166449785232,
      "num_tokens": 25679082.0,
      "step": 13775
    },
    {
      "entropy": 5.858726739883423,
      "epoch": 1.2168844931119747,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048575730766329384,
      "loss": 5.4297,
      "mean_token_accuracy": 0.1935584530234337,
      "num_tokens": 25688408.0,
      "step": 13780
    },
    {
      "entropy": 5.808913040161133,
      "epoch": 1.217326033203815,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048574627931182704,
      "loss": 5.4032,
      "mean_token_accuracy": 0.20316696912050247,
      "num_tokens": 25697415.0,
      "step": 13785
    },
    {
      "entropy": 5.828490257263184,
      "epoch": 1.2177675732956552,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048573524683196047,
      "loss": 5.4758,
      "mean_token_accuracy": 0.1910461187362671,
      "num_tokens": 25706330.0,
      "step": 13790
    },
    {
      "entropy": 5.843984746932984,
      "epoch": 1.2182091133874955,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004857242102239102,
      "loss": 5.4232,
      "mean_token_accuracy": 0.1978504702448845,
      "num_tokens": 25716198.0,
      "step": 13795
    },
    {
      "entropy": 5.848089075088501,
      "epoch": 1.218650653479336,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004857131694878926,
      "loss": 5.4156,
      "mean_token_accuracy": 0.19115940928459169,
      "num_tokens": 25726486.0,
      "step": 13800
    },
    {
      "entropy": 5.811126375198365,
      "epoch": 1.2190921935711763,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048570212462412376,
      "loss": 5.3525,
      "mean_token_accuracy": 0.20375996381044387,
      "num_tokens": 25735559.0,
      "step": 13805
    },
    {
      "entropy": 5.820326900482177,
      "epoch": 1.2195337336630165,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004856910756328202,
      "loss": 5.4127,
      "mean_token_accuracy": 0.19793334752321243,
      "num_tokens": 25744781.0,
      "step": 13810
    },
    {
      "entropy": 5.76948356628418,
      "epoch": 1.219975273754857,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048568002251419835,
      "loss": 5.4159,
      "mean_token_accuracy": 0.20813079923391342,
      "num_tokens": 25753984.0,
      "step": 13815
    },
    {
      "entropy": 5.809472942352295,
      "epoch": 1.2204168138466973,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004856689652684746,
      "loss": 5.4952,
      "mean_token_accuracy": 0.19576571583747865,
      "num_tokens": 25763378.0,
      "step": 13820
    },
    {
      "entropy": 5.833141756057739,
      "epoch": 1.2208583539385376,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004856579038958656,
      "loss": 5.4169,
      "mean_token_accuracy": 0.19592608362436295,
      "num_tokens": 25771907.0,
      "step": 13825
    },
    {
      "entropy": 5.880925941467285,
      "epoch": 1.2212998940303779,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048564683839658805,
      "loss": 5.4578,
      "mean_token_accuracy": 0.20164652019739152,
      "num_tokens": 25780464.0,
      "step": 13830
    },
    {
      "entropy": 5.850726366043091,
      "epoch": 1.2217414341222184,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048563576877085887,
      "loss": 5.468,
      "mean_token_accuracy": 0.20249118357896806,
      "num_tokens": 25790300.0,
      "step": 13835
    },
    {
      "entropy": 5.788461065292358,
      "epoch": 1.2221829742140586,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004856246950188947,
      "loss": 5.3905,
      "mean_token_accuracy": 0.19526033848524094,
      "num_tokens": 25799103.0,
      "step": 13840
    },
    {
      "entropy": 5.857665348052978,
      "epoch": 1.222624514305899,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048561361714091257,
      "loss": 5.4535,
      "mean_token_accuracy": 0.19842367470264435,
      "num_tokens": 25808506.0,
      "step": 13845
    },
    {
      "entropy": 5.91711916923523,
      "epoch": 1.2230660543977394,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048560253513712945,
      "loss": 5.4255,
      "mean_token_accuracy": 0.19719596207141876,
      "num_tokens": 25817965.0,
      "step": 13850
    },
    {
      "entropy": 5.78755373954773,
      "epoch": 1.2235075944895797,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004855914490077624,
      "loss": 5.4066,
      "mean_token_accuracy": 0.19725599884986877,
      "num_tokens": 25827874.0,
      "step": 13855
    },
    {
      "entropy": 5.878895664215088,
      "epoch": 1.22394913458142,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048558035875302864,
      "loss": 5.5216,
      "mean_token_accuracy": 0.19557360112667083,
      "num_tokens": 25836840.0,
      "step": 13860
    },
    {
      "entropy": 5.84828290939331,
      "epoch": 1.2243906746732602,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004855692643731455,
      "loss": 5.4176,
      "mean_token_accuracy": 0.19867638498544693,
      "num_tokens": 25846344.0,
      "step": 13865
    },
    {
      "entropy": 5.772578763961792,
      "epoch": 1.2248322147651007,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048555816586833006,
      "loss": 5.3914,
      "mean_token_accuracy": 0.2070172682404518,
      "num_tokens": 25855336.0,
      "step": 13870
    },
    {
      "entropy": 5.882452440261841,
      "epoch": 1.225273754856941,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048554706323880007,
      "loss": 5.4558,
      "mean_token_accuracy": 0.1981833428144455,
      "num_tokens": 25864402.0,
      "step": 13875
    },
    {
      "entropy": 5.807854795455933,
      "epoch": 1.2257152949487813,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004855359564847728,
      "loss": 5.4173,
      "mean_token_accuracy": 0.20308926999568938,
      "num_tokens": 25873326.0,
      "step": 13880
    },
    {
      "entropy": 5.80874810218811,
      "epoch": 1.2261568350406218,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048552484560646585,
      "loss": 5.5079,
      "mean_token_accuracy": 0.18121581822633742,
      "num_tokens": 25882521.0,
      "step": 13885
    },
    {
      "entropy": 5.800187349319458,
      "epoch": 1.226598375132462,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048551373060409694,
      "loss": 5.3209,
      "mean_token_accuracy": 0.21059587299823762,
      "num_tokens": 25891087.0,
      "step": 13890
    },
    {
      "entropy": 5.84920916557312,
      "epoch": 1.2270399152243023,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048550261147788375,
      "loss": 5.3967,
      "mean_token_accuracy": 0.19528005123138428,
      "num_tokens": 25900757.0,
      "step": 13895
    },
    {
      "entropy": 5.8095996379852295,
      "epoch": 1.2274814553161426,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048549148822804415,
      "loss": 5.3965,
      "mean_token_accuracy": 0.19915473014116286,
      "num_tokens": 25909801.0,
      "step": 13900
    },
    {
      "entropy": 5.852218198776245,
      "epoch": 1.227922995407983,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000485480360854796,
      "loss": 5.523,
      "mean_token_accuracy": 0.18142039179801941,
      "num_tokens": 25919142.0,
      "step": 13905
    },
    {
      "entropy": 5.858808708190918,
      "epoch": 1.2283645354998234,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004854692293583573,
      "loss": 5.419,
      "mean_token_accuracy": 0.20021142810583115,
      "num_tokens": 25927560.0,
      "step": 13910
    },
    {
      "entropy": 5.9525049209594725,
      "epoch": 1.2288060755916637,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048545809373894605,
      "loss": 5.5223,
      "mean_token_accuracy": 0.1962352693080902,
      "num_tokens": 25937184.0,
      "step": 13915
    },
    {
      "entropy": 5.9449995994567875,
      "epoch": 1.2292476156835042,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048544695399678044,
      "loss": 5.533,
      "mean_token_accuracy": 0.18230526745319367,
      "num_tokens": 25946641.0,
      "step": 13920
    },
    {
      "entropy": 5.928172445297241,
      "epoch": 1.2296891557753444,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004854358101320787,
      "loss": 5.59,
      "mean_token_accuracy": 0.1917898952960968,
      "num_tokens": 25955910.0,
      "step": 13925
    },
    {
      "entropy": 5.818526697158814,
      "epoch": 1.2301306958671847,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048542466214505914,
      "loss": 5.4504,
      "mean_token_accuracy": 0.20166580975055695,
      "num_tokens": 25966580.0,
      "step": 13930
    },
    {
      "entropy": 5.803793811798096,
      "epoch": 1.230572235959025,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048541351003594015,
      "loss": 5.3308,
      "mean_token_accuracy": 0.2071426510810852,
      "num_tokens": 25975815.0,
      "step": 13935
    },
    {
      "entropy": 5.809789323806763,
      "epoch": 1.2310137760508655,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004854023538049401,
      "loss": 5.4045,
      "mean_token_accuracy": 0.20217425525188445,
      "num_tokens": 25984784.0,
      "step": 13940
    },
    {
      "entropy": 5.766007328033448,
      "epoch": 1.2314553161427058,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048539119345227774,
      "loss": 5.3754,
      "mean_token_accuracy": 0.20669020116329193,
      "num_tokens": 25994125.0,
      "step": 13945
    },
    {
      "entropy": 5.859079170227051,
      "epoch": 1.231896856234546,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004853800289781714,
      "loss": 5.3981,
      "mean_token_accuracy": 0.20159138441085817,
      "num_tokens": 26002876.0,
      "step": 13950
    },
    {
      "entropy": 5.79793529510498,
      "epoch": 1.2323383963263865,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048536886038284,
      "loss": 5.4347,
      "mean_token_accuracy": 0.2025657996535301,
      "num_tokens": 26012591.0,
      "step": 13955
    },
    {
      "entropy": 5.913652467727661,
      "epoch": 1.2327799364182268,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048535768766650225,
      "loss": 5.5556,
      "mean_token_accuracy": 0.1827597662806511,
      "num_tokens": 26022107.0,
      "step": 13960
    },
    {
      "entropy": 5.863278150558472,
      "epoch": 1.233221476510067,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000485346510829377,
      "loss": 5.5235,
      "mean_token_accuracy": 0.20024748295545577,
      "num_tokens": 26031035.0,
      "step": 13965
    },
    {
      "entropy": 5.81165657043457,
      "epoch": 1.2336630166019074,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048533532987168334,
      "loss": 5.3259,
      "mean_token_accuracy": 0.21129488497972487,
      "num_tokens": 26039980.0,
      "step": 13970
    },
    {
      "entropy": 5.906415891647339,
      "epoch": 1.2341045566937479,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048532414479364016,
      "loss": 5.5538,
      "mean_token_accuracy": 0.18444538563489915,
      "num_tokens": 26049389.0,
      "step": 13975
    },
    {
      "entropy": 5.827001476287842,
      "epoch": 1.2345460967855881,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004853129555954666,
      "loss": 5.383,
      "mean_token_accuracy": 0.20637040138244628,
      "num_tokens": 26058868.0,
      "step": 13980
    },
    {
      "entropy": 5.871132802963257,
      "epoch": 1.2349876368774284,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004853017622773818,
      "loss": 5.4582,
      "mean_token_accuracy": 0.19470107704401016,
      "num_tokens": 26068025.0,
      "step": 13985
    },
    {
      "entropy": 5.895386600494385,
      "epoch": 1.235429176969269,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004852905648396051,
      "loss": 5.4353,
      "mean_token_accuracy": 0.1970718115568161,
      "num_tokens": 26076410.0,
      "step": 13990
    },
    {
      "entropy": 5.795963191986084,
      "epoch": 1.2358707170611092,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048527936328235583,
      "loss": 5.3736,
      "mean_token_accuracy": 0.20549770444631577,
      "num_tokens": 26085624.0,
      "step": 13995
    },
    {
      "entropy": 5.846537828445435,
      "epoch": 1.2363122571529495,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004852681576058534,
      "loss": 5.4878,
      "mean_token_accuracy": 0.19413530230522155,
      "num_tokens": 26094898.0,
      "step": 14000
    },
    {
      "entropy": 5.866560268402099,
      "epoch": 1.2367537972447897,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004852569478103174,
      "loss": 5.4536,
      "mean_token_accuracy": 0.1948859751224518,
      "num_tokens": 26105041.0,
      "step": 14005
    },
    {
      "entropy": 5.862928771972657,
      "epoch": 1.2371953373366302,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048524573389596736,
      "loss": 5.4019,
      "mean_token_accuracy": 0.20501069873571395,
      "num_tokens": 26114140.0,
      "step": 14010
    },
    {
      "entropy": 5.724464416503906,
      "epoch": 1.2376368774284705,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048523451586302295,
      "loss": 5.2575,
      "mean_token_accuracy": 0.2127821758389473,
      "num_tokens": 26122611.0,
      "step": 14015
    },
    {
      "entropy": 5.726232576370239,
      "epoch": 1.2380784175203108,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000485223293711704,
      "loss": 5.3401,
      "mean_token_accuracy": 0.20355012714862825,
      "num_tokens": 26131548.0,
      "step": 14020
    },
    {
      "entropy": 5.867557430267334,
      "epoch": 1.2385199576121513,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048521206744223024,
      "loss": 5.3474,
      "mean_token_accuracy": 0.1960148885846138,
      "num_tokens": 26139806.0,
      "step": 14025
    },
    {
      "entropy": 5.912551498413086,
      "epoch": 1.2389614977039916,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048520083705482165,
      "loss": 5.4706,
      "mean_token_accuracy": 0.191637521982193,
      "num_tokens": 26148960.0,
      "step": 14030
    },
    {
      "entropy": 5.831079769134521,
      "epoch": 1.2394030377958318,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004851896025496982,
      "loss": 5.4644,
      "mean_token_accuracy": 0.19882910400629045,
      "num_tokens": 26158498.0,
      "step": 14035
    },
    {
      "entropy": 5.810053777694702,
      "epoch": 1.239844577887672,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048517836392707994,
      "loss": 5.3223,
      "mean_token_accuracy": 0.20472958981990813,
      "num_tokens": 26167319.0,
      "step": 14040
    },
    {
      "entropy": 5.78435640335083,
      "epoch": 1.2402861179795126,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048516712118718713,
      "loss": 5.3455,
      "mean_token_accuracy": 0.2067430794239044,
      "num_tokens": 26176059.0,
      "step": 14045
    },
    {
      "entropy": 5.787687063217163,
      "epoch": 1.2407276580713529,
      "grad_norm": 1.0,
      "learning_rate": 0.0004851558743302399,
      "loss": 5.3743,
      "mean_token_accuracy": 0.20237334966659545,
      "num_tokens": 26185102.0,
      "step": 14050
    },
    {
      "entropy": 5.801526641845703,
      "epoch": 1.2411691981631932,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004851446233564586,
      "loss": 5.3667,
      "mean_token_accuracy": 0.20010706633329392,
      "num_tokens": 26194873.0,
      "step": 14055
    },
    {
      "entropy": 5.849619340896607,
      "epoch": 1.2416107382550337,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004851333682660637,
      "loss": 5.4191,
      "mean_token_accuracy": 0.19707386046648026,
      "num_tokens": 26205362.0,
      "step": 14060
    },
    {
      "entropy": 5.77091875076294,
      "epoch": 1.242052278346874,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048512210905927555,
      "loss": 5.3298,
      "mean_token_accuracy": 0.20205124616622924,
      "num_tokens": 26214266.0,
      "step": 14065
    },
    {
      "entropy": 5.8740599155426025,
      "epoch": 1.2424938184387142,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004851108457363149,
      "loss": 5.491,
      "mean_token_accuracy": 0.1976918712258339,
      "num_tokens": 26223069.0,
      "step": 14070
    },
    {
      "entropy": 5.791577863693237,
      "epoch": 1.2429353585305545,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004850995782974022,
      "loss": 5.4352,
      "mean_token_accuracy": 0.1926703244447708,
      "num_tokens": 26232547.0,
      "step": 14075
    },
    {
      "entropy": 5.906997108459473,
      "epoch": 1.243376898622395,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004850883067427582,
      "loss": 5.4107,
      "mean_token_accuracy": 0.1996164157986641,
      "num_tokens": 26241014.0,
      "step": 14080
    },
    {
      "entropy": 5.804035615921021,
      "epoch": 1.2438184387142353,
      "grad_norm": 0.890625,
      "learning_rate": 0.00048507703107260384,
      "loss": 5.4503,
      "mean_token_accuracy": 0.20347474366426468,
      "num_tokens": 26250784.0,
      "step": 14085
    },
    {
      "entropy": 5.815296602249146,
      "epoch": 1.2442599788060755,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048506575128715984,
      "loss": 5.3905,
      "mean_token_accuracy": 0.20030366480350495,
      "num_tokens": 26259769.0,
      "step": 14090
    },
    {
      "entropy": 5.847494602203369,
      "epoch": 1.244701518897916,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004850544673866474,
      "loss": 5.4149,
      "mean_token_accuracy": 0.20297379940748214,
      "num_tokens": 26269038.0,
      "step": 14095
    },
    {
      "entropy": 5.837285900115967,
      "epoch": 1.2451430589897563,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004850431793712873,
      "loss": 5.3808,
      "mean_token_accuracy": 0.19335044771432877,
      "num_tokens": 26278657.0,
      "step": 14100
    },
    {
      "entropy": 5.765714168548584,
      "epoch": 1.2455845990815966,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048503188724130075,
      "loss": 5.3864,
      "mean_token_accuracy": 0.20418491065502167,
      "num_tokens": 26287458.0,
      "step": 14105
    },
    {
      "entropy": 5.852335691452026,
      "epoch": 1.2460261391734369,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000485020590996909,
      "loss": 5.4659,
      "mean_token_accuracy": 0.19273151755332946,
      "num_tokens": 26296324.0,
      "step": 14110
    },
    {
      "entropy": 5.834042358398437,
      "epoch": 1.2464676792652774,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048500929063833334,
      "loss": 5.3778,
      "mean_token_accuracy": 0.20434997975826263,
      "num_tokens": 26304799.0,
      "step": 14115
    },
    {
      "entropy": 5.866523027420044,
      "epoch": 1.2469092193571176,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048499798616579505,
      "loss": 5.4835,
      "mean_token_accuracy": 0.19575916081666947,
      "num_tokens": 26314223.0,
      "step": 14120
    },
    {
      "entropy": 5.9105627059936525,
      "epoch": 1.247350759448958,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00048498667757951574,
      "loss": 5.4739,
      "mean_token_accuracy": 0.18705244809389115,
      "num_tokens": 26323599.0,
      "step": 14125
    },
    {
      "entropy": 5.8942108154296875,
      "epoch": 1.2477922995407984,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004849753648797168,
      "loss": 5.5336,
      "mean_token_accuracy": 0.18914407938718797,
      "num_tokens": 26331929.0,
      "step": 14130
    },
    {
      "entropy": 5.80018048286438,
      "epoch": 1.2482338396326387,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004849640480666199,
      "loss": 5.4078,
      "mean_token_accuracy": 0.19961749017238617,
      "num_tokens": 26340929.0,
      "step": 14135
    },
    {
      "entropy": 5.8703961849212645,
      "epoch": 1.248675379724479,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004849527271404467,
      "loss": 5.3918,
      "mean_token_accuracy": 0.19771150201559068,
      "num_tokens": 26350393.0,
      "step": 14140
    },
    {
      "entropy": 5.902312231063843,
      "epoch": 1.2491169198163192,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048494140210141897,
      "loss": 5.5819,
      "mean_token_accuracy": 0.18880074769258498,
      "num_tokens": 26360709.0,
      "step": 14145
    },
    {
      "entropy": 5.790923738479615,
      "epoch": 1.2495584599081597,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048493007294975855,
      "loss": 5.4037,
      "mean_token_accuracy": 0.19863016456365584,
      "num_tokens": 26369447.0,
      "step": 14150
    },
    {
      "entropy": 5.875974655151367,
      "epoch": 1.25,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048491873968568746,
      "loss": 5.3604,
      "mean_token_accuracy": 0.21115702241659165,
      "num_tokens": 26379281.0,
      "step": 14155
    },
    {
      "entropy": 5.7076949119567875,
      "epoch": 1.2504415400918403,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004849074023094276,
      "loss": 5.3159,
      "mean_token_accuracy": 0.21273223906755448,
      "num_tokens": 26387664.0,
      "step": 14160
    },
    {
      "entropy": 5.753626775741577,
      "epoch": 1.2508830801836806,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048489606082120106,
      "loss": 5.3486,
      "mean_token_accuracy": 0.20376153737306596,
      "num_tokens": 26397485.0,
      "step": 14165
    },
    {
      "entropy": 5.865854978561401,
      "epoch": 1.251324620275521,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004848847152212301,
      "loss": 5.3105,
      "mean_token_accuracy": 0.20952945202589035,
      "num_tokens": 26406307.0,
      "step": 14170
    },
    {
      "entropy": 5.885340976715088,
      "epoch": 1.2517661603673613,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000484873365509737,
      "loss": 5.5197,
      "mean_token_accuracy": 0.1915242224931717,
      "num_tokens": 26416155.0,
      "step": 14175
    },
    {
      "entropy": 5.837534999847412,
      "epoch": 1.2522077004592016,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048486201168694397,
      "loss": 5.4173,
      "mean_token_accuracy": 0.19935398697853088,
      "num_tokens": 26424147.0,
      "step": 14180
    },
    {
      "entropy": 5.829942417144776,
      "epoch": 1.252649240551042,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048485065375307353,
      "loss": 5.417,
      "mean_token_accuracy": 0.19690395146608353,
      "num_tokens": 26433249.0,
      "step": 14185
    },
    {
      "entropy": 5.879388236999512,
      "epoch": 1.2530907806428824,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048483929170834805,
      "loss": 5.4617,
      "mean_token_accuracy": 0.1976969927549362,
      "num_tokens": 26442346.0,
      "step": 14190
    },
    {
      "entropy": 5.861908054351806,
      "epoch": 1.2535323207347226,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048482792555299027,
      "loss": 5.4262,
      "mean_token_accuracy": 0.19566735327243806,
      "num_tokens": 26450885.0,
      "step": 14195
    },
    {
      "entropy": 5.791489171981811,
      "epoch": 1.2539738608265631,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004848165552872227,
      "loss": 5.4648,
      "mean_token_accuracy": 0.20064617842435836,
      "num_tokens": 26459827.0,
      "step": 14200
    },
    {
      "entropy": 5.7747917652130125,
      "epoch": 1.2544154009184034,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004848051809112682,
      "loss": 5.4551,
      "mean_token_accuracy": 0.1935664877295494,
      "num_tokens": 26468357.0,
      "step": 14205
    },
    {
      "entropy": 5.862593698501587,
      "epoch": 1.2548569410102437,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048479380242534955,
      "loss": 5.5092,
      "mean_token_accuracy": 0.19768529385328293,
      "num_tokens": 26477396.0,
      "step": 14210
    },
    {
      "entropy": 5.893388271331787,
      "epoch": 1.2552984811020842,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004847824198296896,
      "loss": 5.481,
      "mean_token_accuracy": 0.19868139773607255,
      "num_tokens": 26486468.0,
      "step": 14215
    },
    {
      "entropy": 5.891446828842163,
      "epoch": 1.2557400211939245,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004847710331245113,
      "loss": 5.4485,
      "mean_token_accuracy": 0.19702454507350922,
      "num_tokens": 26496157.0,
      "step": 14220
    },
    {
      "entropy": 5.759126615524292,
      "epoch": 1.2561815612857647,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048475964231003776,
      "loss": 5.3202,
      "mean_token_accuracy": 0.2099949225783348,
      "num_tokens": 26504573.0,
      "step": 14225
    },
    {
      "entropy": 5.724131107330322,
      "epoch": 1.256623101377605,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004847482473864922,
      "loss": 5.2861,
      "mean_token_accuracy": 0.2155895709991455,
      "num_tokens": 26513441.0,
      "step": 14230
    },
    {
      "entropy": 5.8794779777526855,
      "epoch": 1.2570646414694453,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004847368483540977,
      "loss": 5.4995,
      "mean_token_accuracy": 0.1886053577065468,
      "num_tokens": 26522911.0,
      "step": 14235
    },
    {
      "entropy": 5.779993534088135,
      "epoch": 1.2575061815612858,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048472544521307765,
      "loss": 5.3569,
      "mean_token_accuracy": 0.20827997028827666,
      "num_tokens": 26531388.0,
      "step": 14240
    },
    {
      "entropy": 5.836145830154419,
      "epoch": 1.257947721653126,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004847140379636554,
      "loss": 5.3593,
      "mean_token_accuracy": 0.1983882948756218,
      "num_tokens": 26540500.0,
      "step": 14245
    },
    {
      "entropy": 5.845391941070557,
      "epoch": 1.2583892617449663,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004847026266060545,
      "loss": 5.3709,
      "mean_token_accuracy": 0.198971189558506,
      "num_tokens": 26550041.0,
      "step": 14250
    },
    {
      "entropy": 5.791977787017823,
      "epoch": 1.2588308018368068,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004846912111404983,
      "loss": 5.3313,
      "mean_token_accuracy": 0.20202200710773469,
      "num_tokens": 26559594.0,
      "step": 14255
    },
    {
      "entropy": 5.762013292312622,
      "epoch": 1.2592723419286471,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048467979156721055,
      "loss": 5.3989,
      "mean_token_accuracy": 0.20650634914636612,
      "num_tokens": 26569143.0,
      "step": 14260
    },
    {
      "entropy": 5.770469379425049,
      "epoch": 1.2597138820204874,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048466836788641496,
      "loss": 5.3815,
      "mean_token_accuracy": 0.20351646989583969,
      "num_tokens": 26578040.0,
      "step": 14265
    },
    {
      "entropy": 5.801880645751953,
      "epoch": 1.260155422112328,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048465694009833525,
      "loss": 5.4462,
      "mean_token_accuracy": 0.20380374491214753,
      "num_tokens": 26588283.0,
      "step": 14270
    },
    {
      "entropy": 5.890912818908691,
      "epoch": 1.2605969622041682,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048464550820319537,
      "loss": 5.4648,
      "mean_token_accuracy": 0.19421533048152922,
      "num_tokens": 26597815.0,
      "step": 14275
    },
    {
      "entropy": 5.805597448348999,
      "epoch": 1.2610385022960084,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004846340722012192,
      "loss": 5.4091,
      "mean_token_accuracy": 0.19901443123817444,
      "num_tokens": 26606752.0,
      "step": 14280
    },
    {
      "entropy": 5.821068000793457,
      "epoch": 1.261480042387849,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004846226320926308,
      "loss": 5.4364,
      "mean_token_accuracy": 0.1993491008877754,
      "num_tokens": 26615312.0,
      "step": 14285
    },
    {
      "entropy": 5.857545471191406,
      "epoch": 1.2619215824796892,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004846111878776542,
      "loss": 5.4277,
      "mean_token_accuracy": 0.19909490197896956,
      "num_tokens": 26624642.0,
      "step": 14290
    },
    {
      "entropy": 5.769836473464966,
      "epoch": 1.2623631225715295,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004845997395565137,
      "loss": 5.266,
      "mean_token_accuracy": 0.21139654517173767,
      "num_tokens": 26633250.0,
      "step": 14295
    },
    {
      "entropy": 5.746859359741211,
      "epoch": 1.2628046626633698,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004845882871294335,
      "loss": 5.4249,
      "mean_token_accuracy": 0.1992125391960144,
      "num_tokens": 26642422.0,
      "step": 14300
    },
    {
      "entropy": 5.830981349945068,
      "epoch": 1.26324620275521,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004845768305966379,
      "loss": 5.4677,
      "mean_token_accuracy": 0.19054748564958573,
      "num_tokens": 26651978.0,
      "step": 14305
    },
    {
      "entropy": 5.855709075927734,
      "epoch": 1.2636877428470505,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004845653699583515,
      "loss": 5.4805,
      "mean_token_accuracy": 0.19336138367652894,
      "num_tokens": 26662075.0,
      "step": 14310
    },
    {
      "entropy": 5.833240413665772,
      "epoch": 1.2641292829388908,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048455390521479855,
      "loss": 5.3843,
      "mean_token_accuracy": 0.20019437968730927,
      "num_tokens": 26670813.0,
      "step": 14315
    },
    {
      "entropy": 5.8653655529022215,
      "epoch": 1.264570823030731,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048454243636620383,
      "loss": 5.4179,
      "mean_token_accuracy": 0.19402478486299515,
      "num_tokens": 26679867.0,
      "step": 14320
    },
    {
      "entropy": 5.770459222793579,
      "epoch": 1.2650123631225716,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000484530963412792,
      "loss": 5.4271,
      "mean_token_accuracy": 0.19590177685022353,
      "num_tokens": 26688523.0,
      "step": 14325
    },
    {
      "entropy": 5.84591908454895,
      "epoch": 1.2654539032144119,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048451948635478765,
      "loss": 5.4876,
      "mean_token_accuracy": 0.19447792023420335,
      "num_tokens": 26697922.0,
      "step": 14330
    },
    {
      "entropy": 5.927173662185669,
      "epoch": 1.2658954433062521,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004845080051924158,
      "loss": 5.4424,
      "mean_token_accuracy": 0.18839099556207656,
      "num_tokens": 26707058.0,
      "step": 14335
    },
    {
      "entropy": 5.855650472640991,
      "epoch": 1.2663369833980926,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004844965199259013,
      "loss": 5.4508,
      "mean_token_accuracy": 0.19742105156183243,
      "num_tokens": 26715873.0,
      "step": 14340
    },
    {
      "entropy": 5.800871706008911,
      "epoch": 1.266778523489933,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004844850305554692,
      "loss": 5.4298,
      "mean_token_accuracy": 0.2012576714158058,
      "num_tokens": 26726253.0,
      "step": 14345
    },
    {
      "entropy": 5.825993347167969,
      "epoch": 1.2672200635817732,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048447353708134426,
      "loss": 5.4285,
      "mean_token_accuracy": 0.19938729107379913,
      "num_tokens": 26737077.0,
      "step": 14350
    },
    {
      "entropy": 5.819729804992676,
      "epoch": 1.2676616036736137,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000484462039503752,
      "loss": 5.4525,
      "mean_token_accuracy": 0.19648347347974776,
      "num_tokens": 26746603.0,
      "step": 14355
    },
    {
      "entropy": 5.778798151016235,
      "epoch": 1.268103143765454,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048445053782291753,
      "loss": 5.4264,
      "mean_token_accuracy": 0.20044244974851608,
      "num_tokens": 26756174.0,
      "step": 14360
    },
    {
      "entropy": 5.824472713470459,
      "epoch": 1.2685446838572942,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048443903203906614,
      "loss": 5.4238,
      "mean_token_accuracy": 0.2045733779668808,
      "num_tokens": 26765124.0,
      "step": 14365
    },
    {
      "entropy": 5.766734218597412,
      "epoch": 1.2689862239491345,
      "grad_norm": 1.0,
      "learning_rate": 0.0004844275221524232,
      "loss": 5.3989,
      "mean_token_accuracy": 0.20548185259103774,
      "num_tokens": 26774438.0,
      "step": 14370
    },
    {
      "entropy": 5.907923460006714,
      "epoch": 1.2694277640409748,
      "grad_norm": 1.0,
      "learning_rate": 0.00048441600816321424,
      "loss": 5.4406,
      "mean_token_accuracy": 0.2006345048546791,
      "num_tokens": 26783613.0,
      "step": 14375
    },
    {
      "entropy": 5.8193682670593265,
      "epoch": 1.2698693041328153,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004844044900716647,
      "loss": 5.2824,
      "mean_token_accuracy": 0.21384110301733017,
      "num_tokens": 26792136.0,
      "step": 14380
    },
    {
      "entropy": 5.701772880554199,
      "epoch": 1.2703108442246556,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004843929678780004,
      "loss": 5.3074,
      "mean_token_accuracy": 0.21431955248117446,
      "num_tokens": 26801481.0,
      "step": 14385
    },
    {
      "entropy": 5.767381381988526,
      "epoch": 1.2707523843164958,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004843814415824469,
      "loss": 5.4005,
      "mean_token_accuracy": 0.20618583559989928,
      "num_tokens": 26810503.0,
      "step": 14390
    },
    {
      "entropy": 5.903650236129761,
      "epoch": 1.2711939244083363,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048436991118523003,
      "loss": 5.4147,
      "mean_token_accuracy": 0.20471629798412322,
      "num_tokens": 26820007.0,
      "step": 14395
    },
    {
      "entropy": 5.845257520675659,
      "epoch": 1.2716354645001766,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048435837668657565,
      "loss": 5.3517,
      "mean_token_accuracy": 0.20296451300382615,
      "num_tokens": 26829088.0,
      "step": 14400
    },
    {
      "entropy": 5.81347689628601,
      "epoch": 1.272077004592017,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004843468380867098,
      "loss": 5.4452,
      "mean_token_accuracy": 0.20157098770141602,
      "num_tokens": 26838619.0,
      "step": 14405
    },
    {
      "entropy": 5.819590997695923,
      "epoch": 1.2725185446838574,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004843352953858585,
      "loss": 5.4645,
      "mean_token_accuracy": 0.1941637635231018,
      "num_tokens": 26848728.0,
      "step": 14410
    },
    {
      "entropy": 5.848436975479126,
      "epoch": 1.2729600847756977,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004843237485842477,
      "loss": 5.3689,
      "mean_token_accuracy": 0.2138514906167984,
      "num_tokens": 26859293.0,
      "step": 14415
    },
    {
      "entropy": 5.810101270675659,
      "epoch": 1.273401624867538,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00048431219768210374,
      "loss": 5.4086,
      "mean_token_accuracy": 0.20079730898141862,
      "num_tokens": 26868818.0,
      "step": 14420
    },
    {
      "entropy": 5.862733459472656,
      "epoch": 1.2738431649593784,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004843006426796529,
      "loss": 5.5097,
      "mean_token_accuracy": 0.18794319778680801,
      "num_tokens": 26879245.0,
      "step": 14425
    },
    {
      "entropy": 5.844163131713867,
      "epoch": 1.2742847050512187,
      "grad_norm": 1.0,
      "learning_rate": 0.0004842890835771215,
      "loss": 5.386,
      "mean_token_accuracy": 0.20195904672145842,
      "num_tokens": 26888074.0,
      "step": 14430
    },
    {
      "entropy": 5.851729869842529,
      "epoch": 1.274726245143059,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000484277520374736,
      "loss": 5.4255,
      "mean_token_accuracy": 0.19620058536529542,
      "num_tokens": 26897324.0,
      "step": 14435
    },
    {
      "entropy": 5.910194635391235,
      "epoch": 1.2751677852348993,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048426595307272285,
      "loss": 5.527,
      "mean_token_accuracy": 0.19152487367391585,
      "num_tokens": 26906797.0,
      "step": 14440
    },
    {
      "entropy": 5.864139080047607,
      "epoch": 1.2756093253267395,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048425438167130875,
      "loss": 5.4448,
      "mean_token_accuracy": 0.1987533837556839,
      "num_tokens": 26915737.0,
      "step": 14445
    },
    {
      "entropy": 5.765225744247436,
      "epoch": 1.27605086541858,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048424280617072026,
      "loss": 5.412,
      "mean_token_accuracy": 0.19996086806058883,
      "num_tokens": 26925364.0,
      "step": 14450
    },
    {
      "entropy": 5.831092691421508,
      "epoch": 1.2764924055104203,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00048423122657118424,
      "loss": 5.4144,
      "mean_token_accuracy": 0.2001582309603691,
      "num_tokens": 26935188.0,
      "step": 14455
    },
    {
      "entropy": 5.8068177700042725,
      "epoch": 1.2769339456022606,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048421964287292745,
      "loss": 5.3796,
      "mean_token_accuracy": 0.19895507395267487,
      "num_tokens": 26944606.0,
      "step": 14460
    },
    {
      "entropy": 5.887812280654908,
      "epoch": 1.277375485694101,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004842080550761769,
      "loss": 5.5068,
      "mean_token_accuracy": 0.1923213228583336,
      "num_tokens": 26955147.0,
      "step": 14465
    },
    {
      "entropy": 5.839098834991455,
      "epoch": 1.2778170257859414,
      "grad_norm": 1.125,
      "learning_rate": 0.0004841964631811595,
      "loss": 5.4364,
      "mean_token_accuracy": 0.20264236479997635,
      "num_tokens": 26964243.0,
      "step": 14470
    },
    {
      "entropy": 5.819336223602295,
      "epoch": 1.2782585658777816,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048418486718810244,
      "loss": 5.4033,
      "mean_token_accuracy": 0.20000549256801606,
      "num_tokens": 26973284.0,
      "step": 14475
    },
    {
      "entropy": 5.841389942169189,
      "epoch": 1.2787001059696221,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004841732670972327,
      "loss": 5.5313,
      "mean_token_accuracy": 0.19234911054372789,
      "num_tokens": 26982757.0,
      "step": 14480
    },
    {
      "entropy": 5.873554134368897,
      "epoch": 1.2791416460614624,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004841616629087776,
      "loss": 5.4393,
      "mean_token_accuracy": 0.19488273859024047,
      "num_tokens": 26991784.0,
      "step": 14485
    },
    {
      "entropy": 5.883336687088013,
      "epoch": 1.2795831861533027,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048415005462296455,
      "loss": 5.4549,
      "mean_token_accuracy": 0.18665586411952972,
      "num_tokens": 27000700.0,
      "step": 14490
    },
    {
      "entropy": 5.838480710983276,
      "epoch": 1.2800247262451432,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004841384422400209,
      "loss": 5.4159,
      "mean_token_accuracy": 0.2099954754114151,
      "num_tokens": 27010089.0,
      "step": 14495
    },
    {
      "entropy": 5.810708618164062,
      "epoch": 1.2804662663369835,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004841268257601741,
      "loss": 5.4194,
      "mean_token_accuracy": 0.20429598540067673,
      "num_tokens": 27018961.0,
      "step": 14500
    },
    {
      "entropy": 5.803251171112061,
      "epoch": 1.2809078064288237,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004841152051836517,
      "loss": 5.4793,
      "mean_token_accuracy": 0.19944901019334793,
      "num_tokens": 27027905.0,
      "step": 14505
    },
    {
      "entropy": 5.829074668884277,
      "epoch": 1.281349346520664,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048410358051068135,
      "loss": 5.3932,
      "mean_token_accuracy": 0.19561435729265214,
      "num_tokens": 27036969.0,
      "step": 14510
    },
    {
      "entropy": 5.755847215652466,
      "epoch": 1.2817908866125043,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004840919517414908,
      "loss": 5.3428,
      "mean_token_accuracy": 0.20895552933216094,
      "num_tokens": 27046654.0,
      "step": 14515
    },
    {
      "entropy": 5.8790534973144535,
      "epoch": 1.2822324267043448,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048408031887630775,
      "loss": 5.4448,
      "mean_token_accuracy": 0.19693035781383514,
      "num_tokens": 27058492.0,
      "step": 14520
    },
    {
      "entropy": 5.867789602279663,
      "epoch": 1.282673966796185,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004840686819153604,
      "loss": 5.4529,
      "mean_token_accuracy": 0.19810487031936647,
      "num_tokens": 27068863.0,
      "step": 14525
    },
    {
      "entropy": 5.802721881866455,
      "epoch": 1.2831155068880253,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004840570408588762,
      "loss": 5.4063,
      "mean_token_accuracy": 0.20112985819578172,
      "num_tokens": 27078793.0,
      "step": 14530
    },
    {
      "entropy": 5.76865701675415,
      "epoch": 1.2835570469798658,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004840453957070836,
      "loss": 5.2506,
      "mean_token_accuracy": 0.21420191824436188,
      "num_tokens": 27087093.0,
      "step": 14535
    },
    {
      "entropy": 5.771711683273315,
      "epoch": 1.283998587071706,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048403374646021057,
      "loss": 5.3317,
      "mean_token_accuracy": 0.20668103098869323,
      "num_tokens": 27096601.0,
      "step": 14540
    },
    {
      "entropy": 5.813027095794678,
      "epoch": 1.2844401271635464,
      "grad_norm": 1.0,
      "learning_rate": 0.00048402209311848533,
      "loss": 5.3975,
      "mean_token_accuracy": 0.20317106544971467,
      "num_tokens": 27105041.0,
      "step": 14545
    },
    {
      "entropy": 5.838200426101684,
      "epoch": 1.2848816672553869,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048401043568213614,
      "loss": 5.4626,
      "mean_token_accuracy": 0.19362209141254424,
      "num_tokens": 27114252.0,
      "step": 14550
    },
    {
      "entropy": 5.802340459823609,
      "epoch": 1.2853232073472272,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004839987741513914,
      "loss": 5.3686,
      "mean_token_accuracy": 0.19881146401166916,
      "num_tokens": 27122480.0,
      "step": 14555
    },
    {
      "entropy": 5.820025205612183,
      "epoch": 1.2857647474390674,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048398710852647954,
      "loss": 5.4209,
      "mean_token_accuracy": 0.19590677171945572,
      "num_tokens": 27132322.0,
      "step": 14560
    },
    {
      "entropy": 5.753691530227661,
      "epoch": 1.286206287530908,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048397543880762905,
      "loss": 5.3121,
      "mean_token_accuracy": 0.21430548578500747,
      "num_tokens": 27140782.0,
      "step": 14565
    },
    {
      "entropy": 5.774447584152222,
      "epoch": 1.2866478276227482,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048396376499506866,
      "loss": 5.3487,
      "mean_token_accuracy": 0.2051307216286659,
      "num_tokens": 27150058.0,
      "step": 14570
    },
    {
      "entropy": 5.7292139530181885,
      "epoch": 1.2870893677145885,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048395208708902686,
      "loss": 5.3925,
      "mean_token_accuracy": 0.20420772433280945,
      "num_tokens": 27159574.0,
      "step": 14575
    },
    {
      "entropy": 5.819472742080689,
      "epoch": 1.2875309078064288,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048394040508973257,
      "loss": 5.3937,
      "mean_token_accuracy": 0.20418492704629898,
      "num_tokens": 27168025.0,
      "step": 14580
    },
    {
      "entropy": 5.773526811599732,
      "epoch": 1.287972447898269,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004839287189974145,
      "loss": 5.3705,
      "mean_token_accuracy": 0.203872212767601,
      "num_tokens": 27177108.0,
      "step": 14585
    },
    {
      "entropy": 5.801486015319824,
      "epoch": 1.2884139879901095,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048391702881230177,
      "loss": 5.4562,
      "mean_token_accuracy": 0.19423876255750655,
      "num_tokens": 27187486.0,
      "step": 14590
    },
    {
      "entropy": 5.919055414199829,
      "epoch": 1.2888555280819498,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004839053345346232,
      "loss": 5.4422,
      "mean_token_accuracy": 0.19504236727952956,
      "num_tokens": 27195962.0,
      "step": 14595
    },
    {
      "entropy": 5.866123962402344,
      "epoch": 1.28929706817379,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004838936361646079,
      "loss": 5.443,
      "mean_token_accuracy": 0.20055250972509384,
      "num_tokens": 27206354.0,
      "step": 14600
    },
    {
      "entropy": 5.732411909103393,
      "epoch": 1.2897386082656306,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004838819337024852,
      "loss": 5.3143,
      "mean_token_accuracy": 0.20928743332624436,
      "num_tokens": 27215326.0,
      "step": 14605
    },
    {
      "entropy": 5.786622333526611,
      "epoch": 1.2901801483574709,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048387022714848406,
      "loss": 5.4083,
      "mean_token_accuracy": 0.20155924260616304,
      "num_tokens": 27224253.0,
      "step": 14610
    },
    {
      "entropy": 5.81986436843872,
      "epoch": 1.2906216884493111,
      "grad_norm": 0.9375,
      "learning_rate": 0.000483858516502834,
      "loss": 5.4091,
      "mean_token_accuracy": 0.201575630903244,
      "num_tokens": 27233007.0,
      "step": 14615
    },
    {
      "entropy": 5.809497880935669,
      "epoch": 1.2910632285411516,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004838468017657644,
      "loss": 5.3373,
      "mean_token_accuracy": 0.20555283278226852,
      "num_tokens": 27242127.0,
      "step": 14620
    },
    {
      "entropy": 5.830899858474732,
      "epoch": 1.291504768632992,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048383508293750474,
      "loss": 5.4886,
      "mean_token_accuracy": 0.18978028148412704,
      "num_tokens": 27252753.0,
      "step": 14625
    },
    {
      "entropy": 5.904895639419555,
      "epoch": 1.2919463087248322,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004838233600182846,
      "loss": 5.4698,
      "mean_token_accuracy": 0.19388115257024766,
      "num_tokens": 27262498.0,
      "step": 14630
    },
    {
      "entropy": 5.9419198513031,
      "epoch": 1.2923878488166727,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004838116330083336,
      "loss": 5.4736,
      "mean_token_accuracy": 0.19612549543380736,
      "num_tokens": 27271524.0,
      "step": 14635
    },
    {
      "entropy": 5.708930015563965,
      "epoch": 1.292829388908513,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004837999019078814,
      "loss": 5.3106,
      "mean_token_accuracy": 0.2147142231464386,
      "num_tokens": 27280649.0,
      "step": 14640
    },
    {
      "entropy": 5.815315055847168,
      "epoch": 1.2932709290003532,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004837881667171579,
      "loss": 5.4314,
      "mean_token_accuracy": 0.19222755432128907,
      "num_tokens": 27289593.0,
      "step": 14645
    },
    {
      "entropy": 5.841908502578735,
      "epoch": 1.2937124690921935,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048377642743639297,
      "loss": 5.4191,
      "mean_token_accuracy": 0.20031570494174958,
      "num_tokens": 27298070.0,
      "step": 14650
    },
    {
      "entropy": 5.7707356929779055,
      "epoch": 1.2941540091840338,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004837646840658166,
      "loss": 5.3396,
      "mean_token_accuracy": 0.20702026784420013,
      "num_tokens": 27306738.0,
      "step": 14655
    },
    {
      "entropy": 5.829840850830078,
      "epoch": 1.2945955492758743,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048375293660565875,
      "loss": 5.3435,
      "mean_token_accuracy": 0.2051900789141655,
      "num_tokens": 27315243.0,
      "step": 14660
    },
    {
      "entropy": 5.793229389190674,
      "epoch": 1.2950370893677146,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004837411850561496,
      "loss": 5.3472,
      "mean_token_accuracy": 0.20633992403745652,
      "num_tokens": 27325327.0,
      "step": 14665
    },
    {
      "entropy": 5.82656078338623,
      "epoch": 1.2954786294595548,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048372942941751935,
      "loss": 5.3592,
      "mean_token_accuracy": 0.19919373393058776,
      "num_tokens": 27333436.0,
      "step": 14670
    },
    {
      "entropy": 5.870306777954101,
      "epoch": 1.2959201695513953,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004837176696899984,
      "loss": 5.4502,
      "mean_token_accuracy": 0.19145621359348297,
      "num_tokens": 27342689.0,
      "step": 14675
    },
    {
      "entropy": 5.885797643661499,
      "epoch": 1.2963617096432356,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048370590587381684,
      "loss": 5.4521,
      "mean_token_accuracy": 0.20088817924261093,
      "num_tokens": 27352347.0,
      "step": 14680
    },
    {
      "entropy": 5.858293867111206,
      "epoch": 1.2968032497350759,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048369413796920535,
      "loss": 5.4544,
      "mean_token_accuracy": 0.1977135181427002,
      "num_tokens": 27362153.0,
      "step": 14685
    },
    {
      "entropy": 5.7391194820404055,
      "epoch": 1.2972447898269164,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004836823659763944,
      "loss": 5.3834,
      "mean_token_accuracy": 0.19741292893886567,
      "num_tokens": 27371126.0,
      "step": 14690
    },
    {
      "entropy": 5.774929523468018,
      "epoch": 1.2976863299187567,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048367058989561457,
      "loss": 5.3248,
      "mean_token_accuracy": 0.20115848630666733,
      "num_tokens": 27380727.0,
      "step": 14695
    },
    {
      "entropy": 5.8485064029693605,
      "epoch": 1.298127870010597,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004836588097270966,
      "loss": 5.3699,
      "mean_token_accuracy": 0.2064995586872101,
      "num_tokens": 27389970.0,
      "step": 14700
    },
    {
      "entropy": 5.849681043624878,
      "epoch": 1.2985694101024374,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048364702547107113,
      "loss": 5.3321,
      "mean_token_accuracy": 0.19991912543773652,
      "num_tokens": 27399177.0,
      "step": 14705
    },
    {
      "entropy": 5.836047792434693,
      "epoch": 1.2990109501942777,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004836352371277692,
      "loss": 5.3987,
      "mean_token_accuracy": 0.2044302687048912,
      "num_tokens": 27408776.0,
      "step": 14710
    },
    {
      "entropy": 5.81238169670105,
      "epoch": 1.299452490286118,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004836234446974216,
      "loss": 5.457,
      "mean_token_accuracy": 0.196170312166214,
      "num_tokens": 27418708.0,
      "step": 14715
    },
    {
      "entropy": 5.8453374862670895,
      "epoch": 1.2998940303779583,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004836116481802593,
      "loss": 5.3727,
      "mean_token_accuracy": 0.202563051879406,
      "num_tokens": 27428027.0,
      "step": 14720
    },
    {
      "entropy": 5.795147848129273,
      "epoch": 1.3003355704697985,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048359984757651356,
      "loss": 5.3831,
      "mean_token_accuracy": 0.19861516654491423,
      "num_tokens": 27438214.0,
      "step": 14725
    },
    {
      "entropy": 5.820910692214966,
      "epoch": 1.300777110561639,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048358804288641536,
      "loss": 5.3191,
      "mean_token_accuracy": 0.20154762715101243,
      "num_tokens": 27447413.0,
      "step": 14730
    },
    {
      "entropy": 5.79666838645935,
      "epoch": 1.3012186506534793,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048357623411019607,
      "loss": 5.3446,
      "mean_token_accuracy": 0.21003749072551728,
      "num_tokens": 27456316.0,
      "step": 14735
    },
    {
      "entropy": 5.7720255851745605,
      "epoch": 1.3016601907453196,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000483564421248087,
      "loss": 5.4006,
      "mean_token_accuracy": 0.20428019016981125,
      "num_tokens": 27465346.0,
      "step": 14740
    },
    {
      "entropy": 5.762890577316284,
      "epoch": 1.30210173083716,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048355260430031947,
      "loss": 5.2803,
      "mean_token_accuracy": 0.20367016047239303,
      "num_tokens": 27473367.0,
      "step": 14745
    },
    {
      "entropy": 5.87225456237793,
      "epoch": 1.3025432709290004,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000483540783267125,
      "loss": 5.4654,
      "mean_token_accuracy": 0.20448017865419388,
      "num_tokens": 27482187.0,
      "step": 14750
    },
    {
      "entropy": 5.7717162609100345,
      "epoch": 1.3029848110208406,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048352895814873526,
      "loss": 5.3673,
      "mean_token_accuracy": 0.20130963772535324,
      "num_tokens": 27492406.0,
      "step": 14755
    },
    {
      "entropy": 5.817604494094849,
      "epoch": 1.3034263511126811,
      "grad_norm": 1.0,
      "learning_rate": 0.0004835171289453818,
      "loss": 5.4517,
      "mean_token_accuracy": 0.1971517637372017,
      "num_tokens": 27501929.0,
      "step": 14760
    },
    {
      "entropy": 5.814182710647583,
      "epoch": 1.3038678912045214,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004835052956572964,
      "loss": 5.3128,
      "mean_token_accuracy": 0.21751547753810882,
      "num_tokens": 27510657.0,
      "step": 14765
    },
    {
      "entropy": 5.813910007476807,
      "epoch": 1.3043094312963617,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048349345828471084,
      "loss": 5.4617,
      "mean_token_accuracy": 0.1915316805243492,
      "num_tokens": 27519726.0,
      "step": 14770
    },
    {
      "entropy": 5.758480262756348,
      "epoch": 1.3047509713882022,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048348161682785704,
      "loss": 5.3638,
      "mean_token_accuracy": 0.20772658437490463,
      "num_tokens": 27528876.0,
      "step": 14775
    },
    {
      "entropy": 5.843346357345581,
      "epoch": 1.3051925114800424,
      "grad_norm": 1.0,
      "learning_rate": 0.0004834697712869669,
      "loss": 5.4125,
      "mean_token_accuracy": 0.19880662858486176,
      "num_tokens": 27537782.0,
      "step": 14780
    },
    {
      "entropy": 5.834340810775757,
      "epoch": 1.3056340515718827,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004834579216622725,
      "loss": 5.4365,
      "mean_token_accuracy": 0.19541001170873643,
      "num_tokens": 27547322.0,
      "step": 14785
    },
    {
      "entropy": 5.821652841567993,
      "epoch": 1.306075591663723,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000483446067954006,
      "loss": 5.3946,
      "mean_token_accuracy": 0.19541295021772384,
      "num_tokens": 27556070.0,
      "step": 14790
    },
    {
      "entropy": 5.753350639343262,
      "epoch": 1.3065171317555633,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048343421016239945,
      "loss": 5.2822,
      "mean_token_accuracy": 0.20969274044036865,
      "num_tokens": 27565951.0,
      "step": 14795
    },
    {
      "entropy": 5.823746204376221,
      "epoch": 1.3069586718474038,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004834223482876854,
      "loss": 5.3518,
      "mean_token_accuracy": 0.2040032207965851,
      "num_tokens": 27575102.0,
      "step": 14800
    },
    {
      "entropy": 5.863844966888427,
      "epoch": 1.307400211939244,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004834104823300961,
      "loss": 5.4472,
      "mean_token_accuracy": 0.19539563953876496,
      "num_tokens": 27584854.0,
      "step": 14805
    },
    {
      "entropy": 5.815785455703735,
      "epoch": 1.3078417520310843,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048339861228986395,
      "loss": 5.4101,
      "mean_token_accuracy": 0.2083549067378044,
      "num_tokens": 27593403.0,
      "step": 14810
    },
    {
      "entropy": 5.882075691223145,
      "epoch": 1.3082832921229248,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048338673816722145,
      "loss": 5.458,
      "mean_token_accuracy": 0.20046875178813933,
      "num_tokens": 27602162.0,
      "step": 14815
    },
    {
      "entropy": 5.817253637313843,
      "epoch": 1.308724832214765,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048337485996240127,
      "loss": 5.46,
      "mean_token_accuracy": 0.19198155850172044,
      "num_tokens": 27611433.0,
      "step": 14820
    },
    {
      "entropy": 5.785982418060303,
      "epoch": 1.3091663723066054,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004833629776756362,
      "loss": 5.2838,
      "mean_token_accuracy": 0.20872964859008789,
      "num_tokens": 27620443.0,
      "step": 14825
    },
    {
      "entropy": 5.880227565765381,
      "epoch": 1.3096079123984459,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048335109130715875,
      "loss": 5.364,
      "mean_token_accuracy": 0.20242440700531006,
      "num_tokens": 27629507.0,
      "step": 14830
    },
    {
      "entropy": 5.820908832550049,
      "epoch": 1.3100494524902861,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004833392008572021,
      "loss": 5.3751,
      "mean_token_accuracy": 0.20181168913841246,
      "num_tokens": 27638355.0,
      "step": 14835
    },
    {
      "entropy": 5.874566602706909,
      "epoch": 1.3104909925821264,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048332730632599883,
      "loss": 5.4643,
      "mean_token_accuracy": 0.19332274347543715,
      "num_tokens": 27649074.0,
      "step": 14840
    },
    {
      "entropy": 5.839673137664795,
      "epoch": 1.310932532673967,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004833154077137821,
      "loss": 5.4399,
      "mean_token_accuracy": 0.19462324380874635,
      "num_tokens": 27658564.0,
      "step": 14845
    },
    {
      "entropy": 5.8184332847595215,
      "epoch": 1.3113740727658072,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048330350502078515,
      "loss": 5.3781,
      "mean_token_accuracy": 0.20303722470998764,
      "num_tokens": 27667963.0,
      "step": 14850
    },
    {
      "entropy": 5.831605005264282,
      "epoch": 1.3118156128576475,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048329159824724087,
      "loss": 5.4442,
      "mean_token_accuracy": 0.196271875500679,
      "num_tokens": 27677528.0,
      "step": 14855
    },
    {
      "entropy": 5.852317953109742,
      "epoch": 1.3122571529494877,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048327968739338275,
      "loss": 5.3646,
      "mean_token_accuracy": 0.20680750906467438,
      "num_tokens": 27687659.0,
      "step": 14860
    },
    {
      "entropy": 5.761237287521363,
      "epoch": 1.312698693041328,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048326777245944396,
      "loss": 5.3298,
      "mean_token_accuracy": 0.2123411312699318,
      "num_tokens": 27697222.0,
      "step": 14865
    },
    {
      "entropy": 5.798789739608765,
      "epoch": 1.3131402331331685,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000483255853445658,
      "loss": 5.3964,
      "mean_token_accuracy": 0.20664604902267455,
      "num_tokens": 27705664.0,
      "step": 14870
    },
    {
      "entropy": 5.752559185028076,
      "epoch": 1.3135817732250088,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048324393035225824,
      "loss": 5.3445,
      "mean_token_accuracy": 0.20538181066513062,
      "num_tokens": 27714879.0,
      "step": 14875
    },
    {
      "entropy": 5.8446681022644045,
      "epoch": 1.314023313316849,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004832320031794783,
      "loss": 5.4747,
      "mean_token_accuracy": 0.1937073811888695,
      "num_tokens": 27724749.0,
      "step": 14880
    },
    {
      "entropy": 5.851373910903931,
      "epoch": 1.3144648534086896,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004832200719275519,
      "loss": 5.4121,
      "mean_token_accuracy": 0.19959086924791336,
      "num_tokens": 27733109.0,
      "step": 14885
    },
    {
      "entropy": 5.642890977859497,
      "epoch": 1.3149063935005298,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004832081365967128,
      "loss": 5.2137,
      "mean_token_accuracy": 0.21851505488157272,
      "num_tokens": 27742079.0,
      "step": 14890
    },
    {
      "entropy": 5.874283123016357,
      "epoch": 1.3153479335923701,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048319619718719455,
      "loss": 5.485,
      "mean_token_accuracy": 0.19008238315582277,
      "num_tokens": 27750900.0,
      "step": 14895
    },
    {
      "entropy": 5.858968925476074,
      "epoch": 1.3157894736842106,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048318425369923133,
      "loss": 5.3506,
      "mean_token_accuracy": 0.20756099969148636,
      "num_tokens": 27760392.0,
      "step": 14900
    },
    {
      "entropy": 5.851071166992187,
      "epoch": 1.316231013776051,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004831723061330569,
      "loss": 5.4097,
      "mean_token_accuracy": 0.2031494364142418,
      "num_tokens": 27768805.0,
      "step": 14905
    },
    {
      "entropy": 5.866053915023803,
      "epoch": 1.3166725538678912,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048316035448890546,
      "loss": 5.5032,
      "mean_token_accuracy": 0.19315202683210372,
      "num_tokens": 27777165.0,
      "step": 14910
    },
    {
      "entropy": 5.897108268737793,
      "epoch": 1.3171140939597317,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000483148398767011,
      "loss": 5.4208,
      "mean_token_accuracy": 0.20184513926506042,
      "num_tokens": 27786709.0,
      "step": 14915
    },
    {
      "entropy": 5.79182014465332,
      "epoch": 1.317555634051572,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048313643896760776,
      "loss": 5.3721,
      "mean_token_accuracy": 0.19768935292959214,
      "num_tokens": 27795977.0,
      "step": 14920
    },
    {
      "entropy": 5.723812103271484,
      "epoch": 1.3179971741434122,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048312447509093017,
      "loss": 5.3736,
      "mean_token_accuracy": 0.1965114951133728,
      "num_tokens": 27804838.0,
      "step": 14925
    },
    {
      "entropy": 5.808836793899536,
      "epoch": 1.3184387142352525,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004831125071372124,
      "loss": 5.3478,
      "mean_token_accuracy": 0.20593465566635133,
      "num_tokens": 27813245.0,
      "step": 14930
    },
    {
      "entropy": 5.889650917053222,
      "epoch": 1.3188802543270928,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004831005351066889,
      "loss": 5.4948,
      "mean_token_accuracy": 0.19392691999673844,
      "num_tokens": 27822509.0,
      "step": 14935
    },
    {
      "entropy": 5.794647741317749,
      "epoch": 1.3193217944189333,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004830885589995944,
      "loss": 5.2938,
      "mean_token_accuracy": 0.21173144578933717,
      "num_tokens": 27831140.0,
      "step": 14940
    },
    {
      "entropy": 5.87294454574585,
      "epoch": 1.3197633345107735,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048307657881616333,
      "loss": 5.414,
      "mean_token_accuracy": 0.20045745968818665,
      "num_tokens": 27839537.0,
      "step": 14945
    },
    {
      "entropy": 5.830281543731689,
      "epoch": 1.3202048746026138,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048306459455663045,
      "loss": 5.3789,
      "mean_token_accuracy": 0.20270113795995712,
      "num_tokens": 27848888.0,
      "step": 14950
    },
    {
      "entropy": 5.858214712142944,
      "epoch": 1.3206464146944543,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004830526062212305,
      "loss": 5.4358,
      "mean_token_accuracy": 0.19450279623270034,
      "num_tokens": 27858910.0,
      "step": 14955
    },
    {
      "entropy": 5.749471426010132,
      "epoch": 1.3210879547862946,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048304061381019826,
      "loss": 5.336,
      "mean_token_accuracy": 0.21143770664930345,
      "num_tokens": 27867967.0,
      "step": 14960
    },
    {
      "entropy": 5.775195741653443,
      "epoch": 1.3215294948781349,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048302861732376875,
      "loss": 5.3548,
      "mean_token_accuracy": 0.2055211618542671,
      "num_tokens": 27877317.0,
      "step": 14965
    },
    {
      "entropy": 5.871386289596558,
      "epoch": 1.3219710349699754,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048301661676217693,
      "loss": 5.4188,
      "mean_token_accuracy": 0.1976461797952652,
      "num_tokens": 27886400.0,
      "step": 14970
    },
    {
      "entropy": 5.8330635070800785,
      "epoch": 1.3224125750618156,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004830046121256579,
      "loss": 5.3992,
      "mean_token_accuracy": 0.19833968579769135,
      "num_tokens": 27895816.0,
      "step": 14975
    },
    {
      "entropy": 5.876082181930542,
      "epoch": 1.322854115153656,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004829926034144468,
      "loss": 5.4319,
      "mean_token_accuracy": 0.19784627556800843,
      "num_tokens": 27905614.0,
      "step": 14980
    },
    {
      "entropy": 5.830043506622315,
      "epoch": 1.3232956552454964,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004829805906287789,
      "loss": 5.4393,
      "mean_token_accuracy": 0.19370155036449432,
      "num_tokens": 27914714.0,
      "step": 14985
    },
    {
      "entropy": 5.782928419113159,
      "epoch": 1.3237371953373367,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004829685737688895,
      "loss": 5.3692,
      "mean_token_accuracy": 0.20484673529863356,
      "num_tokens": 27924821.0,
      "step": 14990
    },
    {
      "entropy": 5.920098209381104,
      "epoch": 1.324178735429177,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048295655283501403,
      "loss": 5.4454,
      "mean_token_accuracy": 0.1988457813858986,
      "num_tokens": 27934028.0,
      "step": 14995
    },
    {
      "entropy": 5.854908895492554,
      "epoch": 1.3246202755210172,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000482944527827388,
      "loss": 5.3684,
      "mean_token_accuracy": 0.19951189011335374,
      "num_tokens": 27943550.0,
      "step": 15000
    },
    {
      "epoch": 1.3246202755210172,
      "eval_entropy": 5.613093136769054,
      "eval_loss": 5.462072372436523,
      "eval_mean_token_accuracy": 0.20655342451553044,
      "eval_num_tokens": 27943550.0,
      "eval_runtime": 20.3613,
      "eval_samples_per_second": 1734.415,
      "eval_steps_per_second": 216.833,
      "step": 15000
    },
    {
      "entropy": 5.84151930809021,
      "epoch": 1.3250618156128575,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004829324987462469,
      "loss": 5.3709,
      "mean_token_accuracy": 0.19965676069259644,
      "num_tokens": 27952752.0,
      "step": 15005
    },
    {
      "entropy": 5.850358676910401,
      "epoch": 1.325503355704698,
      "grad_norm": 1.0,
      "learning_rate": 0.0004829204655918264,
      "loss": 5.4534,
      "mean_token_accuracy": 0.19576463252305984,
      "num_tokens": 27961535.0,
      "step": 15010
    },
    {
      "entropy": 5.83775749206543,
      "epoch": 1.3259448957965383,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048290842836436226,
      "loss": 5.3779,
      "mean_token_accuracy": 0.20184564888477324,
      "num_tokens": 27969644.0,
      "step": 15015
    },
    {
      "entropy": 5.861055469512939,
      "epoch": 1.3263864358883786,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004828963870640902,
      "loss": 5.4925,
      "mean_token_accuracy": 0.19010561108589172,
      "num_tokens": 27978859.0,
      "step": 15020
    },
    {
      "entropy": 5.857515335083008,
      "epoch": 1.326827975980219,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048288434169124625,
      "loss": 5.4397,
      "mean_token_accuracy": 0.19789244681596757,
      "num_tokens": 27988693.0,
      "step": 15025
    },
    {
      "entropy": 5.8798247337341305,
      "epoch": 1.3272695160720593,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004828722922460662,
      "loss": 5.3666,
      "mean_token_accuracy": 0.20213449597358704,
      "num_tokens": 27998196.0,
      "step": 15030
    },
    {
      "entropy": 5.864420604705811,
      "epoch": 1.3277110561638996,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048286023872878614,
      "loss": 5.5026,
      "mean_token_accuracy": 0.19181415140628816,
      "num_tokens": 28008743.0,
      "step": 15035
    },
    {
      "entropy": 5.875790739059449,
      "epoch": 1.3281525962557401,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004828481811396423,
      "loss": 5.4687,
      "mean_token_accuracy": 0.18993117660284042,
      "num_tokens": 28017684.0,
      "step": 15040
    },
    {
      "entropy": 5.78806357383728,
      "epoch": 1.3285941363475804,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004828361194788708,
      "loss": 5.4342,
      "mean_token_accuracy": 0.2010320618748665,
      "num_tokens": 28027992.0,
      "step": 15045
    },
    {
      "entropy": 5.909226417541504,
      "epoch": 1.3290356764394207,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048282405374670785,
      "loss": 5.5007,
      "mean_token_accuracy": 0.19204058051109313,
      "num_tokens": 28037457.0,
      "step": 15050
    },
    {
      "entropy": 5.914190483093262,
      "epoch": 1.3294772165312612,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048281198394338986,
      "loss": 5.477,
      "mean_token_accuracy": 0.19456910640001296,
      "num_tokens": 28046881.0,
      "step": 15055
    },
    {
      "entropy": 5.846889543533325,
      "epoch": 1.3299187566231014,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048279991006915335,
      "loss": 5.3257,
      "mean_token_accuracy": 0.20549699366092683,
      "num_tokens": 28056932.0,
      "step": 15060
    },
    {
      "entropy": 5.843326091766357,
      "epoch": 1.3303602967149417,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048278783212423484,
      "loss": 5.5046,
      "mean_token_accuracy": 0.19454202502965928,
      "num_tokens": 28066843.0,
      "step": 15065
    },
    {
      "entropy": 5.850677394866944,
      "epoch": 1.330801836806782,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004827757501088708,
      "loss": 5.413,
      "mean_token_accuracy": 0.19561834186315535,
      "num_tokens": 28076329.0,
      "step": 15070
    },
    {
      "entropy": 5.887483263015747,
      "epoch": 1.3312433768986223,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048276366402329803,
      "loss": 5.4739,
      "mean_token_accuracy": 0.19989312887191774,
      "num_tokens": 28086402.0,
      "step": 15075
    },
    {
      "entropy": 5.866004419326782,
      "epoch": 1.3316849169904628,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004827515738677532,
      "loss": 5.3964,
      "mean_token_accuracy": 0.19703940600156783,
      "num_tokens": 28095030.0,
      "step": 15080
    },
    {
      "entropy": 5.790664720535278,
      "epoch": 1.332126457082303,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004827394796424732,
      "loss": 5.3914,
      "mean_token_accuracy": 0.19742633551359176,
      "num_tokens": 28105138.0,
      "step": 15085
    },
    {
      "entropy": 5.811975955963135,
      "epoch": 1.3325679971741433,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048272738134769503,
      "loss": 5.4257,
      "mean_token_accuracy": 0.19294812083244323,
      "num_tokens": 28114359.0,
      "step": 15090
    },
    {
      "entropy": 5.839499759674072,
      "epoch": 1.3330095372659838,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048271527898365545,
      "loss": 5.3737,
      "mean_token_accuracy": 0.21175361126661302,
      "num_tokens": 28124045.0,
      "step": 15095
    },
    {
      "entropy": 5.823828506469726,
      "epoch": 1.333451077357824,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004827031725505917,
      "loss": 5.4192,
      "mean_token_accuracy": 0.2025573283433914,
      "num_tokens": 28132990.0,
      "step": 15100
    },
    {
      "entropy": 5.829469156265259,
      "epoch": 1.3338926174496644,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004826910620487411,
      "loss": 5.3975,
      "mean_token_accuracy": 0.20342407673597335,
      "num_tokens": 28142504.0,
      "step": 15105
    },
    {
      "entropy": 5.860262537002564,
      "epoch": 1.3343341575415049,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004826789474783405,
      "loss": 5.4865,
      "mean_token_accuracy": 0.19261177778244018,
      "num_tokens": 28151227.0,
      "step": 15110
    },
    {
      "entropy": 5.869546890258789,
      "epoch": 1.3347756976333451,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004826668288396276,
      "loss": 5.4061,
      "mean_token_accuracy": 0.19601559787988662,
      "num_tokens": 28160244.0,
      "step": 15115
    },
    {
      "entropy": 5.810975837707519,
      "epoch": 1.3352172377251854,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048265470613283954,
      "loss": 5.4119,
      "mean_token_accuracy": 0.19679898917675018,
      "num_tokens": 28169295.0,
      "step": 15120
    },
    {
      "entropy": 5.92709321975708,
      "epoch": 1.335658777817026,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048264257935821394,
      "loss": 5.4294,
      "mean_token_accuracy": 0.19897546470165253,
      "num_tokens": 28179199.0,
      "step": 15125
    },
    {
      "entropy": 5.867468976974488,
      "epoch": 1.3361003179088662,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004826304485159882,
      "loss": 5.4658,
      "mean_token_accuracy": 0.20500693321228028,
      "num_tokens": 28188315.0,
      "step": 15130
    },
    {
      "entropy": 5.831170654296875,
      "epoch": 1.3365418580007065,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048261831360640016,
      "loss": 5.4588,
      "mean_token_accuracy": 0.1936066895723343,
      "num_tokens": 28197776.0,
      "step": 15135
    },
    {
      "entropy": 5.849354600906372,
      "epoch": 1.3369833980925467,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048260617462968735,
      "loss": 5.4519,
      "mean_token_accuracy": 0.1989843487739563,
      "num_tokens": 28207985.0,
      "step": 15140
    },
    {
      "entropy": 5.800958681106567,
      "epoch": 1.337424938184387,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048259403158608773,
      "loss": 5.2568,
      "mean_token_accuracy": 0.2158833384513855,
      "num_tokens": 28216321.0,
      "step": 15145
    },
    {
      "entropy": 5.82542724609375,
      "epoch": 1.3378664782762275,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004825818844758391,
      "loss": 5.4859,
      "mean_token_accuracy": 0.19664452522993087,
      "num_tokens": 28226763.0,
      "step": 15150
    },
    {
      "entropy": 5.89154281616211,
      "epoch": 1.3383080183680678,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048256973329917943,
      "loss": 5.392,
      "mean_token_accuracy": 0.20025351643562317,
      "num_tokens": 28236123.0,
      "step": 15155
    },
    {
      "entropy": 5.916205167770386,
      "epoch": 1.338749558459908,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048255757805634665,
      "loss": 5.4752,
      "mean_token_accuracy": 0.19791246205568314,
      "num_tokens": 28245848.0,
      "step": 15160
    },
    {
      "entropy": 5.7232136726379395,
      "epoch": 1.3391910985517486,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048254541874757895,
      "loss": 5.4534,
      "mean_token_accuracy": 0.20464010238647462,
      "num_tokens": 28255414.0,
      "step": 15165
    },
    {
      "entropy": 5.812811946868896,
      "epoch": 1.3396326386435888,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048253325537311463,
      "loss": 5.3255,
      "mean_token_accuracy": 0.20433811247348785,
      "num_tokens": 28264162.0,
      "step": 15170
    },
    {
      "entropy": 5.843650054931641,
      "epoch": 1.3400741787354291,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048252108793319184,
      "loss": 5.3874,
      "mean_token_accuracy": 0.2048656165599823,
      "num_tokens": 28273146.0,
      "step": 15175
    },
    {
      "entropy": 5.854528188705444,
      "epoch": 1.3405157188272696,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004825089164280489,
      "loss": 5.3403,
      "mean_token_accuracy": 0.19782265722751619,
      "num_tokens": 28281865.0,
      "step": 15180
    },
    {
      "entropy": 5.778171062469482,
      "epoch": 1.3409572589191099,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048249674085792433,
      "loss": 5.3031,
      "mean_token_accuracy": 0.20337386280298234,
      "num_tokens": 28291197.0,
      "step": 15185
    },
    {
      "entropy": 5.8210736274719235,
      "epoch": 1.3413987990109502,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048248456122305667,
      "loss": 5.487,
      "mean_token_accuracy": 0.1921490639448166,
      "num_tokens": 28300283.0,
      "step": 15190
    },
    {
      "entropy": 5.9150454044342045,
      "epoch": 1.3418403391027907,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004824723775236844,
      "loss": 5.5433,
      "mean_token_accuracy": 0.19164900481700897,
      "num_tokens": 28309121.0,
      "step": 15195
    },
    {
      "entropy": 5.874976348876953,
      "epoch": 1.342281879194631,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004824601897600463,
      "loss": 5.3485,
      "mean_token_accuracy": 0.2040175050497055,
      "num_tokens": 28318377.0,
      "step": 15200
    },
    {
      "entropy": 5.851177406311035,
      "epoch": 1.3427234192864712,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048244799793238104,
      "loss": 5.3507,
      "mean_token_accuracy": 0.19854596853256226,
      "num_tokens": 28328341.0,
      "step": 15205
    },
    {
      "entropy": 5.853946399688721,
      "epoch": 1.3431649593783115,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004824358020409275,
      "loss": 5.5451,
      "mean_token_accuracy": 0.1914572909474373,
      "num_tokens": 28337340.0,
      "step": 15210
    },
    {
      "entropy": 5.756021308898926,
      "epoch": 1.3436064994701518,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004824236020859246,
      "loss": 5.4735,
      "mean_token_accuracy": 0.1991165727376938,
      "num_tokens": 28347653.0,
      "step": 15215
    },
    {
      "entropy": 5.92853627204895,
      "epoch": 1.3440480395619923,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048241139806761124,
      "loss": 5.5124,
      "mean_token_accuracy": 0.19249071627855302,
      "num_tokens": 28356596.0,
      "step": 15220
    },
    {
      "entropy": 5.910774898529053,
      "epoch": 1.3444895796538325,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048239918998622657,
      "loss": 5.4885,
      "mean_token_accuracy": 0.19089748859405517,
      "num_tokens": 28366253.0,
      "step": 15225
    },
    {
      "entropy": 5.760193538665772,
      "epoch": 1.3449311197456728,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004823869778420097,
      "loss": 5.3199,
      "mean_token_accuracy": 0.20183297097682953,
      "num_tokens": 28374815.0,
      "step": 15230
    },
    {
      "entropy": 5.813390016555786,
      "epoch": 1.3453726598375133,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048237476163519994,
      "loss": 5.4291,
      "mean_token_accuracy": 0.19437561482191085,
      "num_tokens": 28384221.0,
      "step": 15235
    },
    {
      "entropy": 5.823602247238159,
      "epoch": 1.3458141999293536,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004823625413660365,
      "loss": 5.4171,
      "mean_token_accuracy": 0.20221185237169265,
      "num_tokens": 28393453.0,
      "step": 15240
    },
    {
      "entropy": 5.761607027053833,
      "epoch": 1.3462557400211939,
      "grad_norm": 1.0,
      "learning_rate": 0.00048235031703475884,
      "loss": 5.3468,
      "mean_token_accuracy": 0.20481251925230026,
      "num_tokens": 28402343.0,
      "step": 15245
    },
    {
      "entropy": 5.791515302658081,
      "epoch": 1.3466972801130344,
      "grad_norm": 1.0,
      "learning_rate": 0.0004823380886416064,
      "loss": 5.3749,
      "mean_token_accuracy": 0.2084643617272377,
      "num_tokens": 28411527.0,
      "step": 15250
    },
    {
      "entropy": 5.849292278289795,
      "epoch": 1.3471388202048746,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004823258561868187,
      "loss": 5.4545,
      "mean_token_accuracy": 0.20024822801351547,
      "num_tokens": 28422518.0,
      "step": 15255
    },
    {
      "entropy": 5.889913558959961,
      "epoch": 1.347580360296715,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048231361967063534,
      "loss": 5.4154,
      "mean_token_accuracy": 0.19866528809070588,
      "num_tokens": 28431692.0,
      "step": 15260
    },
    {
      "entropy": 5.851217794418335,
      "epoch": 1.3480219003885554,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004823013790932961,
      "loss": 5.4589,
      "mean_token_accuracy": 0.19662691056728362,
      "num_tokens": 28440304.0,
      "step": 15265
    },
    {
      "entropy": 5.881387567520141,
      "epoch": 1.3484634404803957,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004822891344550408,
      "loss": 5.5074,
      "mean_token_accuracy": 0.18970640897750854,
      "num_tokens": 28450265.0,
      "step": 15270
    },
    {
      "entropy": 5.822591161727905,
      "epoch": 1.348904980572236,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048227688575610915,
      "loss": 5.3915,
      "mean_token_accuracy": 0.20024269670248032,
      "num_tokens": 28459699.0,
      "step": 15275
    },
    {
      "entropy": 5.808038568496704,
      "epoch": 1.3493465206640762,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004822646329967413,
      "loss": 5.4117,
      "mean_token_accuracy": 0.19073693603277206,
      "num_tokens": 28468523.0,
      "step": 15280
    },
    {
      "entropy": 5.9093286991119385,
      "epoch": 1.3497880607559165,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004822523761771771,
      "loss": 5.4613,
      "mean_token_accuracy": 0.1921578973531723,
      "num_tokens": 28477629.0,
      "step": 15285
    },
    {
      "entropy": 5.795480585098266,
      "epoch": 1.350229600847757,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004822401152976568,
      "loss": 5.3612,
      "mean_token_accuracy": 0.20787472426891326,
      "num_tokens": 28486517.0,
      "step": 15290
    },
    {
      "entropy": 5.8463352680206295,
      "epoch": 1.3506711409395973,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048222785035842045,
      "loss": 5.472,
      "mean_token_accuracy": 0.19570790082216263,
      "num_tokens": 28496103.0,
      "step": 15295
    },
    {
      "entropy": 5.816338348388672,
      "epoch": 1.3511126810314376,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048221558135970834,
      "loss": 5.4246,
      "mean_token_accuracy": 0.2021983817219734,
      "num_tokens": 28505538.0,
      "step": 15300
    },
    {
      "entropy": 5.730299663543701,
      "epoch": 1.351554221123278,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048220330830176086,
      "loss": 5.2925,
      "mean_token_accuracy": 0.20554286241531372,
      "num_tokens": 28514322.0,
      "step": 15305
    },
    {
      "entropy": 5.902645683288574,
      "epoch": 1.3519957612151183,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004821910311848184,
      "loss": 5.4004,
      "mean_token_accuracy": 0.20479389876127244,
      "num_tokens": 28524298.0,
      "step": 15310
    },
    {
      "entropy": 5.882126235961914,
      "epoch": 1.3524373013069586,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004821787500091215,
      "loss": 5.4384,
      "mean_token_accuracy": 0.1970038443803787,
      "num_tokens": 28533567.0,
      "step": 15315
    },
    {
      "entropy": 5.78786997795105,
      "epoch": 1.352878841398799,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048216646477491074,
      "loss": 5.4105,
      "mean_token_accuracy": 0.19877147078514099,
      "num_tokens": 28541746.0,
      "step": 15320
    },
    {
      "entropy": 5.820613098144531,
      "epoch": 1.3533203814906394,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004821541754824267,
      "loss": 5.4475,
      "mean_token_accuracy": 0.19675872772932052,
      "num_tokens": 28551685.0,
      "step": 15325
    },
    {
      "entropy": 5.890408897399903,
      "epoch": 1.3537619215824797,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004821418821319102,
      "loss": 5.48,
      "mean_token_accuracy": 0.19745792001485823,
      "num_tokens": 28560935.0,
      "step": 15330
    },
    {
      "entropy": 5.794356346130371,
      "epoch": 1.3542034616743202,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004821295847236021,
      "loss": 5.3207,
      "mean_token_accuracy": 0.2062166601419449,
      "num_tokens": 28569506.0,
      "step": 15335
    },
    {
      "entropy": 5.800172233581543,
      "epoch": 1.3546450017661604,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004821172832577431,
      "loss": 5.3741,
      "mean_token_accuracy": 0.20243366211652755,
      "num_tokens": 28579792.0,
      "step": 15340
    },
    {
      "entropy": 5.892455387115478,
      "epoch": 1.3550865418580007,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004821049777345744,
      "loss": 5.4835,
      "mean_token_accuracy": 0.19847968816757203,
      "num_tokens": 28589756.0,
      "step": 15345
    },
    {
      "entropy": 5.8243976593017575,
      "epoch": 1.355528081949841,
      "grad_norm": 0.984375,
      "learning_rate": 0.000482092668154337,
      "loss": 5.4099,
      "mean_token_accuracy": 0.2011106237769127,
      "num_tokens": 28599853.0,
      "step": 15350
    },
    {
      "entropy": 5.785915994644165,
      "epoch": 1.3559696220416813,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048208035451727195,
      "loss": 5.4119,
      "mean_token_accuracy": 0.19574137032032013,
      "num_tokens": 28608576.0,
      "step": 15355
    },
    {
      "entropy": 5.867419624328614,
      "epoch": 1.3564111621335218,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004820680368236206,
      "loss": 5.4618,
      "mean_token_accuracy": 0.19275035262107848,
      "num_tokens": 28618820.0,
      "step": 15360
    },
    {
      "entropy": 5.886230325698852,
      "epoch": 1.356852702225362,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004820557150736241,
      "loss": 5.3857,
      "mean_token_accuracy": 0.20416640788316726,
      "num_tokens": 28628092.0,
      "step": 15365
    },
    {
      "entropy": 5.89946608543396,
      "epoch": 1.3572942423172023,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048204338926752396,
      "loss": 5.4637,
      "mean_token_accuracy": 0.1959486111998558,
      "num_tokens": 28637849.0,
      "step": 15370
    },
    {
      "entropy": 5.76736364364624,
      "epoch": 1.3577357824090428,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004820310594055616,
      "loss": 5.3645,
      "mean_token_accuracy": 0.20419100373983384,
      "num_tokens": 28646957.0,
      "step": 15375
    },
    {
      "entropy": 5.886047315597534,
      "epoch": 1.358177322500883,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004820187254879784,
      "loss": 5.459,
      "mean_token_accuracy": 0.20299917906522752,
      "num_tokens": 28655768.0,
      "step": 15380
    },
    {
      "entropy": 5.8470703125,
      "epoch": 1.3586188625927234,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048200638751501616,
      "loss": 5.4191,
      "mean_token_accuracy": 0.1953986570239067,
      "num_tokens": 28664719.0,
      "step": 15385
    },
    {
      "entropy": 5.806668853759765,
      "epoch": 1.3590604026845639,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004819940454869166,
      "loss": 5.3995,
      "mean_token_accuracy": 0.1941104605793953,
      "num_tokens": 28673568.0,
      "step": 15390
    },
    {
      "entropy": 5.817203044891357,
      "epoch": 1.3595019427764041,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004819816994039213,
      "loss": 5.3822,
      "mean_token_accuracy": 0.20431563556194304,
      "num_tokens": 28682558.0,
      "step": 15395
    },
    {
      "entropy": 5.883448076248169,
      "epoch": 1.3599434828682444,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048196934926627227,
      "loss": 5.4692,
      "mean_token_accuracy": 0.1973310723900795,
      "num_tokens": 28691965.0,
      "step": 15400
    },
    {
      "entropy": 5.796695947647095,
      "epoch": 1.360385022960085,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004819569950742114,
      "loss": 5.3362,
      "mean_token_accuracy": 0.21011397689580918,
      "num_tokens": 28700805.0,
      "step": 15405
    },
    {
      "entropy": 5.797438716888427,
      "epoch": 1.3608265630519252,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004819446368279808,
      "loss": 5.3644,
      "mean_token_accuracy": 0.20311632007360458,
      "num_tokens": 28710169.0,
      "step": 15410
    },
    {
      "entropy": 5.92458667755127,
      "epoch": 1.3612681031437655,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048193227452782237,
      "loss": 5.5127,
      "mean_token_accuracy": 0.19419682323932647,
      "num_tokens": 28718656.0,
      "step": 15415
    },
    {
      "entropy": 5.90979380607605,
      "epoch": 1.3617096432356057,
      "grad_norm": 1.0,
      "learning_rate": 0.0004819199081739783,
      "loss": 5.464,
      "mean_token_accuracy": 0.19747065156698226,
      "num_tokens": 28728546.0,
      "step": 15420
    },
    {
      "entropy": 5.87447772026062,
      "epoch": 1.362151183327446,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000481907537766691,
      "loss": 5.341,
      "mean_token_accuracy": 0.20292591601610183,
      "num_tokens": 28737267.0,
      "step": 15425
    },
    {
      "entropy": 5.776773118972779,
      "epoch": 1.3625927234192865,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004818951633062027,
      "loss": 5.4487,
      "mean_token_accuracy": 0.20051742792129518,
      "num_tokens": 28746721.0,
      "step": 15430
    },
    {
      "entropy": 5.88642725944519,
      "epoch": 1.3630342635111268,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048188278479275584,
      "loss": 5.4724,
      "mean_token_accuracy": 0.1973584532737732,
      "num_tokens": 28756099.0,
      "step": 15435
    },
    {
      "entropy": 5.8891956329345705,
      "epoch": 1.363475803602967,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004818704022265928,
      "loss": 5.4233,
      "mean_token_accuracy": 0.1957658648490906,
      "num_tokens": 28765463.0,
      "step": 15440
    },
    {
      "entropy": 5.7847511768341064,
      "epoch": 1.3639173436948075,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048185801560795627,
      "loss": 5.3661,
      "mean_token_accuracy": 0.2016372039914131,
      "num_tokens": 28775284.0,
      "step": 15445
    },
    {
      "entropy": 5.875983715057373,
      "epoch": 1.3643588837866478,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048184562493708895,
      "loss": 5.4169,
      "mean_token_accuracy": 0.20593764781951904,
      "num_tokens": 28784524.0,
      "step": 15450
    },
    {
      "entropy": 5.799777412414551,
      "epoch": 1.364800423878488,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048183323021423334,
      "loss": 5.37,
      "mean_token_accuracy": 0.2009094089269638,
      "num_tokens": 28792942.0,
      "step": 15455
    },
    {
      "entropy": 5.820668077468872,
      "epoch": 1.3652419639703286,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048182083143963247,
      "loss": 5.3778,
      "mean_token_accuracy": 0.20059644281864167,
      "num_tokens": 28801746.0,
      "step": 15460
    },
    {
      "entropy": 5.839265966415406,
      "epoch": 1.3656835040621689,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004818084286135291,
      "loss": 5.4454,
      "mean_token_accuracy": 0.19646372199058532,
      "num_tokens": 28810276.0,
      "step": 15465
    },
    {
      "entropy": 5.876328182220459,
      "epoch": 1.3661250441540091,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004817960217361663,
      "loss": 5.4282,
      "mean_token_accuracy": 0.1994615763425827,
      "num_tokens": 28819545.0,
      "step": 15470
    },
    {
      "entropy": 5.847974491119385,
      "epoch": 1.3665665842458496,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000481783610807787,
      "loss": 5.319,
      "mean_token_accuracy": 0.2033898949623108,
      "num_tokens": 28828163.0,
      "step": 15475
    },
    {
      "entropy": 5.8331139087677,
      "epoch": 1.36700812433769,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004817711958286343,
      "loss": 5.4122,
      "mean_token_accuracy": 0.20001124441623688,
      "num_tokens": 28837952.0,
      "step": 15480
    },
    {
      "entropy": 5.846285486221314,
      "epoch": 1.3674496644295302,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004817587767989516,
      "loss": 5.4752,
      "mean_token_accuracy": 0.19427438229322433,
      "num_tokens": 28847702.0,
      "step": 15485
    },
    {
      "entropy": 5.885765027999878,
      "epoch": 1.3678912045213705,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048174635371898197,
      "loss": 5.4316,
      "mean_token_accuracy": 0.1955796003341675,
      "num_tokens": 28856936.0,
      "step": 15490
    },
    {
      "entropy": 5.840321588516235,
      "epoch": 1.3683327446132107,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004817339265889689,
      "loss": 5.3616,
      "mean_token_accuracy": 0.20821680575609208,
      "num_tokens": 28867072.0,
      "step": 15495
    },
    {
      "entropy": 5.72699875831604,
      "epoch": 1.3687742847050512,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004817214954091557,
      "loss": 5.3109,
      "mean_token_accuracy": 0.20816775858402253,
      "num_tokens": 28877546.0,
      "step": 15500
    },
    {
      "entropy": 5.853149509429931,
      "epoch": 1.3692158247968915,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048170906017978605,
      "loss": 5.4027,
      "mean_token_accuracy": 0.2011694148182869,
      "num_tokens": 28886804.0,
      "step": 15505
    },
    {
      "entropy": 5.871610164642334,
      "epoch": 1.3696573648887318,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004816966209011034,
      "loss": 5.4448,
      "mean_token_accuracy": 0.19629024118185043,
      "num_tokens": 28896476.0,
      "step": 15510
    },
    {
      "entropy": 5.905157041549683,
      "epoch": 1.3700989049805723,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048168417757335155,
      "loss": 5.4373,
      "mean_token_accuracy": 0.19920352846384048,
      "num_tokens": 28906269.0,
      "step": 15515
    },
    {
      "entropy": 5.782048845291138,
      "epoch": 1.3705404450724126,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004816717301967743,
      "loss": 5.3629,
      "mean_token_accuracy": 0.20571523308753967,
      "num_tokens": 28915792.0,
      "step": 15520
    },
    {
      "entropy": 5.8342865943908695,
      "epoch": 1.3709819851642528,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004816592787716152,
      "loss": 5.4243,
      "mean_token_accuracy": 0.19673479050397874,
      "num_tokens": 28925065.0,
      "step": 15525
    },
    {
      "entropy": 5.843438053131104,
      "epoch": 1.3714235252560933,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048164682329811855,
      "loss": 5.4206,
      "mean_token_accuracy": 0.19803667068481445,
      "num_tokens": 28933879.0,
      "step": 15530
    },
    {
      "entropy": 5.76366868019104,
      "epoch": 1.3718650653479336,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048163436377652804,
      "loss": 5.2675,
      "mean_token_accuracy": 0.21203935593366624,
      "num_tokens": 28943176.0,
      "step": 15535
    },
    {
      "entropy": 5.910560131072998,
      "epoch": 1.372306605439774,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004816219002070879,
      "loss": 5.4624,
      "mean_token_accuracy": 0.1993066370487213,
      "num_tokens": 28952770.0,
      "step": 15540
    },
    {
      "entropy": 5.737296867370605,
      "epoch": 1.3727481455316144,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004816094325900422,
      "loss": 5.282,
      "mean_token_accuracy": 0.20551921278238297,
      "num_tokens": 28961186.0,
      "step": 15545
    },
    {
      "entropy": 5.821661472320557,
      "epoch": 1.3731896856234547,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048159696092563533,
      "loss": 5.414,
      "mean_token_accuracy": 0.2063349574804306,
      "num_tokens": 28971787.0,
      "step": 15550
    },
    {
      "entropy": 5.872710752487182,
      "epoch": 1.373631225715295,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004815844852141114,
      "loss": 5.5081,
      "mean_token_accuracy": 0.19428355395793914,
      "num_tokens": 28982095.0,
      "step": 15555
    },
    {
      "entropy": 5.867133617401123,
      "epoch": 1.3740727658071352,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048157200545571497,
      "loss": 5.4204,
      "mean_token_accuracy": 0.2025141552090645,
      "num_tokens": 28991084.0,
      "step": 15560
    },
    {
      "entropy": 5.830327415466309,
      "epoch": 1.3745143058989755,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004815595216506903,
      "loss": 5.4007,
      "mean_token_accuracy": 0.20169737935066223,
      "num_tokens": 29000544.0,
      "step": 15565
    },
    {
      "entropy": 5.8615076541900635,
      "epoch": 1.374955845990816,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004815470337992822,
      "loss": 5.3009,
      "mean_token_accuracy": 0.20388926416635514,
      "num_tokens": 29009366.0,
      "step": 15570
    },
    {
      "entropy": 5.818301630020142,
      "epoch": 1.3753973860826563,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048153454190173517,
      "loss": 5.4771,
      "mean_token_accuracy": 0.19607685506343842,
      "num_tokens": 29018730.0,
      "step": 15575
    },
    {
      "entropy": 5.873378038406372,
      "epoch": 1.3758389261744965,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048152204595829394,
      "loss": 5.4973,
      "mean_token_accuracy": 0.18955930918455124,
      "num_tokens": 29030255.0,
      "step": 15580
    },
    {
      "entropy": 5.902130937576294,
      "epoch": 1.376280466266337,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004815095459692032,
      "loss": 5.3443,
      "mean_token_accuracy": 0.19729165881872177,
      "num_tokens": 29039030.0,
      "step": 15585
    },
    {
      "entropy": 5.861922359466552,
      "epoch": 1.3767220063581773,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000481497041934708,
      "loss": 5.3839,
      "mean_token_accuracy": 0.20678475350141526,
      "num_tokens": 29047504.0,
      "step": 15590
    },
    {
      "entropy": 5.859125852584839,
      "epoch": 1.3771635464500176,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048148453385505317,
      "loss": 5.4769,
      "mean_token_accuracy": 0.2015798032283783,
      "num_tokens": 29056488.0,
      "step": 15595
    },
    {
      "entropy": 5.829116582870483,
      "epoch": 1.377605086541858,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048147202173048366,
      "loss": 5.4136,
      "mean_token_accuracy": 0.21554109305143357,
      "num_tokens": 29066067.0,
      "step": 15600
    },
    {
      "entropy": 5.915482711791992,
      "epoch": 1.3780466266336984,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004814595055612447,
      "loss": 5.4197,
      "mean_token_accuracy": 0.19795806407928468,
      "num_tokens": 29075136.0,
      "step": 15605
    },
    {
      "entropy": 5.772874355316162,
      "epoch": 1.3784881667255386,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048144698534758153,
      "loss": 5.3638,
      "mean_token_accuracy": 0.20789763033390046,
      "num_tokens": 29084970.0,
      "step": 15610
    },
    {
      "entropy": 5.668832015991211,
      "epoch": 1.3789297068173791,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048143446108973924,
      "loss": 5.3126,
      "mean_token_accuracy": 0.20958250313997268,
      "num_tokens": 29093234.0,
      "step": 15615
    },
    {
      "entropy": 5.777319002151489,
      "epoch": 1.3793712469092194,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048142193278796335,
      "loss": 5.3653,
      "mean_token_accuracy": 0.20201972424983977,
      "num_tokens": 29102115.0,
      "step": 15620
    },
    {
      "entropy": 5.875241136550903,
      "epoch": 1.3798127870010597,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004814094004424992,
      "loss": 5.4186,
      "mean_token_accuracy": 0.19845666438341142,
      "num_tokens": 29111065.0,
      "step": 15625
    },
    {
      "entropy": 5.814388656616211,
      "epoch": 1.3802543270929,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004813968640535922,
      "loss": 5.4358,
      "mean_token_accuracy": 0.1984635293483734,
      "num_tokens": 29120238.0,
      "step": 15630
    },
    {
      "entropy": 5.819582605361939,
      "epoch": 1.3806958671847402,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000481384323621488,
      "loss": 5.3498,
      "mean_token_accuracy": 0.19713763296604156,
      "num_tokens": 29129062.0,
      "step": 15635
    },
    {
      "entropy": 5.820842027664185,
      "epoch": 1.3811374072765807,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004813717791464324,
      "loss": 5.475,
      "mean_token_accuracy": 0.19261003732681276,
      "num_tokens": 29138990.0,
      "step": 15640
    },
    {
      "entropy": 5.821529245376587,
      "epoch": 1.381578947368421,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048135923062867094,
      "loss": 5.3442,
      "mean_token_accuracy": 0.20845245122909545,
      "num_tokens": 29147507.0,
      "step": 15645
    },
    {
      "entropy": 5.80028886795044,
      "epoch": 1.3820204874602613,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004813466780684495,
      "loss": 5.4679,
      "mean_token_accuracy": 0.19808721244335176,
      "num_tokens": 29156277.0,
      "step": 15650
    },
    {
      "entropy": 5.883765363693238,
      "epoch": 1.3824620275521018,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004813341214660141,
      "loss": 5.5728,
      "mean_token_accuracy": 0.1865738958120346,
      "num_tokens": 29165925.0,
      "step": 15655
    },
    {
      "entropy": 5.96726016998291,
      "epoch": 1.382903567643942,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004813215608216105,
      "loss": 5.4953,
      "mean_token_accuracy": 0.1907041296362877,
      "num_tokens": 29175811.0,
      "step": 15660
    },
    {
      "entropy": 5.873224258422852,
      "epoch": 1.3833451077357823,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048130899613548487,
      "loss": 5.4619,
      "mean_token_accuracy": 0.2033369868993759,
      "num_tokens": 29185100.0,
      "step": 15665
    },
    {
      "entropy": 5.836682558059692,
      "epoch": 1.3837866478276228,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048129642740788337,
      "loss": 5.4429,
      "mean_token_accuracy": 0.19301197975873946,
      "num_tokens": 29194430.0,
      "step": 15670
    },
    {
      "entropy": 5.799141454696655,
      "epoch": 1.3842281879194631,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004812838546390521,
      "loss": 5.3786,
      "mean_token_accuracy": 0.201407028734684,
      "num_tokens": 29203897.0,
      "step": 15675
    },
    {
      "entropy": 5.851935958862304,
      "epoch": 1.3846697280113034,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004812712778292375,
      "loss": 5.5135,
      "mean_token_accuracy": 0.19677763432264328,
      "num_tokens": 29213474.0,
      "step": 15680
    },
    {
      "entropy": 5.938237094879151,
      "epoch": 1.3851112681031439,
      "grad_norm": 1.03125,
      "learning_rate": 0.000481258696978686,
      "loss": 5.4701,
      "mean_token_accuracy": 0.1967015489935875,
      "num_tokens": 29222176.0,
      "step": 15685
    },
    {
      "entropy": 5.785627746582032,
      "epoch": 1.3855528081949842,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048124611208764375,
      "loss": 5.3687,
      "mean_token_accuracy": 0.19589827209711075,
      "num_tokens": 29231885.0,
      "step": 15690
    },
    {
      "entropy": 5.8095043182373045,
      "epoch": 1.3859943482868244,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004812335231563576,
      "loss": 5.3928,
      "mean_token_accuracy": 0.20141471773386002,
      "num_tokens": 29241921.0,
      "step": 15695
    },
    {
      "entropy": 5.859870910644531,
      "epoch": 1.3864358883786647,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048122093018507396,
      "loss": 5.4818,
      "mean_token_accuracy": 0.19717176556587218,
      "num_tokens": 29251308.0,
      "step": 15700
    },
    {
      "entropy": 5.847552108764648,
      "epoch": 1.386877428470505,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004812083331740396,
      "loss": 5.4057,
      "mean_token_accuracy": 0.2079196572303772,
      "num_tokens": 29260628.0,
      "step": 15705
    },
    {
      "entropy": 5.880796718597412,
      "epoch": 1.3873189685623455,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048119573212350127,
      "loss": 5.4223,
      "mean_token_accuracy": 0.20460325926542283,
      "num_tokens": 29269530.0,
      "step": 15710
    },
    {
      "entropy": 5.852338266372681,
      "epoch": 1.3877605086541858,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048118312703370576,
      "loss": 5.4423,
      "mean_token_accuracy": 0.19531672596931457,
      "num_tokens": 29279846.0,
      "step": 15715
    },
    {
      "entropy": 5.890779876708985,
      "epoch": 1.388202048746026,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048117051790490007,
      "loss": 5.4318,
      "mean_token_accuracy": 0.19866774827241898,
      "num_tokens": 29289822.0,
      "step": 15720
    },
    {
      "entropy": 5.767150974273681,
      "epoch": 1.3886435888378665,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048115790473733124,
      "loss": 5.4258,
      "mean_token_accuracy": 0.19821067303419113,
      "num_tokens": 29299454.0,
      "step": 15725
    },
    {
      "entropy": 5.836018514633179,
      "epoch": 1.3890851289297068,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048114528753124625,
      "loss": 5.4575,
      "mean_token_accuracy": 0.19333788752555847,
      "num_tokens": 29308398.0,
      "step": 15730
    },
    {
      "entropy": 5.838931560516357,
      "epoch": 1.389526669021547,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00048113266628689236,
      "loss": 5.4201,
      "mean_token_accuracy": 0.2065933644771576,
      "num_tokens": 29318968.0,
      "step": 15735
    },
    {
      "entropy": 5.8998829364776615,
      "epoch": 1.3899682091133876,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004811200410045167,
      "loss": 5.4955,
      "mean_token_accuracy": 0.18485084027051926,
      "num_tokens": 29329745.0,
      "step": 15740
    },
    {
      "entropy": 5.811136388778687,
      "epoch": 1.3904097492052279,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004811074116843667,
      "loss": 5.3131,
      "mean_token_accuracy": 0.20991339683532714,
      "num_tokens": 29338438.0,
      "step": 15745
    },
    {
      "entropy": 5.779156637191773,
      "epoch": 1.3908512892970681,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048109477832668977,
      "loss": 5.3881,
      "mean_token_accuracy": 0.19703128039836884,
      "num_tokens": 29347783.0,
      "step": 15750
    },
    {
      "entropy": 5.817676305770874,
      "epoch": 1.3912928293889086,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004810821409317334,
      "loss": 5.451,
      "mean_token_accuracy": 0.20315441936254502,
      "num_tokens": 29357062.0,
      "step": 15755
    },
    {
      "entropy": 5.853411912918091,
      "epoch": 1.391734369480749,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048106949949974494,
      "loss": 5.4146,
      "mean_token_accuracy": 0.20485300123691558,
      "num_tokens": 29366361.0,
      "step": 15760
    },
    {
      "entropy": 5.779572868347168,
      "epoch": 1.3921759095725892,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004810568540309723,
      "loss": 5.4461,
      "mean_token_accuracy": 0.19637374430894852,
      "num_tokens": 29375992.0,
      "step": 15765
    },
    {
      "entropy": 5.84677357673645,
      "epoch": 1.3926174496644295,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048104420452566303,
      "loss": 5.4957,
      "mean_token_accuracy": 0.19725880324840545,
      "num_tokens": 29386408.0,
      "step": 15770
    },
    {
      "entropy": 5.929966163635254,
      "epoch": 1.3930589897562697,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048103155098406505,
      "loss": 5.5096,
      "mean_token_accuracy": 0.1918272316455841,
      "num_tokens": 29395881.0,
      "step": 15775
    },
    {
      "entropy": 5.824849271774292,
      "epoch": 1.3935005298481102,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048101889340642604,
      "loss": 5.3238,
      "mean_token_accuracy": 0.20882944762706757,
      "num_tokens": 29404397.0,
      "step": 15780
    },
    {
      "entropy": 5.768926668167114,
      "epoch": 1.3939420699399505,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004810062317929941,
      "loss": 5.3853,
      "mean_token_accuracy": 0.20247988849878312,
      "num_tokens": 29413219.0,
      "step": 15785
    },
    {
      "entropy": 5.858049106597901,
      "epoch": 1.3943836100317908,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004809935661440172,
      "loss": 5.4772,
      "mean_token_accuracy": 0.19423361569643022,
      "num_tokens": 29422727.0,
      "step": 15790
    },
    {
      "entropy": 5.911684560775757,
      "epoch": 1.3948251501236313,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004809808964597436,
      "loss": 5.4717,
      "mean_token_accuracy": 0.19965349435806273,
      "num_tokens": 29432183.0,
      "step": 15795
    },
    {
      "entropy": 5.823192358016968,
      "epoch": 1.3952666902154716,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048096822274042145,
      "loss": 5.3866,
      "mean_token_accuracy": 0.2023373946547508,
      "num_tokens": 29440821.0,
      "step": 15800
    },
    {
      "entropy": 5.864645719528198,
      "epoch": 1.3957082303073118,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048095554498629885,
      "loss": 5.4923,
      "mean_token_accuracy": 0.19196839183568953,
      "num_tokens": 29450128.0,
      "step": 15805
    },
    {
      "entropy": 5.82134780883789,
      "epoch": 1.3961497703991523,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048094286319762426,
      "loss": 5.3799,
      "mean_token_accuracy": 0.2021815672516823,
      "num_tokens": 29459271.0,
      "step": 15810
    },
    {
      "entropy": 5.82841420173645,
      "epoch": 1.3965913104909926,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048093017737464613,
      "loss": 5.3105,
      "mean_token_accuracy": 0.21012148708105088,
      "num_tokens": 29468517.0,
      "step": 15815
    },
    {
      "entropy": 5.800443077087403,
      "epoch": 1.3970328505828329,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048091748751761295,
      "loss": 5.5048,
      "mean_token_accuracy": 0.19221622943878175,
      "num_tokens": 29478330.0,
      "step": 15820
    },
    {
      "entropy": 5.881490993499756,
      "epoch": 1.3974743906746734,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004809047936267732,
      "loss": 5.4419,
      "mean_token_accuracy": 0.19671426713466644,
      "num_tokens": 29487190.0,
      "step": 15825
    },
    {
      "entropy": 5.894149541854858,
      "epoch": 1.3979159307665137,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048089209570237573,
      "loss": 5.5076,
      "mean_token_accuracy": 0.1914274886250496,
      "num_tokens": 29495885.0,
      "step": 15830
    },
    {
      "entropy": 5.834230709075928,
      "epoch": 1.398357470858354,
      "grad_norm": 1.0,
      "learning_rate": 0.0004808793937446692,
      "loss": 5.3744,
      "mean_token_accuracy": 0.19707161337137222,
      "num_tokens": 29505524.0,
      "step": 15835
    },
    {
      "entropy": 5.8079897403717045,
      "epoch": 1.3987990109501942,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004808666877539025,
      "loss": 5.3666,
      "mean_token_accuracy": 0.20344207137823106,
      "num_tokens": 29514003.0,
      "step": 15840
    },
    {
      "entropy": 5.819286441802978,
      "epoch": 1.3992405510420345,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048085397773032433,
      "loss": 5.3936,
      "mean_token_accuracy": 0.20126733928918839,
      "num_tokens": 29523236.0,
      "step": 15845
    },
    {
      "entropy": 5.875626707077027,
      "epoch": 1.399682091133875,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004808412636741839,
      "loss": 5.4526,
      "mean_token_accuracy": 0.20521328896284102,
      "num_tokens": 29532852.0,
      "step": 15850
    },
    {
      "entropy": 5.809407520294189,
      "epoch": 1.4001236312257153,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048082854558573017,
      "loss": 5.4183,
      "mean_token_accuracy": 0.20285482555627823,
      "num_tokens": 29542035.0,
      "step": 15855
    },
    {
      "entropy": 5.804487419128418,
      "epoch": 1.4005651713175555,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048081582346521236,
      "loss": 5.3698,
      "mean_token_accuracy": 0.20402562767267227,
      "num_tokens": 29551978.0,
      "step": 15860
    },
    {
      "entropy": 5.9571596622467045,
      "epoch": 1.401006711409396,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004808030973128795,
      "loss": 5.5206,
      "mean_token_accuracy": 0.19067658483982086,
      "num_tokens": 29562011.0,
      "step": 15865
    },
    {
      "entropy": 5.859301424026489,
      "epoch": 1.4014482515012363,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048079036712898106,
      "loss": 5.398,
      "mean_token_accuracy": 0.204546457529068,
      "num_tokens": 29571498.0,
      "step": 15870
    },
    {
      "entropy": 5.846553802490234,
      "epoch": 1.4018897915930766,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048077763291376637,
      "loss": 5.353,
      "mean_token_accuracy": 0.21147865504026414,
      "num_tokens": 29580704.0,
      "step": 15875
    },
    {
      "entropy": 5.852332878112793,
      "epoch": 1.402331331684917,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004807648946674849,
      "loss": 5.3271,
      "mean_token_accuracy": 0.20979054421186447,
      "num_tokens": 29590066.0,
      "step": 15880
    },
    {
      "entropy": 5.84045147895813,
      "epoch": 1.4027728717767574,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004807521523903862,
      "loss": 5.4466,
      "mean_token_accuracy": 0.19593202024698259,
      "num_tokens": 29599473.0,
      "step": 15885
    },
    {
      "entropy": 5.80124797821045,
      "epoch": 1.4032144118685976,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048073940608271974,
      "loss": 5.4356,
      "mean_token_accuracy": 0.1963476449251175,
      "num_tokens": 29609371.0,
      "step": 15890
    },
    {
      "entropy": 5.835998010635376,
      "epoch": 1.4036559519604381,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004807266557447354,
      "loss": 5.2955,
      "mean_token_accuracy": 0.20974289923906325,
      "num_tokens": 29618821.0,
      "step": 15895
    },
    {
      "entropy": 5.832048559188843,
      "epoch": 1.4040974920522784,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004807139013766829,
      "loss": 5.3281,
      "mean_token_accuracy": 0.21409380435943604,
      "num_tokens": 29627491.0,
      "step": 15900
    },
    {
      "entropy": 5.778929615020752,
      "epoch": 1.4045390321441187,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048070114297881205,
      "loss": 5.4345,
      "mean_token_accuracy": 0.20080992728471755,
      "num_tokens": 29637699.0,
      "step": 15905
    },
    {
      "entropy": 5.86026930809021,
      "epoch": 1.404980572235959,
      "grad_norm": 1.0,
      "learning_rate": 0.0004806883805513728,
      "loss": 5.3928,
      "mean_token_accuracy": 0.20685179382562638,
      "num_tokens": 29646439.0,
      "step": 15910
    },
    {
      "entropy": 5.87608232498169,
      "epoch": 1.4054221123277992,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004806756140946151,
      "loss": 5.4643,
      "mean_token_accuracy": 0.19283811748027802,
      "num_tokens": 29655704.0,
      "step": 15915
    },
    {
      "entropy": 5.909402322769165,
      "epoch": 1.4058636524196397,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048066284360878913,
      "loss": 5.5232,
      "mean_token_accuracy": 0.19277155250310898,
      "num_tokens": 29665163.0,
      "step": 15920
    },
    {
      "entropy": 5.838028860092163,
      "epoch": 1.40630519251148,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048065006909414507,
      "loss": 5.4532,
      "mean_token_accuracy": 0.2018750011920929,
      "num_tokens": 29674803.0,
      "step": 15925
    },
    {
      "entropy": 5.85578031539917,
      "epoch": 1.4067467326033203,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004806372905509331,
      "loss": 5.4448,
      "mean_token_accuracy": 0.19554970562458038,
      "num_tokens": 29685149.0,
      "step": 15930
    },
    {
      "entropy": 5.747699022293091,
      "epoch": 1.4071882726951608,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048062450797940347,
      "loss": 5.2908,
      "mean_token_accuracy": 0.21219534277915955,
      "num_tokens": 29694224.0,
      "step": 15935
    },
    {
      "entropy": 5.819090509414673,
      "epoch": 1.407629812787001,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004806117213798068,
      "loss": 5.4124,
      "mean_token_accuracy": 0.20134179890155793,
      "num_tokens": 29704512.0,
      "step": 15940
    },
    {
      "entropy": 5.799300861358643,
      "epoch": 1.4080713528788413,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048059893075239334,
      "loss": 5.3279,
      "mean_token_accuracy": 0.20564483851194382,
      "num_tokens": 29712990.0,
      "step": 15945
    },
    {
      "entropy": 5.856165552139283,
      "epoch": 1.4085128929706818,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048058613609741374,
      "loss": 5.4645,
      "mean_token_accuracy": 0.19290683269500733,
      "num_tokens": 29722844.0,
      "step": 15950
    },
    {
      "entropy": 5.8596423149108885,
      "epoch": 1.408954433062522,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004805733374151188,
      "loss": 5.3404,
      "mean_token_accuracy": 0.19993604123592376,
      "num_tokens": 29731593.0,
      "step": 15955
    },
    {
      "entropy": 5.820978832244873,
      "epoch": 1.4093959731543624,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048056053470575894,
      "loss": 5.4446,
      "mean_token_accuracy": 0.19267890006303787,
      "num_tokens": 29741052.0,
      "step": 15960
    },
    {
      "entropy": 5.846028423309326,
      "epoch": 1.4098375132462029,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048054772796958517,
      "loss": 5.429,
      "mean_token_accuracy": 0.197341625392437,
      "num_tokens": 29750187.0,
      "step": 15965
    },
    {
      "entropy": 5.770731258392334,
      "epoch": 1.4102790533380432,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048053491720684836,
      "loss": 5.3562,
      "mean_token_accuracy": 0.19872645139694214,
      "num_tokens": 29759574.0,
      "step": 15970
    },
    {
      "entropy": 5.780177783966065,
      "epoch": 1.4107205934298834,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004805221024177994,
      "loss": 5.4024,
      "mean_token_accuracy": 0.20029570162296295,
      "num_tokens": 29769549.0,
      "step": 15975
    },
    {
      "entropy": 5.880591773986817,
      "epoch": 1.4111621335217237,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004805092836026893,
      "loss": 5.4477,
      "mean_token_accuracy": 0.1947784662246704,
      "num_tokens": 29778643.0,
      "step": 15980
    },
    {
      "entropy": 5.864433097839355,
      "epoch": 1.411603673613564,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004804964607617693,
      "loss": 5.391,
      "mean_token_accuracy": 0.20077617317438126,
      "num_tokens": 29788685.0,
      "step": 15985
    },
    {
      "entropy": 5.836127805709839,
      "epoch": 1.4120452137054045,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048048363389529045,
      "loss": 5.3648,
      "mean_token_accuracy": 0.2051248475909233,
      "num_tokens": 29798648.0,
      "step": 15990
    },
    {
      "entropy": 5.820440435409546,
      "epoch": 1.4124867537972448,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000480470803003504,
      "loss": 5.4195,
      "mean_token_accuracy": 0.1922850042581558,
      "num_tokens": 29807582.0,
      "step": 15995
    },
    {
      "entropy": 5.840159368515015,
      "epoch": 1.412928293889085,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004804579680866614,
      "loss": 5.4258,
      "mean_token_accuracy": 0.1995154470205307,
      "num_tokens": 29817443.0,
      "step": 16000
    },
    {
      "entropy": 5.875101613998413,
      "epoch": 1.4133698339809255,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004804451291450142,
      "loss": 5.4363,
      "mean_token_accuracy": 0.20084422528743745,
      "num_tokens": 29825796.0,
      "step": 16005
    },
    {
      "entropy": 5.837054777145386,
      "epoch": 1.4138113740727658,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048043228617881365,
      "loss": 5.3585,
      "mean_token_accuracy": 0.20034700334072114,
      "num_tokens": 29834874.0,
      "step": 16010
    },
    {
      "entropy": 5.848668813705444,
      "epoch": 1.414252914164606,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004804194391883114,
      "loss": 5.4117,
      "mean_token_accuracy": 0.19753310233354568,
      "num_tokens": 29843884.0,
      "step": 16015
    },
    {
      "entropy": 5.919211912155151,
      "epoch": 1.4146944542564466,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048040658817375927,
      "loss": 5.5199,
      "mean_token_accuracy": 0.1868203714489937,
      "num_tokens": 29853180.0,
      "step": 16020
    },
    {
      "entropy": 5.853323316574096,
      "epoch": 1.4151359943482869,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048039373313540885,
      "loss": 5.384,
      "mean_token_accuracy": 0.19485014379024507,
      "num_tokens": 29862926.0,
      "step": 16025
    },
    {
      "entropy": 5.773474931716919,
      "epoch": 1.4155775344401271,
      "grad_norm": 1.109375,
      "learning_rate": 0.000480380874073512,
      "loss": 5.3709,
      "mean_token_accuracy": 0.20578792691230774,
      "num_tokens": 29871450.0,
      "step": 16030
    },
    {
      "entropy": 5.770298099517822,
      "epoch": 1.4160190745319676,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048036801098832067,
      "loss": 5.3539,
      "mean_token_accuracy": 0.20704010576009751,
      "num_tokens": 29880142.0,
      "step": 16035
    },
    {
      "entropy": 5.808156871795655,
      "epoch": 1.416460614623808,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004803551438800868,
      "loss": 5.4037,
      "mean_token_accuracy": 0.20299105495214462,
      "num_tokens": 29888668.0,
      "step": 16040
    },
    {
      "entropy": 5.906285238265991,
      "epoch": 1.4169021547156482,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048034227274906235,
      "loss": 5.5293,
      "mean_token_accuracy": 0.19299840480089187,
      "num_tokens": 29898727.0,
      "step": 16045
    },
    {
      "entropy": 5.859495830535889,
      "epoch": 1.4173436948074885,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048032939759549964,
      "loss": 5.441,
      "mean_token_accuracy": 0.18627241849899293,
      "num_tokens": 29909089.0,
      "step": 16050
    },
    {
      "entropy": 5.883098411560058,
      "epoch": 1.4177852348993287,
      "grad_norm": 1.0,
      "learning_rate": 0.00048031651841965085,
      "loss": 5.3967,
      "mean_token_accuracy": 0.2024922713637352,
      "num_tokens": 29918169.0,
      "step": 16055
    },
    {
      "entropy": 5.800869464874268,
      "epoch": 1.4182267749911692,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004803036352217682,
      "loss": 5.4241,
      "mean_token_accuracy": 0.2045391783118248,
      "num_tokens": 29928160.0,
      "step": 16060
    },
    {
      "entropy": 5.808574485778808,
      "epoch": 1.4186683150830095,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048029074800210413,
      "loss": 5.4323,
      "mean_token_accuracy": 0.1945517599582672,
      "num_tokens": 29937695.0,
      "step": 16065
    },
    {
      "entropy": 5.9447166442871096,
      "epoch": 1.4191098551748498,
      "grad_norm": 0.953125,
      "learning_rate": 0.000480277856760911,
      "loss": 5.4366,
      "mean_token_accuracy": 0.19815507382154465,
      "num_tokens": 29947380.0,
      "step": 16070
    },
    {
      "entropy": 5.945921468734741,
      "epoch": 1.4195513952666903,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048026496149844146,
      "loss": 5.4953,
      "mean_token_accuracy": 0.1980571538209915,
      "num_tokens": 29956313.0,
      "step": 16075
    },
    {
      "entropy": 5.783976030349732,
      "epoch": 1.4199929353585306,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048025206221494806,
      "loss": 5.3398,
      "mean_token_accuracy": 0.20875429213047028,
      "num_tokens": 29965502.0,
      "step": 16080
    },
    {
      "entropy": 5.830239534378052,
      "epoch": 1.4204344754503708,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004802391589106835,
      "loss": 5.3564,
      "mean_token_accuracy": 0.20533740669488906,
      "num_tokens": 29974243.0,
      "step": 16085
    },
    {
      "entropy": 5.792404699325561,
      "epoch": 1.4208760155422113,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004802262515859005,
      "loss": 5.3209,
      "mean_token_accuracy": 0.21151241958141326,
      "num_tokens": 29983966.0,
      "step": 16090
    },
    {
      "entropy": 5.828254079818725,
      "epoch": 1.4213175556340516,
      "grad_norm": 1.0,
      "learning_rate": 0.0004802133402408519,
      "loss": 5.4248,
      "mean_token_accuracy": 0.20044248104095458,
      "num_tokens": 29993365.0,
      "step": 16095
    },
    {
      "entropy": 5.9269355773925785,
      "epoch": 1.4217590957258919,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048020042487579076,
      "loss": 5.4326,
      "mean_token_accuracy": 0.2006969228386879,
      "num_tokens": 30001919.0,
      "step": 16100
    },
    {
      "entropy": 5.782835054397583,
      "epoch": 1.4222006358177324,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048018750549097,
      "loss": 5.3833,
      "mean_token_accuracy": 0.20162815004587173,
      "num_tokens": 30011353.0,
      "step": 16105
    },
    {
      "entropy": 5.840275001525879,
      "epoch": 1.4226421759095726,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048017458208664265,
      "loss": 5.4483,
      "mean_token_accuracy": 0.19493011832237245,
      "num_tokens": 30019994.0,
      "step": 16110
    },
    {
      "entropy": 5.760211086273193,
      "epoch": 1.423083716001413,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000480161654663062,
      "loss": 5.3076,
      "mean_token_accuracy": 0.20146130621433259,
      "num_tokens": 30028934.0,
      "step": 16115
    },
    {
      "entropy": 5.8535082817077635,
      "epoch": 1.4235252560932532,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004801487232204811,
      "loss": 5.3229,
      "mean_token_accuracy": 0.1995198518037796,
      "num_tokens": 30038069.0,
      "step": 16120
    },
    {
      "entropy": 5.909199523925781,
      "epoch": 1.4239667961850935,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004801357877591535,
      "loss": 5.4848,
      "mean_token_accuracy": 0.19611310362815856,
      "num_tokens": 30047307.0,
      "step": 16125
    },
    {
      "entropy": 5.845629167556763,
      "epoch": 1.424408336276934,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048012284827933245,
      "loss": 5.4188,
      "mean_token_accuracy": 0.20130494832992554,
      "num_tokens": 30057429.0,
      "step": 16130
    },
    {
      "entropy": 5.752436351776123,
      "epoch": 1.4248498763687742,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004801099047812715,
      "loss": 5.3537,
      "mean_token_accuracy": 0.20477347522974015,
      "num_tokens": 30067327.0,
      "step": 16135
    },
    {
      "entropy": 5.880493450164795,
      "epoch": 1.4252914164606145,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004800969572652241,
      "loss": 5.4739,
      "mean_token_accuracy": 0.19458369463682174,
      "num_tokens": 30077103.0,
      "step": 16140
    },
    {
      "entropy": 5.926301002502441,
      "epoch": 1.425732956552455,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048008400573144394,
      "loss": 5.5277,
      "mean_token_accuracy": 0.1887054920196533,
      "num_tokens": 30087539.0,
      "step": 16145
    },
    {
      "entropy": 5.919013166427613,
      "epoch": 1.4261744966442953,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048007105018018487,
      "loss": 5.3959,
      "mean_token_accuracy": 0.20030292570590974,
      "num_tokens": 30097153.0,
      "step": 16150
    },
    {
      "entropy": 5.850428581237793,
      "epoch": 1.4266160367361356,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048005809061170044,
      "loss": 5.4753,
      "mean_token_accuracy": 0.1989583507180214,
      "num_tokens": 30106291.0,
      "step": 16155
    },
    {
      "entropy": 5.855349159240722,
      "epoch": 1.427057576827976,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004800451270262447,
      "loss": 5.4798,
      "mean_token_accuracy": 0.19883015006780624,
      "num_tokens": 30116441.0,
      "step": 16160
    },
    {
      "entropy": 5.824636077880859,
      "epoch": 1.4274991169198163,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004800321594240714,
      "loss": 5.3588,
      "mean_token_accuracy": 0.21407222151756286,
      "num_tokens": 30125040.0,
      "step": 16165
    },
    {
      "entropy": 5.8391622543334964,
      "epoch": 1.4279406570116566,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004800191878054349,
      "loss": 5.3444,
      "mean_token_accuracy": 0.20534966886043549,
      "num_tokens": 30134284.0,
      "step": 16170
    },
    {
      "entropy": 5.86709623336792,
      "epoch": 1.4283821971034971,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048000621217058904,
      "loss": 5.3328,
      "mean_token_accuracy": 0.20978496223688126,
      "num_tokens": 30143679.0,
      "step": 16175
    },
    {
      "entropy": 5.7082648277282715,
      "epoch": 1.4288237371953374,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047999323251978805,
      "loss": 5.2711,
      "mean_token_accuracy": 0.21223915368318558,
      "num_tokens": 30152011.0,
      "step": 16180
    },
    {
      "entropy": 5.734315299987793,
      "epoch": 1.4292652772871777,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00047998024885328625,
      "loss": 5.3649,
      "mean_token_accuracy": 0.1998768299818039,
      "num_tokens": 30161949.0,
      "step": 16185
    },
    {
      "entropy": 5.853041696548462,
      "epoch": 1.429706817379018,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047996726117133795,
      "loss": 5.3856,
      "mean_token_accuracy": 0.2079288735985756,
      "num_tokens": 30172003.0,
      "step": 16190
    },
    {
      "entropy": 5.744950151443481,
      "epoch": 1.4301483574708582,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004799542694741975,
      "loss": 5.3959,
      "mean_token_accuracy": 0.20896569937467574,
      "num_tokens": 30181918.0,
      "step": 16195
    },
    {
      "entropy": 5.811241388320923,
      "epoch": 1.4305898975626987,
      "grad_norm": 1.0,
      "learning_rate": 0.00047994127376211953,
      "loss": 5.3457,
      "mean_token_accuracy": 0.20625263452529907,
      "num_tokens": 30191032.0,
      "step": 16200
    },
    {
      "entropy": 5.8363714694976805,
      "epoch": 1.431031437654539,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004799282740353586,
      "loss": 5.3662,
      "mean_token_accuracy": 0.20566608756780624,
      "num_tokens": 30200276.0,
      "step": 16205
    },
    {
      "entropy": 5.909934043884277,
      "epoch": 1.4314729777463793,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047991527029416924,
      "loss": 5.4271,
      "mean_token_accuracy": 0.20007840394973755,
      "num_tokens": 30208574.0,
      "step": 16210
    },
    {
      "entropy": 5.809680318832397,
      "epoch": 1.4319145178382198,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004799022625388064,
      "loss": 5.4345,
      "mean_token_accuracy": 0.20435420274734498,
      "num_tokens": 30218912.0,
      "step": 16215
    },
    {
      "entropy": 5.885587930679321,
      "epoch": 1.43235605793006,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004798892507695247,
      "loss": 5.5081,
      "mean_token_accuracy": 0.19482046067714692,
      "num_tokens": 30228325.0,
      "step": 16220
    },
    {
      "entropy": 5.860631465911865,
      "epoch": 1.4327975980219003,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004798762349865791,
      "loss": 5.3984,
      "mean_token_accuracy": 0.19849643260240554,
      "num_tokens": 30237169.0,
      "step": 16225
    },
    {
      "entropy": 5.825541543960571,
      "epoch": 1.4332391381137408,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004798632151902246,
      "loss": 5.3692,
      "mean_token_accuracy": 0.20470220744609832,
      "num_tokens": 30246452.0,
      "step": 16230
    },
    {
      "entropy": 5.887117719650268,
      "epoch": 1.433680678205581,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00047985019138071616,
      "loss": 5.4256,
      "mean_token_accuracy": 0.20170413553714753,
      "num_tokens": 30256055.0,
      "step": 16235
    },
    {
      "entropy": 5.784398937225342,
      "epoch": 1.4341222182974214,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047983716355830903,
      "loss": 5.2876,
      "mean_token_accuracy": 0.20524161607027053,
      "num_tokens": 30265764.0,
      "step": 16240
    },
    {
      "entropy": 5.830113935470581,
      "epoch": 1.4345637583892619,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047982413172325845,
      "loss": 5.4367,
      "mean_token_accuracy": 0.20573489218950272,
      "num_tokens": 30274490.0,
      "step": 16245
    },
    {
      "entropy": 5.811930418014526,
      "epoch": 1.4350052984811021,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047981109587581945,
      "loss": 5.3746,
      "mean_token_accuracy": 0.20781485587358475,
      "num_tokens": 30283330.0,
      "step": 16250
    },
    {
      "entropy": 5.882568454742431,
      "epoch": 1.4354468385729424,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004797980560162477,
      "loss": 5.5341,
      "mean_token_accuracy": 0.1854379415512085,
      "num_tokens": 30293542.0,
      "step": 16255
    },
    {
      "entropy": 5.853444290161133,
      "epoch": 1.4358883786647827,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004797850121447984,
      "loss": 5.4736,
      "mean_token_accuracy": 0.2044121265411377,
      "num_tokens": 30302122.0,
      "step": 16260
    },
    {
      "entropy": 5.838605976104736,
      "epoch": 1.436329918756623,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047977196426172725,
      "loss": 5.5062,
      "mean_token_accuracy": 0.2023172616958618,
      "num_tokens": 30311490.0,
      "step": 16265
    },
    {
      "entropy": 5.813236713409424,
      "epoch": 1.4367714588484635,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047975891236728973,
      "loss": 5.3439,
      "mean_token_accuracy": 0.2104988604784012,
      "num_tokens": 30321199.0,
      "step": 16270
    },
    {
      "entropy": 5.876508617401123,
      "epoch": 1.4372129989403037,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047974585646174165,
      "loss": 5.4044,
      "mean_token_accuracy": 0.1991390511393547,
      "num_tokens": 30330114.0,
      "step": 16275
    },
    {
      "entropy": 5.832000017166138,
      "epoch": 1.437654539032144,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047973279654533866,
      "loss": 5.4317,
      "mean_token_accuracy": 0.20468027740716935,
      "num_tokens": 30339290.0,
      "step": 16280
    },
    {
      "entropy": 5.803961992263794,
      "epoch": 1.4380960791239845,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004797197326183365,
      "loss": 5.3772,
      "mean_token_accuracy": 0.20148180723190307,
      "num_tokens": 30349595.0,
      "step": 16285
    },
    {
      "entropy": 5.9002259254455565,
      "epoch": 1.4385376192158248,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004797066646809914,
      "loss": 5.4182,
      "mean_token_accuracy": 0.1969104915857315,
      "num_tokens": 30358010.0,
      "step": 16290
    },
    {
      "entropy": 5.88068585395813,
      "epoch": 1.438979159307665,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047969359273355907,
      "loss": 5.4968,
      "mean_token_accuracy": 0.19931793510913848,
      "num_tokens": 30367408.0,
      "step": 16295
    },
    {
      "entropy": 5.829802513122559,
      "epoch": 1.4394206993995056,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047968051677629565,
      "loss": 5.4762,
      "mean_token_accuracy": 0.20065601766109467,
      "num_tokens": 30376475.0,
      "step": 16300
    },
    {
      "entropy": 5.870932626724243,
      "epoch": 1.4398622394913458,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047966743680945734,
      "loss": 5.3926,
      "mean_token_accuracy": 0.19485222548246384,
      "num_tokens": 30385361.0,
      "step": 16305
    },
    {
      "entropy": 5.937409687042236,
      "epoch": 1.4403037795831861,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004796543528333004,
      "loss": 5.4216,
      "mean_token_accuracy": 0.200754514336586,
      "num_tokens": 30393594.0,
      "step": 16310
    },
    {
      "entropy": 5.845773601531983,
      "epoch": 1.4407453196750266,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000479641264848081,
      "loss": 5.406,
      "mean_token_accuracy": 0.20125416219234465,
      "num_tokens": 30403444.0,
      "step": 16315
    },
    {
      "entropy": 5.743726015090942,
      "epoch": 1.441186859766867,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004796281728540556,
      "loss": 5.2932,
      "mean_token_accuracy": 0.21228694766759873,
      "num_tokens": 30412485.0,
      "step": 16320
    },
    {
      "entropy": 5.808514165878296,
      "epoch": 1.4416283998587072,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047961507685148077,
      "loss": 5.3553,
      "mean_token_accuracy": 0.20754000097513198,
      "num_tokens": 30421104.0,
      "step": 16325
    },
    {
      "entropy": 5.804881477355957,
      "epoch": 1.4420699399505474,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047960197684061287,
      "loss": 5.3779,
      "mean_token_accuracy": 0.20005034208297728,
      "num_tokens": 30430061.0,
      "step": 16330
    },
    {
      "entropy": 5.856679153442383,
      "epoch": 1.4425114800423877,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004795888728217086,
      "loss": 5.3118,
      "mean_token_accuracy": 0.20085889101028442,
      "num_tokens": 30437645.0,
      "step": 16335
    },
    {
      "entropy": 5.7644130229949955,
      "epoch": 1.4429530201342282,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00047957576479502466,
      "loss": 5.2905,
      "mean_token_accuracy": 0.21740503013134002,
      "num_tokens": 30447805.0,
      "step": 16340
    },
    {
      "entropy": 5.870812320709229,
      "epoch": 1.4433945602260685,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004795626527608179,
      "loss": 5.4356,
      "mean_token_accuracy": 0.19969286918640136,
      "num_tokens": 30457524.0,
      "step": 16345
    },
    {
      "entropy": 5.907554388046265,
      "epoch": 1.4438361003179088,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004795495367193451,
      "loss": 5.4057,
      "mean_token_accuracy": 0.2024523138999939,
      "num_tokens": 30467082.0,
      "step": 16350
    },
    {
      "entropy": 5.876122283935547,
      "epoch": 1.4442776404097493,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047953641667086315,
      "loss": 5.456,
      "mean_token_accuracy": 0.20295267701148986,
      "num_tokens": 30476251.0,
      "step": 16355
    },
    {
      "entropy": 5.787760448455811,
      "epoch": 1.4447191805015895,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047952329261562914,
      "loss": 5.3478,
      "mean_token_accuracy": 0.20849206298589706,
      "num_tokens": 30485997.0,
      "step": 16360
    },
    {
      "entropy": 5.886525630950928,
      "epoch": 1.4451607205934298,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047951016455390023,
      "loss": 5.4898,
      "mean_token_accuracy": 0.1977571427822113,
      "num_tokens": 30494886.0,
      "step": 16365
    },
    {
      "entropy": 5.880224132537842,
      "epoch": 1.4456022606852703,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004794970324859334,
      "loss": 5.4754,
      "mean_token_accuracy": 0.19239621758460998,
      "num_tokens": 30504194.0,
      "step": 16370
    },
    {
      "entropy": 5.791299486160279,
      "epoch": 1.4460438007771106,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047948389641198615,
      "loss": 5.343,
      "mean_token_accuracy": 0.20522153228521348,
      "num_tokens": 30513077.0,
      "step": 16375
    },
    {
      "entropy": 5.824699258804321,
      "epoch": 1.4464853408689509,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047947075633231553,
      "loss": 5.4307,
      "mean_token_accuracy": 0.19653559178113938,
      "num_tokens": 30521999.0,
      "step": 16380
    },
    {
      "entropy": 5.914393568038941,
      "epoch": 1.4469268809607914,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004794576122471791,
      "loss": 5.502,
      "mean_token_accuracy": 0.19254199117422105,
      "num_tokens": 30530138.0,
      "step": 16385
    },
    {
      "entropy": 5.813262462615967,
      "epoch": 1.4473684210526316,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004794444641568344,
      "loss": 5.3256,
      "mean_token_accuracy": 0.2041580557823181,
      "num_tokens": 30538850.0,
      "step": 16390
    },
    {
      "entropy": 5.8017425537109375,
      "epoch": 1.447809961144472,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004794313120615388,
      "loss": 5.4109,
      "mean_token_accuracy": 0.19747596085071564,
      "num_tokens": 30547593.0,
      "step": 16395
    },
    {
      "entropy": 5.81569275856018,
      "epoch": 1.4482515012363122,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004794181559615502,
      "loss": 5.375,
      "mean_token_accuracy": 0.20397378653287887,
      "num_tokens": 30557154.0,
      "step": 16400
    },
    {
      "entropy": 5.920955467224121,
      "epoch": 1.4486930413281525,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047940499585712615,
      "loss": 5.3749,
      "mean_token_accuracy": 0.20880068987607955,
      "num_tokens": 30566379.0,
      "step": 16405
    },
    {
      "entropy": 5.820366382598877,
      "epoch": 1.449134581419993,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047939183174852446,
      "loss": 5.3284,
      "mean_token_accuracy": 0.21193971037864684,
      "num_tokens": 30575146.0,
      "step": 16410
    },
    {
      "entropy": 5.720854997634888,
      "epoch": 1.4495761215118332,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047937866363600295,
      "loss": 5.4111,
      "mean_token_accuracy": 0.20401481986045839,
      "num_tokens": 30584444.0,
      "step": 16415
    },
    {
      "entropy": 5.8042151927948,
      "epoch": 1.4500176616036735,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004793654915198198,
      "loss": 5.3861,
      "mean_token_accuracy": 0.21130629181861876,
      "num_tokens": 30593779.0,
      "step": 16420
    },
    {
      "entropy": 5.881678485870362,
      "epoch": 1.450459201695514,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004793523154002328,
      "loss": 5.424,
      "mean_token_accuracy": 0.2025240734219551,
      "num_tokens": 30603762.0,
      "step": 16425
    },
    {
      "entropy": 5.867666101455688,
      "epoch": 1.4509007417873543,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047933913527750015,
      "loss": 5.3971,
      "mean_token_accuracy": 0.19658233374357223,
      "num_tokens": 30612907.0,
      "step": 16430
    },
    {
      "entropy": 5.781331539154053,
      "epoch": 1.4513422818791946,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004793259511518801,
      "loss": 5.3337,
      "mean_token_accuracy": 0.21251970529556274,
      "num_tokens": 30621537.0,
      "step": 16435
    },
    {
      "entropy": 5.820788431167602,
      "epoch": 1.451783821971035,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004793127630236309,
      "loss": 5.4593,
      "mean_token_accuracy": 0.19122121036052703,
      "num_tokens": 30632143.0,
      "step": 16440
    },
    {
      "entropy": 5.90724024772644,
      "epoch": 1.4522253620628753,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047929957089301085,
      "loss": 5.4049,
      "mean_token_accuracy": 0.19678455889225005,
      "num_tokens": 30641145.0,
      "step": 16445
    },
    {
      "entropy": 5.860531234741211,
      "epoch": 1.4526669021547156,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047928637476027836,
      "loss": 5.415,
      "mean_token_accuracy": 0.20139241963624954,
      "num_tokens": 30650036.0,
      "step": 16450
    },
    {
      "entropy": 5.873429489135742,
      "epoch": 1.453108442246556,
      "grad_norm": 0.921875,
      "learning_rate": 0.00047927317462569206,
      "loss": 5.3963,
      "mean_token_accuracy": 0.21078886091709137,
      "num_tokens": 30659552.0,
      "step": 16455
    },
    {
      "entropy": 5.8764019966125485,
      "epoch": 1.4535499823383964,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004792599704895103,
      "loss": 5.3763,
      "mean_token_accuracy": 0.20104623436927796,
      "num_tokens": 30668652.0,
      "step": 16460
    },
    {
      "entropy": 5.844933605194091,
      "epoch": 1.4539915224302367,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000479246762351992,
      "loss": 5.4471,
      "mean_token_accuracy": 0.19569310247898103,
      "num_tokens": 30678613.0,
      "step": 16465
    },
    {
      "entropy": 5.840695190429687,
      "epoch": 1.454433062522077,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047923355021339576,
      "loss": 5.4022,
      "mean_token_accuracy": 0.20952788591384888,
      "num_tokens": 30688259.0,
      "step": 16470
    },
    {
      "entropy": 5.840805149078369,
      "epoch": 1.4548746026139172,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004792203340739805,
      "loss": 5.4196,
      "mean_token_accuracy": 0.19956467747688295,
      "num_tokens": 30697309.0,
      "step": 16475
    },
    {
      "entropy": 5.902931213378906,
      "epoch": 1.4553161427057577,
      "grad_norm": 1.0,
      "learning_rate": 0.0004792071139340049,
      "loss": 5.4687,
      "mean_token_accuracy": 0.19744110852479935,
      "num_tokens": 30707278.0,
      "step": 16480
    },
    {
      "entropy": 5.849676942825317,
      "epoch": 1.455757682797598,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004791938897937281,
      "loss": 5.3574,
      "mean_token_accuracy": 0.20679961442947387,
      "num_tokens": 30716840.0,
      "step": 16485
    },
    {
      "entropy": 5.816055059432983,
      "epoch": 1.4561992228894383,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047918066165340926,
      "loss": 5.449,
      "mean_token_accuracy": 0.19277482032775878,
      "num_tokens": 30727120.0,
      "step": 16490
    },
    {
      "entropy": 5.786322355270386,
      "epoch": 1.4566407629812788,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004791674295133072,
      "loss": 5.344,
      "mean_token_accuracy": 0.20551401525735855,
      "num_tokens": 30735543.0,
      "step": 16495
    },
    {
      "entropy": 5.892681026458741,
      "epoch": 1.457082303073119,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047915419337368146,
      "loss": 5.4388,
      "mean_token_accuracy": 0.19324137419462203,
      "num_tokens": 30744971.0,
      "step": 16500
    },
    {
      "entropy": 5.79856390953064,
      "epoch": 1.4575238431649593,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047914095323479115,
      "loss": 5.2939,
      "mean_token_accuracy": 0.2104108974337578,
      "num_tokens": 30753444.0,
      "step": 16505
    },
    {
      "entropy": 5.828524255752564,
      "epoch": 1.4579653832567998,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004791277090968956,
      "loss": 5.3969,
      "mean_token_accuracy": 0.20633172988891602,
      "num_tokens": 30761469.0,
      "step": 16510
    },
    {
      "entropy": 5.801257085800171,
      "epoch": 1.45840692334864,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004791144609602544,
      "loss": 5.333,
      "mean_token_accuracy": 0.20391583293676377,
      "num_tokens": 30770773.0,
      "step": 16515
    },
    {
      "entropy": 5.832178211212158,
      "epoch": 1.4588484634404804,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047910120882512696,
      "loss": 5.4515,
      "mean_token_accuracy": 0.19713431894779204,
      "num_tokens": 30779624.0,
      "step": 16520
    },
    {
      "entropy": 5.818113851547241,
      "epoch": 1.4592900035323209,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004790879526917729,
      "loss": 5.348,
      "mean_token_accuracy": 0.19848013520240784,
      "num_tokens": 30788671.0,
      "step": 16525
    },
    {
      "entropy": 5.8809247493743895,
      "epoch": 1.4597315436241611,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004790746925604519,
      "loss": 5.4282,
      "mean_token_accuracy": 0.20235230326652526,
      "num_tokens": 30797470.0,
      "step": 16530
    },
    {
      "entropy": 5.871769428253174,
      "epoch": 1.4601730837160014,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047906142843142383,
      "loss": 5.4105,
      "mean_token_accuracy": 0.20621410459280015,
      "num_tokens": 30807209.0,
      "step": 16535
    },
    {
      "entropy": 5.897931671142578,
      "epoch": 1.4606146238078417,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047904816030494834,
      "loss": 5.4344,
      "mean_token_accuracy": 0.1966389685869217,
      "num_tokens": 30816156.0,
      "step": 16540
    },
    {
      "entropy": 5.78818564414978,
      "epoch": 1.461056163899682,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004790348881812855,
      "loss": 5.3509,
      "mean_token_accuracy": 0.21043713688850402,
      "num_tokens": 30825439.0,
      "step": 16545
    },
    {
      "entropy": 5.749357604980469,
      "epoch": 1.4614977039915225,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047902161206069514,
      "loss": 5.3694,
      "mean_token_accuracy": 0.20392918437719346,
      "num_tokens": 30834953.0,
      "step": 16550
    },
    {
      "entropy": 5.8226083278656,
      "epoch": 1.4619392440833627,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047900833194343764,
      "loss": 5.4254,
      "mean_token_accuracy": 0.1996046185493469,
      "num_tokens": 30843534.0,
      "step": 16555
    },
    {
      "entropy": 5.79872236251831,
      "epoch": 1.462380784175203,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004789950478297727,
      "loss": 5.3252,
      "mean_token_accuracy": 0.21067256182432176,
      "num_tokens": 30851869.0,
      "step": 16560
    },
    {
      "entropy": 5.835751295089722,
      "epoch": 1.4628223242670435,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047898175971996095,
      "loss": 5.4571,
      "mean_token_accuracy": 0.19869925528764726,
      "num_tokens": 30862262.0,
      "step": 16565
    },
    {
      "entropy": 5.849185752868652,
      "epoch": 1.4632638643588838,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047896846761426255,
      "loss": 5.4567,
      "mean_token_accuracy": 0.19498667269945144,
      "num_tokens": 30872223.0,
      "step": 16570
    },
    {
      "entropy": 5.840133619308472,
      "epoch": 1.463705404450724,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004789551715129378,
      "loss": 5.3781,
      "mean_token_accuracy": 0.19859335124492644,
      "num_tokens": 30881834.0,
      "step": 16575
    },
    {
      "entropy": 5.875336408615112,
      "epoch": 1.4641469445425646,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004789418714162472,
      "loss": 5.4652,
      "mean_token_accuracy": 0.19814420938491822,
      "num_tokens": 30893414.0,
      "step": 16580
    },
    {
      "entropy": 5.881783103942871,
      "epoch": 1.4645884846344048,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004789285673244514,
      "loss": 5.4899,
      "mean_token_accuracy": 0.19113173335790634,
      "num_tokens": 30902345.0,
      "step": 16585
    },
    {
      "entropy": 5.699627304077149,
      "epoch": 1.465030024726245,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004789152592378109,
      "loss": 5.2066,
      "mean_token_accuracy": 0.22669571340084077,
      "num_tokens": 30911619.0,
      "step": 16590
    },
    {
      "entropy": 5.832030534744263,
      "epoch": 1.4654715648180856,
      "grad_norm": 0.9375,
      "learning_rate": 0.00047890194715658645,
      "loss": 5.4191,
      "mean_token_accuracy": 0.1936587542295456,
      "num_tokens": 30921515.0,
      "step": 16595
    },
    {
      "entropy": 5.884548616409302,
      "epoch": 1.4659131049099259,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004788886310810389,
      "loss": 5.3857,
      "mean_token_accuracy": 0.20652967393398286,
      "num_tokens": 30930807.0,
      "step": 16600
    },
    {
      "entropy": 5.923854541778565,
      "epoch": 1.4663546450017662,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047887531101142893,
      "loss": 5.4646,
      "mean_token_accuracy": 0.19024052768945693,
      "num_tokens": 30939266.0,
      "step": 16605
    },
    {
      "entropy": 5.85533881187439,
      "epoch": 1.4667961850936064,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047886198694801764,
      "loss": 5.363,
      "mean_token_accuracy": 0.19712022840976715,
      "num_tokens": 30948119.0,
      "step": 16610
    },
    {
      "entropy": 5.741694116592408,
      "epoch": 1.4672377251854467,
      "grad_norm": 1.0,
      "learning_rate": 0.00047884865889106593,
      "loss": 5.2957,
      "mean_token_accuracy": 0.21649323254823685,
      "num_tokens": 30957005.0,
      "step": 16615
    },
    {
      "entropy": 5.720598888397217,
      "epoch": 1.4676792652772872,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047883532684083493,
      "loss": 5.3568,
      "mean_token_accuracy": 0.2143753722310066,
      "num_tokens": 30967144.0,
      "step": 16620
    },
    {
      "entropy": 5.863882112503052,
      "epoch": 1.4681208053691275,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004788219907975858,
      "loss": 5.3128,
      "mean_token_accuracy": 0.2041093736886978,
      "num_tokens": 30975315.0,
      "step": 16625
    },
    {
      "entropy": 5.909586763381958,
      "epoch": 1.4685623454609678,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004788086507615798,
      "loss": 5.4839,
      "mean_token_accuracy": 0.19469848722219468,
      "num_tokens": 30984705.0,
      "step": 16630
    },
    {
      "entropy": 5.834614658355713,
      "epoch": 1.4690038855528083,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004787953067330782,
      "loss": 5.4708,
      "mean_token_accuracy": 0.19439245462417604,
      "num_tokens": 30994936.0,
      "step": 16635
    },
    {
      "entropy": 5.760412549972534,
      "epoch": 1.4694454256446485,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004787819587123425,
      "loss": 5.3862,
      "mean_token_accuracy": 0.19505118429660798,
      "num_tokens": 31004011.0,
      "step": 16640
    },
    {
      "entropy": 5.916839361190796,
      "epoch": 1.4698869657364888,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004787686066996341,
      "loss": 5.4285,
      "mean_token_accuracy": 0.19813136905431747,
      "num_tokens": 31013852.0,
      "step": 16645
    },
    {
      "entropy": 5.831754207611084,
      "epoch": 1.4703285058283293,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047875525069521457,
      "loss": 5.4023,
      "mean_token_accuracy": 0.20186276137828826,
      "num_tokens": 31023456.0,
      "step": 16650
    },
    {
      "entropy": 5.887347459793091,
      "epoch": 1.4707700459201696,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004787418906993456,
      "loss": 5.4488,
      "mean_token_accuracy": 0.1886138692498207,
      "num_tokens": 31032828.0,
      "step": 16655
    },
    {
      "entropy": 5.810969734191895,
      "epoch": 1.4712115860120099,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00047872852671228884,
      "loss": 5.3751,
      "mean_token_accuracy": 0.21169993728399278,
      "num_tokens": 31043017.0,
      "step": 16660
    },
    {
      "entropy": 5.74207067489624,
      "epoch": 1.4716531261038504,
      "grad_norm": 0.921875,
      "learning_rate": 0.00047871515873430614,
      "loss": 5.2896,
      "mean_token_accuracy": 0.21016721278429032,
      "num_tokens": 31051491.0,
      "step": 16665
    },
    {
      "entropy": 5.761693859100342,
      "epoch": 1.4720946661956906,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004787017867656593,
      "loss": 5.2881,
      "mean_token_accuracy": 0.2228364810347557,
      "num_tokens": 31060375.0,
      "step": 16670
    },
    {
      "entropy": 5.9136261463165285,
      "epoch": 1.472536206287531,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004786884108066103,
      "loss": 5.5058,
      "mean_token_accuracy": 0.18943930566310882,
      "num_tokens": 31071410.0,
      "step": 16675
    },
    {
      "entropy": 5.833731985092163,
      "epoch": 1.4729777463793712,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004786750308574212,
      "loss": 5.2514,
      "mean_token_accuracy": 0.22078309804201127,
      "num_tokens": 31080123.0,
      "step": 16680
    },
    {
      "entropy": 5.823306083679199,
      "epoch": 1.4734192864712115,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004786616469183541,
      "loss": 5.4124,
      "mean_token_accuracy": 0.20207696706056594,
      "num_tokens": 31089870.0,
      "step": 16685
    },
    {
      "entropy": 5.803851175308227,
      "epoch": 1.473860826563052,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004786482589896711,
      "loss": 5.4401,
      "mean_token_accuracy": 0.19819240123033524,
      "num_tokens": 31098682.0,
      "step": 16690
    },
    {
      "entropy": 5.885953378677368,
      "epoch": 1.4743023666548922,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004786348670716346,
      "loss": 5.4537,
      "mean_token_accuracy": 0.1934850126504898,
      "num_tokens": 31107785.0,
      "step": 16695
    },
    {
      "entropy": 5.81049337387085,
      "epoch": 1.4747439067467325,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004786214711645068,
      "loss": 5.2774,
      "mean_token_accuracy": 0.21126044541597366,
      "num_tokens": 31116853.0,
      "step": 16700
    },
    {
      "entropy": 5.820914459228516,
      "epoch": 1.475185446838573,
      "grad_norm": 1.0,
      "learning_rate": 0.0004786080712685503,
      "loss": 5.3776,
      "mean_token_accuracy": 0.20030050873756408,
      "num_tokens": 31126260.0,
      "step": 16705
    },
    {
      "entropy": 5.85260272026062,
      "epoch": 1.4756269869304133,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047859466738402743,
      "loss": 5.4482,
      "mean_token_accuracy": 0.19973967522382735,
      "num_tokens": 31135047.0,
      "step": 16710
    },
    {
      "entropy": 5.844044208526611,
      "epoch": 1.4760685270222536,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047858125951120073,
      "loss": 5.376,
      "mean_token_accuracy": 0.2031434953212738,
      "num_tokens": 31144177.0,
      "step": 16715
    },
    {
      "entropy": 5.85214319229126,
      "epoch": 1.476510067114094,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000478567847650333,
      "loss": 5.464,
      "mean_token_accuracy": 0.19586363881826402,
      "num_tokens": 31153586.0,
      "step": 16720
    },
    {
      "entropy": 5.929026174545288,
      "epoch": 1.4769516072059343,
      "grad_norm": 1.046875,
      "learning_rate": 0.000478554431801687,
      "loss": 5.5425,
      "mean_token_accuracy": 0.19936016350984573,
      "num_tokens": 31163252.0,
      "step": 16725
    },
    {
      "entropy": 5.7872968196868895,
      "epoch": 1.4773931472977746,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004785410119655253,
      "loss": 5.3365,
      "mean_token_accuracy": 0.20308424681425094,
      "num_tokens": 31171947.0,
      "step": 16730
    },
    {
      "entropy": 5.809985160827637,
      "epoch": 1.477834687389615,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047852758814211104,
      "loss": 5.3693,
      "mean_token_accuracy": 0.20936822295188903,
      "num_tokens": 31180502.0,
      "step": 16735
    },
    {
      "entropy": 5.8745352268219,
      "epoch": 1.4782762274814554,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047851416033170704,
      "loss": 5.4681,
      "mean_token_accuracy": 0.19506845772266387,
      "num_tokens": 31190744.0,
      "step": 16740
    },
    {
      "entropy": 5.9168616771698,
      "epoch": 1.4787177675732956,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004785007285345764,
      "loss": 5.4904,
      "mean_token_accuracy": 0.19723411202430724,
      "num_tokens": 31200869.0,
      "step": 16745
    },
    {
      "entropy": 5.8352379322052,
      "epoch": 1.479159307665136,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004784872927509822,
      "loss": 5.3846,
      "mean_token_accuracy": 0.20750546008348464,
      "num_tokens": 31210046.0,
      "step": 16750
    },
    {
      "entropy": 5.818807458877563,
      "epoch": 1.4796008477569762,
      "grad_norm": 1.125,
      "learning_rate": 0.0004784738529811877,
      "loss": 5.3184,
      "mean_token_accuracy": 0.20589302629232406,
      "num_tokens": 31219140.0,
      "step": 16755
    },
    {
      "entropy": 5.808164453506469,
      "epoch": 1.4800423878488167,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047846040922545604,
      "loss": 5.3339,
      "mean_token_accuracy": 0.20964421331882477,
      "num_tokens": 31227993.0,
      "step": 16760
    },
    {
      "entropy": 5.819470500946045,
      "epoch": 1.480483927940657,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047844696148405074,
      "loss": 5.3803,
      "mean_token_accuracy": 0.20366198271512986,
      "num_tokens": 31236959.0,
      "step": 16765
    },
    {
      "entropy": 5.843044471740723,
      "epoch": 1.4809254680324972,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047843350975723523,
      "loss": 5.291,
      "mean_token_accuracy": 0.21459991186857225,
      "num_tokens": 31246373.0,
      "step": 16770
    },
    {
      "entropy": 5.809376668930054,
      "epoch": 1.4813670081243377,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004784200540452729,
      "loss": 5.4085,
      "mean_token_accuracy": 0.20238426178693772,
      "num_tokens": 31255556.0,
      "step": 16775
    },
    {
      "entropy": 5.785092353820801,
      "epoch": 1.481808548216178,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004784065943484274,
      "loss": 5.3466,
      "mean_token_accuracy": 0.20924828052520753,
      "num_tokens": 31264824.0,
      "step": 16780
    },
    {
      "entropy": 5.776709032058716,
      "epoch": 1.4822500883080183,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047839313066696246,
      "loss": 5.3098,
      "mean_token_accuracy": 0.2110050603747368,
      "num_tokens": 31273946.0,
      "step": 16785
    },
    {
      "entropy": 5.8087320804595945,
      "epoch": 1.4826916283998588,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047837966300114164,
      "loss": 5.3888,
      "mean_token_accuracy": 0.1979956343770027,
      "num_tokens": 31283492.0,
      "step": 16790
    },
    {
      "entropy": 5.894581937789917,
      "epoch": 1.483133168491699,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00047836619135122906,
      "loss": 5.4219,
      "mean_token_accuracy": 0.20325595438480376,
      "num_tokens": 31292250.0,
      "step": 16795
    },
    {
      "entropy": 5.889112043380737,
      "epoch": 1.4835747085835393,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047835271571748834,
      "loss": 5.5386,
      "mean_token_accuracy": 0.18938992619514466,
      "num_tokens": 31302868.0,
      "step": 16800
    },
    {
      "entropy": 5.839573812484741,
      "epoch": 1.4840162486753798,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004783392361001836,
      "loss": 5.3218,
      "mean_token_accuracy": 0.20803227424621581,
      "num_tokens": 31311918.0,
      "step": 16805
    },
    {
      "entropy": 5.7997101783752445,
      "epoch": 1.4844577887672201,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004783257524995788,
      "loss": 5.3219,
      "mean_token_accuracy": 0.20955585688352585,
      "num_tokens": 31321339.0,
      "step": 16810
    },
    {
      "entropy": 5.81311264038086,
      "epoch": 1.4848993288590604,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004783122649159383,
      "loss": 5.3268,
      "mean_token_accuracy": 0.20057355910539626,
      "num_tokens": 31330498.0,
      "step": 16815
    },
    {
      "entropy": 5.801687622070313,
      "epoch": 1.4853408689509007,
      "grad_norm": 1.0625,
      "learning_rate": 0.000478298773349526,
      "loss": 5.3472,
      "mean_token_accuracy": 0.2093326896429062,
      "num_tokens": 31339676.0,
      "step": 16820
    },
    {
      "entropy": 5.912374877929688,
      "epoch": 1.485782409042741,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004782852778006064,
      "loss": 5.4837,
      "mean_token_accuracy": 0.19155890196561814,
      "num_tokens": 31350072.0,
      "step": 16825
    },
    {
      "entropy": 5.8931622982025145,
      "epoch": 1.4862239491345814,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047827177826944396,
      "loss": 5.3733,
      "mean_token_accuracy": 0.20686384290456772,
      "num_tokens": 31359992.0,
      "step": 16830
    },
    {
      "entropy": 5.859904432296753,
      "epoch": 1.4866654892264217,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047825827475630285,
      "loss": 5.3776,
      "mean_token_accuracy": 0.2051248475909233,
      "num_tokens": 31369434.0,
      "step": 16835
    },
    {
      "entropy": 5.804741477966308,
      "epoch": 1.487107029318262,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047824476726144775,
      "loss": 5.3757,
      "mean_token_accuracy": 0.20290971994400026,
      "num_tokens": 31378773.0,
      "step": 16840
    },
    {
      "entropy": 5.8703711986541744,
      "epoch": 1.4875485694101025,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047823125578514325,
      "loss": 5.4194,
      "mean_token_accuracy": 0.2034597411751747,
      "num_tokens": 31388219.0,
      "step": 16845
    },
    {
      "entropy": 5.788036298751831,
      "epoch": 1.4879901095019428,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047821774032765407,
      "loss": 5.3301,
      "mean_token_accuracy": 0.20831093341112136,
      "num_tokens": 31397216.0,
      "step": 16850
    },
    {
      "entropy": 5.775175380706787,
      "epoch": 1.488431649593783,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004782042208892449,
      "loss": 5.3165,
      "mean_token_accuracy": 0.21075984090566635,
      "num_tokens": 31406861.0,
      "step": 16855
    },
    {
      "entropy": 5.803642892837525,
      "epoch": 1.4888731896856235,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047819069747018065,
      "loss": 5.3967,
      "mean_token_accuracy": 0.20211237967014312,
      "num_tokens": 31416387.0,
      "step": 16860
    },
    {
      "entropy": 5.804005718231201,
      "epoch": 1.4893147297774638,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004781771700707262,
      "loss": 5.299,
      "mean_token_accuracy": 0.20730532258749007,
      "num_tokens": 31425056.0,
      "step": 16865
    },
    {
      "entropy": 5.893209075927734,
      "epoch": 1.489756269869304,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004781636386911465,
      "loss": 5.5122,
      "mean_token_accuracy": 0.19564283192157744,
      "num_tokens": 31435746.0,
      "step": 16870
    },
    {
      "entropy": 5.914508438110351,
      "epoch": 1.4901978099611446,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004781501033317067,
      "loss": 5.5373,
      "mean_token_accuracy": 0.1911771312355995,
      "num_tokens": 31444808.0,
      "step": 16875
    },
    {
      "entropy": 5.8947224617004395,
      "epoch": 1.4906393500529849,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004781365639926718,
      "loss": 5.4274,
      "mean_token_accuracy": 0.20052673369646073,
      "num_tokens": 31454205.0,
      "step": 16880
    },
    {
      "entropy": 5.874072456359864,
      "epoch": 1.4910808901448251,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004781230206743073,
      "loss": 5.4409,
      "mean_token_accuracy": 0.19775310009717942,
      "num_tokens": 31463446.0,
      "step": 16885
    },
    {
      "entropy": 5.805663919448852,
      "epoch": 1.4915224302366654,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004781094733768783,
      "loss": 5.3927,
      "mean_token_accuracy": 0.20490449219942092,
      "num_tokens": 31473168.0,
      "step": 16890
    },
    {
      "entropy": 5.82881269454956,
      "epoch": 1.4919639703285057,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047809592210065017,
      "loss": 5.3025,
      "mean_token_accuracy": 0.20378587692975997,
      "num_tokens": 31482899.0,
      "step": 16895
    },
    {
      "entropy": 5.861395931243896,
      "epoch": 1.4924055104203462,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047808236684588847,
      "loss": 5.2961,
      "mean_token_accuracy": 0.20091051012277603,
      "num_tokens": 31492503.0,
      "step": 16900
    },
    {
      "entropy": 5.72953085899353,
      "epoch": 1.4928470505121865,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004780688076128587,
      "loss": 5.2802,
      "mean_token_accuracy": 0.21831628531217576,
      "num_tokens": 31501520.0,
      "step": 16905
    },
    {
      "entropy": 5.8094563484191895,
      "epoch": 1.4932885906040267,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047805524440182654,
      "loss": 5.433,
      "mean_token_accuracy": 0.20691922903060914,
      "num_tokens": 31510821.0,
      "step": 16910
    },
    {
      "entropy": 5.851106786727906,
      "epoch": 1.4937301306958672,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047804167721305756,
      "loss": 5.3937,
      "mean_token_accuracy": 0.20294977724552155,
      "num_tokens": 31520472.0,
      "step": 16915
    },
    {
      "entropy": 5.836772918701172,
      "epoch": 1.4941716707877075,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047802810604681757,
      "loss": 5.4105,
      "mean_token_accuracy": 0.20710429102182387,
      "num_tokens": 31529129.0,
      "step": 16920
    },
    {
      "entropy": 5.7587299823760985,
      "epoch": 1.4946132108795478,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047801453090337256,
      "loss": 5.3587,
      "mean_token_accuracy": 0.20782049894332885,
      "num_tokens": 31537897.0,
      "step": 16925
    },
    {
      "entropy": 5.860333633422852,
      "epoch": 1.4950547509713883,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004780009517829882,
      "loss": 5.3624,
      "mean_token_accuracy": 0.2069771721959114,
      "num_tokens": 31547656.0,
      "step": 16930
    },
    {
      "entropy": 5.825451326370239,
      "epoch": 1.4954962910632286,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004779873686859308,
      "loss": 5.3736,
      "mean_token_accuracy": 0.20781843662261962,
      "num_tokens": 31557303.0,
      "step": 16935
    },
    {
      "entropy": 5.906974411010742,
      "epoch": 1.4959378311550688,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047797378161246613,
      "loss": 5.3926,
      "mean_token_accuracy": 0.1985720783472061,
      "num_tokens": 31565749.0,
      "step": 16940
    },
    {
      "entropy": 5.854325914382935,
      "epoch": 1.4963793712469093,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047796019056286067,
      "loss": 5.4327,
      "mean_token_accuracy": 0.20241083800792695,
      "num_tokens": 31575286.0,
      "step": 16945
    },
    {
      "entropy": 5.887064981460571,
      "epoch": 1.4968209113387496,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004779465955373804,
      "loss": 5.375,
      "mean_token_accuracy": 0.20670202523469924,
      "num_tokens": 31583723.0,
      "step": 16950
    },
    {
      "entropy": 5.742811679840088,
      "epoch": 1.49726245143059,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004779329965362918,
      "loss": 5.2181,
      "mean_token_accuracy": 0.22568588852882385,
      "num_tokens": 31593171.0,
      "step": 16955
    },
    {
      "entropy": 5.792586898803711,
      "epoch": 1.4977039915224302,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047791939355986123,
      "loss": 5.4014,
      "mean_token_accuracy": 0.19927307814359665,
      "num_tokens": 31603958.0,
      "step": 16960
    },
    {
      "entropy": 5.823423433303833,
      "epoch": 1.4981455316142704,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004779057866083551,
      "loss": 5.3901,
      "mean_token_accuracy": 0.19900340884923934,
      "num_tokens": 31612460.0,
      "step": 16965
    },
    {
      "entropy": 5.917697954177856,
      "epoch": 1.498587071706111,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047789217568204004,
      "loss": 5.446,
      "mean_token_accuracy": 0.19984558671712876,
      "num_tokens": 31622114.0,
      "step": 16970
    },
    {
      "entropy": 5.877597141265869,
      "epoch": 1.4990286117979512,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047787856078118266,
      "loss": 5.3875,
      "mean_token_accuracy": 0.20543920993804932,
      "num_tokens": 31632272.0,
      "step": 16975
    },
    {
      "entropy": 5.819973516464233,
      "epoch": 1.4994701518897915,
      "grad_norm": 1.0,
      "learning_rate": 0.0004778649419060496,
      "loss": 5.352,
      "mean_token_accuracy": 0.20504435002803803,
      "num_tokens": 31642235.0,
      "step": 16980
    },
    {
      "entropy": 5.876267766952514,
      "epoch": 1.499911691981632,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004778513190569077,
      "loss": 5.4591,
      "mean_token_accuracy": 0.1948750615119934,
      "num_tokens": 31652773.0,
      "step": 16985
    },
    {
      "entropy": 5.9093646049499515,
      "epoch": 1.5003532320734723,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004778376922340239,
      "loss": 5.4106,
      "mean_token_accuracy": 0.2044292151927948,
      "num_tokens": 31662681.0,
      "step": 16990
    },
    {
      "entropy": 5.7939286708831785,
      "epoch": 1.5007947721653125,
      "grad_norm": 1.0625,
      "learning_rate": 0.000477824061437665,
      "loss": 5.3156,
      "mean_token_accuracy": 0.20705605000257493,
      "num_tokens": 31671576.0,
      "step": 16995
    },
    {
      "entropy": 5.883949708938599,
      "epoch": 1.501236312257153,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047781042666809816,
      "loss": 5.554,
      "mean_token_accuracy": 0.18697230219841005,
      "num_tokens": 31681149.0,
      "step": 17000
    },
    {
      "entropy": 5.8254352569580075,
      "epoch": 1.5016778523489933,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004777967879255904,
      "loss": 5.3622,
      "mean_token_accuracy": 0.20532894134521484,
      "num_tokens": 31690007.0,
      "step": 17005
    },
    {
      "entropy": 5.958875370025635,
      "epoch": 1.5021193924408336,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004777831452104088,
      "loss": 5.4937,
      "mean_token_accuracy": 0.19368649274110794,
      "num_tokens": 31700055.0,
      "step": 17010
    },
    {
      "entropy": 5.853254556655884,
      "epoch": 1.502560932532674,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004777694985228208,
      "loss": 5.3876,
      "mean_token_accuracy": 0.20470905303955078,
      "num_tokens": 31708255.0,
      "step": 17015
    },
    {
      "entropy": 5.821598052978516,
      "epoch": 1.5030024726245141,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047775584786309365,
      "loss": 5.4398,
      "mean_token_accuracy": 0.19863502830266952,
      "num_tokens": 31716855.0,
      "step": 17020
    },
    {
      "entropy": 5.816057014465332,
      "epoch": 1.5034440127163546,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047774219323149474,
      "loss": 5.3706,
      "mean_token_accuracy": 0.2109145477414131,
      "num_tokens": 31726153.0,
      "step": 17025
    },
    {
      "entropy": 5.812814950942993,
      "epoch": 1.5038855528081951,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004777285346282916,
      "loss": 5.3558,
      "mean_token_accuracy": 0.2040833353996277,
      "num_tokens": 31736935.0,
      "step": 17030
    },
    {
      "entropy": 5.881922674179077,
      "epoch": 1.5043270929000352,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004777148720537517,
      "loss": 5.4544,
      "mean_token_accuracy": 0.1992572620511055,
      "num_tokens": 31746402.0,
      "step": 17035
    },
    {
      "entropy": 5.831804084777832,
      "epoch": 1.5047686329918757,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047770120550814276,
      "loss": 5.2661,
      "mean_token_accuracy": 0.2085268184542656,
      "num_tokens": 31755615.0,
      "step": 17040
    },
    {
      "entropy": 5.788525199890136,
      "epoch": 1.505210173083716,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004776875349917325,
      "loss": 5.4364,
      "mean_token_accuracy": 0.2040776476264,
      "num_tokens": 31765701.0,
      "step": 17045
    },
    {
      "entropy": 5.850245380401612,
      "epoch": 1.5056517131755562,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004776738605047887,
      "loss": 5.3964,
      "mean_token_accuracy": 0.20048863142728807,
      "num_tokens": 31774674.0,
      "step": 17050
    },
    {
      "entropy": 5.85404896736145,
      "epoch": 1.5060932532673967,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004776601820475793,
      "loss": 5.3733,
      "mean_token_accuracy": 0.20120461881160737,
      "num_tokens": 31784871.0,
      "step": 17055
    },
    {
      "entropy": 5.8014312267303465,
      "epoch": 1.506534793359237,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004776464996203721,
      "loss": 5.3261,
      "mean_token_accuracy": 0.21462402045726775,
      "num_tokens": 31793274.0,
      "step": 17060
    },
    {
      "entropy": 5.838164901733398,
      "epoch": 1.5069763334510773,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004776328132234353,
      "loss": 5.5001,
      "mean_token_accuracy": 0.19429777264595033,
      "num_tokens": 31803543.0,
      "step": 17065
    },
    {
      "entropy": 5.971688938140869,
      "epoch": 1.5074178735429178,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004776191228570368,
      "loss": 5.4657,
      "mean_token_accuracy": 0.19177936017513275,
      "num_tokens": 31814529.0,
      "step": 17070
    },
    {
      "entropy": 5.94747486114502,
      "epoch": 1.507859413634758,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004776054285214451,
      "loss": 5.4404,
      "mean_token_accuracy": 0.20195518881082536,
      "num_tokens": 31824064.0,
      "step": 17075
    },
    {
      "entropy": 5.798696613311767,
      "epoch": 1.5083009537265983,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047759173021692813,
      "loss": 5.4117,
      "mean_token_accuracy": 0.1961733281612396,
      "num_tokens": 31833881.0,
      "step": 17080
    },
    {
      "entropy": 5.822088432312012,
      "epoch": 1.5087424938184388,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004775780279437545,
      "loss": 5.3575,
      "mean_token_accuracy": 0.21328300088644028,
      "num_tokens": 31842987.0,
      "step": 17085
    },
    {
      "entropy": 5.8510472774505615,
      "epoch": 1.5091840339102789,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00047756432170219243,
      "loss": 5.3193,
      "mean_token_accuracy": 0.20450857579708098,
      "num_tokens": 31851806.0,
      "step": 17090
    },
    {
      "entropy": 5.8785346984863285,
      "epoch": 1.5096255740021194,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004775506114925106,
      "loss": 5.4773,
      "mean_token_accuracy": 0.19472805857658387,
      "num_tokens": 31861857.0,
      "step": 17095
    },
    {
      "entropy": 5.862979936599731,
      "epoch": 1.5100671140939599,
      "grad_norm": 1.0,
      "learning_rate": 0.00047753689731497734,
      "loss": 5.3781,
      "mean_token_accuracy": 0.1985655054450035,
      "num_tokens": 31871252.0,
      "step": 17100
    },
    {
      "entropy": 5.912623882293701,
      "epoch": 1.5105086541858,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004775231791698616,
      "loss": 5.4435,
      "mean_token_accuracy": 0.18940001875162124,
      "num_tokens": 31880734.0,
      "step": 17105
    },
    {
      "entropy": 5.855919408798218,
      "epoch": 1.5109501942776404,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047750945705743187,
      "loss": 5.3634,
      "mean_token_accuracy": 0.19766008257865905,
      "num_tokens": 31889304.0,
      "step": 17110
    },
    {
      "entropy": 5.807578039169312,
      "epoch": 1.5113917343694807,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004774957309779571,
      "loss": 5.4,
      "mean_token_accuracy": 0.19846741408109664,
      "num_tokens": 31898586.0,
      "step": 17115
    },
    {
      "entropy": 5.90633111000061,
      "epoch": 1.511833274461321,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004774820009317061,
      "loss": 5.4523,
      "mean_token_accuracy": 0.19630279690027236,
      "num_tokens": 31908153.0,
      "step": 17120
    },
    {
      "entropy": 5.8275305271148685,
      "epoch": 1.5122748145531615,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047746826691894787,
      "loss": 5.2505,
      "mean_token_accuracy": 0.21836305111646653,
      "num_tokens": 31916467.0,
      "step": 17125
    },
    {
      "entropy": 5.788586235046386,
      "epoch": 1.5127163546450018,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047745452893995145,
      "loss": 5.3643,
      "mean_token_accuracy": 0.19612461030483247,
      "num_tokens": 31925469.0,
      "step": 17130
    },
    {
      "entropy": 5.793550443649292,
      "epoch": 1.513157894736842,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047744078699498587,
      "loss": 5.4631,
      "mean_token_accuracy": 0.1943300947546959,
      "num_tokens": 31934747.0,
      "step": 17135
    },
    {
      "entropy": 5.818152809143067,
      "epoch": 1.5135994348286825,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047742704108432046,
      "loss": 5.3534,
      "mean_token_accuracy": 0.1993481770157814,
      "num_tokens": 31943773.0,
      "step": 17140
    },
    {
      "entropy": 5.8493987083435055,
      "epoch": 1.5140409749205228,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00047741329120822446,
      "loss": 5.4224,
      "mean_token_accuracy": 0.20505952835083008,
      "num_tokens": 31952699.0,
      "step": 17145
    },
    {
      "entropy": 5.840079307556152,
      "epoch": 1.514482515012363,
      "grad_norm": 1.125,
      "learning_rate": 0.0004773995373669672,
      "loss": 5.3944,
      "mean_token_accuracy": 0.19767861813306808,
      "num_tokens": 31962845.0,
      "step": 17150
    },
    {
      "entropy": 5.8120654106140135,
      "epoch": 1.5149240551042036,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004773857795608181,
      "loss": 5.2894,
      "mean_token_accuracy": 0.21276037842035295,
      "num_tokens": 31972437.0,
      "step": 17155
    },
    {
      "entropy": 5.875831747055054,
      "epoch": 1.5153655951960436,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047737201779004667,
      "loss": 5.4996,
      "mean_token_accuracy": 0.1989943951368332,
      "num_tokens": 31981652.0,
      "step": 17160
    },
    {
      "entropy": 5.855569887161255,
      "epoch": 1.5158071352878841,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004773582520549226,
      "loss": 5.3873,
      "mean_token_accuracy": 0.20434178709983825,
      "num_tokens": 31990088.0,
      "step": 17165
    },
    {
      "entropy": 5.870453596115112,
      "epoch": 1.5162486753797246,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004773444823557153,
      "loss": 5.4056,
      "mean_token_accuracy": 0.20224764496088027,
      "num_tokens": 31999386.0,
      "step": 17170
    },
    {
      "entropy": 5.831242275238037,
      "epoch": 1.5166902154715647,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004773307086926948,
      "loss": 5.3313,
      "mean_token_accuracy": 0.20481727570295333,
      "num_tokens": 32007695.0,
      "step": 17175
    },
    {
      "entropy": 5.8826985359191895,
      "epoch": 1.5171317555634052,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047731693106613075,
      "loss": 5.4619,
      "mean_token_accuracy": 0.2010740280151367,
      "num_tokens": 32018300.0,
      "step": 17180
    },
    {
      "entropy": 5.893084335327148,
      "epoch": 1.5175732956552455,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004773031494762931,
      "loss": 5.448,
      "mean_token_accuracy": 0.1982845038175583,
      "num_tokens": 32028386.0,
      "step": 17185
    },
    {
      "entropy": 5.907029438018799,
      "epoch": 1.5180148357470857,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004772893639234518,
      "loss": 5.4529,
      "mean_token_accuracy": 0.1987977385520935,
      "num_tokens": 32037260.0,
      "step": 17190
    },
    {
      "entropy": 5.8555513381958,
      "epoch": 1.5184563758389262,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004772755744078769,
      "loss": 5.4685,
      "mean_token_accuracy": 0.19956057518720627,
      "num_tokens": 32047654.0,
      "step": 17195
    },
    {
      "entropy": 5.840859317779541,
      "epoch": 1.5188979159307665,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004772617809298385,
      "loss": 5.3707,
      "mean_token_accuracy": 0.20469409376382827,
      "num_tokens": 32056791.0,
      "step": 17200
    },
    {
      "entropy": 5.886715650558472,
      "epoch": 1.5193394560226068,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004772479834896069,
      "loss": 5.3741,
      "mean_token_accuracy": 0.20050286501646042,
      "num_tokens": 32066561.0,
      "step": 17205
    },
    {
      "entropy": 5.923398065567016,
      "epoch": 1.5197809961144473,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047723418208745226,
      "loss": 5.422,
      "mean_token_accuracy": 0.1947730541229248,
      "num_tokens": 32075801.0,
      "step": 17210
    },
    {
      "entropy": 5.81295075416565,
      "epoch": 1.5202225362062876,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004772203767236451,
      "loss": 5.3556,
      "mean_token_accuracy": 0.20861087292432784,
      "num_tokens": 32085274.0,
      "step": 17215
    },
    {
      "entropy": 5.9020044803619385,
      "epoch": 1.5206640762981278,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047720656739845573,
      "loss": 5.3902,
      "mean_token_accuracy": 0.2032547190785408,
      "num_tokens": 32094481.0,
      "step": 17220
    },
    {
      "entropy": 5.80965485572815,
      "epoch": 1.5211056163899683,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004771927541121547,
      "loss": 5.3669,
      "mean_token_accuracy": 0.2061656892299652,
      "num_tokens": 32103094.0,
      "step": 17225
    },
    {
      "entropy": 5.789358472824096,
      "epoch": 1.5215471564818084,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004771789368650126,
      "loss": 5.356,
      "mean_token_accuracy": 0.20408538430929185,
      "num_tokens": 32112230.0,
      "step": 17230
    },
    {
      "entropy": 5.845158672332763,
      "epoch": 1.5219886965736489,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047716511565730013,
      "loss": 5.3794,
      "mean_token_accuracy": 0.2095283031463623,
      "num_tokens": 32121472.0,
      "step": 17235
    },
    {
      "entropy": 5.850609827041626,
      "epoch": 1.5224302366654894,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000477151290489288,
      "loss": 5.4185,
      "mean_token_accuracy": 0.200775046646595,
      "num_tokens": 32130849.0,
      "step": 17240
    },
    {
      "entropy": 5.839935541152954,
      "epoch": 1.5228717767573294,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004771374613612471,
      "loss": 5.3251,
      "mean_token_accuracy": 0.20684054791927337,
      "num_tokens": 32140874.0,
      "step": 17245
    },
    {
      "entropy": 5.850006771087647,
      "epoch": 1.52331331684917,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004771236282734482,
      "loss": 5.3445,
      "mean_token_accuracy": 0.20047370046377183,
      "num_tokens": 32150774.0,
      "step": 17250
    },
    {
      "entropy": 5.805947971343994,
      "epoch": 1.5237548569410102,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047710979122616247,
      "loss": 5.3327,
      "mean_token_accuracy": 0.20334020704030992,
      "num_tokens": 32159825.0,
      "step": 17255
    },
    {
      "entropy": 5.812037086486816,
      "epoch": 1.5241963970328505,
      "grad_norm": 1.125,
      "learning_rate": 0.0004770959502196609,
      "loss": 5.3886,
      "mean_token_accuracy": 0.19752940088510512,
      "num_tokens": 32168443.0,
      "step": 17260
    },
    {
      "entropy": 5.835806941986084,
      "epoch": 1.524637937124691,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047708210525421445,
      "loss": 5.3436,
      "mean_token_accuracy": 0.20664990395307542,
      "num_tokens": 32178065.0,
      "step": 17265
    },
    {
      "entropy": 5.792875337600708,
      "epoch": 1.5250794772165313,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047706825633009466,
      "loss": 5.2512,
      "mean_token_accuracy": 0.211979740858078,
      "num_tokens": 32187274.0,
      "step": 17270
    },
    {
      "entropy": 5.822767353057861,
      "epoch": 1.5255210173083715,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047705440344757264,
      "loss": 5.2235,
      "mean_token_accuracy": 0.2181275174021721,
      "num_tokens": 32195811.0,
      "step": 17275
    },
    {
      "entropy": 5.754204750061035,
      "epoch": 1.525962557400212,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047704054660691964,
      "loss": 5.3172,
      "mean_token_accuracy": 0.20618013441562652,
      "num_tokens": 32204958.0,
      "step": 17280
    },
    {
      "entropy": 5.796719312667847,
      "epoch": 1.5264040974920523,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004770266858084074,
      "loss": 5.351,
      "mean_token_accuracy": 0.20292788743972778,
      "num_tokens": 32214299.0,
      "step": 17285
    },
    {
      "entropy": 5.878060579299927,
      "epoch": 1.5268456375838926,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047701282105230716,
      "loss": 5.4198,
      "mean_token_accuracy": 0.2011943593621254,
      "num_tokens": 32223353.0,
      "step": 17290
    },
    {
      "entropy": 5.852122783660889,
      "epoch": 1.527287177675733,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004769989523388906,
      "loss": 5.4203,
      "mean_token_accuracy": 0.18991311341524125,
      "num_tokens": 32232641.0,
      "step": 17295
    },
    {
      "entropy": 5.8938398361206055,
      "epoch": 1.5277287177675731,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004769850796684296,
      "loss": 5.4508,
      "mean_token_accuracy": 0.199055977165699,
      "num_tokens": 32243028.0,
      "step": 17300
    },
    {
      "entropy": 5.875230169296264,
      "epoch": 1.5281702578594136,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004769712030411957,
      "loss": 5.3962,
      "mean_token_accuracy": 0.20871222019195557,
      "num_tokens": 32253660.0,
      "step": 17305
    },
    {
      "entropy": 5.822922468185425,
      "epoch": 1.5286117979512541,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047695732245746076,
      "loss": 5.3859,
      "mean_token_accuracy": 0.2028472825884819,
      "num_tokens": 32263267.0,
      "step": 17310
    },
    {
      "entropy": 5.848821353912354,
      "epoch": 1.5290533380430942,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004769434379174967,
      "loss": 5.3469,
      "mean_token_accuracy": 0.20837787985801698,
      "num_tokens": 32273929.0,
      "step": 17315
    },
    {
      "entropy": 5.809392261505127,
      "epoch": 1.5294948781349347,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004769295494215757,
      "loss": 5.366,
      "mean_token_accuracy": 0.20906019657850267,
      "num_tokens": 32282717.0,
      "step": 17320
    },
    {
      "entropy": 5.8649170875549315,
      "epoch": 1.529936418226775,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047691565696996954,
      "loss": 5.3914,
      "mean_token_accuracy": 0.20684265494346618,
      "num_tokens": 32293225.0,
      "step": 17325
    },
    {
      "entropy": 5.890305137634277,
      "epoch": 1.5303779583186152,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004769017605629505,
      "loss": 5.3779,
      "mean_token_accuracy": 0.20241788625717164,
      "num_tokens": 32301582.0,
      "step": 17330
    },
    {
      "entropy": 5.8151891231536865,
      "epoch": 1.5308194984104557,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004768878602007909,
      "loss": 5.3187,
      "mean_token_accuracy": 0.20678892135620117,
      "num_tokens": 32310131.0,
      "step": 17335
    },
    {
      "entropy": 5.791430807113647,
      "epoch": 1.531261038502296,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004768739558837628,
      "loss": 5.3816,
      "mean_token_accuracy": 0.20074218064546584,
      "num_tokens": 32320038.0,
      "step": 17340
    },
    {
      "entropy": 5.834036922454834,
      "epoch": 1.5317025785941363,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004768600476121388,
      "loss": 5.4501,
      "mean_token_accuracy": 0.19252036958932878,
      "num_tokens": 32329568.0,
      "step": 17345
    },
    {
      "entropy": 5.9029418468475345,
      "epoch": 1.5321441186859768,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047684613538619125,
      "loss": 5.3819,
      "mean_token_accuracy": 0.2056812733411789,
      "num_tokens": 32338940.0,
      "step": 17350
    },
    {
      "entropy": 5.857225322723389,
      "epoch": 1.532585658777817,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004768322192061927,
      "loss": 5.3345,
      "mean_token_accuracy": 0.20741656720638274,
      "num_tokens": 32347757.0,
      "step": 17355
    },
    {
      "entropy": 5.781398344039917,
      "epoch": 1.5330271988696573,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004768182990724158,
      "loss": 5.2749,
      "mean_token_accuracy": 0.2075476586818695,
      "num_tokens": 32357099.0,
      "step": 17360
    },
    {
      "entropy": 5.823176717758178,
      "epoch": 1.5334687389614978,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047680437498513317,
      "loss": 5.3983,
      "mean_token_accuracy": 0.1995120480656624,
      "num_tokens": 32366548.0,
      "step": 17365
    },
    {
      "entropy": 5.9196630954742435,
      "epoch": 1.5339102790533379,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047679044694461765,
      "loss": 5.5132,
      "mean_token_accuracy": 0.19590681344270705,
      "num_tokens": 32377435.0,
      "step": 17370
    },
    {
      "entropy": 5.92530927658081,
      "epoch": 1.5343518191451784,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000476776514951142,
      "loss": 5.406,
      "mean_token_accuracy": 0.19295999258756638,
      "num_tokens": 32386922.0,
      "step": 17375
    },
    {
      "entropy": 5.840206813812256,
      "epoch": 1.5347933592370189,
      "grad_norm": 1.0,
      "learning_rate": 0.00047676257900497915,
      "loss": 5.4434,
      "mean_token_accuracy": 0.19346087723970412,
      "num_tokens": 32396350.0,
      "step": 17380
    },
    {
      "entropy": 5.754315328598023,
      "epoch": 1.535234899328859,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047674863910640217,
      "loss": 5.2431,
      "mean_token_accuracy": 0.2165575474500656,
      "num_tokens": 32405678.0,
      "step": 17385
    },
    {
      "entropy": 5.845928001403808,
      "epoch": 1.5356764394206994,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047673469525568406,
      "loss": 5.3411,
      "mean_token_accuracy": 0.20299955904483796,
      "num_tokens": 32415793.0,
      "step": 17390
    },
    {
      "entropy": 5.806760215759278,
      "epoch": 1.5361179795125397,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047672074745309804,
      "loss": 5.2636,
      "mean_token_accuracy": 0.2122264862060547,
      "num_tokens": 32424735.0,
      "step": 17395
    },
    {
      "entropy": 5.790989017486572,
      "epoch": 1.53655951960438,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004767067956989172,
      "loss": 5.405,
      "mean_token_accuracy": 0.20501989275217056,
      "num_tokens": 32435213.0,
      "step": 17400
    },
    {
      "entropy": 5.788079929351807,
      "epoch": 1.5370010596962205,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004766928399934151,
      "loss": 5.2675,
      "mean_token_accuracy": 0.20899792164564132,
      "num_tokens": 32443932.0,
      "step": 17405
    },
    {
      "entropy": 5.792341327667236,
      "epoch": 1.5374425997880607,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004766788803368649,
      "loss": 5.2905,
      "mean_token_accuracy": 0.20316837579011918,
      "num_tokens": 32453851.0,
      "step": 17410
    },
    {
      "entropy": 5.819828987121582,
      "epoch": 1.537884139879901,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004766649167295401,
      "loss": 5.3785,
      "mean_token_accuracy": 0.20670698285102845,
      "num_tokens": 32462906.0,
      "step": 17415
    },
    {
      "entropy": 5.788566398620605,
      "epoch": 1.5383256799717415,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004766509491717144,
      "loss": 5.3314,
      "mean_token_accuracy": 0.2036026418209076,
      "num_tokens": 32471840.0,
      "step": 17420
    },
    {
      "entropy": 5.799122381210327,
      "epoch": 1.5387672200635818,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047663697766366114,
      "loss": 5.2938,
      "mean_token_accuracy": 0.20713522881269456,
      "num_tokens": 32481111.0,
      "step": 17425
    },
    {
      "entropy": 5.855897712707519,
      "epoch": 1.539208760155422,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004766230022056542,
      "loss": 5.4008,
      "mean_token_accuracy": 0.19948508590459824,
      "num_tokens": 32489827.0,
      "step": 17430
    },
    {
      "entropy": 5.82600245475769,
      "epoch": 1.5396503002472626,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047660902279796736,
      "loss": 5.3174,
      "mean_token_accuracy": 0.21005891263484955,
      "num_tokens": 32499106.0,
      "step": 17435
    },
    {
      "entropy": 5.971836709976197,
      "epoch": 1.5400918403391026,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004765950394408744,
      "loss": 5.5817,
      "mean_token_accuracy": 0.18005635142326354,
      "num_tokens": 32507858.0,
      "step": 17440
    },
    {
      "entropy": 5.840009021759033,
      "epoch": 1.5405333804309431,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004765810521346493,
      "loss": 5.3555,
      "mean_token_accuracy": 0.21056333929300308,
      "num_tokens": 32516761.0,
      "step": 17445
    },
    {
      "entropy": 5.819193315505982,
      "epoch": 1.5409749205227836,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047656706087956604,
      "loss": 5.3778,
      "mean_token_accuracy": 0.20764204114675522,
      "num_tokens": 32526585.0,
      "step": 17450
    },
    {
      "entropy": 5.870976734161377,
      "epoch": 1.5414164606146237,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047655306567589864,
      "loss": 5.3791,
      "mean_token_accuracy": 0.20405740588903426,
      "num_tokens": 32535607.0,
      "step": 17455
    },
    {
      "entropy": 5.774896669387817,
      "epoch": 1.5418580007064642,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047653906652392135,
      "loss": 5.2853,
      "mean_token_accuracy": 0.21170148849487305,
      "num_tokens": 32545401.0,
      "step": 17460
    },
    {
      "entropy": 5.836571121215821,
      "epoch": 1.5422995407983044,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004765250634239083,
      "loss": 5.3645,
      "mean_token_accuracy": 0.20433194935321808,
      "num_tokens": 32554421.0,
      "step": 17465
    },
    {
      "entropy": 5.869767808914185,
      "epoch": 1.5427410808901447,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047651105637613395,
      "loss": 5.4561,
      "mean_token_accuracy": 0.19872623682022095,
      "num_tokens": 32563652.0,
      "step": 17470
    },
    {
      "entropy": 5.901002073287964,
      "epoch": 1.5431826209819852,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004764970453808726,
      "loss": 5.4422,
      "mean_token_accuracy": 0.20603567808866502,
      "num_tokens": 32572812.0,
      "step": 17475
    },
    {
      "entropy": 5.7957408905029295,
      "epoch": 1.5436241610738255,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047648303043839875,
      "loss": 5.3615,
      "mean_token_accuracy": 0.20554557889699937,
      "num_tokens": 32581746.0,
      "step": 17480
    },
    {
      "entropy": 5.906637954711914,
      "epoch": 1.5440657011656658,
      "grad_norm": 1.0,
      "learning_rate": 0.0004764690115489869,
      "loss": 5.3766,
      "mean_token_accuracy": 0.2001782089471817,
      "num_tokens": 32591596.0,
      "step": 17485
    },
    {
      "entropy": 5.860601902008057,
      "epoch": 1.5445072412575063,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047645498871291163,
      "loss": 5.3757,
      "mean_token_accuracy": 0.203918893635273,
      "num_tokens": 32600610.0,
      "step": 17490
    },
    {
      "entropy": 5.841221666336059,
      "epoch": 1.5449487813493465,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004764409619304478,
      "loss": 5.3928,
      "mean_token_accuracy": 0.20123014003038406,
      "num_tokens": 32609827.0,
      "step": 17495
    },
    {
      "entropy": 5.781210422515869,
      "epoch": 1.5453903214411868,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047642693120187007,
      "loss": 5.4205,
      "mean_token_accuracy": 0.2035561054944992,
      "num_tokens": 32618258.0,
      "step": 17500
    },
    {
      "entropy": 5.886530780792237,
      "epoch": 1.5458318615330273,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004764128965274533,
      "loss": 5.3906,
      "mean_token_accuracy": 0.19994398057460785,
      "num_tokens": 32628155.0,
      "step": 17505
    },
    {
      "entropy": 5.939802551269532,
      "epoch": 1.5462734016248674,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047639885790747244,
      "loss": 5.4503,
      "mean_token_accuracy": 0.2002129688858986,
      "num_tokens": 32638347.0,
      "step": 17510
    },
    {
      "entropy": 5.838560485839844,
      "epoch": 1.5467149417167079,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047638481534220244,
      "loss": 5.3168,
      "mean_token_accuracy": 0.20974133610725404,
      "num_tokens": 32648636.0,
      "step": 17515
    },
    {
      "entropy": 5.791447496414184,
      "epoch": 1.5471564818085484,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004763707688319185,
      "loss": 5.3922,
      "mean_token_accuracy": 0.20038572549819947,
      "num_tokens": 32657762.0,
      "step": 17520
    },
    {
      "entropy": 5.830028676986695,
      "epoch": 1.5475980219003884,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004763567183768958,
      "loss": 5.3519,
      "mean_token_accuracy": 0.2025613531470299,
      "num_tokens": 32668265.0,
      "step": 17525
    },
    {
      "entropy": 5.824889755249023,
      "epoch": 1.548039561992229,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047634266397740934,
      "loss": 5.2836,
      "mean_token_accuracy": 0.20865013748407363,
      "num_tokens": 32678051.0,
      "step": 17530
    },
    {
      "entropy": 5.940329456329346,
      "epoch": 1.5484811020840692,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047632860563373466,
      "loss": 5.4788,
      "mean_token_accuracy": 0.20264360904693604,
      "num_tokens": 32687186.0,
      "step": 17535
    },
    {
      "entropy": 5.863598728179932,
      "epoch": 1.5489226421759095,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004763145433461472,
      "loss": 5.3882,
      "mean_token_accuracy": 0.21008908301591872,
      "num_tokens": 32697082.0,
      "step": 17540
    },
    {
      "entropy": 5.857601547241211,
      "epoch": 1.54936418226775,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004763004771149222,
      "loss": 5.3782,
      "mean_token_accuracy": 0.1941447600722313,
      "num_tokens": 32706502.0,
      "step": 17545
    },
    {
      "entropy": 5.860434532165527,
      "epoch": 1.5498057223595902,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004762864069403354,
      "loss": 5.3662,
      "mean_token_accuracy": 0.20193805396556855,
      "num_tokens": 32715035.0,
      "step": 17550
    },
    {
      "entropy": 5.836504745483398,
      "epoch": 1.5502472624514305,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047627233282266236,
      "loss": 5.3839,
      "mean_token_accuracy": 0.20119669586420058,
      "num_tokens": 32724713.0,
      "step": 17555
    },
    {
      "entropy": 5.850826597213745,
      "epoch": 1.550688802543271,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004762582547621788,
      "loss": 5.3756,
      "mean_token_accuracy": 0.20523784756660463,
      "num_tokens": 32734584.0,
      "step": 17560
    },
    {
      "entropy": 5.87843599319458,
      "epoch": 1.5511303426351113,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047624417275916045,
      "loss": 5.3417,
      "mean_token_accuracy": 0.2070808857679367,
      "num_tokens": 32742744.0,
      "step": 17565
    },
    {
      "entropy": 5.776598167419434,
      "epoch": 1.5515718827269516,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047623008681388324,
      "loss": 5.2969,
      "mean_token_accuracy": 0.21753205358982086,
      "num_tokens": 32752051.0,
      "step": 17570
    },
    {
      "entropy": 5.817727661132812,
      "epoch": 1.552013422818792,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047621599692662306,
      "loss": 5.3734,
      "mean_token_accuracy": 0.2033172905445099,
      "num_tokens": 32761740.0,
      "step": 17575
    },
    {
      "entropy": 5.770384740829468,
      "epoch": 1.5524549629106321,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047620190309765595,
      "loss": 5.3229,
      "mean_token_accuracy": 0.2133523106575012,
      "num_tokens": 32770809.0,
      "step": 17580
    },
    {
      "entropy": 5.879725694656372,
      "epoch": 1.5528965030024726,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047618780532725805,
      "loss": 5.3917,
      "mean_token_accuracy": 0.20065007656812667,
      "num_tokens": 32780614.0,
      "step": 17585
    },
    {
      "entropy": 5.905575323104858,
      "epoch": 1.5533380430943131,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047617370361570534,
      "loss": 5.3893,
      "mean_token_accuracy": 0.19946929812431335,
      "num_tokens": 32790336.0,
      "step": 17590
    },
    {
      "entropy": 5.77630763053894,
      "epoch": 1.5537795831861532,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004761595979632743,
      "loss": 5.3561,
      "mean_token_accuracy": 0.20518322438001632,
      "num_tokens": 32798991.0,
      "step": 17595
    },
    {
      "entropy": 5.906645965576172,
      "epoch": 1.5542211232779937,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004761454883702411,
      "loss": 5.4053,
      "mean_token_accuracy": 0.20123518705368043,
      "num_tokens": 32807616.0,
      "step": 17600
    },
    {
      "entropy": 5.8880298137664795,
      "epoch": 1.554662663369834,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004761313748368821,
      "loss": 5.4445,
      "mean_token_accuracy": 0.19556560218334199,
      "num_tokens": 32817107.0,
      "step": 17605
    },
    {
      "entropy": 5.789643144607544,
      "epoch": 1.5551042034616742,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004761172573634739,
      "loss": 5.3752,
      "mean_token_accuracy": 0.20864083021879196,
      "num_tokens": 32827205.0,
      "step": 17610
    },
    {
      "entropy": 5.868758249282837,
      "epoch": 1.5555457435535147,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004761031359502931,
      "loss": 5.3161,
      "mean_token_accuracy": 0.21065206825733185,
      "num_tokens": 32836102.0,
      "step": 17615
    },
    {
      "entropy": 5.892943382263184,
      "epoch": 1.555987283645355,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047608901059761617,
      "loss": 5.4211,
      "mean_token_accuracy": 0.19518836587667465,
      "num_tokens": 32845389.0,
      "step": 17620
    },
    {
      "entropy": 5.777494430541992,
      "epoch": 1.5564288237371953,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047607488130571997,
      "loss": 5.3258,
      "mean_token_accuracy": 0.20757322162389755,
      "num_tokens": 32854186.0,
      "step": 17625
    },
    {
      "entropy": 5.78282356262207,
      "epoch": 1.5568703638290358,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004760607480748811,
      "loss": 5.3271,
      "mean_token_accuracy": 0.20399378687143327,
      "num_tokens": 32863498.0,
      "step": 17630
    },
    {
      "entropy": 5.853580284118652,
      "epoch": 1.557311903920876,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047604661090537657,
      "loss": 5.4329,
      "mean_token_accuracy": 0.19530525505542756,
      "num_tokens": 32872483.0,
      "step": 17635
    },
    {
      "entropy": 5.897101497650146,
      "epoch": 1.5577534440127163,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004760324697974832,
      "loss": 5.4635,
      "mean_token_accuracy": 0.19322464615106583,
      "num_tokens": 32882739.0,
      "step": 17640
    },
    {
      "entropy": 5.912864351272583,
      "epoch": 1.5581949841045568,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004760183247514782,
      "loss": 5.513,
      "mean_token_accuracy": 0.193892902135849,
      "num_tokens": 32892707.0,
      "step": 17645
    },
    {
      "entropy": 5.890587854385376,
      "epoch": 1.5586365241963969,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00047600417576763843,
      "loss": 5.4084,
      "mean_token_accuracy": 0.20191147923469543,
      "num_tokens": 32901916.0,
      "step": 17650
    },
    {
      "entropy": 5.823947334289551,
      "epoch": 1.5590780642882374,
      "grad_norm": 1.0,
      "learning_rate": 0.0004759900228462412,
      "loss": 5.3467,
      "mean_token_accuracy": 0.2064092352986336,
      "num_tokens": 32910656.0,
      "step": 17655
    },
    {
      "entropy": 5.923448896408081,
      "epoch": 1.5595196043800779,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004759758659875638,
      "loss": 5.4651,
      "mean_token_accuracy": 0.20218765288591384,
      "num_tokens": 32919579.0,
      "step": 17660
    },
    {
      "entropy": 5.895467233657837,
      "epoch": 1.559961144471918,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047596170519188335,
      "loss": 5.4146,
      "mean_token_accuracy": 0.19703866839408873,
      "num_tokens": 32928271.0,
      "step": 17665
    },
    {
      "entropy": 5.855054187774658,
      "epoch": 1.5604026845637584,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004759475404594776,
      "loss": 5.3159,
      "mean_token_accuracy": 0.2074134573340416,
      "num_tokens": 32937671.0,
      "step": 17670
    },
    {
      "entropy": 5.869637680053711,
      "epoch": 1.5608442246555987,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004759333717906236,
      "loss": 5.4886,
      "mean_token_accuracy": 0.19632164835929872,
      "num_tokens": 32946896.0,
      "step": 17675
    },
    {
      "entropy": 5.911066055297852,
      "epoch": 1.561285764747439,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047591919918559927,
      "loss": 5.4229,
      "mean_token_accuracy": 0.2013778120279312,
      "num_tokens": 32956706.0,
      "step": 17680
    },
    {
      "entropy": 5.8082053661346436,
      "epoch": 1.5617273048392795,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004759050226446821,
      "loss": 5.2845,
      "mean_token_accuracy": 0.2147594213485718,
      "num_tokens": 32965990.0,
      "step": 17685
    },
    {
      "entropy": 5.7986860275268555,
      "epoch": 1.5621688449311197,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004758908421681497,
      "loss": 5.2748,
      "mean_token_accuracy": 0.21321889013051987,
      "num_tokens": 32975311.0,
      "step": 17690
    },
    {
      "entropy": 5.865177249908447,
      "epoch": 1.56261038502296,
      "grad_norm": 1.0,
      "learning_rate": 0.00047587665775627997,
      "loss": 5.374,
      "mean_token_accuracy": 0.2026787981390953,
      "num_tokens": 32984641.0,
      "step": 17695
    },
    {
      "entropy": 5.866060256958008,
      "epoch": 1.5630519251148005,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00047586246940935087,
      "loss": 5.4386,
      "mean_token_accuracy": 0.1965424045920372,
      "num_tokens": 32994696.0,
      "step": 17700
    },
    {
      "entropy": 5.878091669082641,
      "epoch": 1.5634934652066408,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047584827712764013,
      "loss": 5.33,
      "mean_token_accuracy": 0.20413600504398347,
      "num_tokens": 33003392.0,
      "step": 17705
    },
    {
      "entropy": 5.839124488830566,
      "epoch": 1.563935005298481,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004758340809114259,
      "loss": 5.3866,
      "mean_token_accuracy": 0.19434957653284074,
      "num_tokens": 33012896.0,
      "step": 17710
    },
    {
      "entropy": 5.8867676734924315,
      "epoch": 1.5643765453903216,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004758198807609862,
      "loss": 5.4388,
      "mean_token_accuracy": 0.19910742938518525,
      "num_tokens": 33022710.0,
      "step": 17715
    },
    {
      "entropy": 5.917848587036133,
      "epoch": 1.5648180854821618,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004758056766765993,
      "loss": 5.364,
      "mean_token_accuracy": 0.20309918969869614,
      "num_tokens": 33031982.0,
      "step": 17720
    },
    {
      "entropy": 5.893701696395874,
      "epoch": 1.5652596255740021,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047579146865854333,
      "loss": 5.4731,
      "mean_token_accuracy": 0.19704148918390274,
      "num_tokens": 33041496.0,
      "step": 17725
    },
    {
      "entropy": 5.842383813858032,
      "epoch": 1.5657011656658426,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047577725670709675,
      "loss": 5.3805,
      "mean_token_accuracy": 0.20649636387825013,
      "num_tokens": 33050764.0,
      "step": 17730
    },
    {
      "entropy": 5.847108793258667,
      "epoch": 1.5661427057576827,
      "grad_norm": 0.96875,
      "learning_rate": 0.000475763040822538,
      "loss": 5.4108,
      "mean_token_accuracy": 0.20037575215101242,
      "num_tokens": 33060809.0,
      "step": 17735
    },
    {
      "entropy": 5.770617151260376,
      "epoch": 1.5665842458495232,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004757488210051453,
      "loss": 5.2755,
      "mean_token_accuracy": 0.20983420461416244,
      "num_tokens": 33070155.0,
      "step": 17740
    },
    {
      "entropy": 5.8552123546600345,
      "epoch": 1.5670257859413634,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004757345972551974,
      "loss": 5.4023,
      "mean_token_accuracy": 0.20433754473924637,
      "num_tokens": 33079354.0,
      "step": 17745
    },
    {
      "entropy": 5.899097728729248,
      "epoch": 1.5674673260332037,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004757203695729729,
      "loss": 5.331,
      "mean_token_accuracy": 0.2078401491045952,
      "num_tokens": 33087877.0,
      "step": 17750
    },
    {
      "entropy": 5.837959575653076,
      "epoch": 1.5679088661250442,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047570613795875055,
      "loss": 5.4258,
      "mean_token_accuracy": 0.19665904343128204,
      "num_tokens": 33097375.0,
      "step": 17755
    },
    {
      "entropy": 5.874375104904175,
      "epoch": 1.5683504062168845,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004756919024128091,
      "loss": 5.3893,
      "mean_token_accuracy": 0.20355249643325807,
      "num_tokens": 33106336.0,
      "step": 17760
    },
    {
      "entropy": 5.779546785354614,
      "epoch": 1.5687919463087248,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004756776629354274,
      "loss": 5.2528,
      "mean_token_accuracy": 0.21987343728542327,
      "num_tokens": 33115753.0,
      "step": 17765
    },
    {
      "entropy": 5.876402950286865,
      "epoch": 1.5692334864005653,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004756634195268844,
      "loss": 5.4202,
      "mean_token_accuracy": 0.19458318799734114,
      "num_tokens": 33125205.0,
      "step": 17770
    },
    {
      "entropy": 5.792306280136108,
      "epoch": 1.5696750264924055,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004756491721874592,
      "loss": 5.3319,
      "mean_token_accuracy": 0.21207024306058883,
      "num_tokens": 33134470.0,
      "step": 17775
    },
    {
      "entropy": 5.799787712097168,
      "epoch": 1.5701165665842458,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047563492091743075,
      "loss": 5.3122,
      "mean_token_accuracy": 0.215229594707489,
      "num_tokens": 33143346.0,
      "step": 17780
    },
    {
      "entropy": 5.792463970184326,
      "epoch": 1.5705581066760863,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047562066571707836,
      "loss": 5.3202,
      "mean_token_accuracy": 0.20465569496154784,
      "num_tokens": 33153291.0,
      "step": 17785
    },
    {
      "entropy": 5.948005247116089,
      "epoch": 1.5709996467679266,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047560640658668126,
      "loss": 5.5063,
      "mean_token_accuracy": 0.19285668730735778,
      "num_tokens": 33163642.0,
      "step": 17790
    },
    {
      "entropy": 5.909404325485229,
      "epoch": 1.5714411868597669,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047559214352651863,
      "loss": 5.3406,
      "mean_token_accuracy": 0.20744375586509706,
      "num_tokens": 33173529.0,
      "step": 17795
    },
    {
      "entropy": 5.874653053283692,
      "epoch": 1.5718827269516074,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047557787653687014,
      "loss": 5.3306,
      "mean_token_accuracy": 0.20597842633724212,
      "num_tokens": 33180877.0,
      "step": 17800
    },
    {
      "entropy": 5.81123719215393,
      "epoch": 1.5723242670434474,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000475563605618015,
      "loss": 5.4059,
      "mean_token_accuracy": 0.20450567007064818,
      "num_tokens": 33190831.0,
      "step": 17805
    },
    {
      "entropy": 5.8893884181976315,
      "epoch": 1.572765807135288,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000475549330770233,
      "loss": 5.4117,
      "mean_token_accuracy": 0.19570378214120865,
      "num_tokens": 33200313.0,
      "step": 17810
    },
    {
      "entropy": 5.896144580841065,
      "epoch": 1.5732073472271282,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004755350519938036,
      "loss": 5.3941,
      "mean_token_accuracy": 0.20024604946374894,
      "num_tokens": 33209272.0,
      "step": 17815
    },
    {
      "entropy": 5.822745084762573,
      "epoch": 1.5736488873189685,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047552076928900663,
      "loss": 5.3276,
      "mean_token_accuracy": 0.21161558032035827,
      "num_tokens": 33218712.0,
      "step": 17820
    },
    {
      "entropy": 5.8529187679290775,
      "epoch": 1.574090427410809,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004755064826561217,
      "loss": 5.4518,
      "mean_token_accuracy": 0.20207060724496842,
      "num_tokens": 33227480.0,
      "step": 17825
    },
    {
      "entropy": 5.821786403656006,
      "epoch": 1.5745319675026492,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004754921920954289,
      "loss": 5.3239,
      "mean_token_accuracy": 0.20574539750814438,
      "num_tokens": 33236044.0,
      "step": 17830
    },
    {
      "entropy": 5.816183853149414,
      "epoch": 1.5749735075944895,
      "grad_norm": 1.125,
      "learning_rate": 0.0004754778976072081,
      "loss": 5.3981,
      "mean_token_accuracy": 0.20449062287807465,
      "num_tokens": 33244770.0,
      "step": 17835
    },
    {
      "entropy": 5.8485435962677,
      "epoch": 1.57541504768633,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047546359919173933,
      "loss": 5.305,
      "mean_token_accuracy": 0.20976659506559373,
      "num_tokens": 33254178.0,
      "step": 17840
    },
    {
      "entropy": 5.775556182861328,
      "epoch": 1.5758565877781703,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004754492968493026,
      "loss": 5.3698,
      "mean_token_accuracy": 0.2095904141664505,
      "num_tokens": 33264210.0,
      "step": 17845
    },
    {
      "entropy": 5.73330717086792,
      "epoch": 1.5762981278700106,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004754349905801782,
      "loss": 5.2925,
      "mean_token_accuracy": 0.21290307343006135,
      "num_tokens": 33273348.0,
      "step": 17850
    },
    {
      "entropy": 5.810392284393311,
      "epoch": 1.576739667961851,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047542068038464625,
      "loss": 5.3244,
      "mean_token_accuracy": 0.2157539740204811,
      "num_tokens": 33281906.0,
      "step": 17855
    },
    {
      "entropy": 5.941177701950073,
      "epoch": 1.5771812080536913,
      "grad_norm": 1.125,
      "learning_rate": 0.0004754063662629872,
      "loss": 5.4449,
      "mean_token_accuracy": 0.19736992567777634,
      "num_tokens": 33291363.0,
      "step": 17860
    },
    {
      "entropy": 5.865746259689331,
      "epoch": 1.5776227481455316,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047539204821548146,
      "loss": 5.3827,
      "mean_token_accuracy": 0.20610157400369644,
      "num_tokens": 33299785.0,
      "step": 17865
    },
    {
      "entropy": 5.907922983169556,
      "epoch": 1.578064288237372,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004753777262424094,
      "loss": 5.4795,
      "mean_token_accuracy": 0.19846486300230026,
      "num_tokens": 33309655.0,
      "step": 17870
    },
    {
      "entropy": 5.830729198455811,
      "epoch": 1.5785058283292122,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00047536340034405166,
      "loss": 5.4502,
      "mean_token_accuracy": 0.2007785901427269,
      "num_tokens": 33320158.0,
      "step": 17875
    },
    {
      "entropy": 5.783361387252808,
      "epoch": 1.5789473684210527,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047534907052068886,
      "loss": 5.3196,
      "mean_token_accuracy": 0.2107610061764717,
      "num_tokens": 33328959.0,
      "step": 17880
    },
    {
      "entropy": 5.907349157333374,
      "epoch": 1.579388908512893,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004753347367726017,
      "loss": 5.4308,
      "mean_token_accuracy": 0.2007957726716995,
      "num_tokens": 33338624.0,
      "step": 17885
    },
    {
      "entropy": 5.863383054733276,
      "epoch": 1.5798304486047332,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00047532039910007097,
      "loss": 5.3984,
      "mean_token_accuracy": 0.20311308205127715,
      "num_tokens": 33348083.0,
      "step": 17890
    },
    {
      "entropy": 5.839151430130005,
      "epoch": 1.5802719886965737,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047530605750337765,
      "loss": 5.328,
      "mean_token_accuracy": 0.2022930294275284,
      "num_tokens": 33358226.0,
      "step": 17895
    },
    {
      "entropy": 5.888965082168579,
      "epoch": 1.580713528788414,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047529171198280244,
      "loss": 5.4721,
      "mean_token_accuracy": 0.19955771863460542,
      "num_tokens": 33367997.0,
      "step": 17900
    },
    {
      "entropy": 5.886206912994385,
      "epoch": 1.5811550688802543,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047527736253862664,
      "loss": 5.4247,
      "mean_token_accuracy": 0.19584109187126159,
      "num_tokens": 33377574.0,
      "step": 17905
    },
    {
      "entropy": 5.845288419723511,
      "epoch": 1.5815966089720948,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047526300917113116,
      "loss": 5.4399,
      "mean_token_accuracy": 0.20210800617933272,
      "num_tokens": 33387016.0,
      "step": 17910
    },
    {
      "entropy": 5.871938276290893,
      "epoch": 1.582038149063935,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004752486518805972,
      "loss": 5.3612,
      "mean_token_accuracy": 0.20363385379314422,
      "num_tokens": 33396239.0,
      "step": 17915
    },
    {
      "entropy": 5.851468086242676,
      "epoch": 1.5824796891557753,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047523429066730603,
      "loss": 5.2824,
      "mean_token_accuracy": 0.21232890486717224,
      "num_tokens": 33404738.0,
      "step": 17920
    },
    {
      "entropy": 5.79391655921936,
      "epoch": 1.5829212292476158,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047521992553153903,
      "loss": 5.2687,
      "mean_token_accuracy": 0.21605887711048127,
      "num_tokens": 33413670.0,
      "step": 17925
    },
    {
      "entropy": 5.802855110168457,
      "epoch": 1.583362769339456,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004752055564735775,
      "loss": 5.3973,
      "mean_token_accuracy": 0.19847149699926375,
      "num_tokens": 33422841.0,
      "step": 17930
    },
    {
      "entropy": 5.92512845993042,
      "epoch": 1.5838043094312964,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047519118349370306,
      "loss": 5.4156,
      "mean_token_accuracy": 0.2049744799733162,
      "num_tokens": 33432023.0,
      "step": 17935
    },
    {
      "entropy": 5.865643548965454,
      "epoch": 1.5842458495231369,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004751768065921971,
      "loss": 5.3967,
      "mean_token_accuracy": 0.19575154781341553,
      "num_tokens": 33442181.0,
      "step": 17940
    },
    {
      "entropy": 5.928330183029175,
      "epoch": 1.584687389614977,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047516242576934144,
      "loss": 5.454,
      "mean_token_accuracy": 0.19018019288778304,
      "num_tokens": 33451847.0,
      "step": 17945
    },
    {
      "entropy": 5.887420177459717,
      "epoch": 1.5851289297068174,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047514804102541765,
      "loss": 5.4281,
      "mean_token_accuracy": 0.20378214567899705,
      "num_tokens": 33461587.0,
      "step": 17950
    },
    {
      "entropy": 5.919276428222656,
      "epoch": 1.5855704697986577,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047513365236070753,
      "loss": 5.4536,
      "mean_token_accuracy": 0.19712756872177123,
      "num_tokens": 33471553.0,
      "step": 17955
    },
    {
      "entropy": 5.7478841781616214,
      "epoch": 1.586012009890498,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047511925977549303,
      "loss": 5.2852,
      "mean_token_accuracy": 0.21106737554073335,
      "num_tokens": 33480984.0,
      "step": 17960
    },
    {
      "entropy": 5.8340246200561525,
      "epoch": 1.5864535499823385,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047510486327005604,
      "loss": 5.4365,
      "mean_token_accuracy": 0.2043757900595665,
      "num_tokens": 33491196.0,
      "step": 17965
    },
    {
      "entropy": 5.801984882354736,
      "epoch": 1.5868950900741787,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047509046284467855,
      "loss": 5.2733,
      "mean_token_accuracy": 0.20755961239337922,
      "num_tokens": 33500404.0,
      "step": 17970
    },
    {
      "entropy": 5.937586784362793,
      "epoch": 1.587336630166019,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047507605849964274,
      "loss": 5.4694,
      "mean_token_accuracy": 0.19570547789335252,
      "num_tokens": 33509281.0,
      "step": 17975
    },
    {
      "entropy": 5.924002933502197,
      "epoch": 1.5877781702578595,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004750616502352307,
      "loss": 5.3458,
      "mean_token_accuracy": 0.20356533229351043,
      "num_tokens": 33518556.0,
      "step": 17980
    },
    {
      "entropy": 5.91500301361084,
      "epoch": 1.5882197103496998,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047504723805172474,
      "loss": 5.5362,
      "mean_token_accuracy": 0.1891017362475395,
      "num_tokens": 33529266.0,
      "step": 17985
    },
    {
      "entropy": 5.849728584289551,
      "epoch": 1.58866125044154,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047503282194940715,
      "loss": 5.4006,
      "mean_token_accuracy": 0.20334022492170334,
      "num_tokens": 33538972.0,
      "step": 17990
    },
    {
      "entropy": 5.850444889068603,
      "epoch": 1.5891027905333805,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004750184019285603,
      "loss": 5.3433,
      "mean_token_accuracy": 0.2019957810640335,
      "num_tokens": 33549528.0,
      "step": 17995
    },
    {
      "entropy": 5.809695053100586,
      "epoch": 1.5895443306252208,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004750039779894668,
      "loss": 5.2668,
      "mean_token_accuracy": 0.20987723767757416,
      "num_tokens": 33557828.0,
      "step": 18000
    },
    {
      "epoch": 1.5895443306252208,
      "eval_entropy": 5.575396425510722,
      "eval_loss": 5.42067813873291,
      "eval_mean_token_accuracy": 0.21030604640277165,
      "eval_num_tokens": 33557828.0,
      "eval_runtime": 20.1472,
      "eval_samples_per_second": 1752.847,
      "eval_steps_per_second": 219.137,
      "step": 18000
    },
    {
      "entropy": 5.881388092041016,
      "epoch": 1.589985870717061,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047498955013240907,
      "loss": 5.4308,
      "mean_token_accuracy": 0.197257362306118,
      "num_tokens": 33567087.0,
      "step": 18005
    },
    {
      "entropy": 5.878854417800904,
      "epoch": 1.5904274108089016,
      "grad_norm": 1.0,
      "learning_rate": 0.00047497511835766984,
      "loss": 5.4012,
      "mean_token_accuracy": 0.19553140550851822,
      "num_tokens": 33577442.0,
      "step": 18010
    },
    {
      "entropy": 5.893789529800415,
      "epoch": 1.5908689509007417,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047496068266553165,
      "loss": 5.4126,
      "mean_token_accuracy": 0.2019266277551651,
      "num_tokens": 33585655.0,
      "step": 18015
    },
    {
      "entropy": 5.873789072036743,
      "epoch": 1.5913104909925821,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047494624305627754,
      "loss": 5.414,
      "mean_token_accuracy": 0.20742429941892623,
      "num_tokens": 33594808.0,
      "step": 18020
    },
    {
      "entropy": 5.933353424072266,
      "epoch": 1.5917520310844224,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004749317995301902,
      "loss": 5.4607,
      "mean_token_accuracy": 0.19954823702573776,
      "num_tokens": 33603924.0,
      "step": 18025
    },
    {
      "entropy": 5.84439697265625,
      "epoch": 1.5921935711762627,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004749173520875525,
      "loss": 5.3446,
      "mean_token_accuracy": 0.20497927963733673,
      "num_tokens": 33613870.0,
      "step": 18030
    },
    {
      "entropy": 5.817981767654419,
      "epoch": 1.5926351112681032,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047490290072864773,
      "loss": 5.382,
      "mean_token_accuracy": 0.20596250593662263,
      "num_tokens": 33622938.0,
      "step": 18035
    },
    {
      "entropy": 5.865955924987793,
      "epoch": 1.5930766513599435,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047488844545375886,
      "loss": 5.4436,
      "mean_token_accuracy": 0.20010052770376205,
      "num_tokens": 33631520.0,
      "step": 18040
    },
    {
      "entropy": 5.925880241394043,
      "epoch": 1.5935181914517837,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004748739862631689,
      "loss": 5.4307,
      "mean_token_accuracy": 0.20103561133146286,
      "num_tokens": 33641365.0,
      "step": 18045
    },
    {
      "entropy": 5.831374645233154,
      "epoch": 1.5939597315436242,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047485952315716126,
      "loss": 5.2374,
      "mean_token_accuracy": 0.2163265123963356,
      "num_tokens": 33650425.0,
      "step": 18050
    },
    {
      "entropy": 5.857035827636719,
      "epoch": 1.5944012716354645,
      "grad_norm": 1.125,
      "learning_rate": 0.00047484505613601917,
      "loss": 5.446,
      "mean_token_accuracy": 0.19701125025749205,
      "num_tokens": 33659928.0,
      "step": 18055
    },
    {
      "entropy": 5.838379144668579,
      "epoch": 1.5948428117273048,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004748305852000262,
      "loss": 5.4421,
      "mean_token_accuracy": 0.20323508083820344,
      "num_tokens": 33668860.0,
      "step": 18060
    },
    {
      "entropy": 5.901949787139893,
      "epoch": 1.5952843518191453,
      "grad_norm": 1.0,
      "learning_rate": 0.00047481611034946565,
      "loss": 5.3716,
      "mean_token_accuracy": 0.19717168658971787,
      "num_tokens": 33679343.0,
      "step": 18065
    },
    {
      "entropy": 5.805941200256347,
      "epoch": 1.5957258919109856,
      "grad_norm": 1.0,
      "learning_rate": 0.0004748016315846212,
      "loss": 5.3419,
      "mean_token_accuracy": 0.2030827969312668,
      "num_tokens": 33688749.0,
      "step": 18070
    },
    {
      "entropy": 5.857469367980957,
      "epoch": 1.5961674320028258,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047478714890577633,
      "loss": 5.4197,
      "mean_token_accuracy": 0.19789186418056487,
      "num_tokens": 33698359.0,
      "step": 18075
    },
    {
      "entropy": 5.925813961029053,
      "epoch": 1.5966089720946663,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047477266231321484,
      "loss": 5.3982,
      "mean_token_accuracy": 0.19987578094005584,
      "num_tokens": 33707402.0,
      "step": 18080
    },
    {
      "entropy": 5.869510984420776,
      "epoch": 1.5970505121865064,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00047475817180722056,
      "loss": 5.3421,
      "mean_token_accuracy": 0.2070999413728714,
      "num_tokens": 33717043.0,
      "step": 18085
    },
    {
      "entropy": 5.750817203521729,
      "epoch": 1.597492052278347,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00047474367738807734,
      "loss": 5.3132,
      "mean_token_accuracy": 0.2143101140856743,
      "num_tokens": 33726527.0,
      "step": 18090
    },
    {
      "entropy": 5.792326211929321,
      "epoch": 1.5979335923701872,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047472917905606903,
      "loss": 5.3179,
      "mean_token_accuracy": 0.20932081192731858,
      "num_tokens": 33735357.0,
      "step": 18095
    },
    {
      "entropy": 5.7805774211883545,
      "epoch": 1.5983751324620274,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047471467681147974,
      "loss": 5.209,
      "mean_token_accuracy": 0.22694893032312394,
      "num_tokens": 33743660.0,
      "step": 18100
    },
    {
      "entropy": 5.848110008239746,
      "epoch": 1.598816672553868,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004747001706545935,
      "loss": 5.3991,
      "mean_token_accuracy": 0.201560440659523,
      "num_tokens": 33752899.0,
      "step": 18105
    },
    {
      "entropy": 5.867289304733276,
      "epoch": 1.5992582126457082,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004746856605856944,
      "loss": 5.3977,
      "mean_token_accuracy": 0.2011600747704506,
      "num_tokens": 33761333.0,
      "step": 18110
    },
    {
      "entropy": 5.876346731185913,
      "epoch": 1.5996997527375485,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047467114660506687,
      "loss": 5.4301,
      "mean_token_accuracy": 0.19527532607316972,
      "num_tokens": 33770507.0,
      "step": 18115
    },
    {
      "entropy": 5.765473651885986,
      "epoch": 1.600141292829389,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004746566287129951,
      "loss": 5.3256,
      "mean_token_accuracy": 0.20503395348787307,
      "num_tokens": 33781658.0,
      "step": 18120
    },
    {
      "entropy": 5.878257131576538,
      "epoch": 1.6005828329212293,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004746421069097636,
      "loss": 5.4533,
      "mean_token_accuracy": 0.1909184128046036,
      "num_tokens": 33792229.0,
      "step": 18125
    },
    {
      "entropy": 5.928519010543823,
      "epoch": 1.6010243730130695,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047462758119565663,
      "loss": 5.4091,
      "mean_token_accuracy": 0.20041222125291824,
      "num_tokens": 33801674.0,
      "step": 18130
    },
    {
      "entropy": 5.8398048877716064,
      "epoch": 1.60146591310491,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000474613051570959,
      "loss": 5.345,
      "mean_token_accuracy": 0.21145925521850586,
      "num_tokens": 33812066.0,
      "step": 18135
    },
    {
      "entropy": 5.8932593822479244,
      "epoch": 1.6019074531967503,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047459851803595524,
      "loss": 5.468,
      "mean_token_accuracy": 0.19464504420757295,
      "num_tokens": 33822206.0,
      "step": 18140
    },
    {
      "entropy": 5.913393831253051,
      "epoch": 1.6023489932885906,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00047458398059092995,
      "loss": 5.454,
      "mean_token_accuracy": 0.19806140959262847,
      "num_tokens": 33832075.0,
      "step": 18145
    },
    {
      "entropy": 5.817515468597412,
      "epoch": 1.602790533380431,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047456943923616793,
      "loss": 5.2806,
      "mean_token_accuracy": 0.21167910546064378,
      "num_tokens": 33841523.0,
      "step": 18150
    },
    {
      "entropy": 5.870386791229248,
      "epoch": 1.6032320734722711,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047455489397195416,
      "loss": 5.3393,
      "mean_token_accuracy": 0.207174251973629,
      "num_tokens": 33850756.0,
      "step": 18155
    },
    {
      "entropy": 5.868167924880981,
      "epoch": 1.6036736135641116,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047454034479857356,
      "loss": 5.4247,
      "mean_token_accuracy": 0.20147025883197783,
      "num_tokens": 33859611.0,
      "step": 18160
    },
    {
      "entropy": 5.852504253387451,
      "epoch": 1.604115153655952,
      "grad_norm": 1.09375,
      "learning_rate": 0.000474525791716311,
      "loss": 5.4122,
      "mean_token_accuracy": 0.20078572481870652,
      "num_tokens": 33867993.0,
      "step": 18165
    },
    {
      "entropy": 5.923367261886597,
      "epoch": 1.6045566937477922,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047451123472545175,
      "loss": 5.4812,
      "mean_token_accuracy": 0.19497620314359665,
      "num_tokens": 33876785.0,
      "step": 18170
    },
    {
      "entropy": 5.877623701095581,
      "epoch": 1.6049982338396327,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047449667382628083,
      "loss": 5.3153,
      "mean_token_accuracy": 0.2080849826335907,
      "num_tokens": 33885729.0,
      "step": 18175
    },
    {
      "entropy": 5.8686470031738285,
      "epoch": 1.605439773931473,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004744821090190835,
      "loss": 5.3477,
      "mean_token_accuracy": 0.20913055837154387,
      "num_tokens": 33894622.0,
      "step": 18180
    },
    {
      "entropy": 5.867928886413575,
      "epoch": 1.6058813140233132,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047446754030414513,
      "loss": 5.4419,
      "mean_token_accuracy": 0.19912338107824326,
      "num_tokens": 33904187.0,
      "step": 18185
    },
    {
      "entropy": 5.8461769104003904,
      "epoch": 1.6063228541151537,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004744529676817511,
      "loss": 5.4355,
      "mean_token_accuracy": 0.2053565114736557,
      "num_tokens": 33914327.0,
      "step": 18190
    },
    {
      "entropy": 5.895674133300782,
      "epoch": 1.606764394206994,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047443839115218685,
      "loss": 5.2759,
      "mean_token_accuracy": 0.20811100602149962,
      "num_tokens": 33923424.0,
      "step": 18195
    },
    {
      "entropy": 5.852606964111328,
      "epoch": 1.6072059342988343,
      "grad_norm": 1.09375,
      "learning_rate": 0.000474423810715738,
      "loss": 5.321,
      "mean_token_accuracy": 0.20784832090139388,
      "num_tokens": 33932145.0,
      "step": 18200
    },
    {
      "entropy": 5.789508914947509,
      "epoch": 1.6076474743906748,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004744092263726901,
      "loss": 5.3339,
      "mean_token_accuracy": 0.2098607525229454,
      "num_tokens": 33942505.0,
      "step": 18205
    },
    {
      "entropy": 5.848215389251709,
      "epoch": 1.608089014482515,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004743946381233288,
      "loss": 5.3973,
      "mean_token_accuracy": 0.20695634037256241,
      "num_tokens": 33952156.0,
      "step": 18210
    },
    {
      "entropy": 5.91343207359314,
      "epoch": 1.6085305545743553,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047438004596794,
      "loss": 5.4195,
      "mean_token_accuracy": 0.19986581057310104,
      "num_tokens": 33962478.0,
      "step": 18215
    },
    {
      "entropy": 5.866097068786621,
      "epoch": 1.6089720946661958,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004743654499068095,
      "loss": 5.3509,
      "mean_token_accuracy": 0.19943293333053588,
      "num_tokens": 33971581.0,
      "step": 18220
    },
    {
      "entropy": 5.816497421264648,
      "epoch": 1.609413634758036,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004743508499402232,
      "loss": 5.3722,
      "mean_token_accuracy": 0.20656681656837464,
      "num_tokens": 33981262.0,
      "step": 18225
    },
    {
      "entropy": 5.899464416503906,
      "epoch": 1.6098551748498764,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004743362460684671,
      "loss": 5.4086,
      "mean_token_accuracy": 0.203243488073349,
      "num_tokens": 33989865.0,
      "step": 18230
    },
    {
      "entropy": 5.893782281875611,
      "epoch": 1.6102967149417167,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047432163829182737,
      "loss": 5.3748,
      "mean_token_accuracy": 0.20195204466581346,
      "num_tokens": 33999261.0,
      "step": 18235
    },
    {
      "entropy": 5.925965452194214,
      "epoch": 1.610738255033557,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047430702661059007,
      "loss": 5.385,
      "mean_token_accuracy": 0.19887196868658066,
      "num_tokens": 34008471.0,
      "step": 18240
    },
    {
      "entropy": 5.780384492874146,
      "epoch": 1.6111797951253974,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047429241102504144,
      "loss": 5.2668,
      "mean_token_accuracy": 0.21342891454696655,
      "num_tokens": 34018305.0,
      "step": 18245
    },
    {
      "entropy": 5.877861070632934,
      "epoch": 1.6116213352172377,
      "grad_norm": 1.0,
      "learning_rate": 0.00047427779153546787,
      "loss": 5.4222,
      "mean_token_accuracy": 0.18988059014081954,
      "num_tokens": 34027393.0,
      "step": 18250
    },
    {
      "entropy": 5.898454761505127,
      "epoch": 1.612062875309078,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047426316814215563,
      "loss": 5.3608,
      "mean_token_accuracy": 0.20616988688707352,
      "num_tokens": 34035835.0,
      "step": 18255
    },
    {
      "entropy": 5.811767196655273,
      "epoch": 1.6125044154009185,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047424854084539134,
      "loss": 5.4164,
      "mean_token_accuracy": 0.20323833525180818,
      "num_tokens": 34045946.0,
      "step": 18260
    },
    {
      "entropy": 5.925228691101074,
      "epoch": 1.6129459554927588,
      "grad_norm": 1.0,
      "learning_rate": 0.00047423390964546145,
      "loss": 5.4678,
      "mean_token_accuracy": 0.19535911977291107,
      "num_tokens": 34055356.0,
      "step": 18265
    },
    {
      "entropy": 5.977757596969605,
      "epoch": 1.613387495584599,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004742192745426525,
      "loss": 5.5347,
      "mean_token_accuracy": 0.18955558091402053,
      "num_tokens": 34065862.0,
      "step": 18270
    },
    {
      "entropy": 5.963193941116333,
      "epoch": 1.6138290356764395,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004742046355372513,
      "loss": 5.4159,
      "mean_token_accuracy": 0.20128547996282578,
      "num_tokens": 34074980.0,
      "step": 18275
    },
    {
      "entropy": 5.8671472549438475,
      "epoch": 1.6142705757682798,
      "grad_norm": 1.0,
      "learning_rate": 0.00047418999262954466,
      "loss": 5.3847,
      "mean_token_accuracy": 0.19881257563829421,
      "num_tokens": 34085501.0,
      "step": 18280
    },
    {
      "entropy": 5.845548009872436,
      "epoch": 1.61471211586012,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004741753458198192,
      "loss": 5.3587,
      "mean_token_accuracy": 0.20400122851133345,
      "num_tokens": 34095564.0,
      "step": 18285
    },
    {
      "entropy": 5.896277904510498,
      "epoch": 1.6151536559519606,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047416069510836215,
      "loss": 5.4195,
      "mean_token_accuracy": 0.19548761695623398,
      "num_tokens": 34104571.0,
      "step": 18290
    },
    {
      "entropy": 5.878162956237793,
      "epoch": 1.6155951960438006,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004741460404954602,
      "loss": 5.4421,
      "mean_token_accuracy": 0.19408143907785416,
      "num_tokens": 34114272.0,
      "step": 18295
    },
    {
      "entropy": 5.76869306564331,
      "epoch": 1.6160367361356411,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004741313819814007,
      "loss": 5.352,
      "mean_token_accuracy": 0.20639338493347167,
      "num_tokens": 34125521.0,
      "step": 18300
    },
    {
      "entropy": 5.844787216186523,
      "epoch": 1.6164782762274814,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004741167195664706,
      "loss": 5.4392,
      "mean_token_accuracy": 0.19454080611467361,
      "num_tokens": 34134447.0,
      "step": 18305
    },
    {
      "entropy": 5.8771710872650145,
      "epoch": 1.6169198163193217,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047410205325095725,
      "loss": 5.3178,
      "mean_token_accuracy": 0.20895783901214598,
      "num_tokens": 34143207.0,
      "step": 18310
    },
    {
      "entropy": 5.873680162429809,
      "epoch": 1.6173613564111622,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004740873830351479,
      "loss": 5.3493,
      "mean_token_accuracy": 0.20246138870716096,
      "num_tokens": 34153025.0,
      "step": 18315
    },
    {
      "entropy": 5.865786027908325,
      "epoch": 1.6178028965030025,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047407270891932995,
      "loss": 5.3558,
      "mean_token_accuracy": 0.20315389782190324,
      "num_tokens": 34162793.0,
      "step": 18320
    },
    {
      "entropy": 5.862230348587036,
      "epoch": 1.6182444365948427,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047405803090379083,
      "loss": 5.4092,
      "mean_token_accuracy": 0.1985282376408577,
      "num_tokens": 34171941.0,
      "step": 18325
    },
    {
      "entropy": 5.887270593643189,
      "epoch": 1.6186859766866832,
      "grad_norm": 1.0,
      "learning_rate": 0.0004740433489888182,
      "loss": 5.382,
      "mean_token_accuracy": 0.20298713445663452,
      "num_tokens": 34180993.0,
      "step": 18330
    },
    {
      "entropy": 5.89012336730957,
      "epoch": 1.6191275167785235,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047402866317469944,
      "loss": 5.3029,
      "mean_token_accuracy": 0.21341717392206191,
      "num_tokens": 34190078.0,
      "step": 18335
    },
    {
      "entropy": 5.774645376205444,
      "epoch": 1.6195690568703638,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004740139734617224,
      "loss": 5.3122,
      "mean_token_accuracy": 0.21125834286212922,
      "num_tokens": 34199091.0,
      "step": 18340
    },
    {
      "entropy": 5.848790168762207,
      "epoch": 1.6200105969622043,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004739992798501748,
      "loss": 5.4204,
      "mean_token_accuracy": 0.20708162635564803,
      "num_tokens": 34209042.0,
      "step": 18345
    },
    {
      "entropy": 5.9101039409637455,
      "epoch": 1.6204521370540446,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047398458234034445,
      "loss": 5.3888,
      "mean_token_accuracy": 0.20516490340232849,
      "num_tokens": 34217311.0,
      "step": 18350
    },
    {
      "entropy": 5.782332801818848,
      "epoch": 1.6208936771458848,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047396988093251934,
      "loss": 5.3518,
      "mean_token_accuracy": 0.2104898452758789,
      "num_tokens": 34227001.0,
      "step": 18355
    },
    {
      "entropy": 5.827291965484619,
      "epoch": 1.6213352172377253,
      "grad_norm": 1.0,
      "learning_rate": 0.0004739551756269875,
      "loss": 5.3817,
      "mean_token_accuracy": 0.20240366756916045,
      "num_tokens": 34236214.0,
      "step": 18360
    },
    {
      "entropy": 5.875600099563599,
      "epoch": 1.6217767573295654,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004739404664240368,
      "loss": 5.3561,
      "mean_token_accuracy": 0.2042583554983139,
      "num_tokens": 34244808.0,
      "step": 18365
    },
    {
      "entropy": 5.841067981719971,
      "epoch": 1.6222182974214059,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047392575332395556,
      "loss": 5.3669,
      "mean_token_accuracy": 0.21230745315551758,
      "num_tokens": 34253355.0,
      "step": 18370
    },
    {
      "entropy": 5.9156270027160645,
      "epoch": 1.6226598375132462,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004739110363270319,
      "loss": 5.4646,
      "mean_token_accuracy": 0.19377294778823853,
      "num_tokens": 34262235.0,
      "step": 18375
    },
    {
      "entropy": 5.871703672409057,
      "epoch": 1.6231013776050864,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00047389631543355424,
      "loss": 5.3273,
      "mean_token_accuracy": 0.2104901224374771,
      "num_tokens": 34272174.0,
      "step": 18380
    },
    {
      "entropy": 5.890290784835815,
      "epoch": 1.623542917696927,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004738815906438108,
      "loss": 5.3552,
      "mean_token_accuracy": 0.20510149002075195,
      "num_tokens": 34281153.0,
      "step": 18385
    },
    {
      "entropy": 5.851149082183838,
      "epoch": 1.6239844577887672,
      "grad_norm": 1.0,
      "learning_rate": 0.00047386686195809017,
      "loss": 5.3791,
      "mean_token_accuracy": 0.1983202025294304,
      "num_tokens": 34291045.0,
      "step": 18390
    },
    {
      "entropy": 5.82644214630127,
      "epoch": 1.6244259978806075,
      "grad_norm": 1.125,
      "learning_rate": 0.00047385212937668084,
      "loss": 5.3332,
      "mean_token_accuracy": 0.20592467188835145,
      "num_tokens": 34299234.0,
      "step": 18395
    },
    {
      "entropy": 5.822642421722412,
      "epoch": 1.624867537972448,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047383739289987134,
      "loss": 5.4037,
      "mean_token_accuracy": 0.2028586596250534,
      "num_tokens": 34309084.0,
      "step": 18400
    },
    {
      "entropy": 5.932138252258301,
      "epoch": 1.6253090780642883,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004738226525279504,
      "loss": 5.47,
      "mean_token_accuracy": 0.19281615167856217,
      "num_tokens": 34320039.0,
      "step": 18405
    },
    {
      "entropy": 5.83445372581482,
      "epoch": 1.6257506181561285,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004738079082612068,
      "loss": 5.2778,
      "mean_token_accuracy": 0.2126028507947922,
      "num_tokens": 34329020.0,
      "step": 18410
    },
    {
      "entropy": 5.821759986877441,
      "epoch": 1.626192158247969,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047379316009992927,
      "loss": 5.3986,
      "mean_token_accuracy": 0.19910838752985,
      "num_tokens": 34337953.0,
      "step": 18415
    },
    {
      "entropy": 5.7514075756073,
      "epoch": 1.6266336983398093,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047377840804440686,
      "loss": 5.2854,
      "mean_token_accuracy": 0.21218639463186265,
      "num_tokens": 34346564.0,
      "step": 18420
    },
    {
      "entropy": 5.89910740852356,
      "epoch": 1.6270752384316496,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004737636520949285,
      "loss": 5.5159,
      "mean_token_accuracy": 0.1931287094950676,
      "num_tokens": 34357163.0,
      "step": 18425
    },
    {
      "entropy": 5.906308269500732,
      "epoch": 1.62751677852349,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004737488922517832,
      "loss": 5.4759,
      "mean_token_accuracy": 0.19571054875850677,
      "num_tokens": 34366731.0,
      "step": 18430
    },
    {
      "entropy": 5.814606523513794,
      "epoch": 1.6279583186153301,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047373412851526013,
      "loss": 5.3377,
      "mean_token_accuracy": 0.20792262107133866,
      "num_tokens": 34375845.0,
      "step": 18435
    },
    {
      "entropy": 5.864726448059082,
      "epoch": 1.6283998587071706,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047371936088564856,
      "loss": 5.3398,
      "mean_token_accuracy": 0.21043552607297897,
      "num_tokens": 34385335.0,
      "step": 18440
    },
    {
      "entropy": 5.8956443786621096,
      "epoch": 1.628841398799011,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004737045893632377,
      "loss": 5.4335,
      "mean_token_accuracy": 0.20258648693561554,
      "num_tokens": 34394356.0,
      "step": 18445
    },
    {
      "entropy": 5.829281616210937,
      "epoch": 1.6292829388908512,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047368981394831696,
      "loss": 5.3708,
      "mean_token_accuracy": 0.2046119302511215,
      "num_tokens": 34403727.0,
      "step": 18450
    },
    {
      "entropy": 5.924932670593262,
      "epoch": 1.6297244789826917,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047367503464117573,
      "loss": 5.4336,
      "mean_token_accuracy": 0.19864198565483093,
      "num_tokens": 34413253.0,
      "step": 18455
    },
    {
      "entropy": 5.813119983673095,
      "epoch": 1.630166019074532,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047366025144210356,
      "loss": 5.339,
      "mean_token_accuracy": 0.20886782556772232,
      "num_tokens": 34422430.0,
      "step": 18460
    },
    {
      "entropy": 5.824563121795654,
      "epoch": 1.6306075591663722,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047364546435139004,
      "loss": 5.3796,
      "mean_token_accuracy": 0.2088511675596237,
      "num_tokens": 34431111.0,
      "step": 18465
    },
    {
      "entropy": 5.8530223846435545,
      "epoch": 1.6310490992582127,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047363067336932493,
      "loss": 5.3678,
      "mean_token_accuracy": 0.20952707827091216,
      "num_tokens": 34440044.0,
      "step": 18470
    },
    {
      "entropy": 5.826060247421265,
      "epoch": 1.631490639350053,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004736158784961978,
      "loss": 5.3683,
      "mean_token_accuracy": 0.21185309141874314,
      "num_tokens": 34450315.0,
      "step": 18475
    },
    {
      "entropy": 5.820285654067993,
      "epoch": 1.6319321794418933,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047360107973229865,
      "loss": 5.2941,
      "mean_token_accuracy": 0.21950102746486663,
      "num_tokens": 34459380.0,
      "step": 18480
    },
    {
      "entropy": 5.89492883682251,
      "epoch": 1.6323737195337338,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004735862770779172,
      "loss": 5.4187,
      "mean_token_accuracy": 0.20642026364803315,
      "num_tokens": 34468224.0,
      "step": 18485
    },
    {
      "entropy": 5.780845785140992,
      "epoch": 1.632815259625574,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004735714705333436,
      "loss": 5.2894,
      "mean_token_accuracy": 0.20784697830677032,
      "num_tokens": 34476917.0,
      "step": 18490
    },
    {
      "entropy": 5.784996795654297,
      "epoch": 1.6332567997174143,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047355666009886785,
      "loss": 5.3613,
      "mean_token_accuracy": 0.21024878323078156,
      "num_tokens": 34487192.0,
      "step": 18495
    },
    {
      "entropy": 5.862994718551636,
      "epoch": 1.6336983398092548,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047354184577478,
      "loss": 5.3609,
      "mean_token_accuracy": 0.20948976129293442,
      "num_tokens": 34496658.0,
      "step": 18500
    },
    {
      "entropy": 5.878349256515503,
      "epoch": 1.6341398799010949,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047352702756137035,
      "loss": 5.2955,
      "mean_token_accuracy": 0.21331375241279601,
      "num_tokens": 34505126.0,
      "step": 18505
    },
    {
      "entropy": 5.845956468582154,
      "epoch": 1.6345814199929354,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004735122054589291,
      "loss": 5.3476,
      "mean_token_accuracy": 0.2015761613845825,
      "num_tokens": 34514236.0,
      "step": 18510
    },
    {
      "entropy": 5.735983371734619,
      "epoch": 1.6350229600847757,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004734973794677467,
      "loss": 5.273,
      "mean_token_accuracy": 0.21282364428043365,
      "num_tokens": 34522609.0,
      "step": 18515
    },
    {
      "entropy": 5.772179412841797,
      "epoch": 1.635464500176616,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047348254958811347,
      "loss": 5.3932,
      "mean_token_accuracy": 0.20632438063621522,
      "num_tokens": 34531742.0,
      "step": 18520
    },
    {
      "entropy": 5.805711889266968,
      "epoch": 1.6359060402684564,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047346771582031994,
      "loss": 5.3198,
      "mean_token_accuracy": 0.20902577936649322,
      "num_tokens": 34540821.0,
      "step": 18525
    },
    {
      "entropy": 5.914322185516357,
      "epoch": 1.6363475803602967,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004734528781646568,
      "loss": 5.3986,
      "mean_token_accuracy": 0.20035960674285888,
      "num_tokens": 34549795.0,
      "step": 18530
    },
    {
      "entropy": 5.883692359924316,
      "epoch": 1.636789120452137,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004734380366214146,
      "loss": 5.4239,
      "mean_token_accuracy": 0.19963254034519196,
      "num_tokens": 34559455.0,
      "step": 18535
    },
    {
      "entropy": 5.880136299133301,
      "epoch": 1.6372306605439775,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004734231911908842,
      "loss": 5.4154,
      "mean_token_accuracy": 0.2007180407643318,
      "num_tokens": 34568832.0,
      "step": 18540
    },
    {
      "entropy": 5.80340805053711,
      "epoch": 1.6376722006358178,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00047340834187335626,
      "loss": 5.3684,
      "mean_token_accuracy": 0.207799032330513,
      "num_tokens": 34578630.0,
      "step": 18545
    },
    {
      "entropy": 5.888047599792481,
      "epoch": 1.638113740727658,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047339348866912173,
      "loss": 5.4145,
      "mean_token_accuracy": 0.19899772703647614,
      "num_tokens": 34587782.0,
      "step": 18550
    },
    {
      "entropy": 5.838915061950684,
      "epoch": 1.6385552808194985,
      "grad_norm": 1.125,
      "learning_rate": 0.0004733786315784716,
      "loss": 5.343,
      "mean_token_accuracy": 0.20412253439426423,
      "num_tokens": 34597133.0,
      "step": 18555
    },
    {
      "entropy": 5.820776557922363,
      "epoch": 1.6389968209113388,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047336377060169696,
      "loss": 5.4203,
      "mean_token_accuracy": 0.19890081137418747,
      "num_tokens": 34606697.0,
      "step": 18560
    },
    {
      "entropy": 5.866944885253906,
      "epoch": 1.639438361003179,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004733489057390888,
      "loss": 5.373,
      "mean_token_accuracy": 0.20852273404598237,
      "num_tokens": 34615806.0,
      "step": 18565
    },
    {
      "entropy": 5.891376066207886,
      "epoch": 1.6398799010950196,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004733340369909384,
      "loss": 5.5243,
      "mean_token_accuracy": 0.19144340604543686,
      "num_tokens": 34626563.0,
      "step": 18570
    },
    {
      "entropy": 5.879660797119141,
      "epoch": 1.6403214411868596,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047331916435753706,
      "loss": 5.4042,
      "mean_token_accuracy": 0.20562248826026916,
      "num_tokens": 34636077.0,
      "step": 18575
    },
    {
      "entropy": 5.878395032882691,
      "epoch": 1.6407629812787001,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047330428783917607,
      "loss": 5.4508,
      "mean_token_accuracy": 0.1952090084552765,
      "num_tokens": 34646164.0,
      "step": 18580
    },
    {
      "entropy": 5.882407474517822,
      "epoch": 1.6412045213705404,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047328940743614683,
      "loss": 5.3852,
      "mean_token_accuracy": 0.19838855415582657,
      "num_tokens": 34655577.0,
      "step": 18585
    },
    {
      "entropy": 5.875201749801636,
      "epoch": 1.6416460614623807,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047327452314874094,
      "loss": 5.4252,
      "mean_token_accuracy": 0.192774598300457,
      "num_tokens": 34664520.0,
      "step": 18590
    },
    {
      "entropy": 5.820896291732788,
      "epoch": 1.6420876015542212,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047325963497724983,
      "loss": 5.3218,
      "mean_token_accuracy": 0.205902536213398,
      "num_tokens": 34673528.0,
      "step": 18595
    },
    {
      "entropy": 5.865429306030274,
      "epoch": 1.6425291416460615,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004732447429219652,
      "loss": 5.3969,
      "mean_token_accuracy": 0.20284605920314788,
      "num_tokens": 34682563.0,
      "step": 18600
    },
    {
      "entropy": 5.89749960899353,
      "epoch": 1.6429706817379017,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004732298469831789,
      "loss": 5.3048,
      "mean_token_accuracy": 0.20845038294792176,
      "num_tokens": 34691932.0,
      "step": 18605
    },
    {
      "entropy": 5.887390661239624,
      "epoch": 1.6434122218297422,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047321494716118256,
      "loss": 5.3446,
      "mean_token_accuracy": 0.2051656261086464,
      "num_tokens": 34701376.0,
      "step": 18610
    },
    {
      "entropy": 5.795174598693848,
      "epoch": 1.6438537619215825,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047320004345626816,
      "loss": 5.3459,
      "mean_token_accuracy": 0.20236467868089675,
      "num_tokens": 34711112.0,
      "step": 18615
    },
    {
      "entropy": 5.837432384490967,
      "epoch": 1.6442953020134228,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047318513586872766,
      "loss": 5.351,
      "mean_token_accuracy": 0.21455023884773256,
      "num_tokens": 34720429.0,
      "step": 18620
    },
    {
      "entropy": 5.854586839675903,
      "epoch": 1.6447368421052633,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000473170224398853,
      "loss": 5.3693,
      "mean_token_accuracy": 0.20948509573936464,
      "num_tokens": 34729422.0,
      "step": 18625
    },
    {
      "entropy": 5.915212249755859,
      "epoch": 1.6451783821971036,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047315530904693643,
      "loss": 5.3917,
      "mean_token_accuracy": 0.20037249922752381,
      "num_tokens": 34738231.0,
      "step": 18630
    },
    {
      "entropy": 5.9505791664123535,
      "epoch": 1.6456199222889438,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047314038981327,
      "loss": 5.4269,
      "mean_token_accuracy": 0.20173603296279907,
      "num_tokens": 34747002.0,
      "step": 18635
    },
    {
      "entropy": 5.7558191299438475,
      "epoch": 1.6460614623807843,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000473125466698146,
      "loss": 5.3056,
      "mean_token_accuracy": 0.215034019947052,
      "num_tokens": 34756093.0,
      "step": 18640
    },
    {
      "entropy": 5.770248460769653,
      "epoch": 1.6465030024726244,
      "grad_norm": 1.0,
      "learning_rate": 0.0004731105397018569,
      "loss": 5.3776,
      "mean_token_accuracy": 0.20310365110635759,
      "num_tokens": 34766031.0,
      "step": 18645
    },
    {
      "entropy": 5.918416357040405,
      "epoch": 1.6469445425644649,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004730956088246949,
      "loss": 5.4459,
      "mean_token_accuracy": 0.20414311438798904,
      "num_tokens": 34776487.0,
      "step": 18650
    },
    {
      "entropy": 5.903948783874512,
      "epoch": 1.6473860826563052,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047308067406695255,
      "loss": 5.3058,
      "mean_token_accuracy": 0.2076905757188797,
      "num_tokens": 34785372.0,
      "step": 18655
    },
    {
      "entropy": 5.789320850372315,
      "epoch": 1.6478276227481454,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047306573542892257,
      "loss": 5.3299,
      "mean_token_accuracy": 0.20968879461288453,
      "num_tokens": 34795453.0,
      "step": 18660
    },
    {
      "entropy": 5.8344183444976805,
      "epoch": 1.648269162839986,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047305079291089735,
      "loss": 5.4095,
      "mean_token_accuracy": 0.20453164875507354,
      "num_tokens": 34804818.0,
      "step": 18665
    },
    {
      "entropy": 5.847234725952148,
      "epoch": 1.6487107029318262,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047303584651316985,
      "loss": 5.4392,
      "mean_token_accuracy": 0.19708086997270585,
      "num_tokens": 34814031.0,
      "step": 18670
    },
    {
      "entropy": 5.816784191131592,
      "epoch": 1.6491522430236665,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004730208962360327,
      "loss": 5.3254,
      "mean_token_accuracy": 0.20470742732286454,
      "num_tokens": 34823560.0,
      "step": 18675
    },
    {
      "entropy": 5.883140230178833,
      "epoch": 1.649593783115507,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004730059420797788,
      "loss": 5.3244,
      "mean_token_accuracy": 0.20900787711143493,
      "num_tokens": 34833662.0,
      "step": 18680
    },
    {
      "entropy": 5.897165203094483,
      "epoch": 1.6500353232073472,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004729909840447011,
      "loss": 5.4622,
      "mean_token_accuracy": 0.1925043895840645,
      "num_tokens": 34844264.0,
      "step": 18685
    },
    {
      "entropy": 5.902440738677979,
      "epoch": 1.6504768632991875,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047297602213109265,
      "loss": 5.3679,
      "mean_token_accuracy": 0.20841593146324158,
      "num_tokens": 34853708.0,
      "step": 18690
    },
    {
      "entropy": 5.937035799026489,
      "epoch": 1.650918403391028,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004729610563392465,
      "loss": 5.4398,
      "mean_token_accuracy": 0.19457600265741348,
      "num_tokens": 34861926.0,
      "step": 18695
    },
    {
      "entropy": 5.8199704647064205,
      "epoch": 1.6513599434828683,
      "grad_norm": 1.125,
      "learning_rate": 0.0004729460866694558,
      "loss": 5.3512,
      "mean_token_accuracy": 0.2098531037569046,
      "num_tokens": 34870919.0,
      "step": 18700
    },
    {
      "entropy": 5.851771879196167,
      "epoch": 1.6518014835747086,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000472931113122014,
      "loss": 5.3979,
      "mean_token_accuracy": 0.2007477343082428,
      "num_tokens": 34879801.0,
      "step": 18705
    },
    {
      "entropy": 5.8653754711151125,
      "epoch": 1.652243023666549,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004729161356972141,
      "loss": 5.3702,
      "mean_token_accuracy": 0.20218620598316192,
      "num_tokens": 34888605.0,
      "step": 18710
    },
    {
      "entropy": 5.882697534561157,
      "epoch": 1.6526845637583891,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047290115439534977,
      "loss": 5.4375,
      "mean_token_accuracy": 0.19343627393245696,
      "num_tokens": 34899294.0,
      "step": 18715
    },
    {
      "entropy": 5.9275675296783445,
      "epoch": 1.6531261038502296,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004728861692167143,
      "loss": 5.437,
      "mean_token_accuracy": 0.19946447908878326,
      "num_tokens": 34908624.0,
      "step": 18720
    },
    {
      "entropy": 5.847243595123291,
      "epoch": 1.65356764394207,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004728711801616013,
      "loss": 5.3221,
      "mean_token_accuracy": 0.20005699694156648,
      "num_tokens": 34917390.0,
      "step": 18725
    },
    {
      "entropy": 5.853270483016968,
      "epoch": 1.6540091840339102,
      "grad_norm": 1.0,
      "learning_rate": 0.0004728561872303045,
      "loss": 5.3924,
      "mean_token_accuracy": 0.2000747174024582,
      "num_tokens": 34926652.0,
      "step": 18730
    },
    {
      "entropy": 5.840923500061035,
      "epoch": 1.6544507241257507,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004728411904231175,
      "loss": 5.4278,
      "mean_token_accuracy": 0.19802002906799315,
      "num_tokens": 34936900.0,
      "step": 18735
    },
    {
      "entropy": 5.9109937191009525,
      "epoch": 1.654892264217591,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004728261897403341,
      "loss": 5.4059,
      "mean_token_accuracy": 0.2028878390789032,
      "num_tokens": 34946827.0,
      "step": 18740
    },
    {
      "entropy": 5.887080526351928,
      "epoch": 1.6553338043094312,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047281118518224807,
      "loss": 5.4529,
      "mean_token_accuracy": 0.1989116445183754,
      "num_tokens": 34955987.0,
      "step": 18745
    },
    {
      "entropy": 5.810631990432739,
      "epoch": 1.6557753444012717,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004727961767491535,
      "loss": 5.3405,
      "mean_token_accuracy": 0.20601450204849242,
      "num_tokens": 34965001.0,
      "step": 18750
    },
    {
      "entropy": 5.88202395439148,
      "epoch": 1.656216884493112,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004727811644413443,
      "loss": 5.4249,
      "mean_token_accuracy": 0.19914984703063965,
      "num_tokens": 34974113.0,
      "step": 18755
    },
    {
      "entropy": 5.803858470916748,
      "epoch": 1.6566584245849523,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004727661482591146,
      "loss": 5.2662,
      "mean_token_accuracy": 0.21207729279994963,
      "num_tokens": 34982535.0,
      "step": 18760
    },
    {
      "entropy": 5.839359855651855,
      "epoch": 1.6570999646767928,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004727511282027585,
      "loss": 5.4013,
      "mean_token_accuracy": 0.20235300362110137,
      "num_tokens": 34991828.0,
      "step": 18765
    },
    {
      "entropy": 5.86194076538086,
      "epoch": 1.657541504768633,
      "grad_norm": 1.0,
      "learning_rate": 0.0004727361042725703,
      "loss": 5.3522,
      "mean_token_accuracy": 0.20406592339277269,
      "num_tokens": 35001670.0,
      "step": 18770
    },
    {
      "entropy": 5.82854733467102,
      "epoch": 1.6579830448604733,
      "grad_norm": 1.0,
      "learning_rate": 0.0004727210764688442,
      "loss": 5.3053,
      "mean_token_accuracy": 0.21206227391958238,
      "num_tokens": 35011407.0,
      "step": 18775
    },
    {
      "entropy": 5.759233093261718,
      "epoch": 1.6584245849523138,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004727060447918748,
      "loss": 5.2626,
      "mean_token_accuracy": 0.21727235168218612,
      "num_tokens": 35020753.0,
      "step": 18780
    },
    {
      "entropy": 5.832109642028809,
      "epoch": 1.6588661250441539,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00047269100924195623,
      "loss": 5.3895,
      "mean_token_accuracy": 0.20545273274183273,
      "num_tokens": 35030599.0,
      "step": 18785
    },
    {
      "entropy": 5.880756568908692,
      "epoch": 1.6593076651359944,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047267596981938337,
      "loss": 5.4193,
      "mean_token_accuracy": 0.20258154422044755,
      "num_tokens": 35039098.0,
      "step": 18790
    },
    {
      "entropy": 5.892031860351563,
      "epoch": 1.6597492052278346,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047266092652445066,
      "loss": 5.3551,
      "mean_token_accuracy": 0.20884549170732497,
      "num_tokens": 35048286.0,
      "step": 18795
    },
    {
      "entropy": 5.917185974121094,
      "epoch": 1.660190745319675,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004726458793574528,
      "loss": 5.4458,
      "mean_token_accuracy": 0.19773625135421752,
      "num_tokens": 35057331.0,
      "step": 18800
    },
    {
      "entropy": 5.88164587020874,
      "epoch": 1.6606322854115154,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004726308283186846,
      "loss": 5.399,
      "mean_token_accuracy": 0.1980038672685623,
      "num_tokens": 35066321.0,
      "step": 18805
    },
    {
      "entropy": 5.773690843582154,
      "epoch": 1.6610738255033557,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004726157734084409,
      "loss": 5.339,
      "mean_token_accuracy": 0.20270509272813797,
      "num_tokens": 35075218.0,
      "step": 18810
    },
    {
      "entropy": 5.808038377761841,
      "epoch": 1.661515365595196,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004726007146270164,
      "loss": 5.3983,
      "mean_token_accuracy": 0.20630132853984834,
      "num_tokens": 35084563.0,
      "step": 18815
    },
    {
      "entropy": 5.79873571395874,
      "epoch": 1.6619569056870365,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004725856519747065,
      "loss": 5.2779,
      "mean_token_accuracy": 0.21271708309650422,
      "num_tokens": 35093346.0,
      "step": 18820
    },
    {
      "entropy": 5.857068824768066,
      "epoch": 1.6623984457788767,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004725705854518059,
      "loss": 5.3805,
      "mean_token_accuracy": 0.19910885095596315,
      "num_tokens": 35102501.0,
      "step": 18825
    },
    {
      "entropy": 5.855516576766968,
      "epoch": 1.662839985870717,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047255551505861007,
      "loss": 5.3788,
      "mean_token_accuracy": 0.2015061244368553,
      "num_tokens": 35112083.0,
      "step": 18830
    },
    {
      "entropy": 5.809852600097656,
      "epoch": 1.6632815259625575,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047254044079541387,
      "loss": 5.2996,
      "mean_token_accuracy": 0.2174018293619156,
      "num_tokens": 35121380.0,
      "step": 18835
    },
    {
      "entropy": 5.855141305923462,
      "epoch": 1.6637230660543978,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004725253626625129,
      "loss": 5.4386,
      "mean_token_accuracy": 0.19308264851570128,
      "num_tokens": 35130083.0,
      "step": 18840
    },
    {
      "entropy": 5.848509502410889,
      "epoch": 1.664164606146238,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047251028066020237,
      "loss": 5.3052,
      "mean_token_accuracy": 0.21321660578250884,
      "num_tokens": 35139001.0,
      "step": 18845
    },
    {
      "entropy": 5.81809778213501,
      "epoch": 1.6646061462380786,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047249519478877774,
      "loss": 5.2762,
      "mean_token_accuracy": 0.2110721617937088,
      "num_tokens": 35147701.0,
      "step": 18850
    },
    {
      "entropy": 5.805716037750244,
      "epoch": 1.6650476863299186,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004724801050485346,
      "loss": 5.285,
      "mean_token_accuracy": 0.21251884400844573,
      "num_tokens": 35157334.0,
      "step": 18855
    },
    {
      "entropy": 5.817666625976562,
      "epoch": 1.6654892264217591,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004724650114397685,
      "loss": 5.3325,
      "mean_token_accuracy": 0.2110571950674057,
      "num_tokens": 35166647.0,
      "step": 18860
    },
    {
      "entropy": 5.8946551322937015,
      "epoch": 1.6659307665135994,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047244991396277515,
      "loss": 5.3328,
      "mean_token_accuracy": 0.2022341623902321,
      "num_tokens": 35176345.0,
      "step": 18865
    },
    {
      "entropy": 5.8745075225830075,
      "epoch": 1.6663723066054397,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047243481261785027,
      "loss": 5.3335,
      "mean_token_accuracy": 0.20662084966897964,
      "num_tokens": 35185793.0,
      "step": 18870
    },
    {
      "entropy": 5.837036418914795,
      "epoch": 1.6668138466972802,
      "grad_norm": 1.0,
      "learning_rate": 0.00047241970740528975,
      "loss": 5.4159,
      "mean_token_accuracy": 0.20366771072149276,
      "num_tokens": 35195905.0,
      "step": 18875
    },
    {
      "entropy": 5.760806751251221,
      "epoch": 1.6672553867891204,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047240459832538934,
      "loss": 5.2823,
      "mean_token_accuracy": 0.2141048491001129,
      "num_tokens": 35204340.0,
      "step": 18880
    },
    {
      "entropy": 5.842314195632935,
      "epoch": 1.6676969268809607,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004723894853784452,
      "loss": 5.3789,
      "mean_token_accuracy": 0.20157068520784377,
      "num_tokens": 35213270.0,
      "step": 18885
    },
    {
      "entropy": 5.786710500717163,
      "epoch": 1.6681384669728012,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004723743685647533,
      "loss": 5.3618,
      "mean_token_accuracy": 0.2016805350780487,
      "num_tokens": 35222652.0,
      "step": 18890
    },
    {
      "entropy": 5.872417354583741,
      "epoch": 1.6685800070646415,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004723592478846097,
      "loss": 5.3571,
      "mean_token_accuracy": 0.2074940726161003,
      "num_tokens": 35231684.0,
      "step": 18895
    },
    {
      "entropy": 5.881430053710938,
      "epoch": 1.6690215471564818,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047234412333831073,
      "loss": 5.4161,
      "mean_token_accuracy": 0.20022842437028884,
      "num_tokens": 35241589.0,
      "step": 18900
    },
    {
      "entropy": 5.9737931251525875,
      "epoch": 1.6694630872483223,
      "grad_norm": 1.0,
      "learning_rate": 0.0004723289949261527,
      "loss": 5.5155,
      "mean_token_accuracy": 0.18734343498945236,
      "num_tokens": 35250903.0,
      "step": 18905
    },
    {
      "entropy": 5.850294208526611,
      "epoch": 1.6699046273401625,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004723138626484318,
      "loss": 5.375,
      "mean_token_accuracy": 0.2025665521621704,
      "num_tokens": 35260556.0,
      "step": 18910
    },
    {
      "entropy": 5.778967475891113,
      "epoch": 1.6703461674320028,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047229872650544446,
      "loss": 5.2654,
      "mean_token_accuracy": 0.2170637235045433,
      "num_tokens": 35270237.0,
      "step": 18915
    },
    {
      "entropy": 5.769305229187012,
      "epoch": 1.6707877075238433,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047228358649748745,
      "loss": 5.3051,
      "mean_token_accuracy": 0.2084587976336479,
      "num_tokens": 35279017.0,
      "step": 18920
    },
    {
      "entropy": 5.787815856933594,
      "epoch": 1.6712292476156834,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004722684426248571,
      "loss": 5.3269,
      "mean_token_accuracy": 0.20815094709396362,
      "num_tokens": 35288150.0,
      "step": 18925
    },
    {
      "entropy": 5.880165719985962,
      "epoch": 1.6716707877075239,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047225329488785015,
      "loss": 5.4056,
      "mean_token_accuracy": 0.20196864753961563,
      "num_tokens": 35297463.0,
      "step": 18930
    },
    {
      "entropy": 5.85289740562439,
      "epoch": 1.6721123277993641,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047223814328676336,
      "loss": 5.3692,
      "mean_token_accuracy": 0.20511680692434311,
      "num_tokens": 35307409.0,
      "step": 18935
    },
    {
      "entropy": 5.79543023109436,
      "epoch": 1.6725538678912044,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004722229878218935,
      "loss": 5.3172,
      "mean_token_accuracy": 0.20564032942056656,
      "num_tokens": 35315613.0,
      "step": 18940
    },
    {
      "entropy": 5.8183387279510494,
      "epoch": 1.672995407983045,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047220782849353743,
      "loss": 5.3838,
      "mean_token_accuracy": 0.2033582389354706,
      "num_tokens": 35325297.0,
      "step": 18945
    },
    {
      "entropy": 5.8459968090057375,
      "epoch": 1.6734369480748852,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004721926653019922,
      "loss": 5.3654,
      "mean_token_accuracy": 0.19865651726722716,
      "num_tokens": 35334014.0,
      "step": 18950
    },
    {
      "entropy": 5.82857460975647,
      "epoch": 1.6738784881667255,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004721774982475548,
      "loss": 5.3758,
      "mean_token_accuracy": 0.20930878072977066,
      "num_tokens": 35342784.0,
      "step": 18955
    },
    {
      "entropy": 5.930045795440674,
      "epoch": 1.674320028258566,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047216232733052237,
      "loss": 5.4783,
      "mean_token_accuracy": 0.1991959422826767,
      "num_tokens": 35351985.0,
      "step": 18960
    },
    {
      "entropy": 5.867730951309204,
      "epoch": 1.6747615683504062,
      "grad_norm": 1.125,
      "learning_rate": 0.0004721471525511921,
      "loss": 5.3389,
      "mean_token_accuracy": 0.1984658047556877,
      "num_tokens": 35360857.0,
      "step": 18965
    },
    {
      "entropy": 5.805293941497803,
      "epoch": 1.6752031084422465,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004721319739098611,
      "loss": 5.3304,
      "mean_token_accuracy": 0.2052135795354843,
      "num_tokens": 35370200.0,
      "step": 18970
    },
    {
      "entropy": 5.837771272659301,
      "epoch": 1.675644648534087,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000472116791406827,
      "loss": 5.3998,
      "mean_token_accuracy": 0.20853707939386368,
      "num_tokens": 35379481.0,
      "step": 18975
    },
    {
      "entropy": 5.810681009292603,
      "epoch": 1.6760861886259273,
      "grad_norm": 1.03125,
      "learning_rate": 0.000472101605042387,
      "loss": 5.3705,
      "mean_token_accuracy": 0.211078479886055,
      "num_tokens": 35387951.0,
      "step": 18980
    },
    {
      "entropy": 5.936203479766846,
      "epoch": 1.6765277287177676,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004720864148168387,
      "loss": 5.397,
      "mean_token_accuracy": 0.19801660627126694,
      "num_tokens": 35396200.0,
      "step": 18985
    },
    {
      "entropy": 5.892510652542114,
      "epoch": 1.676969268809608,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004720712207304796,
      "loss": 5.3794,
      "mean_token_accuracy": 0.19763886779546738,
      "num_tokens": 35404474.0,
      "step": 18990
    },
    {
      "entropy": 5.919994831085205,
      "epoch": 1.6774108089014481,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004720560227836073,
      "loss": 5.4726,
      "mean_token_accuracy": 0.19684788435697556,
      "num_tokens": 35414419.0,
      "step": 18995
    },
    {
      "entropy": 5.822895193099976,
      "epoch": 1.6778523489932886,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004720408209765197,
      "loss": 5.3023,
      "mean_token_accuracy": 0.21042004972696304,
      "num_tokens": 35423727.0,
      "step": 19000
    },
    {
      "entropy": 5.760235261917114,
      "epoch": 1.6782938890851289,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00047202561530951443,
      "loss": 5.2702,
      "mean_token_accuracy": 0.21885372698307037,
      "num_tokens": 35433364.0,
      "step": 19005
    },
    {
      "entropy": 5.919628286361695,
      "epoch": 1.6787354291769692,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004720104057828894,
      "loss": 5.4346,
      "mean_token_accuracy": 0.19773211181163788,
      "num_tokens": 35442665.0,
      "step": 19010
    },
    {
      "entropy": 5.8532500743865965,
      "epoch": 1.6791769692688097,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047199519239694257,
      "loss": 5.2658,
      "mean_token_accuracy": 0.21819208264350892,
      "num_tokens": 35451443.0,
      "step": 19015
    },
    {
      "entropy": 5.762861824035644,
      "epoch": 1.67961850936065,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047197997515197195,
      "loss": 5.2665,
      "mean_token_accuracy": 0.22124966681003572,
      "num_tokens": 35460622.0,
      "step": 19020
    },
    {
      "entropy": 5.757262182235718,
      "epoch": 1.6800600494524902,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047196475404827565,
      "loss": 5.409,
      "mean_token_accuracy": 0.20361010432243348,
      "num_tokens": 35470345.0,
      "step": 19025
    },
    {
      "entropy": 5.923474836349487,
      "epoch": 1.6805015895443307,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047194952908615183,
      "loss": 5.4904,
      "mean_token_accuracy": 0.19669502973556519,
      "num_tokens": 35480460.0,
      "step": 19030
    },
    {
      "entropy": 5.9418315410614015,
      "epoch": 1.680943129636171,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004719343002658987,
      "loss": 5.4182,
      "mean_token_accuracy": 0.1977526530623436,
      "num_tokens": 35489983.0,
      "step": 19035
    },
    {
      "entropy": 5.886590909957886,
      "epoch": 1.6813846697280113,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047191906758781475,
      "loss": 5.3489,
      "mean_token_accuracy": 0.20787034481763839,
      "num_tokens": 35498963.0,
      "step": 19040
    },
    {
      "entropy": 5.824855089187622,
      "epoch": 1.6818262098198518,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004719038310521982,
      "loss": 5.3827,
      "mean_token_accuracy": 0.2082638829946518,
      "num_tokens": 35509054.0,
      "step": 19045
    },
    {
      "entropy": 5.876221179962158,
      "epoch": 1.682267749911692,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004718885906593476,
      "loss": 5.5138,
      "mean_token_accuracy": 0.19023756682872772,
      "num_tokens": 35518093.0,
      "step": 19050
    },
    {
      "entropy": 5.888548374176025,
      "epoch": 1.6827092900035323,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004718733464095614,
      "loss": 5.3959,
      "mean_token_accuracy": 0.20310216248035431,
      "num_tokens": 35527330.0,
      "step": 19055
    },
    {
      "entropy": 5.882849168777466,
      "epoch": 1.6831508300953728,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047185809830313844,
      "loss": 5.3672,
      "mean_token_accuracy": 0.20466699302196503,
      "num_tokens": 35538147.0,
      "step": 19060
    },
    {
      "entropy": 5.962263298034668,
      "epoch": 1.6835923701872129,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047184284634037727,
      "loss": 5.4588,
      "mean_token_accuracy": 0.19409112334251405,
      "num_tokens": 35547930.0,
      "step": 19065
    },
    {
      "entropy": 5.858351755142212,
      "epoch": 1.6840339102790534,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047182759052157665,
      "loss": 5.2959,
      "mean_token_accuracy": 0.2073899418115616,
      "num_tokens": 35557111.0,
      "step": 19070
    },
    {
      "entropy": 5.8369776725769045,
      "epoch": 1.6844754503708936,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004718123308470355,
      "loss": 5.4761,
      "mean_token_accuracy": 0.19234734028577805,
      "num_tokens": 35565739.0,
      "step": 19075
    },
    {
      "entropy": 5.918671321868897,
      "epoch": 1.684916990462734,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047179706731705263,
      "loss": 5.3051,
      "mean_token_accuracy": 0.20207255184650422,
      "num_tokens": 35574164.0,
      "step": 19080
    },
    {
      "entropy": 5.855103445053101,
      "epoch": 1.6853585305545744,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047178179993192723,
      "loss": 5.3752,
      "mean_token_accuracy": 0.19642841666936875,
      "num_tokens": 35582813.0,
      "step": 19085
    },
    {
      "entropy": 5.870343112945557,
      "epoch": 1.6858000706464147,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047176652869195825,
      "loss": 5.3455,
      "mean_token_accuracy": 0.1986942782998085,
      "num_tokens": 35591884.0,
      "step": 19090
    },
    {
      "entropy": 5.872808456420898,
      "epoch": 1.686241610738255,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047175125359744487,
      "loss": 5.3648,
      "mean_token_accuracy": 0.201996086537838,
      "num_tokens": 35601824.0,
      "step": 19095
    },
    {
      "entropy": 5.8208756923675535,
      "epoch": 1.6866831508300955,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047173597464868634,
      "loss": 5.3101,
      "mean_token_accuracy": 0.2113776221871376,
      "num_tokens": 35610756.0,
      "step": 19100
    },
    {
      "entropy": 5.795255994796753,
      "epoch": 1.6871246909219357,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000471720691845982,
      "loss": 5.3149,
      "mean_token_accuracy": 0.20665600448846816,
      "num_tokens": 35620160.0,
      "step": 19105
    },
    {
      "entropy": 5.824592638015747,
      "epoch": 1.687566231013776,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047170540518963114,
      "loss": 5.4617,
      "mean_token_accuracy": 0.19557576924562453,
      "num_tokens": 35630487.0,
      "step": 19110
    },
    {
      "entropy": 5.980372428894043,
      "epoch": 1.6880077711056165,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004716901146799333,
      "loss": 5.5166,
      "mean_token_accuracy": 0.1869860053062439,
      "num_tokens": 35641924.0,
      "step": 19115
    },
    {
      "entropy": 5.903148031234741,
      "epoch": 1.6884493111974568,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004716748203171879,
      "loss": 5.3355,
      "mean_token_accuracy": 0.2022925317287445,
      "num_tokens": 35651575.0,
      "step": 19120
    },
    {
      "entropy": 5.825525188446045,
      "epoch": 1.688890851289297,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047165952210169475,
      "loss": 5.3211,
      "mean_token_accuracy": 0.20450648814439773,
      "num_tokens": 35661481.0,
      "step": 19125
    },
    {
      "entropy": 5.7702141284942625,
      "epoch": 1.6893323913811376,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004716442200337533,
      "loss": 5.2953,
      "mean_token_accuracy": 0.20387887954711914,
      "num_tokens": 35672049.0,
      "step": 19130
    },
    {
      "entropy": 5.893923759460449,
      "epoch": 1.6897739314729776,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047162891411366343,
      "loss": 5.3663,
      "mean_token_accuracy": 0.2059311419725418,
      "num_tokens": 35681653.0,
      "step": 19135
    },
    {
      "entropy": 5.908714342117309,
      "epoch": 1.690215471564818,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047161360434172506,
      "loss": 5.429,
      "mean_token_accuracy": 0.21284251362085344,
      "num_tokens": 35690868.0,
      "step": 19140
    },
    {
      "entropy": 5.822940111160278,
      "epoch": 1.6906570116566584,
      "grad_norm": 1.125,
      "learning_rate": 0.0004715982907182379,
      "loss": 5.3511,
      "mean_token_accuracy": 0.2021711513400078,
      "num_tokens": 35699062.0,
      "step": 19145
    },
    {
      "entropy": 5.790659379959107,
      "epoch": 1.6910985517484987,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004715829732435021,
      "loss": 5.3988,
      "mean_token_accuracy": 0.20533424168825148,
      "num_tokens": 35708411.0,
      "step": 19150
    },
    {
      "entropy": 5.82406325340271,
      "epoch": 1.6915400918403392,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047156765191781763,
      "loss": 5.3191,
      "mean_token_accuracy": 0.2091885730624199,
      "num_tokens": 35717937.0,
      "step": 19155
    },
    {
      "entropy": 5.8888551712036135,
      "epoch": 1.6919816319321794,
      "grad_norm": 1.0,
      "learning_rate": 0.00047155232674148465,
      "loss": 5.3308,
      "mean_token_accuracy": 0.20362054705619811,
      "num_tokens": 35727712.0,
      "step": 19160
    },
    {
      "entropy": 5.902427530288696,
      "epoch": 1.6924231720240197,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047153699771480335,
      "loss": 5.4092,
      "mean_token_accuracy": 0.2020666107535362,
      "num_tokens": 35737076.0,
      "step": 19165
    },
    {
      "entropy": 5.824070167541504,
      "epoch": 1.6928647121158602,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000471521664838074,
      "loss": 5.2314,
      "mean_token_accuracy": 0.21509755849838258,
      "num_tokens": 35746436.0,
      "step": 19170
    },
    {
      "entropy": 5.830689811706543,
      "epoch": 1.6933062522077005,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004715063281115971,
      "loss": 5.4214,
      "mean_token_accuracy": 0.20655142813920974,
      "num_tokens": 35755786.0,
      "step": 19175
    },
    {
      "entropy": 5.871500778198242,
      "epoch": 1.6937477922995408,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000471490987535673,
      "loss": 5.398,
      "mean_token_accuracy": 0.20298216193914415,
      "num_tokens": 35764424.0,
      "step": 19180
    },
    {
      "entropy": 5.887312316894532,
      "epoch": 1.6941893323913813,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047147564311060216,
      "loss": 5.3626,
      "mean_token_accuracy": 0.20557391196489333,
      "num_tokens": 35773792.0,
      "step": 19185
    },
    {
      "entropy": 5.825941038131714,
      "epoch": 1.6946308724832215,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047146029483668517,
      "loss": 5.3065,
      "mean_token_accuracy": 0.20931385159492494,
      "num_tokens": 35783285.0,
      "step": 19190
    },
    {
      "entropy": 5.8302113056182865,
      "epoch": 1.6950724125750618,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004714449427142228,
      "loss": 5.4321,
      "mean_token_accuracy": 0.19831449091434478,
      "num_tokens": 35792476.0,
      "step": 19195
    },
    {
      "entropy": 5.861258459091187,
      "epoch": 1.6955139526669023,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047142958674351563,
      "loss": 5.3686,
      "mean_token_accuracy": 0.20720222294330598,
      "num_tokens": 35801640.0,
      "step": 19200
    },
    {
      "entropy": 5.8794529914855955,
      "epoch": 1.6959554927587424,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047141422692486466,
      "loss": 5.3574,
      "mean_token_accuracy": 0.21005713939666748,
      "num_tokens": 35811073.0,
      "step": 19205
    },
    {
      "entropy": 5.860421895980835,
      "epoch": 1.6963970328505829,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047139886325857074,
      "loss": 5.3364,
      "mean_token_accuracy": 0.21036551147699356,
      "num_tokens": 35819814.0,
      "step": 19210
    },
    {
      "entropy": 5.8425548553466795,
      "epoch": 1.6968385729424231,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004713834957449347,
      "loss": 5.3207,
      "mean_token_accuracy": 0.20893237739801407,
      "num_tokens": 35828363.0,
      "step": 19215
    },
    {
      "entropy": 5.867175626754761,
      "epoch": 1.6972801130342634,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004713681243842577,
      "loss": 5.2849,
      "mean_token_accuracy": 0.20875922739505767,
      "num_tokens": 35836868.0,
      "step": 19220
    },
    {
      "entropy": 5.859682321548462,
      "epoch": 1.697721653126104,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047135274917684087,
      "loss": 5.3439,
      "mean_token_accuracy": 0.2039727121591568,
      "num_tokens": 35846030.0,
      "step": 19225
    },
    {
      "entropy": 5.782294034957886,
      "epoch": 1.6981631932179442,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004713373701229852,
      "loss": 5.2554,
      "mean_token_accuracy": 0.2128818228840828,
      "num_tokens": 35854862.0,
      "step": 19230
    },
    {
      "entropy": 5.820282459259033,
      "epoch": 1.6986047333097845,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004713219872229923,
      "loss": 5.3034,
      "mean_token_accuracy": 0.21242024898529052,
      "num_tokens": 35864238.0,
      "step": 19235
    },
    {
      "entropy": 5.845459508895874,
      "epoch": 1.699046273401625,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004713066004771633,
      "loss": 5.3233,
      "mean_token_accuracy": 0.20124825537204744,
      "num_tokens": 35872542.0,
      "step": 19240
    },
    {
      "entropy": 5.882920169830323,
      "epoch": 1.6994878134934652,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047129120988579953,
      "loss": 5.4001,
      "mean_token_accuracy": 0.20187266319990158,
      "num_tokens": 35881899.0,
      "step": 19245
    },
    {
      "entropy": 5.870782089233399,
      "epoch": 1.6999293535853055,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047127581544920263,
      "loss": 5.3855,
      "mean_token_accuracy": 0.19976069480180741,
      "num_tokens": 35891970.0,
      "step": 19250
    },
    {
      "entropy": 5.813725709915161,
      "epoch": 1.700370893677146,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004712604171676742,
      "loss": 5.2545,
      "mean_token_accuracy": 0.20687177926301956,
      "num_tokens": 35901286.0,
      "step": 19255
    },
    {
      "entropy": 5.792936611175537,
      "epoch": 1.7008124337689863,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047124501504151574,
      "loss": 5.478,
      "mean_token_accuracy": 0.19672661423683166,
      "num_tokens": 35912301.0,
      "step": 19260
    },
    {
      "entropy": 5.8657081604003904,
      "epoch": 1.7012539738608266,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047122960907102906,
      "loss": 5.3591,
      "mean_token_accuracy": 0.2061811089515686,
      "num_tokens": 35920767.0,
      "step": 19265
    },
    {
      "entropy": 5.921802234649658,
      "epoch": 1.701695513952667,
      "grad_norm": 1.03125,
      "learning_rate": 0.000471214199256516,
      "loss": 5.4451,
      "mean_token_accuracy": 0.20307040959596634,
      "num_tokens": 35930235.0,
      "step": 19270
    },
    {
      "entropy": 5.912538194656372,
      "epoch": 1.702137054044507,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047119878559827825,
      "loss": 5.3888,
      "mean_token_accuracy": 0.1973385512828827,
      "num_tokens": 35939213.0,
      "step": 19275
    },
    {
      "entropy": 5.912663698196411,
      "epoch": 1.7025785941363476,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047118336809661793,
      "loss": 5.4631,
      "mean_token_accuracy": 0.1904297724366188,
      "num_tokens": 35948221.0,
      "step": 19280
    },
    {
      "entropy": 5.872070264816284,
      "epoch": 1.7030201342281879,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004711679467518371,
      "loss": 5.3659,
      "mean_token_accuracy": 0.1964646816253662,
      "num_tokens": 35957961.0,
      "step": 19285
    },
    {
      "entropy": 5.861671447753906,
      "epoch": 1.7034616743200282,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047115252156423767,
      "loss": 5.4075,
      "mean_token_accuracy": 0.19784276336431503,
      "num_tokens": 35966742.0,
      "step": 19290
    },
    {
      "entropy": 5.852165079116821,
      "epoch": 1.7039032144118686,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004711370925341219,
      "loss": 5.3424,
      "mean_token_accuracy": 0.19969536662101744,
      "num_tokens": 35976312.0,
      "step": 19295
    },
    {
      "entropy": 5.88031039237976,
      "epoch": 1.704344754503709,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047112165966179206,
      "loss": 5.4497,
      "mean_token_accuracy": 0.19996579140424728,
      "num_tokens": 35985537.0,
      "step": 19300
    },
    {
      "entropy": 5.875399875640869,
      "epoch": 1.7047862945955492,
      "grad_norm": 1.0,
      "learning_rate": 0.00047110622294755045,
      "loss": 5.352,
      "mean_token_accuracy": 0.20590488314628602,
      "num_tokens": 35995004.0,
      "step": 19305
    },
    {
      "entropy": 5.860301446914673,
      "epoch": 1.7052278346873897,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047109078239169946,
      "loss": 5.3819,
      "mean_token_accuracy": 0.19609652310609818,
      "num_tokens": 36004786.0,
      "step": 19310
    },
    {
      "entropy": 5.814460706710816,
      "epoch": 1.70566937477923,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004710753379945415,
      "loss": 5.3075,
      "mean_token_accuracy": 0.2170813649892807,
      "num_tokens": 36014047.0,
      "step": 19315
    },
    {
      "entropy": 5.8558917999267575,
      "epoch": 1.7061109148710702,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047105988975637927,
      "loss": 5.3342,
      "mean_token_accuracy": 0.21030336767435073,
      "num_tokens": 36022921.0,
      "step": 19320
    },
    {
      "entropy": 5.8279869079589846,
      "epoch": 1.7065524549629107,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004710444376775153,
      "loss": 5.3297,
      "mean_token_accuracy": 0.20479645133018493,
      "num_tokens": 36032713.0,
      "step": 19325
    },
    {
      "entropy": 5.837489318847656,
      "epoch": 1.706993995054751,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047102898175825224,
      "loss": 5.3658,
      "mean_token_accuracy": 0.20159162431955338,
      "num_tokens": 36041595.0,
      "step": 19330
    },
    {
      "entropy": 5.861054944992065,
      "epoch": 1.7074355351465913,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047101352199889295,
      "loss": 5.3502,
      "mean_token_accuracy": 0.19800026714801788,
      "num_tokens": 36051993.0,
      "step": 19335
    },
    {
      "entropy": 5.784552431106567,
      "epoch": 1.7078770752384318,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004709980583997402,
      "loss": 5.3483,
      "mean_token_accuracy": 0.20705602020025254,
      "num_tokens": 36061806.0,
      "step": 19340
    },
    {
      "entropy": 5.8445803165435795,
      "epoch": 1.7083186153302719,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000470982590961097,
      "loss": 5.3988,
      "mean_token_accuracy": 0.20039329826831817,
      "num_tokens": 36071390.0,
      "step": 19345
    },
    {
      "entropy": 5.84760890007019,
      "epoch": 1.7087601554221123,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047096711968326626,
      "loss": 5.3843,
      "mean_token_accuracy": 0.20336378067731858,
      "num_tokens": 36080561.0,
      "step": 19350
    },
    {
      "entropy": 5.865473937988281,
      "epoch": 1.7092016955139526,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004709516445665512,
      "loss": 5.3233,
      "mean_token_accuracy": 0.20303022712469102,
      "num_tokens": 36089568.0,
      "step": 19355
    },
    {
      "entropy": 5.825430583953858,
      "epoch": 1.709643235605793,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004709361656112547,
      "loss": 5.2948,
      "mean_token_accuracy": 0.213546322286129,
      "num_tokens": 36098881.0,
      "step": 19360
    },
    {
      "entropy": 5.835883045196534,
      "epoch": 1.7100847756976334,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004709206828176804,
      "loss": 5.3588,
      "mean_token_accuracy": 0.21046178936958312,
      "num_tokens": 36108209.0,
      "step": 19365
    },
    {
      "entropy": 5.8493904113769535,
      "epoch": 1.7105263157894737,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004709051961861311,
      "loss": 5.3865,
      "mean_token_accuracy": 0.20557094365358353,
      "num_tokens": 36117573.0,
      "step": 19370
    },
    {
      "entropy": 5.861238622665406,
      "epoch": 1.710967855881314,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004708897057169105,
      "loss": 5.3747,
      "mean_token_accuracy": 0.20894356369972228,
      "num_tokens": 36127405.0,
      "step": 19375
    },
    {
      "entropy": 5.935452270507812,
      "epoch": 1.7114093959731544,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004708742114103221,
      "loss": 5.3765,
      "mean_token_accuracy": 0.2082043021917343,
      "num_tokens": 36136148.0,
      "step": 19380
    },
    {
      "entropy": 5.837888240814209,
      "epoch": 1.7118509360649947,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004708587132666692,
      "loss": 5.375,
      "mean_token_accuracy": 0.20259551852941513,
      "num_tokens": 36146164.0,
      "step": 19385
    },
    {
      "entropy": 5.863293027877807,
      "epoch": 1.712292476156835,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047084321128625545,
      "loss": 5.3689,
      "mean_token_accuracy": 0.1976170539855957,
      "num_tokens": 36155313.0,
      "step": 19390
    },
    {
      "entropy": 5.822325849533081,
      "epoch": 1.7127340162486755,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047082770546938457,
      "loss": 5.3739,
      "mean_token_accuracy": 0.2107451856136322,
      "num_tokens": 36164902.0,
      "step": 19395
    },
    {
      "entropy": 5.869547176361084,
      "epoch": 1.7131755563405158,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004708121958163604,
      "loss": 5.3986,
      "mean_token_accuracy": 0.21294075548648833,
      "num_tokens": 36174128.0,
      "step": 19400
    },
    {
      "entropy": 5.898516464233398,
      "epoch": 1.713617096432356,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004707966823274866,
      "loss": 5.4469,
      "mean_token_accuracy": 0.19630146622657776,
      "num_tokens": 36183084.0,
      "step": 19405
    },
    {
      "entropy": 5.932181167602539,
      "epoch": 1.7140586365241965,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047078116500306723,
      "loss": 5.4008,
      "mean_token_accuracy": 0.20068621933460234,
      "num_tokens": 36191650.0,
      "step": 19410
    },
    {
      "entropy": 5.870257472991943,
      "epoch": 1.7145001766160366,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047076564384340616,
      "loss": 5.3869,
      "mean_token_accuracy": 0.2046503245830536,
      "num_tokens": 36200712.0,
      "step": 19415
    },
    {
      "entropy": 5.818039178848267,
      "epoch": 1.714941716707877,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004707501188488074,
      "loss": 5.3581,
      "mean_token_accuracy": 0.2052019789814949,
      "num_tokens": 36209347.0,
      "step": 19420
    },
    {
      "entropy": 5.869766092300415,
      "epoch": 1.7153832567997174,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047073459001957524,
      "loss": 5.3546,
      "mean_token_accuracy": 0.20100276470184325,
      "num_tokens": 36218233.0,
      "step": 19425
    },
    {
      "entropy": 5.930176305770874,
      "epoch": 1.7158247968915576,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004707190573560137,
      "loss": 5.4033,
      "mean_token_accuracy": 0.2007612481713295,
      "num_tokens": 36227627.0,
      "step": 19430
    },
    {
      "entropy": 5.843554735183716,
      "epoch": 1.7162663369833981,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004707035208584272,
      "loss": 5.3669,
      "mean_token_accuracy": 0.2091110333800316,
      "num_tokens": 36236298.0,
      "step": 19435
    },
    {
      "entropy": 5.8809840202331545,
      "epoch": 1.7167078770752384,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047068798052711996,
      "loss": 5.4268,
      "mean_token_accuracy": 0.20471113473176955,
      "num_tokens": 36245623.0,
      "step": 19440
    },
    {
      "entropy": 5.8725379467010494,
      "epoch": 1.7171494171670787,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004706724363623965,
      "loss": 5.2879,
      "mean_token_accuracy": 0.21204689890146255,
      "num_tokens": 36255117.0,
      "step": 19445
    },
    {
      "entropy": 5.809520244598389,
      "epoch": 1.7175909572589192,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047065688836456134,
      "loss": 5.3386,
      "mean_token_accuracy": 0.21131264716386794,
      "num_tokens": 36264535.0,
      "step": 19450
    },
    {
      "entropy": 5.877058744430542,
      "epoch": 1.7180324973507595,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047064133653391895,
      "loss": 5.3843,
      "mean_token_accuracy": 0.20071642249822616,
      "num_tokens": 36274432.0,
      "step": 19455
    },
    {
      "entropy": 5.84521894454956,
      "epoch": 1.7184740374425997,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004706257808707741,
      "loss": 5.2955,
      "mean_token_accuracy": 0.21847237795591354,
      "num_tokens": 36282644.0,
      "step": 19460
    },
    {
      "entropy": 5.839371538162231,
      "epoch": 1.7189155775344402,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047061022137543144,
      "loss": 5.3538,
      "mean_token_accuracy": 0.20643906444311141,
      "num_tokens": 36290885.0,
      "step": 19465
    },
    {
      "entropy": 5.913707494735718,
      "epoch": 1.7193571176262805,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047059465804819576,
      "loss": 5.4586,
      "mean_token_accuracy": 0.20248591303825378,
      "num_tokens": 36299477.0,
      "step": 19470
    },
    {
      "entropy": 5.84574408531189,
      "epoch": 1.7197986577181208,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004705790908893721,
      "loss": 5.4272,
      "mean_token_accuracy": 0.20245751589536667,
      "num_tokens": 36309638.0,
      "step": 19475
    },
    {
      "entropy": 5.9297737121582035,
      "epoch": 1.7202401978099613,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004705635198992651,
      "loss": 5.4782,
      "mean_token_accuracy": 0.19718853384256363,
      "num_tokens": 36319027.0,
      "step": 19480
    },
    {
      "entropy": 5.856172370910644,
      "epoch": 1.7206817379018013,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047054794507818,
      "loss": 5.3223,
      "mean_token_accuracy": 0.19970888942480086,
      "num_tokens": 36327936.0,
      "step": 19485
    },
    {
      "entropy": 5.900039005279541,
      "epoch": 1.7211232779936418,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047053236642642193,
      "loss": 5.5262,
      "mean_token_accuracy": 0.19295619279146195,
      "num_tokens": 36337162.0,
      "step": 19490
    },
    {
      "entropy": 5.8792760372161865,
      "epoch": 1.7215648180854821,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000470516783944296,
      "loss": 5.3747,
      "mean_token_accuracy": 0.2089792400598526,
      "num_tokens": 36345587.0,
      "step": 19495
    },
    {
      "entropy": 5.868822526931763,
      "epoch": 1.7220063581773224,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047050119763210747,
      "loss": 5.4098,
      "mean_token_accuracy": 0.2047387808561325,
      "num_tokens": 36354536.0,
      "step": 19500
    },
    {
      "entropy": 5.872997760772705,
      "epoch": 1.722447898269163,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047048560749016167,
      "loss": 5.4632,
      "mean_token_accuracy": 0.19283899962902068,
      "num_tokens": 36363691.0,
      "step": 19505
    },
    {
      "entropy": 5.9692363262176515,
      "epoch": 1.7228894383610032,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047047001351876403,
      "loss": 5.4985,
      "mean_token_accuracy": 0.19898141473531722,
      "num_tokens": 36373898.0,
      "step": 19510
    },
    {
      "entropy": 5.9161478042602536,
      "epoch": 1.7233309784528434,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004704544157182199,
      "loss": 5.4788,
      "mean_token_accuracy": 0.19759699404239656,
      "num_tokens": 36384345.0,
      "step": 19515
    },
    {
      "entropy": 5.849001693725586,
      "epoch": 1.723772518544684,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047043881408883497,
      "loss": 5.361,
      "mean_token_accuracy": 0.2048543244600296,
      "num_tokens": 36393021.0,
      "step": 19520
    },
    {
      "entropy": 5.940569639205933,
      "epoch": 1.7242140586365242,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004704232086309148,
      "loss": 5.357,
      "mean_token_accuracy": 0.20328376740217208,
      "num_tokens": 36402729.0,
      "step": 19525
    },
    {
      "entropy": 5.796016836166382,
      "epoch": 1.7246555987283645,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047040759934476517,
      "loss": 5.2884,
      "mean_token_accuracy": 0.20681367069482803,
      "num_tokens": 36412360.0,
      "step": 19530
    },
    {
      "entropy": 5.779270648956299,
      "epoch": 1.725097138820205,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004703919862306918,
      "loss": 5.3021,
      "mean_token_accuracy": 0.20544631779193878,
      "num_tokens": 36420245.0,
      "step": 19535
    },
    {
      "entropy": 5.821827411651611,
      "epoch": 1.7255386789120453,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004703763692890006,
      "loss": 5.2665,
      "mean_token_accuracy": 0.21545811444520951,
      "num_tokens": 36428758.0,
      "step": 19540
    },
    {
      "entropy": 5.86518120765686,
      "epoch": 1.7259802190038855,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004703607485199973,
      "loss": 5.3507,
      "mean_token_accuracy": 0.20525264889001846,
      "num_tokens": 36437883.0,
      "step": 19545
    },
    {
      "entropy": 5.83111162185669,
      "epoch": 1.726421759095726,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047034512392398815,
      "loss": 5.3521,
      "mean_token_accuracy": 0.20012767761945724,
      "num_tokens": 36447950.0,
      "step": 19550
    },
    {
      "entropy": 5.891711902618408,
      "epoch": 1.726863299187566,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004703294955012791,
      "loss": 5.3968,
      "mean_token_accuracy": 0.20180790722370148,
      "num_tokens": 36456993.0,
      "step": 19555
    },
    {
      "entropy": 5.908048009872436,
      "epoch": 1.7273048392794066,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004703138632521762,
      "loss": 5.4438,
      "mean_token_accuracy": 0.1943171054124832,
      "num_tokens": 36465942.0,
      "step": 19560
    },
    {
      "entropy": 5.8576130867004395,
      "epoch": 1.7277463793712469,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000470298227176986,
      "loss": 5.4374,
      "mean_token_accuracy": 0.19999451488256453,
      "num_tokens": 36474906.0,
      "step": 19565
    },
    {
      "entropy": 5.904548978805542,
      "epoch": 1.7281879194630871,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004702825872760145,
      "loss": 5.3558,
      "mean_token_accuracy": 0.20032448470592498,
      "num_tokens": 36483579.0,
      "step": 19570
    },
    {
      "entropy": 5.862172698974609,
      "epoch": 1.7286294595549276,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004702669435495681,
      "loss": 5.3593,
      "mean_token_accuracy": 0.20729408860206605,
      "num_tokens": 36492541.0,
      "step": 19575
    },
    {
      "entropy": 5.898200178146363,
      "epoch": 1.729070999646768,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004702512959979535,
      "loss": 5.4942,
      "mean_token_accuracy": 0.19613170325756074,
      "num_tokens": 36502833.0,
      "step": 19580
    },
    {
      "entropy": 5.887970066070556,
      "epoch": 1.7295125397386082,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047023564462147697,
      "loss": 5.3655,
      "mean_token_accuracy": 0.2045273721218109,
      "num_tokens": 36513091.0,
      "step": 19585
    },
    {
      "entropy": 5.830621767044067,
      "epoch": 1.7299540798304487,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047021998942044517,
      "loss": 5.3695,
      "mean_token_accuracy": 0.20839569270610808,
      "num_tokens": 36522911.0,
      "step": 19590
    },
    {
      "entropy": 5.887117099761963,
      "epoch": 1.730395619922289,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047020433039516484,
      "loss": 5.4355,
      "mean_token_accuracy": 0.20266944468021392,
      "num_tokens": 36531934.0,
      "step": 19595
    },
    {
      "entropy": 5.859900617599488,
      "epoch": 1.7308371600141292,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004701886675459427,
      "loss": 5.3646,
      "mean_token_accuracy": 0.20867136120796204,
      "num_tokens": 36541823.0,
      "step": 19600
    },
    {
      "entropy": 5.910891008377075,
      "epoch": 1.7312787001059697,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047017300087308557,
      "loss": 5.4687,
      "mean_token_accuracy": 0.1963314786553383,
      "num_tokens": 36550954.0,
      "step": 19605
    },
    {
      "entropy": 5.871105909347534,
      "epoch": 1.73172024019781,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004701573303769004,
      "loss": 5.3638,
      "mean_token_accuracy": 0.20587648451328278,
      "num_tokens": 36560689.0,
      "step": 19610
    },
    {
      "entropy": 5.8392587184906,
      "epoch": 1.7321617802896503,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000470141656057694,
      "loss": 5.3521,
      "mean_token_accuracy": 0.2004813715815544,
      "num_tokens": 36569587.0,
      "step": 19615
    },
    {
      "entropy": 5.83994312286377,
      "epoch": 1.7326033203814908,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00047012597791577364,
      "loss": 5.3763,
      "mean_token_accuracy": 0.2069721922278404,
      "num_tokens": 36579609.0,
      "step": 19620
    },
    {
      "entropy": 5.857770109176636,
      "epoch": 1.7330448604733308,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004701102959514463,
      "loss": 5.3849,
      "mean_token_accuracy": 0.20199847519397734,
      "num_tokens": 36589685.0,
      "step": 19625
    },
    {
      "entropy": 5.89018497467041,
      "epoch": 1.7334864005651713,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047009461016501926,
      "loss": 5.339,
      "mean_token_accuracy": 0.19970999509096146,
      "num_tokens": 36598225.0,
      "step": 19630
    },
    {
      "entropy": 5.759757232666016,
      "epoch": 1.7339279406570116,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047007892055679976,
      "loss": 5.2741,
      "mean_token_accuracy": 0.2128346160054207,
      "num_tokens": 36607481.0,
      "step": 19635
    },
    {
      "entropy": 5.864145040512085,
      "epoch": 1.7343694807488519,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047006322712709513,
      "loss": 5.3642,
      "mean_token_accuracy": 0.2028595194220543,
      "num_tokens": 36616707.0,
      "step": 19640
    },
    {
      "entropy": 5.926380443572998,
      "epoch": 1.7348110208406924,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047004752987621286,
      "loss": 5.4997,
      "mean_token_accuracy": 0.19786589741706848,
      "num_tokens": 36626290.0,
      "step": 19645
    },
    {
      "entropy": 5.85288290977478,
      "epoch": 1.7352525609325327,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004700318288044603,
      "loss": 5.3845,
      "mean_token_accuracy": 0.20945847332477568,
      "num_tokens": 36635155.0,
      "step": 19650
    },
    {
      "entropy": 5.878585720062256,
      "epoch": 1.735694101024373,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047001612391214527,
      "loss": 5.3789,
      "mean_token_accuracy": 0.20974762439727784,
      "num_tokens": 36644632.0,
      "step": 19655
    },
    {
      "entropy": 5.898300361633301,
      "epoch": 1.7361356411162134,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004700004151995753,
      "loss": 5.4212,
      "mean_token_accuracy": 0.19723953157663346,
      "num_tokens": 36654818.0,
      "step": 19660
    },
    {
      "entropy": 5.8098838329315186,
      "epoch": 1.7365771812080537,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046998470266705795,
      "loss": 5.2586,
      "mean_token_accuracy": 0.2128022387623787,
      "num_tokens": 36664149.0,
      "step": 19665
    },
    {
      "entropy": 5.85724892616272,
      "epoch": 1.737018721299894,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046996898631490125,
      "loss": 5.3207,
      "mean_token_accuracy": 0.21209775507450104,
      "num_tokens": 36673447.0,
      "step": 19670
    },
    {
      "entropy": 5.989524602890015,
      "epoch": 1.7374602613917345,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000469953266143413,
      "loss": 5.5064,
      "mean_token_accuracy": 0.1909137487411499,
      "num_tokens": 36683426.0,
      "step": 19675
    },
    {
      "entropy": 5.9506079196929935,
      "epoch": 1.7379018014835748,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046993754215290116,
      "loss": 5.3937,
      "mean_token_accuracy": 0.2005843475461006,
      "num_tokens": 36692708.0,
      "step": 19680
    },
    {
      "entropy": 5.7777101516723635,
      "epoch": 1.738343341575415,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004699218143436737,
      "loss": 5.2959,
      "mean_token_accuracy": 0.20974681824445723,
      "num_tokens": 36701527.0,
      "step": 19685
    },
    {
      "entropy": 5.869098091125489,
      "epoch": 1.7387848816672555,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046990608271603874,
      "loss": 5.4436,
      "mean_token_accuracy": 0.201623073220253,
      "num_tokens": 36711458.0,
      "step": 19690
    },
    {
      "entropy": 5.8512145519256595,
      "epoch": 1.7392264217590956,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004698903472703045,
      "loss": 5.3629,
      "mean_token_accuracy": 0.20296697765588761,
      "num_tokens": 36721260.0,
      "step": 19695
    },
    {
      "entropy": 5.909906721115112,
      "epoch": 1.739667961850936,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046987460800677915,
      "loss": 5.3755,
      "mean_token_accuracy": 0.2055405154824257,
      "num_tokens": 36730496.0,
      "step": 19700
    },
    {
      "entropy": 5.840519809722901,
      "epoch": 1.7401095019427764,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046985886492577106,
      "loss": 5.337,
      "mean_token_accuracy": 0.20332951694726945,
      "num_tokens": 36739016.0,
      "step": 19705
    },
    {
      "entropy": 5.922931909561157,
      "epoch": 1.7405510420346166,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004698431180275887,
      "loss": 5.4611,
      "mean_token_accuracy": 0.19510802030563354,
      "num_tokens": 36748259.0,
      "step": 19710
    },
    {
      "entropy": 6.0068436622619625,
      "epoch": 1.7409925821264571,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046982736731254044,
      "loss": 5.4818,
      "mean_token_accuracy": 0.19545356631278993,
      "num_tokens": 36758174.0,
      "step": 19715
    },
    {
      "entropy": 5.9150248050689695,
      "epoch": 1.7414341222182974,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046981161278093487,
      "loss": 5.4696,
      "mean_token_accuracy": 0.19996329993009568,
      "num_tokens": 36767286.0,
      "step": 19720
    },
    {
      "entropy": 5.8429395198822025,
      "epoch": 1.7418756623101377,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046979585443308056,
      "loss": 5.3046,
      "mean_token_accuracy": 0.2088826209306717,
      "num_tokens": 36775818.0,
      "step": 19725
    },
    {
      "entropy": 5.806203556060791,
      "epoch": 1.7423172024019782,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046978009226928633,
      "loss": 5.3612,
      "mean_token_accuracy": 0.20912386924028398,
      "num_tokens": 36785489.0,
      "step": 19730
    },
    {
      "entropy": 5.8916933059692385,
      "epoch": 1.7427587424938185,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004697643262898608,
      "loss": 5.379,
      "mean_token_accuracy": 0.20245468020439147,
      "num_tokens": 36794241.0,
      "step": 19735
    },
    {
      "entropy": 5.917622709274292,
      "epoch": 1.7432002825856587,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004697485564951128,
      "loss": 5.3809,
      "mean_token_accuracy": 0.2006335362792015,
      "num_tokens": 36803477.0,
      "step": 19740
    },
    {
      "entropy": 5.871290874481201,
      "epoch": 1.7436418226774992,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004697327828853515,
      "loss": 5.3756,
      "mean_token_accuracy": 0.20136721283197404,
      "num_tokens": 36811974.0,
      "step": 19745
    },
    {
      "entropy": 5.881715536117554,
      "epoch": 1.7440833627693395,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004697170054608857,
      "loss": 5.4335,
      "mean_token_accuracy": 0.20168333500623703,
      "num_tokens": 36822049.0,
      "step": 19750
    },
    {
      "entropy": 5.893443298339844,
      "epoch": 1.7445249028611798,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004697012242220245,
      "loss": 5.374,
      "mean_token_accuracy": 0.20690135806798934,
      "num_tokens": 36831989.0,
      "step": 19755
    },
    {
      "entropy": 5.902186012268066,
      "epoch": 1.7449664429530203,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046968543916907704,
      "loss": 5.409,
      "mean_token_accuracy": 0.1928035080432892,
      "num_tokens": 36841141.0,
      "step": 19760
    },
    {
      "entropy": 5.828170442581177,
      "epoch": 1.7454079830448603,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046966965030235253,
      "loss": 5.2616,
      "mean_token_accuracy": 0.2130754292011261,
      "num_tokens": 36850101.0,
      "step": 19765
    },
    {
      "entropy": 5.850824928283691,
      "epoch": 1.7458495231367008,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046965385762216024,
      "loss": 5.3773,
      "mean_token_accuracy": 0.2056165024638176,
      "num_tokens": 36859234.0,
      "step": 19770
    },
    {
      "entropy": 5.952357816696167,
      "epoch": 1.746291063228541,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004696380611288096,
      "loss": 5.4591,
      "mean_token_accuracy": 0.20363277941942215,
      "num_tokens": 36868937.0,
      "step": 19775
    },
    {
      "entropy": 5.823459577560425,
      "epoch": 1.7467326033203814,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046962226082261014,
      "loss": 5.2983,
      "mean_token_accuracy": 0.20970505475997925,
      "num_tokens": 36878380.0,
      "step": 19780
    },
    {
      "entropy": 5.855659341812133,
      "epoch": 1.7471741434122219,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004696064567038712,
      "loss": 5.4302,
      "mean_token_accuracy": 0.2020963564515114,
      "num_tokens": 36887702.0,
      "step": 19785
    },
    {
      "entropy": 5.877581596374512,
      "epoch": 1.7476156835040622,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046959064877290246,
      "loss": 5.2773,
      "mean_token_accuracy": 0.2190902128815651,
      "num_tokens": 36896777.0,
      "step": 19790
    },
    {
      "entropy": 5.896066093444825,
      "epoch": 1.7480572235959024,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046957483703001366,
      "loss": 5.3274,
      "mean_token_accuracy": 0.20655572563409805,
      "num_tokens": 36905256.0,
      "step": 19795
    },
    {
      "entropy": 5.838788747787476,
      "epoch": 1.748498763687743,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004695590214755143,
      "loss": 5.3537,
      "mean_token_accuracy": 0.20737936347723007,
      "num_tokens": 36913767.0,
      "step": 19800
    },
    {
      "entropy": 5.82216157913208,
      "epoch": 1.7489403037795832,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046954320210971444,
      "loss": 5.3378,
      "mean_token_accuracy": 0.21220385134220124,
      "num_tokens": 36922838.0,
      "step": 19805
    },
    {
      "entropy": 5.869567346572876,
      "epoch": 1.7493818438714235,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046952737893292396,
      "loss": 5.2205,
      "mean_token_accuracy": 0.21445235013961791,
      "num_tokens": 36931955.0,
      "step": 19810
    },
    {
      "entropy": 5.818287324905396,
      "epoch": 1.749823383963264,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046951155194545265,
      "loss": 5.3198,
      "mean_token_accuracy": 0.20727813988924026,
      "num_tokens": 36941505.0,
      "step": 19815
    },
    {
      "entropy": 5.858857727050781,
      "epoch": 1.7502649240551043,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046949572114761076,
      "loss": 5.4319,
      "mean_token_accuracy": 0.19525412917137147,
      "num_tokens": 36951007.0,
      "step": 19820
    },
    {
      "entropy": 5.891967630386352,
      "epoch": 1.7507064641469445,
      "grad_norm": 0.9375,
      "learning_rate": 0.00046947988653970826,
      "loss": 5.3919,
      "mean_token_accuracy": 0.21053603440523147,
      "num_tokens": 36961111.0,
      "step": 19825
    },
    {
      "entropy": 5.883198499679565,
      "epoch": 1.751148004238785,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004694640481220554,
      "loss": 5.4018,
      "mean_token_accuracy": 0.20545209646224977,
      "num_tokens": 36969605.0,
      "step": 19830
    },
    {
      "entropy": 5.902576494216919,
      "epoch": 1.751589544330625,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004694482058949624,
      "loss": 5.3924,
      "mean_token_accuracy": 0.1992309421300888,
      "num_tokens": 36979051.0,
      "step": 19835
    },
    {
      "entropy": 5.822027969360351,
      "epoch": 1.7520310844224656,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046943235985873975,
      "loss": 5.3414,
      "mean_token_accuracy": 0.20949007272720338,
      "num_tokens": 36988807.0,
      "step": 19840
    },
    {
      "entropy": 5.864878177642822,
      "epoch": 1.7524726245143059,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004694165100136976,
      "loss": 5.2909,
      "mean_token_accuracy": 0.21140781044960022,
      "num_tokens": 36997813.0,
      "step": 19845
    },
    {
      "entropy": 5.89201397895813,
      "epoch": 1.7529141646061461,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046940065636014665,
      "loss": 5.3421,
      "mean_token_accuracy": 0.20457070618867873,
      "num_tokens": 37007849.0,
      "step": 19850
    },
    {
      "entropy": 5.815494537353516,
      "epoch": 1.7533557046979866,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004693847988983974,
      "loss": 5.2468,
      "mean_token_accuracy": 0.21503526866436004,
      "num_tokens": 37016109.0,
      "step": 19855
    },
    {
      "entropy": 5.831973314285278,
      "epoch": 1.753797244789827,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004693689376287605,
      "loss": 5.3692,
      "mean_token_accuracy": 0.19793687313795089,
      "num_tokens": 37025156.0,
      "step": 19860
    },
    {
      "entropy": 5.887754487991333,
      "epoch": 1.7542387848816672,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046935307255154663,
      "loss": 5.4666,
      "mean_token_accuracy": 0.19373612403869628,
      "num_tokens": 37035361.0,
      "step": 19865
    },
    {
      "entropy": 5.93493971824646,
      "epoch": 1.7546803249735077,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004693372036670667,
      "loss": 5.3973,
      "mean_token_accuracy": 0.1965683400630951,
      "num_tokens": 37045249.0,
      "step": 19870
    },
    {
      "entropy": 5.965833044052124,
      "epoch": 1.755121865065348,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046932133097563137,
      "loss": 5.4105,
      "mean_token_accuracy": 0.20748149901628493,
      "num_tokens": 37054637.0,
      "step": 19875
    },
    {
      "entropy": 5.883833026885986,
      "epoch": 1.7555634051571882,
      "grad_norm": 1.0,
      "learning_rate": 0.00046930545447755167,
      "loss": 5.4061,
      "mean_token_accuracy": 0.20418444871902466,
      "num_tokens": 37064557.0,
      "step": 19880
    },
    {
      "entropy": 5.870358991622925,
      "epoch": 1.7560049452490287,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004692895741731387,
      "loss": 5.4176,
      "mean_token_accuracy": 0.1922706186771393,
      "num_tokens": 37073915.0,
      "step": 19885
    },
    {
      "entropy": 5.920988941192627,
      "epoch": 1.756446485340869,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004692736900627034,
      "loss": 5.3856,
      "mean_token_accuracy": 0.19981521517038345,
      "num_tokens": 37083008.0,
      "step": 19890
    },
    {
      "entropy": 5.937457704544068,
      "epoch": 1.7568880254327093,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046925780214655713,
      "loss": 5.4136,
      "mean_token_accuracy": 0.20175595134496688,
      "num_tokens": 37092742.0,
      "step": 19895
    },
    {
      "entropy": 5.934474658966065,
      "epoch": 1.7573295655245498,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004692419104250108,
      "loss": 5.4582,
      "mean_token_accuracy": 0.19153337329626083,
      "num_tokens": 37102242.0,
      "step": 19900
    },
    {
      "entropy": 5.846590995788574,
      "epoch": 1.7577711056163898,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004692260148983761,
      "loss": 5.3376,
      "mean_token_accuracy": 0.21341789662837982,
      "num_tokens": 37111304.0,
      "step": 19905
    },
    {
      "entropy": 5.873947095870972,
      "epoch": 1.7582126457082303,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004692101155669641,
      "loss": 5.3434,
      "mean_token_accuracy": 0.20788217186927796,
      "num_tokens": 37121420.0,
      "step": 19910
    },
    {
      "entropy": 5.9507365226745605,
      "epoch": 1.7586541858000706,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004691942124310865,
      "loss": 5.4119,
      "mean_token_accuracy": 0.20299457162618637,
      "num_tokens": 37131265.0,
      "step": 19915
    },
    {
      "entropy": 5.830027723312378,
      "epoch": 1.7590957258919109,
      "grad_norm": 0.953125,
      "learning_rate": 0.00046917830549105465,
      "loss": 5.3081,
      "mean_token_accuracy": 0.21015240848064423,
      "num_tokens": 37141097.0,
      "step": 19920
    },
    {
      "entropy": 5.721772575378418,
      "epoch": 1.7595372659837514,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004691623947471803,
      "loss": 5.2226,
      "mean_token_accuracy": 0.21801266819238663,
      "num_tokens": 37149687.0,
      "step": 19925
    },
    {
      "entropy": 5.851511716842651,
      "epoch": 1.7599788060755917,
      "grad_norm": 1.0,
      "learning_rate": 0.00046914648019977495,
      "loss": 5.3536,
      "mean_token_accuracy": 0.2111783504486084,
      "num_tokens": 37159219.0,
      "step": 19930
    },
    {
      "entropy": 5.841753673553467,
      "epoch": 1.760420346167432,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046913056184915054,
      "loss": 5.3601,
      "mean_token_accuracy": 0.20732667595148085,
      "num_tokens": 37168681.0,
      "step": 19935
    },
    {
      "entropy": 5.7493514060974125,
      "epoch": 1.7608618862592724,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004691146396956188,
      "loss": 5.2552,
      "mean_token_accuracy": 0.21570795625448227,
      "num_tokens": 37177932.0,
      "step": 19940
    },
    {
      "entropy": 5.844285154342652,
      "epoch": 1.7613034263511127,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004690987137394917,
      "loss": 5.2732,
      "mean_token_accuracy": 0.20587903410196304,
      "num_tokens": 37186783.0,
      "step": 19945
    },
    {
      "entropy": 5.902088165283203,
      "epoch": 1.761744966442953,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004690827839810812,
      "loss": 5.4538,
      "mean_token_accuracy": 0.19667417407035828,
      "num_tokens": 37196290.0,
      "step": 19950
    },
    {
      "entropy": 5.8575023174285885,
      "epoch": 1.7621865065347935,
      "grad_norm": 1.125,
      "learning_rate": 0.00046906685042069927,
      "loss": 5.3296,
      "mean_token_accuracy": 0.20638182163238525,
      "num_tokens": 37205057.0,
      "step": 19955
    },
    {
      "entropy": 5.8945387840271,
      "epoch": 1.7626280466266337,
      "grad_norm": 1.0,
      "learning_rate": 0.0004690509130586582,
      "loss": 5.3757,
      "mean_token_accuracy": 0.20165475010871886,
      "num_tokens": 37214078.0,
      "step": 19960
    },
    {
      "entropy": 5.828520059585571,
      "epoch": 1.763069586718474,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046903497189527006,
      "loss": 5.3298,
      "mean_token_accuracy": 0.2077646389603615,
      "num_tokens": 37223232.0,
      "step": 19965
    },
    {
      "entropy": 5.844703960418701,
      "epoch": 1.7635111268103145,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004690190269308472,
      "loss": 5.3322,
      "mean_token_accuracy": 0.20792658030986785,
      "num_tokens": 37232054.0,
      "step": 19970
    },
    {
      "entropy": 5.786932897567749,
      "epoch": 1.7639526669021546,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004690030781657019,
      "loss": 5.293,
      "mean_token_accuracy": 0.21082770228385925,
      "num_tokens": 37241451.0,
      "step": 19975
    },
    {
      "entropy": 5.85056791305542,
      "epoch": 1.764394206993995,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004689871256001467,
      "loss": 5.3471,
      "mean_token_accuracy": 0.20343182533979415,
      "num_tokens": 37251029.0,
      "step": 19980
    },
    {
      "entropy": 5.8258668899536135,
      "epoch": 1.7648357470858353,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046897116923449403,
      "loss": 5.3484,
      "mean_token_accuracy": 0.2111079141497612,
      "num_tokens": 37259594.0,
      "step": 19985
    },
    {
      "entropy": 5.808517980575561,
      "epoch": 1.7652772871776756,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046895520906905635,
      "loss": 5.2942,
      "mean_token_accuracy": 0.2086445465683937,
      "num_tokens": 37268824.0,
      "step": 19990
    },
    {
      "entropy": 5.873378944396973,
      "epoch": 1.7657188272695161,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046893924510414654,
      "loss": 5.3428,
      "mean_token_accuracy": 0.20504022538661956,
      "num_tokens": 37278494.0,
      "step": 19995
    },
    {
      "entropy": 5.713855457305908,
      "epoch": 1.7661603673613564,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004689232773400772,
      "loss": 5.164,
      "mean_token_accuracy": 0.23042462021112442,
      "num_tokens": 37287648.0,
      "step": 20000
    },
    {
      "entropy": 5.843976974487305,
      "epoch": 1.7666019074531967,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004689073057771611,
      "loss": 5.3266,
      "mean_token_accuracy": 0.2016871228814125,
      "num_tokens": 37295938.0,
      "step": 20005
    },
    {
      "entropy": 5.829558897018432,
      "epoch": 1.7670434475450372,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046889133041571126,
      "loss": 5.3263,
      "mean_token_accuracy": 0.2090665102005005,
      "num_tokens": 37305097.0,
      "step": 20010
    },
    {
      "entropy": 5.809263563156128,
      "epoch": 1.7674849876368774,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004688753512560405,
      "loss": 5.3465,
      "mean_token_accuracy": 0.20816302597522734,
      "num_tokens": 37313672.0,
      "step": 20015
    },
    {
      "entropy": 5.83147964477539,
      "epoch": 1.7679265277287177,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004688593682984619,
      "loss": 5.4079,
      "mean_token_accuracy": 0.20093484371900558,
      "num_tokens": 37323342.0,
      "step": 20020
    },
    {
      "entropy": 5.878684139251709,
      "epoch": 1.7683680678205582,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004688433815432884,
      "loss": 5.3675,
      "mean_token_accuracy": 0.20378257781267167,
      "num_tokens": 37332578.0,
      "step": 20025
    },
    {
      "entropy": 5.928590631484985,
      "epoch": 1.7688096079123985,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046882739099083335,
      "loss": 5.3809,
      "mean_token_accuracy": 0.1996745079755783,
      "num_tokens": 37341769.0,
      "step": 20030
    },
    {
      "entropy": 5.929939413070679,
      "epoch": 1.7692511480042388,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046881139664141006,
      "loss": 5.4369,
      "mean_token_accuracy": 0.19845349639654158,
      "num_tokens": 37351158.0,
      "step": 20035
    },
    {
      "entropy": 5.881866216659546,
      "epoch": 1.7696926880960793,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004687953984953316,
      "loss": 5.3704,
      "mean_token_accuracy": 0.20912910103797913,
      "num_tokens": 37360924.0,
      "step": 20040
    },
    {
      "entropy": 5.798215103149414,
      "epoch": 1.7701342281879193,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004687793965529115,
      "loss": 5.3142,
      "mean_token_accuracy": 0.21615274101495743,
      "num_tokens": 37369295.0,
      "step": 20045
    },
    {
      "entropy": 5.812409973144531,
      "epoch": 1.7705757682797598,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004687633908144633,
      "loss": 5.3617,
      "mean_token_accuracy": 0.20486588031053543,
      "num_tokens": 37379135.0,
      "step": 20050
    },
    {
      "entropy": 5.833855152130127,
      "epoch": 1.7710173083716,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046874738128030045,
      "loss": 5.2546,
      "mean_token_accuracy": 0.21422645896673204,
      "num_tokens": 37387927.0,
      "step": 20055
    },
    {
      "entropy": 5.776106119155884,
      "epoch": 1.7714588484634404,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004687313679507366,
      "loss": 5.2944,
      "mean_token_accuracy": 0.21309812664985656,
      "num_tokens": 37397629.0,
      "step": 20060
    },
    {
      "entropy": 5.78966007232666,
      "epoch": 1.7719003885552809,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00046871535082608544,
      "loss": 5.3182,
      "mean_token_accuracy": 0.20881196856498718,
      "num_tokens": 37407688.0,
      "step": 20065
    },
    {
      "entropy": 5.893298101425171,
      "epoch": 1.7723419286471211,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004686993299066607,
      "loss": 5.2992,
      "mean_token_accuracy": 0.2105863317847252,
      "num_tokens": 37417019.0,
      "step": 20070
    },
    {
      "entropy": 5.867874240875244,
      "epoch": 1.7727834687389614,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004686833051927763,
      "loss": 5.3629,
      "mean_token_accuracy": 0.20231543034315108,
      "num_tokens": 37427685.0,
      "step": 20075
    },
    {
      "entropy": 5.797391891479492,
      "epoch": 1.773225008830802,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046866727668474605,
      "loss": 5.2991,
      "mean_token_accuracy": 0.2138638973236084,
      "num_tokens": 37436844.0,
      "step": 20080
    },
    {
      "entropy": 5.782184791564942,
      "epoch": 1.7736665489226422,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000468651244382884,
      "loss": 5.2698,
      "mean_token_accuracy": 0.21559372395277024,
      "num_tokens": 37445755.0,
      "step": 20085
    },
    {
      "entropy": 5.855338668823242,
      "epoch": 1.7741080890144825,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004686352082875043,
      "loss": 5.4357,
      "mean_token_accuracy": 0.1951846346259117,
      "num_tokens": 37454176.0,
      "step": 20090
    },
    {
      "entropy": 5.871099424362183,
      "epoch": 1.774549629106323,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004686191683989209,
      "loss": 5.2938,
      "mean_token_accuracy": 0.21474916040897368,
      "num_tokens": 37463696.0,
      "step": 20095
    },
    {
      "entropy": 5.90635724067688,
      "epoch": 1.7749911691981632,
      "grad_norm": 1.0,
      "learning_rate": 0.0004686031247174482,
      "loss": 5.3519,
      "mean_token_accuracy": 0.2026623472571373,
      "num_tokens": 37473651.0,
      "step": 20100
    },
    {
      "entropy": 5.868699455261231,
      "epoch": 1.7754327092900035,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004685870772434002,
      "loss": 5.4312,
      "mean_token_accuracy": 0.19730533808469772,
      "num_tokens": 37483137.0,
      "step": 20105
    },
    {
      "entropy": 5.912846851348877,
      "epoch": 1.775874249381844,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004685710259770916,
      "loss": 5.4006,
      "mean_token_accuracy": 0.19890353232622146,
      "num_tokens": 37492029.0,
      "step": 20110
    },
    {
      "entropy": 5.893759250640869,
      "epoch": 1.776315789473684,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046855497091883664,
      "loss": 5.3572,
      "mean_token_accuracy": 0.20875534415245056,
      "num_tokens": 37500421.0,
      "step": 20115
    },
    {
      "entropy": 5.820963430404663,
      "epoch": 1.7767573295655246,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004685389120689499,
      "loss": 5.2993,
      "mean_token_accuracy": 0.20525917112827302,
      "num_tokens": 37510332.0,
      "step": 20120
    },
    {
      "entropy": 5.798265886306763,
      "epoch": 1.7771988696573648,
      "grad_norm": 1.09375,
      "learning_rate": 0.000468522849427746,
      "loss": 5.343,
      "mean_token_accuracy": 0.2129892036318779,
      "num_tokens": 37519802.0,
      "step": 20125
    },
    {
      "entropy": 5.951801538467407,
      "epoch": 1.7776404097492051,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046850678299553943,
      "loss": 5.4744,
      "mean_token_accuracy": 0.1945207715034485,
      "num_tokens": 37529622.0,
      "step": 20130
    },
    {
      "entropy": 5.783419799804688,
      "epoch": 1.7780819498410456,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004684907127726451,
      "loss": 5.2962,
      "mean_token_accuracy": 0.21682336628437043,
      "num_tokens": 37539446.0,
      "step": 20135
    },
    {
      "entropy": 5.807795906066895,
      "epoch": 1.778523489932886,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004684746387593778,
      "loss": 5.2873,
      "mean_token_accuracy": 0.2105346664786339,
      "num_tokens": 37547877.0,
      "step": 20140
    },
    {
      "entropy": 5.8958127975463865,
      "epoch": 1.7789650300247262,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004684585609560523,
      "loss": 5.5018,
      "mean_token_accuracy": 0.1957994669675827,
      "num_tokens": 37557841.0,
      "step": 20145
    },
    {
      "entropy": 5.947286176681518,
      "epoch": 1.7794065701165667,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004684424793629836,
      "loss": 5.4075,
      "mean_token_accuracy": 0.19818983525037764,
      "num_tokens": 37566876.0,
      "step": 20150
    },
    {
      "entropy": 5.906966066360473,
      "epoch": 1.779848110208407,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046842639398048683,
      "loss": 5.4073,
      "mean_token_accuracy": 0.20548618882894515,
      "num_tokens": 37575571.0,
      "step": 20155
    },
    {
      "entropy": 5.896219158172608,
      "epoch": 1.7802896503002472,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004684103048088769,
      "loss": 5.4797,
      "mean_token_accuracy": 0.20168900340795518,
      "num_tokens": 37585972.0,
      "step": 20160
    },
    {
      "entropy": 5.9010529041290285,
      "epoch": 1.7807311903920877,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004683942118484692,
      "loss": 5.4147,
      "mean_token_accuracy": 0.20082128047943115,
      "num_tokens": 37594849.0,
      "step": 20165
    },
    {
      "entropy": 5.890091323852539,
      "epoch": 1.781172730483928,
      "grad_norm": 1.0,
      "learning_rate": 0.00046837811509957883,
      "loss": 5.3469,
      "mean_token_accuracy": 0.2057483211159706,
      "num_tokens": 37603634.0,
      "step": 20170
    },
    {
      "entropy": 5.869364023208618,
      "epoch": 1.7816142705757683,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046836201456252116,
      "loss": 5.2968,
      "mean_token_accuracy": 0.21155764311552047,
      "num_tokens": 37612518.0,
      "step": 20175
    },
    {
      "entropy": 5.8324672222137455,
      "epoch": 1.7820558106676088,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004683459102376116,
      "loss": 5.3033,
      "mean_token_accuracy": 0.20891628116369249,
      "num_tokens": 37621553.0,
      "step": 20180
    },
    {
      "entropy": 5.829755783081055,
      "epoch": 1.7824973507594488,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046832980212516557,
      "loss": 5.4122,
      "mean_token_accuracy": 0.20100240558385848,
      "num_tokens": 37631263.0,
      "step": 20185
    },
    {
      "entropy": 5.879919385910034,
      "epoch": 1.7829388908512893,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046831369022549876,
      "loss": 5.2989,
      "mean_token_accuracy": 0.2082589492201805,
      "num_tokens": 37640810.0,
      "step": 20190
    },
    {
      "entropy": 5.89871563911438,
      "epoch": 1.7833804309431296,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004682975745389267,
      "loss": 5.4003,
      "mean_token_accuracy": 0.2027232214808464,
      "num_tokens": 37650848.0,
      "step": 20195
    },
    {
      "entropy": 5.897505903244019,
      "epoch": 1.7838219710349699,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046828145506576504,
      "loss": 5.3668,
      "mean_token_accuracy": 0.20478703081607819,
      "num_tokens": 37660370.0,
      "step": 20200
    },
    {
      "entropy": 5.817348146438599,
      "epoch": 1.7842635111268104,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004682653318063296,
      "loss": 5.3343,
      "mean_token_accuracy": 0.2099165752530098,
      "num_tokens": 37669570.0,
      "step": 20205
    },
    {
      "entropy": 5.802600765228272,
      "epoch": 1.7847050512186506,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046824920476093624,
      "loss": 5.4042,
      "mean_token_accuracy": 0.20283826887607576,
      "num_tokens": 37679585.0,
      "step": 20210
    },
    {
      "entropy": 5.915673637390137,
      "epoch": 1.785146591310491,
      "grad_norm": 1.125,
      "learning_rate": 0.00046823307392990086,
      "loss": 5.3891,
      "mean_token_accuracy": 0.2046528786420822,
      "num_tokens": 37688736.0,
      "step": 20215
    },
    {
      "entropy": 5.977085590362549,
      "epoch": 1.7855881314023314,
      "grad_norm": 1.609375,
      "learning_rate": 0.00046821693931353947,
      "loss": 5.4315,
      "mean_token_accuracy": 0.19578227698802947,
      "num_tokens": 37699104.0,
      "step": 20220
    },
    {
      "entropy": 5.911003971099854,
      "epoch": 1.7860296714941717,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004682008009121681,
      "loss": 5.4324,
      "mean_token_accuracy": 0.20133746415376663,
      "num_tokens": 37708260.0,
      "step": 20225
    },
    {
      "entropy": 5.814093494415284,
      "epoch": 1.786471211586012,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000468184658726103,
      "loss": 5.3673,
      "mean_token_accuracy": 0.20417275428771972,
      "num_tokens": 37717442.0,
      "step": 20230
    },
    {
      "entropy": 5.884583139419556,
      "epoch": 1.7869127516778525,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004681685127556602,
      "loss": 5.4064,
      "mean_token_accuracy": 0.19773323088884354,
      "num_tokens": 37727063.0,
      "step": 20235
    },
    {
      "entropy": 5.875490999221801,
      "epoch": 1.7873542917696927,
      "grad_norm": 1.0,
      "learning_rate": 0.00046815236300115615,
      "loss": 5.3248,
      "mean_token_accuracy": 0.20776914954185485,
      "num_tokens": 37736441.0,
      "step": 20240
    },
    {
      "entropy": 5.76911768913269,
      "epoch": 1.787795831861533,
      "grad_norm": 1.125,
      "learning_rate": 0.00046813620946290716,
      "loss": 5.2723,
      "mean_token_accuracy": 0.2133665129542351,
      "num_tokens": 37745290.0,
      "step": 20245
    },
    {
      "entropy": 5.900927352905273,
      "epoch": 1.7882373719533735,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004681200521412296,
      "loss": 5.3875,
      "mean_token_accuracy": 0.20325565040111543,
      "num_tokens": 37754786.0,
      "step": 20250
    },
    {
      "entropy": 5.908439207077026,
      "epoch": 1.7886789120452136,
      "grad_norm": 1.0,
      "learning_rate": 0.00046810389103644015,
      "loss": 5.4023,
      "mean_token_accuracy": 0.20222677737474443,
      "num_tokens": 37763513.0,
      "step": 20255
    },
    {
      "entropy": 5.862041330337524,
      "epoch": 1.789120452137054,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004680877261488552,
      "loss": 5.3817,
      "mean_token_accuracy": 0.1998486340045929,
      "num_tokens": 37771752.0,
      "step": 20260
    },
    {
      "entropy": 5.865772724151611,
      "epoch": 1.7895619922288943,
      "grad_norm": 1.0,
      "learning_rate": 0.0004680715574787916,
      "loss": 5.3866,
      "mean_token_accuracy": 0.19422840029001237,
      "num_tokens": 37781531.0,
      "step": 20265
    },
    {
      "entropy": 5.838648986816406,
      "epoch": 1.7900035323207346,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046805538502656596,
      "loss": 5.3202,
      "mean_token_accuracy": 0.20190480947494507,
      "num_tokens": 37791469.0,
      "step": 20270
    },
    {
      "entropy": 5.900450754165649,
      "epoch": 1.7904450724125751,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004680392087924952,
      "loss": 5.3937,
      "mean_token_accuracy": 0.20292436927556992,
      "num_tokens": 37800301.0,
      "step": 20275
    },
    {
      "entropy": 5.8141848087310795,
      "epoch": 1.7908866125044154,
      "grad_norm": 1.078125,
      "learning_rate": 0.000468023028776896,
      "loss": 5.2347,
      "mean_token_accuracy": 0.21848550885915757,
      "num_tokens": 37808849.0,
      "step": 20280
    },
    {
      "entropy": 5.88062219619751,
      "epoch": 1.7913281525962557,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004680068449800855,
      "loss": 5.4048,
      "mean_token_accuracy": 0.2034169316291809,
      "num_tokens": 37818632.0,
      "step": 20285
    },
    {
      "entropy": 5.898212194442749,
      "epoch": 1.7917696926880962,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004679906574023807,
      "loss": 5.3949,
      "mean_token_accuracy": 0.19932047724723817,
      "num_tokens": 37828158.0,
      "step": 20290
    },
    {
      "entropy": 5.8748620510101315,
      "epoch": 1.7922112327799364,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004679744660440987,
      "loss": 5.3289,
      "mean_token_accuracy": 0.20268318355083464,
      "num_tokens": 37836242.0,
      "step": 20295
    },
    {
      "entropy": 5.866003608703613,
      "epoch": 1.7926527728717767,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046795827090555654,
      "loss": 5.3535,
      "mean_token_accuracy": 0.202984955906868,
      "num_tokens": 37844286.0,
      "step": 20300
    },
    {
      "entropy": 5.847696304321289,
      "epoch": 1.7930943129636172,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004679420719870717,
      "loss": 5.3068,
      "mean_token_accuracy": 0.20818886011838914,
      "num_tokens": 37853440.0,
      "step": 20305
    },
    {
      "entropy": 5.872677278518677,
      "epoch": 1.7935358530554575,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004679258692889613,
      "loss": 5.2455,
      "mean_token_accuracy": 0.21110901981592178,
      "num_tokens": 37862511.0,
      "step": 20310
    },
    {
      "entropy": 5.942249822616577,
      "epoch": 1.7939773931472978,
      "grad_norm": 1.046875,
      "learning_rate": 0.000467909662811543,
      "loss": 5.4542,
      "mean_token_accuracy": 0.19627978652715683,
      "num_tokens": 37872797.0,
      "step": 20315
    },
    {
      "entropy": 5.844317817687989,
      "epoch": 1.7944189332391383,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000467893452555134,
      "loss": 5.3134,
      "mean_token_accuracy": 0.21182480156421662,
      "num_tokens": 37881859.0,
      "step": 20320
    },
    {
      "entropy": 5.825922060012817,
      "epoch": 1.7948604733309783,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046787723852005197,
      "loss": 5.3365,
      "mean_token_accuracy": 0.2158793330192566,
      "num_tokens": 37891381.0,
      "step": 20325
    },
    {
      "entropy": 5.741138362884522,
      "epoch": 1.7953020134228188,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004678610207066146,
      "loss": 5.2496,
      "mean_token_accuracy": 0.21238402128219605,
      "num_tokens": 37900554.0,
      "step": 20330
    },
    {
      "entropy": 5.89889121055603,
      "epoch": 1.795743553514659,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004678447991151395,
      "loss": 5.3023,
      "mean_token_accuracy": 0.2121421828866005,
      "num_tokens": 37908965.0,
      "step": 20335
    },
    {
      "entropy": 5.792270231246948,
      "epoch": 1.7961850936064994,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046782857374594447,
      "loss": 5.296,
      "mean_token_accuracy": 0.21209638863801955,
      "num_tokens": 37917555.0,
      "step": 20340
    },
    {
      "entropy": 5.810970449447632,
      "epoch": 1.7966266336983399,
      "grad_norm": 1.125,
      "learning_rate": 0.0004678123445993473,
      "loss": 5.3585,
      "mean_token_accuracy": 0.21037175506353378,
      "num_tokens": 37926334.0,
      "step": 20345
    },
    {
      "entropy": 5.803108406066895,
      "epoch": 1.7970681737901801,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000467796111675666,
      "loss": 5.3096,
      "mean_token_accuracy": 0.2096547305583954,
      "num_tokens": 37935912.0,
      "step": 20350
    },
    {
      "entropy": 5.768332433700562,
      "epoch": 1.7975097138820204,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046777987497521845,
      "loss": 5.3272,
      "mean_token_accuracy": 0.21359478533267975,
      "num_tokens": 37945238.0,
      "step": 20355
    },
    {
      "entropy": 5.833979177474975,
      "epoch": 1.797951253973861,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046776363449832286,
      "loss": 5.3248,
      "mean_token_accuracy": 0.21153734475374222,
      "num_tokens": 37954945.0,
      "step": 20360
    },
    {
      "entropy": 5.851474046707153,
      "epoch": 1.7983927940657012,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046774739024529736,
      "loss": 5.2451,
      "mean_token_accuracy": 0.21519911140203477,
      "num_tokens": 37964114.0,
      "step": 20365
    },
    {
      "entropy": 5.908031940460205,
      "epoch": 1.7988343341575415,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046773114221645995,
      "loss": 5.3995,
      "mean_token_accuracy": 0.1979236423969269,
      "num_tokens": 37973774.0,
      "step": 20370
    },
    {
      "entropy": 5.910662221908569,
      "epoch": 1.799275874249382,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00046771489041212916,
      "loss": 5.3562,
      "mean_token_accuracy": 0.20842334032058715,
      "num_tokens": 37982763.0,
      "step": 20375
    },
    {
      "entropy": 5.931088447570801,
      "epoch": 1.7997174143412222,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004676986348326233,
      "loss": 5.4658,
      "mean_token_accuracy": 0.20048643052577972,
      "num_tokens": 37992302.0,
      "step": 20380
    },
    {
      "entropy": 5.952190208435058,
      "epoch": 1.8001589544330625,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004676823754782607,
      "loss": 5.4894,
      "mean_token_accuracy": 0.19188101589679718,
      "num_tokens": 38001785.0,
      "step": 20385
    },
    {
      "entropy": 5.8138501167297365,
      "epoch": 1.800600494524903,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046766611234936,
      "loss": 5.2994,
      "mean_token_accuracy": 0.20842182636260986,
      "num_tokens": 38010823.0,
      "step": 20390
    },
    {
      "entropy": 5.829443883895874,
      "epoch": 1.801042034616743,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004676498454462398,
      "loss": 5.3183,
      "mean_token_accuracy": 0.214272178709507,
      "num_tokens": 38019718.0,
      "step": 20395
    },
    {
      "entropy": 5.8638537406921385,
      "epoch": 1.8014835747085836,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046763357476921856,
      "loss": 5.3213,
      "mean_token_accuracy": 0.20627929270267487,
      "num_tokens": 38028962.0,
      "step": 20400
    },
    {
      "entropy": 5.863303899765015,
      "epoch": 1.8019251148004238,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004676173003186152,
      "loss": 5.2979,
      "mean_token_accuracy": 0.21785591393709183,
      "num_tokens": 38037824.0,
      "step": 20405
    },
    {
      "entropy": 5.882257127761841,
      "epoch": 1.802366654892264,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046760102209474855,
      "loss": 5.4427,
      "mean_token_accuracy": 0.19779300838708877,
      "num_tokens": 38046764.0,
      "step": 20410
    },
    {
      "entropy": 5.906739568710327,
      "epoch": 1.8028081949841046,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004675847400979373,
      "loss": 5.3682,
      "mean_token_accuracy": 0.2031184047460556,
      "num_tokens": 38055363.0,
      "step": 20415
    },
    {
      "entropy": 5.864423513412476,
      "epoch": 1.8032497350759449,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046756845432850057,
      "loss": 5.409,
      "mean_token_accuracy": 0.20096191316843032,
      "num_tokens": 38065471.0,
      "step": 20420
    },
    {
      "entropy": 5.9324547290802006,
      "epoch": 1.8036912751677852,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004675521647867573,
      "loss": 5.4975,
      "mean_token_accuracy": 0.19525431841611862,
      "num_tokens": 38075448.0,
      "step": 20425
    },
    {
      "entropy": 5.859852504730225,
      "epoch": 1.8041328152596257,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00046753587147302665,
      "loss": 5.4327,
      "mean_token_accuracy": 0.2063135489821434,
      "num_tokens": 38085657.0,
      "step": 20430
    },
    {
      "entropy": 5.919007635116577,
      "epoch": 1.804574355351466,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004675195743876278,
      "loss": 5.3549,
      "mean_token_accuracy": 0.2081198126077652,
      "num_tokens": 38095580.0,
      "step": 20435
    },
    {
      "entropy": 5.965777969360351,
      "epoch": 1.8050158954433062,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046750327353088,
      "loss": 5.4592,
      "mean_token_accuracy": 0.19610353410243989,
      "num_tokens": 38104969.0,
      "step": 20440
    },
    {
      "entropy": 5.905309343338013,
      "epoch": 1.8054574355351467,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004674869689031026,
      "loss": 5.4032,
      "mean_token_accuracy": 0.2008821427822113,
      "num_tokens": 38114536.0,
      "step": 20445
    },
    {
      "entropy": 5.842322158813476,
      "epoch": 1.805898975626987,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046747066050461486,
      "loss": 5.3056,
      "mean_token_accuracy": 0.2088845983147621,
      "num_tokens": 38122969.0,
      "step": 20450
    },
    {
      "entropy": 5.8650373935699465,
      "epoch": 1.8063405157188273,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046745434833573635,
      "loss": 5.3631,
      "mean_token_accuracy": 0.20031063109636307,
      "num_tokens": 38132892.0,
      "step": 20455
    },
    {
      "entropy": 5.780540418624878,
      "epoch": 1.8067820558106678,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004674380323967867,
      "loss": 5.3324,
      "mean_token_accuracy": 0.2102600157260895,
      "num_tokens": 38141950.0,
      "step": 20460
    },
    {
      "entropy": 5.939284229278565,
      "epoch": 1.8072235959025078,
      "grad_norm": 1.125,
      "learning_rate": 0.0004674217126880853,
      "loss": 5.4464,
      "mean_token_accuracy": 0.19473828077316285,
      "num_tokens": 38151314.0,
      "step": 20465
    },
    {
      "entropy": 5.97164945602417,
      "epoch": 1.8076651359943483,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004674053892099521,
      "loss": 5.4392,
      "mean_token_accuracy": 0.20110773295164108,
      "num_tokens": 38159992.0,
      "step": 20470
    },
    {
      "entropy": 5.8675071716308596,
      "epoch": 1.8081066760861886,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046738906196270663,
      "loss": 5.3446,
      "mean_token_accuracy": 0.1976835235953331,
      "num_tokens": 38168436.0,
      "step": 20475
    },
    {
      "entropy": 5.8300371170043945,
      "epoch": 1.8085482161780289,
      "grad_norm": 1.015625,
      "learning_rate": 0.000467372730946669,
      "loss": 5.3449,
      "mean_token_accuracy": 0.20450993329286576,
      "num_tokens": 38177932.0,
      "step": 20480
    },
    {
      "entropy": 5.902409601211548,
      "epoch": 1.8089897562698694,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046735639616215884,
      "loss": 5.3557,
      "mean_token_accuracy": 0.21158699095249175,
      "num_tokens": 38187727.0,
      "step": 20485
    },
    {
      "entropy": 5.843787002563476,
      "epoch": 1.8094312963617096,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046734005760949634,
      "loss": 5.4154,
      "mean_token_accuracy": 0.19625237584114075,
      "num_tokens": 38198768.0,
      "step": 20490
    },
    {
      "entropy": 5.8692399024963375,
      "epoch": 1.80987283645355,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046732371528900154,
      "loss": 5.4124,
      "mean_token_accuracy": 0.2083530083298683,
      "num_tokens": 38207538.0,
      "step": 20495
    },
    {
      "entropy": 5.932307720184326,
      "epoch": 1.8103143765453904,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004673073692009945,
      "loss": 5.449,
      "mean_token_accuracy": 0.19892775416374206,
      "num_tokens": 38216254.0,
      "step": 20500
    },
    {
      "entropy": 5.947169828414917,
      "epoch": 1.8107559166372307,
      "grad_norm": 1.0,
      "learning_rate": 0.0004672910193457955,
      "loss": 5.4194,
      "mean_token_accuracy": 0.19809736758470536,
      "num_tokens": 38225356.0,
      "step": 20505
    },
    {
      "entropy": 5.851454734802246,
      "epoch": 1.811197456729071,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004672746657237248,
      "loss": 5.2808,
      "mean_token_accuracy": 0.21226958483457564,
      "num_tokens": 38234561.0,
      "step": 20510
    },
    {
      "entropy": 5.815788793563843,
      "epoch": 1.8116389968209115,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004672583083351027,
      "loss": 5.3572,
      "mean_token_accuracy": 0.20180919766426086,
      "num_tokens": 38244418.0,
      "step": 20515
    },
    {
      "entropy": 5.874659538269043,
      "epoch": 1.8120805369127517,
      "grad_norm": 1.125,
      "learning_rate": 0.0004672419471802497,
      "loss": 5.3386,
      "mean_token_accuracy": 0.21120245158672332,
      "num_tokens": 38253639.0,
      "step": 20520
    },
    {
      "entropy": 5.896960735321045,
      "epoch": 1.812522077004592,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046722558225948627,
      "loss": 5.2595,
      "mean_token_accuracy": 0.21947896033525466,
      "num_tokens": 38262401.0,
      "step": 20525
    },
    {
      "entropy": 5.8342921257019045,
      "epoch": 1.8129636170964325,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046720921357313303,
      "loss": 5.3504,
      "mean_token_accuracy": 0.21129164844751358,
      "num_tokens": 38271070.0,
      "step": 20530
    },
    {
      "entropy": 5.88353271484375,
      "epoch": 1.8134051571882726,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004671928411215106,
      "loss": 5.3405,
      "mean_token_accuracy": 0.21325495690107346,
      "num_tokens": 38280675.0,
      "step": 20535
    },
    {
      "entropy": 5.879548454284668,
      "epoch": 1.813846697280113,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046717646490493977,
      "loss": 5.4133,
      "mean_token_accuracy": 0.1977847248315811,
      "num_tokens": 38290116.0,
      "step": 20540
    },
    {
      "entropy": 5.905095386505127,
      "epoch": 1.8142882373719533,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004671600849237412,
      "loss": 5.3818,
      "mean_token_accuracy": 0.20388080626726152,
      "num_tokens": 38299415.0,
      "step": 20545
    },
    {
      "entropy": 5.7524528980255125,
      "epoch": 1.8147297774637936,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004671437011782359,
      "loss": 5.2546,
      "mean_token_accuracy": 0.22169750183820724,
      "num_tokens": 38308761.0,
      "step": 20550
    },
    {
      "entropy": 5.912684583663941,
      "epoch": 1.815171317555634,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004671273136687448,
      "loss": 5.4034,
      "mean_token_accuracy": 0.2000546410679817,
      "num_tokens": 38318212.0,
      "step": 20555
    },
    {
      "entropy": 5.923445415496826,
      "epoch": 1.8156128576474744,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004671109223955889,
      "loss": 5.3775,
      "mean_token_accuracy": 0.20213733166456221,
      "num_tokens": 38327233.0,
      "step": 20560
    },
    {
      "entropy": 5.8516336441040036,
      "epoch": 1.8160543977393147,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004670945273590892,
      "loss": 5.2799,
      "mean_token_accuracy": 0.21335710287094117,
      "num_tokens": 38336473.0,
      "step": 20565
    },
    {
      "entropy": 5.808154964447022,
      "epoch": 1.8164959378311551,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004670781285595671,
      "loss": 5.3954,
      "mean_token_accuracy": 0.20852681398391723,
      "num_tokens": 38345286.0,
      "step": 20570
    },
    {
      "entropy": 5.894282007217408,
      "epoch": 1.8169374779229954,
      "grad_norm": 1.0,
      "learning_rate": 0.0004670617259973435,
      "loss": 5.3161,
      "mean_token_accuracy": 0.2049289181828499,
      "num_tokens": 38355191.0,
      "step": 20575
    },
    {
      "entropy": 5.878620290756226,
      "epoch": 1.8173790180148357,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046704531967274007,
      "loss": 5.309,
      "mean_token_accuracy": 0.20789871513843536,
      "num_tokens": 38363902.0,
      "step": 20580
    },
    {
      "entropy": 5.8712554454803465,
      "epoch": 1.8178205581066762,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046702890958607804,
      "loss": 5.4714,
      "mean_token_accuracy": 0.19961811155080794,
      "num_tokens": 38374828.0,
      "step": 20585
    },
    {
      "entropy": 5.814764404296875,
      "epoch": 1.8182620981985165,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004670124957376788,
      "loss": 5.3378,
      "mean_token_accuracy": 0.2025281995534897,
      "num_tokens": 38384918.0,
      "step": 20590
    },
    {
      "entropy": 5.942892074584961,
      "epoch": 1.8187036382903567,
      "grad_norm": 1.0,
      "learning_rate": 0.0004669960781278641,
      "loss": 5.4034,
      "mean_token_accuracy": 0.20002204030752183,
      "num_tokens": 38394005.0,
      "step": 20595
    },
    {
      "entropy": 5.973907041549682,
      "epoch": 1.8191451783821972,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004669796567569553,
      "loss": 5.4208,
      "mean_token_accuracy": 0.20189059376716614,
      "num_tokens": 38402722.0,
      "step": 20600
    },
    {
      "entropy": 5.843851327896118,
      "epoch": 1.8195867184740373,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004669632316252743,
      "loss": 5.3348,
      "mean_token_accuracy": 0.21077610552310944,
      "num_tokens": 38411726.0,
      "step": 20605
    },
    {
      "entropy": 5.899236488342285,
      "epoch": 1.8200282585658778,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046694680273314275,
      "loss": 5.396,
      "mean_token_accuracy": 0.2030399203300476,
      "num_tokens": 38421187.0,
      "step": 20610
    },
    {
      "entropy": 5.877706670761109,
      "epoch": 1.820469798657718,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004669303700808824,
      "loss": 5.4301,
      "mean_token_accuracy": 0.1965902864933014,
      "num_tokens": 38430722.0,
      "step": 20615
    },
    {
      "entropy": 5.840076351165772,
      "epoch": 1.8209113387495584,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004669139336688154,
      "loss": 5.3489,
      "mean_token_accuracy": 0.2044423222541809,
      "num_tokens": 38440763.0,
      "step": 20620
    },
    {
      "entropy": 5.823544597625732,
      "epoch": 1.8213528788413988,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046689749349726353,
      "loss": 5.3848,
      "mean_token_accuracy": 0.2033069536089897,
      "num_tokens": 38449298.0,
      "step": 20625
    },
    {
      "entropy": 5.859510135650635,
      "epoch": 1.8217944189332391,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004668810495665489,
      "loss": 5.3558,
      "mean_token_accuracy": 0.20442309677600862,
      "num_tokens": 38458941.0,
      "step": 20630
    },
    {
      "entropy": 5.876007127761841,
      "epoch": 1.8222359590250794,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004668646018769936,
      "loss": 5.3266,
      "mean_token_accuracy": 0.21604098826646806,
      "num_tokens": 38468381.0,
      "step": 20635
    },
    {
      "entropy": 5.9373369216918945,
      "epoch": 1.82267749911692,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046684815042891993,
      "loss": 5.3783,
      "mean_token_accuracy": 0.202556112408638,
      "num_tokens": 38477767.0,
      "step": 20640
    },
    {
      "entropy": 5.837852525711059,
      "epoch": 1.8231190392087602,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004668316952226501,
      "loss": 5.3002,
      "mean_token_accuracy": 0.2069453090429306,
      "num_tokens": 38486884.0,
      "step": 20645
    },
    {
      "entropy": 5.843544769287109,
      "epoch": 1.8235605793006004,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004668152362585064,
      "loss": 5.4226,
      "mean_token_accuracy": 0.1992058902978897,
      "num_tokens": 38496243.0,
      "step": 20650
    },
    {
      "entropy": 5.778637886047363,
      "epoch": 1.824002119392441,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004667987735368113,
      "loss": 5.294,
      "mean_token_accuracy": 0.2042941391468048,
      "num_tokens": 38504447.0,
      "step": 20655
    },
    {
      "entropy": 5.882816457748413,
      "epoch": 1.8244436594842812,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004667823070578874,
      "loss": 5.4054,
      "mean_token_accuracy": 0.20013791173696518,
      "num_tokens": 38514423.0,
      "step": 20660
    },
    {
      "entropy": 5.884525346755981,
      "epoch": 1.8248851995761215,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046676583682205713,
      "loss": 5.3297,
      "mean_token_accuracy": 0.21381803154945372,
      "num_tokens": 38522765.0,
      "step": 20665
    },
    {
      "entropy": 5.917822170257568,
      "epoch": 1.825326739667962,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046674936282964326,
      "loss": 5.4478,
      "mean_token_accuracy": 0.19618436247110366,
      "num_tokens": 38532074.0,
      "step": 20670
    },
    {
      "entropy": 5.845418119430542,
      "epoch": 1.825768279759802,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004667328850809683,
      "loss": 5.3291,
      "mean_token_accuracy": 0.2020283207297325,
      "num_tokens": 38541423.0,
      "step": 20675
    },
    {
      "entropy": 5.869617366790772,
      "epoch": 1.8262098198516425,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004667164035763552,
      "loss": 5.4233,
      "mean_token_accuracy": 0.20130159556865693,
      "num_tokens": 38552369.0,
      "step": 20680
    },
    {
      "entropy": 5.909629583358765,
      "epoch": 1.8266513599434828,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004666999183161268,
      "loss": 5.3648,
      "mean_token_accuracy": 0.20436142832040788,
      "num_tokens": 38561310.0,
      "step": 20685
    },
    {
      "entropy": 5.919795942306519,
      "epoch": 1.827092900035323,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046668342930060604,
      "loss": 5.4257,
      "mean_token_accuracy": 0.20203170478343963,
      "num_tokens": 38571157.0,
      "step": 20690
    },
    {
      "entropy": 5.862828111648559,
      "epoch": 1.8275344401271636,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046666693653011596,
      "loss": 5.3492,
      "mean_token_accuracy": 0.2039842963218689,
      "num_tokens": 38579862.0,
      "step": 20695
    },
    {
      "entropy": 5.870121860504151,
      "epoch": 1.8279759802190039,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046665044000497954,
      "loss": 5.4234,
      "mean_token_accuracy": 0.20539480149745942,
      "num_tokens": 38589784.0,
      "step": 20700
    },
    {
      "entropy": 5.910328960418701,
      "epoch": 1.8284175203108441,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046663393972551994,
      "loss": 5.3317,
      "mean_token_accuracy": 0.20784323811531066,
      "num_tokens": 38598925.0,
      "step": 20705
    },
    {
      "entropy": 5.888635158538818,
      "epoch": 1.8288590604026846,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00046661743569206056,
      "loss": 5.392,
      "mean_token_accuracy": 0.1957668572664261,
      "num_tokens": 38608245.0,
      "step": 20710
    },
    {
      "entropy": 5.883733415603638,
      "epoch": 1.829300600494525,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004666009279049245,
      "loss": 5.4586,
      "mean_token_accuracy": 0.1983842819929123,
      "num_tokens": 38617586.0,
      "step": 20715
    },
    {
      "entropy": 5.905021858215332,
      "epoch": 1.8297421405863652,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046658441636443526,
      "loss": 5.4061,
      "mean_token_accuracy": 0.2049261063337326,
      "num_tokens": 38626420.0,
      "step": 20720
    },
    {
      "entropy": 5.881510162353516,
      "epoch": 1.8301836806782057,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046656790107091627,
      "loss": 5.3007,
      "mean_token_accuracy": 0.21221004873514177,
      "num_tokens": 38635585.0,
      "step": 20725
    },
    {
      "entropy": 5.830260181427002,
      "epoch": 1.830625220770046,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004665513820246911,
      "loss": 5.3552,
      "mean_token_accuracy": 0.20323756635189055,
      "num_tokens": 38643998.0,
      "step": 20730
    },
    {
      "entropy": 5.841861295700073,
      "epoch": 1.8310667608618862,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004665348592260833,
      "loss": 5.3721,
      "mean_token_accuracy": 0.20905961394309996,
      "num_tokens": 38653018.0,
      "step": 20735
    },
    {
      "entropy": 5.84530348777771,
      "epoch": 1.8315083009537267,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004665183326754165,
      "loss": 5.3195,
      "mean_token_accuracy": 0.2096097707748413,
      "num_tokens": 38662079.0,
      "step": 20740
    },
    {
      "entropy": 5.885212516784668,
      "epoch": 1.8319498410455668,
      "grad_norm": 1.125,
      "learning_rate": 0.0004665018023730144,
      "loss": 5.3803,
      "mean_token_accuracy": 0.20793042033910752,
      "num_tokens": 38669928.0,
      "step": 20745
    },
    {
      "entropy": 5.855934047698975,
      "epoch": 1.8323913811374073,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000466485268319201,
      "loss": 5.3482,
      "mean_token_accuracy": 0.2029734492301941,
      "num_tokens": 38679475.0,
      "step": 20750
    },
    {
      "entropy": 5.873014354705811,
      "epoch": 1.8328329212292476,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004664687305143001,
      "loss": 5.4035,
      "mean_token_accuracy": 0.20414873957633972,
      "num_tokens": 38689057.0,
      "step": 20755
    },
    {
      "entropy": 5.84939866065979,
      "epoch": 1.8332744613210878,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046645218895863564,
      "loss": 5.258,
      "mean_token_accuracy": 0.21808112561702728,
      "num_tokens": 38698362.0,
      "step": 20760
    },
    {
      "entropy": 5.82479043006897,
      "epoch": 1.8337160014129283,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00046643564365253173,
      "loss": 5.2754,
      "mean_token_accuracy": 0.2201460063457489,
      "num_tokens": 38708319.0,
      "step": 20765
    },
    {
      "entropy": 5.890381574630737,
      "epoch": 1.8341575415047686,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004664190945963124,
      "loss": 5.399,
      "mean_token_accuracy": 0.19942618310451507,
      "num_tokens": 38717441.0,
      "step": 20770
    },
    {
      "entropy": 5.956099128723144,
      "epoch": 1.834599081596609,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046640254179030184,
      "loss": 5.3988,
      "mean_token_accuracy": 0.1964491829276085,
      "num_tokens": 38726903.0,
      "step": 20775
    },
    {
      "entropy": 5.838819360733032,
      "epoch": 1.8350406216884494,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004663859852348244,
      "loss": 5.3292,
      "mean_token_accuracy": 0.2094813823699951,
      "num_tokens": 38735983.0,
      "step": 20780
    },
    {
      "entropy": 5.838028144836426,
      "epoch": 1.8354821617802897,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046636942493020437,
      "loss": 5.443,
      "mean_token_accuracy": 0.20704626590013503,
      "num_tokens": 38745017.0,
      "step": 20785
    },
    {
      "entropy": 5.783022260665893,
      "epoch": 1.83592370187213,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004663528608767661,
      "loss": 5.2822,
      "mean_token_accuracy": 0.20643096119165422,
      "num_tokens": 38754756.0,
      "step": 20790
    },
    {
      "entropy": 5.893237066268921,
      "epoch": 1.8363652419639704,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004663362930748341,
      "loss": 5.3616,
      "mean_token_accuracy": 0.2074210524559021,
      "num_tokens": 38763695.0,
      "step": 20795
    },
    {
      "entropy": 5.913727617263794,
      "epoch": 1.8368067820558107,
      "grad_norm": 1.0,
      "learning_rate": 0.00046631972152473295,
      "loss": 5.4402,
      "mean_token_accuracy": 0.20066530108451844,
      "num_tokens": 38773637.0,
      "step": 20800
    },
    {
      "entropy": 5.744051313400268,
      "epoch": 1.837248322147651,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046630314622678724,
      "loss": 5.2374,
      "mean_token_accuracy": 0.21938198655843735,
      "num_tokens": 38783004.0,
      "step": 20805
    },
    {
      "entropy": 5.781368541717529,
      "epoch": 1.8376898622394915,
      "grad_norm": 1.125,
      "learning_rate": 0.0004662865671813217,
      "loss": 5.3663,
      "mean_token_accuracy": 0.20539602339267732,
      "num_tokens": 38791342.0,
      "step": 20810
    },
    {
      "entropy": 5.856006097793579,
      "epoch": 1.8381314023313315,
      "grad_norm": 0.921875,
      "learning_rate": 0.00046626998438866114,
      "loss": 5.31,
      "mean_token_accuracy": 0.20879921317100525,
      "num_tokens": 38801046.0,
      "step": 20815
    },
    {
      "entropy": 5.882837390899658,
      "epoch": 1.838572942423172,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046625339784913035,
      "loss": 5.3695,
      "mean_token_accuracy": 0.2043130800127983,
      "num_tokens": 38810806.0,
      "step": 20820
    },
    {
      "entropy": 5.84970908164978,
      "epoch": 1.8390144825150123,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046623680756305427,
      "loss": 5.4013,
      "mean_token_accuracy": 0.2009470283985138,
      "num_tokens": 38819362.0,
      "step": 20825
    },
    {
      "entropy": 5.897683763504029,
      "epoch": 1.8394560226068526,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004662202135307579,
      "loss": 5.3996,
      "mean_token_accuracy": 0.20360717922449112,
      "num_tokens": 38829337.0,
      "step": 20830
    },
    {
      "entropy": 5.874234962463379,
      "epoch": 1.839897562698693,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004662036157525663,
      "loss": 5.36,
      "mean_token_accuracy": 0.20298291891813278,
      "num_tokens": 38838351.0,
      "step": 20835
    },
    {
      "entropy": 5.895681667327881,
      "epoch": 1.8403391027905334,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046618701422880456,
      "loss": 5.4089,
      "mean_token_accuracy": 0.1998438224196434,
      "num_tokens": 38847830.0,
      "step": 20840
    },
    {
      "entropy": 5.9258873462677,
      "epoch": 1.8407806428823736,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000466170408959798,
      "loss": 5.4135,
      "mean_token_accuracy": 0.1994987055659294,
      "num_tokens": 38856945.0,
      "step": 20845
    },
    {
      "entropy": 5.83818883895874,
      "epoch": 1.8412221829742141,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004661537999458718,
      "loss": 5.3314,
      "mean_token_accuracy": 0.20225901603698732,
      "num_tokens": 38866493.0,
      "step": 20850
    },
    {
      "entropy": 5.7961172580719,
      "epoch": 1.8416637230660544,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046613718718735143,
      "loss": 5.2967,
      "mean_token_accuracy": 0.21373035609722138,
      "num_tokens": 38875653.0,
      "step": 20855
    },
    {
      "entropy": 5.8129431247711185,
      "epoch": 1.8421052631578947,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046612057068456226,
      "loss": 5.229,
      "mean_token_accuracy": 0.21884686946868898,
      "num_tokens": 38884740.0,
      "step": 20860
    },
    {
      "entropy": 5.852499675750733,
      "epoch": 1.8425468032497352,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046610395043782983,
      "loss": 5.3558,
      "mean_token_accuracy": 0.20886240303516387,
      "num_tokens": 38893801.0,
      "step": 20865
    },
    {
      "entropy": 5.85790696144104,
      "epoch": 1.8429883433415755,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046608732644747963,
      "loss": 5.357,
      "mean_token_accuracy": 0.2100034236907959,
      "num_tokens": 38902298.0,
      "step": 20870
    },
    {
      "entropy": 5.854526329040527,
      "epoch": 1.8434298834334157,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046607069871383745,
      "loss": 5.2604,
      "mean_token_accuracy": 0.2157875820994377,
      "num_tokens": 38910904.0,
      "step": 20875
    },
    {
      "entropy": 5.789853858947754,
      "epoch": 1.8438714235252562,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004660540672372289,
      "loss": 5.2504,
      "mean_token_accuracy": 0.2158147305250168,
      "num_tokens": 38920405.0,
      "step": 20880
    },
    {
      "entropy": 5.800595092773437,
      "epoch": 1.8443129636170963,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004660374320179798,
      "loss": 5.2519,
      "mean_token_accuracy": 0.2045004189014435,
      "num_tokens": 38930118.0,
      "step": 20885
    },
    {
      "entropy": 5.828299236297608,
      "epoch": 1.8447545037089368,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046602079305641615,
      "loss": 5.3589,
      "mean_token_accuracy": 0.20638312846422197,
      "num_tokens": 38939388.0,
      "step": 20890
    },
    {
      "entropy": 5.839706134796143,
      "epoch": 1.845196043800777,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046600415035286374,
      "loss": 5.3956,
      "mean_token_accuracy": 0.20893628001213074,
      "num_tokens": 38948632.0,
      "step": 20895
    },
    {
      "entropy": 5.926899003982544,
      "epoch": 1.8456375838926173,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004659875039076487,
      "loss": 5.3824,
      "mean_token_accuracy": 0.20359280109405517,
      "num_tokens": 38957504.0,
      "step": 20900
    },
    {
      "entropy": 5.85803861618042,
      "epoch": 1.8460791239844578,
      "grad_norm": 0.953125,
      "learning_rate": 0.000465970853721097,
      "loss": 5.3212,
      "mean_token_accuracy": 0.20657461285591125,
      "num_tokens": 38967713.0,
      "step": 20905
    },
    {
      "entropy": 5.840960073471069,
      "epoch": 1.8465206640762981,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046595419979353494,
      "loss": 5.3741,
      "mean_token_accuracy": 0.20615524798631668,
      "num_tokens": 38976982.0,
      "step": 20910
    },
    {
      "entropy": 5.816210126876831,
      "epoch": 1.8469622041681384,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004659375421252887,
      "loss": 5.3323,
      "mean_token_accuracy": 0.20973749160766603,
      "num_tokens": 38985479.0,
      "step": 20915
    },
    {
      "entropy": 5.9450376510620115,
      "epoch": 1.8474037442599789,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004659208807166846,
      "loss": 5.4157,
      "mean_token_accuracy": 0.1954493373632431,
      "num_tokens": 38995418.0,
      "step": 20920
    },
    {
      "entropy": 5.858893203735351,
      "epoch": 1.8478452843518192,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004659042155680491,
      "loss": 5.339,
      "mean_token_accuracy": 0.20293375253677368,
      "num_tokens": 39004961.0,
      "step": 20925
    },
    {
      "entropy": 5.779239082336426,
      "epoch": 1.8482868244436594,
      "grad_norm": 1.0,
      "learning_rate": 0.0004658875466797085,
      "loss": 5.3175,
      "mean_token_accuracy": 0.2112161859869957,
      "num_tokens": 39014282.0,
      "step": 20930
    },
    {
      "entropy": 5.800938892364502,
      "epoch": 1.8487283645355,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046587087405198943,
      "loss": 5.3293,
      "mean_token_accuracy": 0.2036822706460953,
      "num_tokens": 39024480.0,
      "step": 20935
    },
    {
      "entropy": 5.891244697570801,
      "epoch": 1.8491699046273402,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00046585419768521854,
      "loss": 5.3606,
      "mean_token_accuracy": 0.20762478113174437,
      "num_tokens": 39034363.0,
      "step": 20940
    },
    {
      "entropy": 5.90705361366272,
      "epoch": 1.8496114447191805,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046583751757972244,
      "loss": 5.3102,
      "mean_token_accuracy": 0.210002301633358,
      "num_tokens": 39043823.0,
      "step": 20945
    },
    {
      "entropy": 5.8491464138031,
      "epoch": 1.850052984811021,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004658208337358279,
      "loss": 5.3048,
      "mean_token_accuracy": 0.20848426967859268,
      "num_tokens": 39052219.0,
      "step": 20950
    },
    {
      "entropy": 5.882801914215088,
      "epoch": 1.850494524902861,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046580414615386175,
      "loss": 5.4307,
      "mean_token_accuracy": 0.20012145340442658,
      "num_tokens": 39062974.0,
      "step": 20955
    },
    {
      "entropy": 5.717336177825928,
      "epoch": 1.8509360649947015,
      "grad_norm": 1.125,
      "learning_rate": 0.00046578745483415087,
      "loss": 5.1892,
      "mean_token_accuracy": 0.22310186475515364,
      "num_tokens": 39072178.0,
      "step": 20960
    },
    {
      "entropy": 5.8452311038970945,
      "epoch": 1.8513776050865418,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004657707597770223,
      "loss": 5.3498,
      "mean_token_accuracy": 0.2042410597205162,
      "num_tokens": 39081494.0,
      "step": 20965
    },
    {
      "entropy": 5.826963758468628,
      "epoch": 1.851819145178382,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000465754060982803,
      "loss": 5.3756,
      "mean_token_accuracy": 0.20149335712194444,
      "num_tokens": 39090783.0,
      "step": 20970
    },
    {
      "entropy": 5.85706934928894,
      "epoch": 1.8522606852702226,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004657373584518201,
      "loss": 5.3728,
      "mean_token_accuracy": 0.20822679102420807,
      "num_tokens": 39100150.0,
      "step": 20975
    },
    {
      "entropy": 5.936109209060669,
      "epoch": 1.8527022253620629,
      "grad_norm": 1.03125,
      "learning_rate": 0.000465720652184401,
      "loss": 5.4202,
      "mean_token_accuracy": 0.20123695582151413,
      "num_tokens": 39109419.0,
      "step": 20980
    },
    {
      "entropy": 5.829092693328858,
      "epoch": 1.8531437654539031,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004657039421808726,
      "loss": 5.2763,
      "mean_token_accuracy": 0.21640390008687974,
      "num_tokens": 39118312.0,
      "step": 20985
    },
    {
      "entropy": 5.874718379974365,
      "epoch": 1.8535853055457436,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046568722844156254,
      "loss": 5.4306,
      "mean_token_accuracy": 0.19490169882774352,
      "num_tokens": 39128169.0,
      "step": 20990
    },
    {
      "entropy": 5.848998928070069,
      "epoch": 1.854026845637584,
      "grad_norm": 1.0,
      "learning_rate": 0.00046567051096679807,
      "loss": 5.3597,
      "mean_token_accuracy": 0.2045652076601982,
      "num_tokens": 39136790.0,
      "step": 20995
    },
    {
      "entropy": 5.858442258834839,
      "epoch": 1.8544683857294242,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004656537897569068,
      "loss": 5.2841,
      "mean_token_accuracy": 0.21445168554782867,
      "num_tokens": 39146974.0,
      "step": 21000
    },
    {
      "epoch": 1.8544683857294242,
      "eval_entropy": 5.64301583370023,
      "eval_loss": 5.393889904022217,
      "eval_mean_token_accuracy": 0.2127020798530746,
      "eval_num_tokens": 39146974.0,
      "eval_runtime": 20.2421,
      "eval_samples_per_second": 1744.631,
      "eval_steps_per_second": 218.11,
      "step": 21000
    },
    {
      "entropy": 5.9857076644897464,
      "epoch": 1.8549099258212647,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004656370648122161,
      "loss": 5.4548,
      "mean_token_accuracy": 0.2057218387722969,
      "num_tokens": 39156910.0,
      "step": 21005
    },
    {
      "entropy": 5.912996578216553,
      "epoch": 1.855351465913105,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046562033613305377,
      "loss": 5.4353,
      "mean_token_accuracy": 0.20232681185007095,
      "num_tokens": 39165790.0,
      "step": 21010
    },
    {
      "entropy": 5.761199951171875,
      "epoch": 1.8557930060049452,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004656036037197475,
      "loss": 5.2633,
      "mean_token_accuracy": 0.22008705139160156,
      "num_tokens": 39174811.0,
      "step": 21015
    },
    {
      "entropy": 5.825755834579468,
      "epoch": 1.8562345460967857,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004655868675726251,
      "loss": 5.385,
      "mean_token_accuracy": 0.20006946474313736,
      "num_tokens": 39184325.0,
      "step": 21020
    },
    {
      "entropy": 5.925671625137329,
      "epoch": 1.8566760861886258,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046557012769201425,
      "loss": 5.3268,
      "mean_token_accuracy": 0.21048913300037383,
      "num_tokens": 39194879.0,
      "step": 21025
    },
    {
      "entropy": 5.843914127349853,
      "epoch": 1.8571176262804663,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000465553384078243,
      "loss": 5.3566,
      "mean_token_accuracy": 0.2091958314180374,
      "num_tokens": 39204702.0,
      "step": 21030
    },
    {
      "entropy": 5.831756210327148,
      "epoch": 1.8575591663723066,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004655366367316394,
      "loss": 5.3749,
      "mean_token_accuracy": 0.21127330064773558,
      "num_tokens": 39214561.0,
      "step": 21035
    },
    {
      "entropy": 5.90690426826477,
      "epoch": 1.8580007064641468,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046551988565253136,
      "loss": 5.4598,
      "mean_token_accuracy": 0.19376136362552643,
      "num_tokens": 39224411.0,
      "step": 21040
    },
    {
      "entropy": 5.877468824386597,
      "epoch": 1.8584422465559873,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004655031308412471,
      "loss": 5.3732,
      "mean_token_accuracy": 0.2032475307583809,
      "num_tokens": 39233422.0,
      "step": 21045
    },
    {
      "entropy": 5.9029398441314695,
      "epoch": 1.8588837866478276,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046548637229811495,
      "loss": 5.3929,
      "mean_token_accuracy": 0.2050471395254135,
      "num_tokens": 39243919.0,
      "step": 21050
    },
    {
      "entropy": 5.936050033569336,
      "epoch": 1.8593253267396679,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046546961002346304,
      "loss": 5.4114,
      "mean_token_accuracy": 0.20312367081642152,
      "num_tokens": 39253682.0,
      "step": 21055
    },
    {
      "entropy": 5.895182180404663,
      "epoch": 1.8597668668315084,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004654528440176198,
      "loss": 5.4499,
      "mean_token_accuracy": 0.20048205107450484,
      "num_tokens": 39263855.0,
      "step": 21060
    },
    {
      "entropy": 5.912688207626343,
      "epoch": 1.8602084069233487,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004654360742809137,
      "loss": 5.2943,
      "mean_token_accuracy": 0.21288205832242965,
      "num_tokens": 39272292.0,
      "step": 21065
    },
    {
      "entropy": 5.888539266586304,
      "epoch": 1.860649947015189,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004654193008136732,
      "loss": 5.3536,
      "mean_token_accuracy": 0.2024380460381508,
      "num_tokens": 39281321.0,
      "step": 21070
    },
    {
      "entropy": 5.936291694641113,
      "epoch": 1.8610914871070294,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046540252361622683,
      "loss": 5.45,
      "mean_token_accuracy": 0.19980594664812087,
      "num_tokens": 39292255.0,
      "step": 21075
    },
    {
      "entropy": 5.905332326889038,
      "epoch": 1.8615330271988697,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046538574268890334,
      "loss": 5.381,
      "mean_token_accuracy": 0.2068076640367508,
      "num_tokens": 39301841.0,
      "step": 21080
    },
    {
      "entropy": 5.877217149734497,
      "epoch": 1.86197456729071,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046536895803203144,
      "loss": 5.3715,
      "mean_token_accuracy": 0.20257276594638823,
      "num_tokens": 39311238.0,
      "step": 21085
    },
    {
      "entropy": 5.896858835220337,
      "epoch": 1.8624161073825505,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046535216964593997,
      "loss": 5.3787,
      "mean_token_accuracy": 0.20414014011621476,
      "num_tokens": 39319945.0,
      "step": 21090
    },
    {
      "entropy": 5.88555679321289,
      "epoch": 1.8628576474743905,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004653353775309576,
      "loss": 5.3437,
      "mean_token_accuracy": 0.20710037499666215,
      "num_tokens": 39329436.0,
      "step": 21095
    },
    {
      "entropy": 5.925942277908325,
      "epoch": 1.863299187566231,
      "grad_norm": 1.125,
      "learning_rate": 0.0004653185816874136,
      "loss": 5.3743,
      "mean_token_accuracy": 0.2059625506401062,
      "num_tokens": 39338447.0,
      "step": 21100
    },
    {
      "entropy": 5.82644624710083,
      "epoch": 1.8637407276580713,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004653017821156368,
      "loss": 5.2958,
      "mean_token_accuracy": 0.21237327456474303,
      "num_tokens": 39346744.0,
      "step": 21105
    },
    {
      "entropy": 5.853572511672974,
      "epoch": 1.8641822677499116,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004652849788159563,
      "loss": 5.406,
      "mean_token_accuracy": 0.19667177200317382,
      "num_tokens": 39357402.0,
      "step": 21110
    },
    {
      "entropy": 5.9240320205688475,
      "epoch": 1.864623807841752,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046526817178870123,
      "loss": 5.3913,
      "mean_token_accuracy": 0.1992943748831749,
      "num_tokens": 39366902.0,
      "step": 21115
    },
    {
      "entropy": 5.9621518611907955,
      "epoch": 1.8650653479335924,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046525136103420093,
      "loss": 5.3677,
      "mean_token_accuracy": 0.20447319000959396,
      "num_tokens": 39376330.0,
      "step": 21120
    },
    {
      "entropy": 5.8133303165435795,
      "epoch": 1.8655068880254326,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046523454655278466,
      "loss": 5.3279,
      "mean_token_accuracy": 0.21593234986066817,
      "num_tokens": 39385926.0,
      "step": 21125
    },
    {
      "entropy": 5.854966020584106,
      "epoch": 1.8659484281172731,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004652177283447818,
      "loss": 5.2904,
      "mean_token_accuracy": 0.20635225772857665,
      "num_tokens": 39395558.0,
      "step": 21130
    },
    {
      "entropy": 5.88421483039856,
      "epoch": 1.8663899682091134,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046520090641052187,
      "loss": 5.3576,
      "mean_token_accuracy": 0.2085113123059273,
      "num_tokens": 39405072.0,
      "step": 21135
    },
    {
      "entropy": 5.926157474517822,
      "epoch": 1.8668315083009537,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046518408075033437,
      "loss": 5.4929,
      "mean_token_accuracy": 0.19987283945083617,
      "num_tokens": 39414714.0,
      "step": 21140
    },
    {
      "entropy": 5.870391607284546,
      "epoch": 1.8672730483927942,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004651672513645488,
      "loss": 5.2767,
      "mean_token_accuracy": 0.21129160225391388,
      "num_tokens": 39423189.0,
      "step": 21145
    },
    {
      "entropy": 5.865139818191528,
      "epoch": 1.8677145884846345,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046515041825349505,
      "loss": 5.4277,
      "mean_token_accuracy": 0.19804446995258332,
      "num_tokens": 39432791.0,
      "step": 21150
    },
    {
      "entropy": 5.8997735500335695,
      "epoch": 1.8681561285764747,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004651335814175027,
      "loss": 5.3748,
      "mean_token_accuracy": 0.20265783816576005,
      "num_tokens": 39442849.0,
      "step": 21155
    },
    {
      "entropy": 5.897530889511108,
      "epoch": 1.8685976686683152,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046511674085690155,
      "loss": 5.3692,
      "mean_token_accuracy": 0.2050998792052269,
      "num_tokens": 39452171.0,
      "step": 21160
    },
    {
      "entropy": 5.926449632644653,
      "epoch": 1.8690392087601553,
      "grad_norm": 1.0,
      "learning_rate": 0.0004650998965720216,
      "loss": 5.3764,
      "mean_token_accuracy": 0.20553335398435593,
      "num_tokens": 39460351.0,
      "step": 21165
    },
    {
      "entropy": 5.940393590927124,
      "epoch": 1.8694807488519958,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046508304856319283,
      "loss": 5.441,
      "mean_token_accuracy": 0.2016812741756439,
      "num_tokens": 39470050.0,
      "step": 21170
    },
    {
      "entropy": 5.8886185646057125,
      "epoch": 1.869922288943836,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046506619683074526,
      "loss": 5.3666,
      "mean_token_accuracy": 0.2059075877070427,
      "num_tokens": 39480192.0,
      "step": 21175
    },
    {
      "entropy": 5.90000433921814,
      "epoch": 1.8703638290356763,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046504934137500903,
      "loss": 5.4128,
      "mean_token_accuracy": 0.20533685088157655,
      "num_tokens": 39490118.0,
      "step": 21180
    },
    {
      "entropy": 5.928255319595337,
      "epoch": 1.8708053691275168,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046503248219631417,
      "loss": 5.4288,
      "mean_token_accuracy": 0.1986628070473671,
      "num_tokens": 39499173.0,
      "step": 21185
    },
    {
      "entropy": 5.79513783454895,
      "epoch": 1.871246909219357,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046501561929499115,
      "loss": 5.2627,
      "mean_token_accuracy": 0.2171308770775795,
      "num_tokens": 39508595.0,
      "step": 21190
    },
    {
      "entropy": 5.8287464618682865,
      "epoch": 1.8716884493111974,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046499875267137016,
      "loss": 5.3347,
      "mean_token_accuracy": 0.2113501623272896,
      "num_tokens": 39519300.0,
      "step": 21195
    },
    {
      "entropy": 5.902194929122925,
      "epoch": 1.8721299894030379,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004649818823257817,
      "loss": 5.3783,
      "mean_token_accuracy": 0.20862430334091187,
      "num_tokens": 39527722.0,
      "step": 21200
    },
    {
      "entropy": 5.890845537185669,
      "epoch": 1.8725715294948782,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004649650082585562,
      "loss": 5.3559,
      "mean_token_accuracy": 0.20628983825445174,
      "num_tokens": 39537063.0,
      "step": 21205
    },
    {
      "entropy": 5.8790748596191404,
      "epoch": 1.8730130695867184,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046494813047002424,
      "loss": 5.3915,
      "mean_token_accuracy": 0.2070586144924164,
      "num_tokens": 39546402.0,
      "step": 21210
    },
    {
      "entropy": 5.90912446975708,
      "epoch": 1.873454609678559,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046493124896051634,
      "loss": 5.38,
      "mean_token_accuracy": 0.20982841402292252,
      "num_tokens": 39555877.0,
      "step": 21215
    },
    {
      "entropy": 5.839294862747193,
      "epoch": 1.8738961497703992,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046491436373036346,
      "loss": 5.3273,
      "mean_token_accuracy": 0.20793818384408952,
      "num_tokens": 39565153.0,
      "step": 21220
    },
    {
      "entropy": 5.8650837421417235,
      "epoch": 1.8743376898622395,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004648974747798961,
      "loss": 5.3464,
      "mean_token_accuracy": 0.21023863106966018,
      "num_tokens": 39573775.0,
      "step": 21225
    },
    {
      "entropy": 5.9614763259887695,
      "epoch": 1.87477922995408,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004648805821094452,
      "loss": 5.4367,
      "mean_token_accuracy": 0.20038348138332368,
      "num_tokens": 39583159.0,
      "step": 21230
    },
    {
      "entropy": 6.000168514251709,
      "epoch": 1.87522077004592,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046486368571934175,
      "loss": 5.4559,
      "mean_token_accuracy": 0.19545531868934632,
      "num_tokens": 39592545.0,
      "step": 21235
    },
    {
      "entropy": 5.721289873123169,
      "epoch": 1.8756623101377605,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004648467856099167,
      "loss": 5.1806,
      "mean_token_accuracy": 0.22161275744438172,
      "num_tokens": 39602059.0,
      "step": 21240
    },
    {
      "entropy": 5.847084808349609,
      "epoch": 1.8761038502296008,
      "grad_norm": 1.015625,
      "learning_rate": 0.000464829881781501,
      "loss": 5.3379,
      "mean_token_accuracy": 0.20980760604143142,
      "num_tokens": 39610777.0,
      "step": 21245
    },
    {
      "entropy": 5.907437419891357,
      "epoch": 1.876545390321441,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046481297423442593,
      "loss": 5.4176,
      "mean_token_accuracy": 0.20618286430835725,
      "num_tokens": 39620782.0,
      "step": 21250
    },
    {
      "entropy": 6.001585054397583,
      "epoch": 1.8769869304132816,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004647960629690226,
      "loss": 5.4581,
      "mean_token_accuracy": 0.19751667380332946,
      "num_tokens": 39631340.0,
      "step": 21255
    },
    {
      "entropy": 5.867327404022217,
      "epoch": 1.8774284705051218,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004647791479856223,
      "loss": 5.2793,
      "mean_token_accuracy": 0.21446926891803741,
      "num_tokens": 39640361.0,
      "step": 21260
    },
    {
      "entropy": 5.942128372192383,
      "epoch": 1.8778700105969621,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004647622292845566,
      "loss": 5.3487,
      "mean_token_accuracy": 0.19976659417152404,
      "num_tokens": 39649726.0,
      "step": 21265
    },
    {
      "entropy": 5.884883880615234,
      "epoch": 1.8783115506888026,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004647453068661566,
      "loss": 5.3979,
      "mean_token_accuracy": 0.20370790958404542,
      "num_tokens": 39660226.0,
      "step": 21270
    },
    {
      "entropy": 5.824184226989746,
      "epoch": 1.878753090780643,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046472838073075397,
      "loss": 5.2184,
      "mean_token_accuracy": 0.21244992017745973,
      "num_tokens": 39668731.0,
      "step": 21275
    },
    {
      "entropy": 5.834587097167969,
      "epoch": 1.8791946308724832,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046471145087868034,
      "loss": 5.3708,
      "mean_token_accuracy": 0.20916611552238465,
      "num_tokens": 39678465.0,
      "step": 21280
    },
    {
      "entropy": 5.821716356277466,
      "epoch": 1.8796361709643237,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046469451731026725,
      "loss": 5.2415,
      "mean_token_accuracy": 0.2200051501393318,
      "num_tokens": 39687057.0,
      "step": 21285
    },
    {
      "entropy": 5.867373943328857,
      "epoch": 1.880077711056164,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004646775800258464,
      "loss": 5.3661,
      "mean_token_accuracy": 0.20868303030729293,
      "num_tokens": 39696387.0,
      "step": 21290
    },
    {
      "entropy": 5.875208377838135,
      "epoch": 1.8805192511480042,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046466063902574973,
      "loss": 5.3156,
      "mean_token_accuracy": 0.21101562827825546,
      "num_tokens": 39705756.0,
      "step": 21295
    },
    {
      "entropy": 5.863362646102905,
      "epoch": 1.8809607912398447,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046464369431030894,
      "loss": 5.3434,
      "mean_token_accuracy": 0.2099301591515541,
      "num_tokens": 39713916.0,
      "step": 21300
    },
    {
      "entropy": 5.8952765464782715,
      "epoch": 1.8814023313316848,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000464626745879856,
      "loss": 5.4328,
      "mean_token_accuracy": 0.197401724755764,
      "num_tokens": 39722920.0,
      "step": 21305
    },
    {
      "entropy": 5.950076246261597,
      "epoch": 1.8818438714235253,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000464609793734723,
      "loss": 5.4163,
      "mean_token_accuracy": 0.199638032913208,
      "num_tokens": 39732217.0,
      "step": 21310
    },
    {
      "entropy": 5.821593332290649,
      "epoch": 1.8822854115153655,
      "grad_norm": 1.0,
      "learning_rate": 0.000464592837875242,
      "loss": 5.2585,
      "mean_token_accuracy": 0.21253640949726105,
      "num_tokens": 39741636.0,
      "step": 21315
    },
    {
      "entropy": 5.897186279296875,
      "epoch": 1.8827269516072058,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004645758783017451,
      "loss": 5.3943,
      "mean_token_accuracy": 0.20481854379177095,
      "num_tokens": 39751457.0,
      "step": 21320
    },
    {
      "entropy": 5.972907638549804,
      "epoch": 1.8831684916990463,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046455891501456456,
      "loss": 5.3531,
      "mean_token_accuracy": 0.20630601793527603,
      "num_tokens": 39760308.0,
      "step": 21325
    },
    {
      "entropy": 5.8535772323608395,
      "epoch": 1.8836100317908866,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046454194801403274,
      "loss": 5.3037,
      "mean_token_accuracy": 0.21301157027482986,
      "num_tokens": 39770002.0,
      "step": 21330
    },
    {
      "entropy": 5.815937948226929,
      "epoch": 1.8840515718827269,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004645249773004819,
      "loss": 5.3369,
      "mean_token_accuracy": 0.20491984188556672,
      "num_tokens": 39779043.0,
      "step": 21335
    },
    {
      "entropy": 5.884713268280029,
      "epoch": 1.8844931119745674,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046450800287424455,
      "loss": 5.3563,
      "mean_token_accuracy": 0.20212042480707168,
      "num_tokens": 39788545.0,
      "step": 21340
    },
    {
      "entropy": 5.865634202957153,
      "epoch": 1.8849346520664076,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004644910247356532,
      "loss": 5.273,
      "mean_token_accuracy": 0.21859100908041001,
      "num_tokens": 39798063.0,
      "step": 21345
    },
    {
      "entropy": 5.872667264938355,
      "epoch": 1.885376192158248,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004644740428850404,
      "loss": 5.2906,
      "mean_token_accuracy": 0.21180068403482438,
      "num_tokens": 39808622.0,
      "step": 21350
    },
    {
      "entropy": 5.9357787609100345,
      "epoch": 1.8858177322500884,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004644570573227389,
      "loss": 5.4309,
      "mean_token_accuracy": 0.19891026765108108,
      "num_tokens": 39817608.0,
      "step": 21355
    },
    {
      "entropy": 5.848351001739502,
      "epoch": 1.8862592723419287,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046444006804908144,
      "loss": 5.2401,
      "mean_token_accuracy": 0.22318484485149384,
      "num_tokens": 39826055.0,
      "step": 21360
    },
    {
      "entropy": 5.762154769897461,
      "epoch": 1.886700812433769,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004644230750644007,
      "loss": 5.2688,
      "mean_token_accuracy": 0.21999183297157288,
      "num_tokens": 39835490.0,
      "step": 21365
    },
    {
      "entropy": 5.827249479293823,
      "epoch": 1.8871423525256095,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046440607836902974,
      "loss": 5.3533,
      "mean_token_accuracy": 0.21116445511579512,
      "num_tokens": 39845613.0,
      "step": 21370
    },
    {
      "entropy": 5.91176815032959,
      "epoch": 1.8875838926174495,
      "grad_norm": 1.125,
      "learning_rate": 0.00046438907796330145,
      "loss": 5.3532,
      "mean_token_accuracy": 0.20562867373228072,
      "num_tokens": 39854495.0,
      "step": 21375
    },
    {
      "entropy": 5.876791667938233,
      "epoch": 1.88802543270929,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046437207384754874,
      "loss": 5.3752,
      "mean_token_accuracy": 0.20743363797664643,
      "num_tokens": 39864528.0,
      "step": 21380
    },
    {
      "entropy": 5.925540733337402,
      "epoch": 1.8884669728011303,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004643550660221048,
      "loss": 5.4246,
      "mean_token_accuracy": 0.20055753737688065,
      "num_tokens": 39873780.0,
      "step": 21385
    },
    {
      "entropy": 5.8628504276275635,
      "epoch": 1.8889085128929706,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046433805448730287,
      "loss": 5.2423,
      "mean_token_accuracy": 0.21738229244947432,
      "num_tokens": 39882529.0,
      "step": 21390
    },
    {
      "entropy": 5.947251176834106,
      "epoch": 1.889350052984811,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004643210392434761,
      "loss": 5.3834,
      "mean_token_accuracy": 0.2069010093808174,
      "num_tokens": 39890688.0,
      "step": 21395
    },
    {
      "entropy": 5.857395315170288,
      "epoch": 1.8897915930766513,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046430402029095784,
      "loss": 5.4122,
      "mean_token_accuracy": 0.19547487795352936,
      "num_tokens": 39900849.0,
      "step": 21400
    },
    {
      "entropy": 5.800002336502075,
      "epoch": 1.8902331331684916,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004642869976300815,
      "loss": 5.3773,
      "mean_token_accuracy": 0.20372939556837083,
      "num_tokens": 39911297.0,
      "step": 21405
    },
    {
      "entropy": 5.818081521987915,
      "epoch": 1.8906746732603321,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004642699712611806,
      "loss": 5.3169,
      "mean_token_accuracy": 0.20706899762153624,
      "num_tokens": 39920312.0,
      "step": 21410
    },
    {
      "entropy": 5.865331077575684,
      "epoch": 1.8911162133521724,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004642529411845885,
      "loss": 5.3976,
      "mean_token_accuracy": 0.20539042204618455,
      "num_tokens": 39930128.0,
      "step": 21415
    },
    {
      "entropy": 5.919701719284058,
      "epoch": 1.8915577534440127,
      "grad_norm": 1.0,
      "learning_rate": 0.0004642359074006389,
      "loss": 5.3389,
      "mean_token_accuracy": 0.21122706234455108,
      "num_tokens": 39939177.0,
      "step": 21420
    },
    {
      "entropy": 5.946272134780884,
      "epoch": 1.8919992935358532,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004642188699096656,
      "loss": 5.431,
      "mean_token_accuracy": 0.19848523736000062,
      "num_tokens": 39948537.0,
      "step": 21425
    },
    {
      "entropy": 5.893461275100708,
      "epoch": 1.8924408336276934,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004642018287120022,
      "loss": 5.3503,
      "mean_token_accuracy": 0.20792061239480972,
      "num_tokens": 39958623.0,
      "step": 21430
    },
    {
      "entropy": 5.926222562789917,
      "epoch": 1.8928823737195337,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046418478380798263,
      "loss": 5.4411,
      "mean_token_accuracy": 0.2013219639658928,
      "num_tokens": 39968736.0,
      "step": 21435
    },
    {
      "entropy": 5.769104671478272,
      "epoch": 1.8933239138113742,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004641677351979407,
      "loss": 5.2944,
      "mean_token_accuracy": 0.20479418337345123,
      "num_tokens": 39977600.0,
      "step": 21440
    },
    {
      "entropy": 5.847313785552979,
      "epoch": 1.8937654539032143,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004641506828822104,
      "loss": 5.402,
      "mean_token_accuracy": 0.1978820577263832,
      "num_tokens": 39987808.0,
      "step": 21445
    },
    {
      "entropy": 5.967379426956176,
      "epoch": 1.8942069939950548,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004641336268611258,
      "loss": 5.5056,
      "mean_token_accuracy": 0.20149057656526564,
      "num_tokens": 39998553.0,
      "step": 21450
    },
    {
      "entropy": 6.031639814376831,
      "epoch": 1.894648534086895,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046411656713502106,
      "loss": 5.4261,
      "mean_token_accuracy": 0.19868351370096207,
      "num_tokens": 40007529.0,
      "step": 21455
    },
    {
      "entropy": 5.936395454406738,
      "epoch": 1.8950900741787353,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004640995037042303,
      "loss": 5.4313,
      "mean_token_accuracy": 0.20063990950584412,
      "num_tokens": 40017116.0,
      "step": 21460
    },
    {
      "entropy": 5.888848066329956,
      "epoch": 1.8955316142705758,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004640824365690878,
      "loss": 5.4385,
      "mean_token_accuracy": 0.20070538073778152,
      "num_tokens": 40026601.0,
      "step": 21465
    },
    {
      "entropy": 5.894359588623047,
      "epoch": 1.895973154362416,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046406536572992795,
      "loss": 5.3302,
      "mean_token_accuracy": 0.21494613885879515,
      "num_tokens": 40036029.0,
      "step": 21470
    },
    {
      "entropy": 5.902152967453003,
      "epoch": 1.8964146944542564,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000464048291187085,
      "loss": 5.3687,
      "mean_token_accuracy": 0.20566665828227998,
      "num_tokens": 40045892.0,
      "step": 21475
    },
    {
      "entropy": 5.9180056095123295,
      "epoch": 1.8968562345460969,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004640312129408936,
      "loss": 5.3119,
      "mean_token_accuracy": 0.21512603461742402,
      "num_tokens": 40055411.0,
      "step": 21480
    },
    {
      "entropy": 5.936558055877685,
      "epoch": 1.8972977746379371,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004640141309916883,
      "loss": 5.405,
      "mean_token_accuracy": 0.2002963423728943,
      "num_tokens": 40063988.0,
      "step": 21485
    },
    {
      "entropy": 5.892999792098999,
      "epoch": 1.8977393147297774,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004639970453398037,
      "loss": 5.354,
      "mean_token_accuracy": 0.2138413041830063,
      "num_tokens": 40072671.0,
      "step": 21490
    },
    {
      "entropy": 5.915765333175659,
      "epoch": 1.898180854821618,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046397995598557434,
      "loss": 5.4451,
      "mean_token_accuracy": 0.20091729909181594,
      "num_tokens": 40082191.0,
      "step": 21495
    },
    {
      "entropy": 5.877042198181153,
      "epoch": 1.8986223949134582,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004639628629293352,
      "loss": 5.4021,
      "mean_token_accuracy": 0.19798944145441055,
      "num_tokens": 40090823.0,
      "step": 21500
    },
    {
      "entropy": 5.910460090637207,
      "epoch": 1.8990639350052985,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000463945766171421,
      "loss": 5.3423,
      "mean_token_accuracy": 0.2063678190112114,
      "num_tokens": 40099825.0,
      "step": 21505
    },
    {
      "entropy": 5.934996223449707,
      "epoch": 1.899505475097139,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004639286657121668,
      "loss": 5.3691,
      "mean_token_accuracy": 0.20978682190179826,
      "num_tokens": 40109589.0,
      "step": 21510
    },
    {
      "entropy": 5.848345184326172,
      "epoch": 1.899947015188979,
      "grad_norm": 1.0,
      "learning_rate": 0.0004639115615519074,
      "loss": 5.3202,
      "mean_token_accuracy": 0.2055730938911438,
      "num_tokens": 40119194.0,
      "step": 21515
    },
    {
      "entropy": 5.886260843276977,
      "epoch": 1.9003885552808195,
      "grad_norm": 1.0625,
      "learning_rate": 0.000463894453690978,
      "loss": 5.4047,
      "mean_token_accuracy": 0.20396253168582917,
      "num_tokens": 40128184.0,
      "step": 21520
    },
    {
      "entropy": 5.89281997680664,
      "epoch": 1.9008300953726598,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046387734212971363,
      "loss": 5.419,
      "mean_token_accuracy": 0.19890098571777343,
      "num_tokens": 40137842.0,
      "step": 21525
    },
    {
      "entropy": 5.902789640426636,
      "epoch": 1.9012716354645,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004638602268684496,
      "loss": 5.3791,
      "mean_token_accuracy": 0.2025577872991562,
      "num_tokens": 40147151.0,
      "step": 21530
    },
    {
      "entropy": 5.902479410171509,
      "epoch": 1.9017131755563406,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046384310790752105,
      "loss": 5.3458,
      "mean_token_accuracy": 0.20398520231246947,
      "num_tokens": 40157170.0,
      "step": 21535
    },
    {
      "entropy": 5.898153638839721,
      "epoch": 1.9021547156481808,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004638259852472635,
      "loss": 5.3335,
      "mean_token_accuracy": 0.20552589893341064,
      "num_tokens": 40166787.0,
      "step": 21540
    },
    {
      "entropy": 5.887764644622803,
      "epoch": 1.9025962557400211,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004638088588880122,
      "loss": 5.2987,
      "mean_token_accuracy": 0.21487546265125274,
      "num_tokens": 40176640.0,
      "step": 21545
    },
    {
      "entropy": 5.88833384513855,
      "epoch": 1.9030377958318616,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004637917288301029,
      "loss": 5.3655,
      "mean_token_accuracy": 0.21044817119836806,
      "num_tokens": 40185732.0,
      "step": 21550
    },
    {
      "entropy": 5.9018439769744875,
      "epoch": 1.9034793359237019,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004637745950738709,
      "loss": 5.2949,
      "mean_token_accuracy": 0.21157171428203583,
      "num_tokens": 40194798.0,
      "step": 21555
    },
    {
      "entropy": 5.92890567779541,
      "epoch": 1.9039208760155422,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004637574576196519,
      "loss": 5.4597,
      "mean_token_accuracy": 0.20000536739826202,
      "num_tokens": 40204507.0,
      "step": 21560
    },
    {
      "entropy": 5.93493914604187,
      "epoch": 1.9043624161073827,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004637403164677817,
      "loss": 5.3832,
      "mean_token_accuracy": 0.1958839178085327,
      "num_tokens": 40213546.0,
      "step": 21565
    },
    {
      "entropy": 5.892339468002319,
      "epoch": 1.904803956199223,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046372317161859605,
      "loss": 5.3211,
      "mean_token_accuracy": 0.20888868272304534,
      "num_tokens": 40222571.0,
      "step": 21570
    },
    {
      "entropy": 5.944922876358032,
      "epoch": 1.9052454962910632,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046370602307243083,
      "loss": 5.4081,
      "mean_token_accuracy": 0.20033878684043885,
      "num_tokens": 40231803.0,
      "step": 21575
    },
    {
      "entropy": 5.9365276336669925,
      "epoch": 1.9056870363829037,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046368887082962187,
      "loss": 5.4632,
      "mean_token_accuracy": 0.20124018490314483,
      "num_tokens": 40241755.0,
      "step": 21580
    },
    {
      "entropy": 5.888432312011719,
      "epoch": 1.9061285764747438,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004636717148905053,
      "loss": 5.3514,
      "mean_token_accuracy": 0.2052123323082924,
      "num_tokens": 40250963.0,
      "step": 21585
    },
    {
      "entropy": 5.93085503578186,
      "epoch": 1.9065701165665843,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004636545552554171,
      "loss": 5.347,
      "mean_token_accuracy": 0.20675989240407944,
      "num_tokens": 40260626.0,
      "step": 21590
    },
    {
      "entropy": 5.904647779464722,
      "epoch": 1.9070116566584245,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004636373919246935,
      "loss": 5.3249,
      "mean_token_accuracy": 0.20984554588794707,
      "num_tokens": 40269852.0,
      "step": 21595
    },
    {
      "entropy": 5.87044882774353,
      "epoch": 1.9074531967502648,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004636202248986707,
      "loss": 5.3358,
      "mean_token_accuracy": 0.2062729373574257,
      "num_tokens": 40279499.0,
      "step": 21600
    },
    {
      "entropy": 5.965744924545288,
      "epoch": 1.9078947368421053,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046360305417768484,
      "loss": 5.4459,
      "mean_token_accuracy": 0.19920772910118104,
      "num_tokens": 40288131.0,
      "step": 21605
    },
    {
      "entropy": 5.970402812957763,
      "epoch": 1.9083362769339456,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004635858797620725,
      "loss": 5.387,
      "mean_token_accuracy": 0.20197357684373857,
      "num_tokens": 40296951.0,
      "step": 21610
    },
    {
      "entropy": 5.882398843765259,
      "epoch": 1.9087778170257859,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046356870165217004,
      "loss": 5.3202,
      "mean_token_accuracy": 0.21134337037801743,
      "num_tokens": 40306250.0,
      "step": 21615
    },
    {
      "entropy": 5.811308526992798,
      "epoch": 1.9092193571176264,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046355151984831395,
      "loss": 5.2959,
      "mean_token_accuracy": 0.21645228415727616,
      "num_tokens": 40314561.0,
      "step": 21620
    },
    {
      "entropy": 5.788237428665161,
      "epoch": 1.9096608972094666,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046353433435084083,
      "loss": 5.2602,
      "mean_token_accuracy": 0.20963333547115326,
      "num_tokens": 40323580.0,
      "step": 21625
    },
    {
      "entropy": 5.838340187072754,
      "epoch": 1.910102437301307,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004635171451600872,
      "loss": 5.3438,
      "mean_token_accuracy": 0.20755885988473893,
      "num_tokens": 40333747.0,
      "step": 21630
    },
    {
      "entropy": 5.965235948562622,
      "epoch": 1.9105439773931474,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046349995227639015,
      "loss": 5.4485,
      "mean_token_accuracy": 0.19505509734153748,
      "num_tokens": 40343435.0,
      "step": 21635
    },
    {
      "entropy": 5.988936710357666,
      "epoch": 1.9109855174849877,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004634827557000861,
      "loss": 5.4062,
      "mean_token_accuracy": 0.20402436703443527,
      "num_tokens": 40353282.0,
      "step": 21640
    },
    {
      "entropy": 5.904768800735473,
      "epoch": 1.911427057576828,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000463465555431512,
      "loss": 5.4503,
      "mean_token_accuracy": 0.1991128444671631,
      "num_tokens": 40363781.0,
      "step": 21645
    },
    {
      "entropy": 5.879390859603882,
      "epoch": 1.9118685976686685,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046344835147100495,
      "loss": 5.3562,
      "mean_token_accuracy": 0.20000342428684234,
      "num_tokens": 40373240.0,
      "step": 21650
    },
    {
      "entropy": 5.894801187515259,
      "epoch": 1.9123101377605085,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046343114381890187,
      "loss": 5.4388,
      "mean_token_accuracy": 0.20514141023159027,
      "num_tokens": 40383088.0,
      "step": 21655
    },
    {
      "entropy": 5.984843826293945,
      "epoch": 1.912751677852349,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004634139324755398,
      "loss": 5.3974,
      "mean_token_accuracy": 0.21017206609249114,
      "num_tokens": 40391654.0,
      "step": 21660
    },
    {
      "entropy": 5.879476881027221,
      "epoch": 1.9131932179441893,
      "grad_norm": 1.125,
      "learning_rate": 0.000463396717441256,
      "loss": 5.2974,
      "mean_token_accuracy": 0.2165357306599617,
      "num_tokens": 40400616.0,
      "step": 21665
    },
    {
      "entropy": 5.811811876296997,
      "epoch": 1.9136347580360296,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004633794987163876,
      "loss": 5.3681,
      "mean_token_accuracy": 0.19875801354646683,
      "num_tokens": 40411271.0,
      "step": 21670
    },
    {
      "entropy": 5.89321117401123,
      "epoch": 1.91407629812787,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046336227630127205,
      "loss": 5.2857,
      "mean_token_accuracy": 0.20427657961845397,
      "num_tokens": 40420019.0,
      "step": 21675
    },
    {
      "entropy": 5.9374107837677,
      "epoch": 1.9145178382197103,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046334505019624656,
      "loss": 5.3641,
      "mean_token_accuracy": 0.20939703285694122,
      "num_tokens": 40429858.0,
      "step": 21680
    },
    {
      "entropy": 5.84266414642334,
      "epoch": 1.9149593783115506,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046332782040164867,
      "loss": 5.3403,
      "mean_token_accuracy": 0.21174548715353012,
      "num_tokens": 40439566.0,
      "step": 21685
    },
    {
      "entropy": 5.845394515991211,
      "epoch": 1.915400918403391,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004633105869178159,
      "loss": 5.2912,
      "mean_token_accuracy": 0.21488560289144515,
      "num_tokens": 40448527.0,
      "step": 21690
    },
    {
      "entropy": 5.875892210006714,
      "epoch": 1.9158424584952314,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004632933497450858,
      "loss": 5.3107,
      "mean_token_accuracy": 0.21092088520526886,
      "num_tokens": 40458076.0,
      "step": 21695
    },
    {
      "entropy": 5.843246698379517,
      "epoch": 1.9162839985870717,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004632761088837961,
      "loss": 5.2335,
      "mean_token_accuracy": 0.22885193526744843,
      "num_tokens": 40467639.0,
      "step": 21700
    },
    {
      "entropy": 5.881030607223511,
      "epoch": 1.9167255386789122,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046325886433428446,
      "loss": 5.3242,
      "mean_token_accuracy": 0.20737940818071365,
      "num_tokens": 40476243.0,
      "step": 21705
    },
    {
      "entropy": 5.891556739807129,
      "epoch": 1.9171670787707524,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046324161609688876,
      "loss": 5.3737,
      "mean_token_accuracy": 0.2093565955758095,
      "num_tokens": 40485789.0,
      "step": 21710
    },
    {
      "entropy": 5.822904968261719,
      "epoch": 1.9176086188625927,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004632243641719469,
      "loss": 5.3087,
      "mean_token_accuracy": 0.20545696020126342,
      "num_tokens": 40494139.0,
      "step": 21715
    },
    {
      "entropy": 5.857991886138916,
      "epoch": 1.9180501589544332,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046320710855979677,
      "loss": 5.3234,
      "mean_token_accuracy": 0.2088524729013443,
      "num_tokens": 40503965.0,
      "step": 21720
    },
    {
      "entropy": 5.859075736999512,
      "epoch": 1.9184916990462733,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046318984926077637,
      "loss": 5.3339,
      "mean_token_accuracy": 0.21351005136966705,
      "num_tokens": 40513140.0,
      "step": 21725
    },
    {
      "entropy": 5.830721712112426,
      "epoch": 1.9189332391381138,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046317258627522393,
      "loss": 5.334,
      "mean_token_accuracy": 0.2193341925740242,
      "num_tokens": 40522858.0,
      "step": 21730
    },
    {
      "entropy": 5.832539844512939,
      "epoch": 1.919374779229954,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004631553196034774,
      "loss": 5.3281,
      "mean_token_accuracy": 0.21606432199478148,
      "num_tokens": 40532050.0,
      "step": 21735
    },
    {
      "entropy": 5.85909743309021,
      "epoch": 1.9198163193217943,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004631380492458753,
      "loss": 5.3193,
      "mean_token_accuracy": 0.20459819585084915,
      "num_tokens": 40540942.0,
      "step": 21740
    },
    {
      "entropy": 5.9041587829589846,
      "epoch": 1.9202578594136348,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004631207752027558,
      "loss": 5.3587,
      "mean_token_accuracy": 0.2107951134443283,
      "num_tokens": 40550158.0,
      "step": 21745
    },
    {
      "entropy": 5.971966934204102,
      "epoch": 1.920699399505475,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004631034974744573,
      "loss": 5.4672,
      "mean_token_accuracy": 0.19536066353321074,
      "num_tokens": 40559267.0,
      "step": 21750
    },
    {
      "entropy": 5.903441667556763,
      "epoch": 1.9211409395973154,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004630862160613183,
      "loss": 5.362,
      "mean_token_accuracy": 0.20459360331296922,
      "num_tokens": 40568503.0,
      "step": 21755
    },
    {
      "entropy": 5.877523279190063,
      "epoch": 1.9215824796891559,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004630689309636772,
      "loss": 5.3063,
      "mean_token_accuracy": 0.21025119721889496,
      "num_tokens": 40578464.0,
      "step": 21760
    },
    {
      "entropy": 5.955185747146606,
      "epoch": 1.9220240197809961,
      "grad_norm": 1.0,
      "learning_rate": 0.0004630516421818728,
      "loss": 5.4656,
      "mean_token_accuracy": 0.1963024139404297,
      "num_tokens": 40588338.0,
      "step": 21765
    },
    {
      "entropy": 5.934965562820435,
      "epoch": 1.9224655598728364,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046303434971624357,
      "loss": 5.3346,
      "mean_token_accuracy": 0.21062635779380798,
      "num_tokens": 40598912.0,
      "step": 21770
    },
    {
      "entropy": 5.812967967987061,
      "epoch": 1.922907099964677,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046301705356712846,
      "loss": 5.2545,
      "mean_token_accuracy": 0.2087301179766655,
      "num_tokens": 40607084.0,
      "step": 21775
    },
    {
      "entropy": 5.886069917678833,
      "epoch": 1.9233486400565172,
      "grad_norm": 1.125,
      "learning_rate": 0.00046299975373486626,
      "loss": 5.3648,
      "mean_token_accuracy": 0.2038731813430786,
      "num_tokens": 40616422.0,
      "step": 21780
    },
    {
      "entropy": 5.845412397384644,
      "epoch": 1.9237901801483575,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004629824502197956,
      "loss": 5.3053,
      "mean_token_accuracy": 0.20799794048070908,
      "num_tokens": 40626189.0,
      "step": 21785
    },
    {
      "entropy": 5.83835654258728,
      "epoch": 1.924231720240198,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046296514302225585,
      "loss": 5.2614,
      "mean_token_accuracy": 0.20808082967996597,
      "num_tokens": 40636053.0,
      "step": 21790
    },
    {
      "entropy": 5.927354049682617,
      "epoch": 1.924673260332038,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004629478321425858,
      "loss": 5.4047,
      "mean_token_accuracy": 0.19723138511180877,
      "num_tokens": 40645159.0,
      "step": 21795
    },
    {
      "entropy": 5.872242498397827,
      "epoch": 1.9251148004238785,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046293051758112457,
      "loss": 5.2596,
      "mean_token_accuracy": 0.2172661915421486,
      "num_tokens": 40653890.0,
      "step": 21800
    },
    {
      "entropy": 5.955357456207276,
      "epoch": 1.9255563405157188,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046291319933821143,
      "loss": 5.4262,
      "mean_token_accuracy": 0.19706215262413024,
      "num_tokens": 40663041.0,
      "step": 21805
    },
    {
      "entropy": 5.833741331100464,
      "epoch": 1.925997880607559,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046289587741418553,
      "loss": 5.2702,
      "mean_token_accuracy": 0.2103172242641449,
      "num_tokens": 40670972.0,
      "step": 21810
    },
    {
      "entropy": 5.877347087860107,
      "epoch": 1.9264394206993996,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046287855180938633,
      "loss": 5.3996,
      "mean_token_accuracy": 0.20953988283872604,
      "num_tokens": 40680503.0,
      "step": 21815
    },
    {
      "entropy": 5.869355058670044,
      "epoch": 1.9268809607912398,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046286122252415303,
      "loss": 5.3944,
      "mean_token_accuracy": 0.2054565280675888,
      "num_tokens": 40691339.0,
      "step": 21820
    },
    {
      "entropy": 5.9580910205841064,
      "epoch": 1.92732250088308,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046284388955882526,
      "loss": 5.3525,
      "mean_token_accuracy": 0.2055668607354164,
      "num_tokens": 40699744.0,
      "step": 21825
    },
    {
      "entropy": 5.965705442428589,
      "epoch": 1.9277640409749206,
      "grad_norm": 1.0,
      "learning_rate": 0.00046282655291374257,
      "loss": 5.4008,
      "mean_token_accuracy": 0.19989280253648758,
      "num_tokens": 40709144.0,
      "step": 21830
    },
    {
      "entropy": 5.842627954483032,
      "epoch": 1.9282055810667609,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00046280921258924455,
      "loss": 5.4256,
      "mean_token_accuracy": 0.20699867308139802,
      "num_tokens": 40719513.0,
      "step": 21835
    },
    {
      "entropy": 5.872743844985962,
      "epoch": 1.9286471211586012,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046279186858567083,
      "loss": 5.3509,
      "mean_token_accuracy": 0.20761857479810714,
      "num_tokens": 40728846.0,
      "step": 21840
    },
    {
      "entropy": 5.857518672943115,
      "epoch": 1.9290886612504416,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004627745209033613,
      "loss": 5.27,
      "mean_token_accuracy": 0.21707218736410142,
      "num_tokens": 40737904.0,
      "step": 21845
    },
    {
      "entropy": 5.819572305679321,
      "epoch": 1.929530201342282,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004627571695426556,
      "loss": 5.2947,
      "mean_token_accuracy": 0.21966431438922882,
      "num_tokens": 40747328.0,
      "step": 21850
    },
    {
      "entropy": 5.912659072875977,
      "epoch": 1.9299717414341222,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046273981450389374,
      "loss": 5.3293,
      "mean_token_accuracy": 0.2070286139845848,
      "num_tokens": 40755567.0,
      "step": 21855
    },
    {
      "entropy": 5.850630807876587,
      "epoch": 1.9304132815259627,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046272245578741567,
      "loss": 5.3404,
      "mean_token_accuracy": 0.20996105968952178,
      "num_tokens": 40764684.0,
      "step": 21860
    },
    {
      "entropy": 5.876852512359619,
      "epoch": 1.9308548216178028,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004627050933935615,
      "loss": 5.3712,
      "mean_token_accuracy": 0.21365745067596437,
      "num_tokens": 40774050.0,
      "step": 21865
    },
    {
      "entropy": 5.871454763412475,
      "epoch": 1.9312963617096433,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046268772732267126,
      "loss": 5.3103,
      "mean_token_accuracy": 0.21163002997636796,
      "num_tokens": 40782722.0,
      "step": 21870
    },
    {
      "entropy": 5.886638879776001,
      "epoch": 1.9317379018014835,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004626703575750852,
      "loss": 5.3838,
      "mean_token_accuracy": 0.20145509093999864,
      "num_tokens": 40791978.0,
      "step": 21875
    },
    {
      "entropy": 5.892842483520508,
      "epoch": 1.9321794418933238,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004626529841511436,
      "loss": 5.3909,
      "mean_token_accuracy": 0.20379009544849397,
      "num_tokens": 40801855.0,
      "step": 21880
    },
    {
      "entropy": 5.903077697753906,
      "epoch": 1.9326209819851643,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046263560705118675,
      "loss": 5.4645,
      "mean_token_accuracy": 0.19780069440603257,
      "num_tokens": 40812025.0,
      "step": 21885
    },
    {
      "entropy": 5.926669549942017,
      "epoch": 1.9330625220770046,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004626182262755551,
      "loss": 5.2774,
      "mean_token_accuracy": 0.20607707649469376,
      "num_tokens": 40820758.0,
      "step": 21890
    },
    {
      "entropy": 5.917462491989136,
      "epoch": 1.9335040621688449,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046260084182458906,
      "loss": 5.3892,
      "mean_token_accuracy": 0.20371164977550507,
      "num_tokens": 40830653.0,
      "step": 21895
    },
    {
      "entropy": 5.888490962982178,
      "epoch": 1.9339456022606853,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046258345369862915,
      "loss": 5.4622,
      "mean_token_accuracy": 0.19466545432806015,
      "num_tokens": 40840370.0,
      "step": 21900
    },
    {
      "entropy": 5.878164577484131,
      "epoch": 1.9343871423525256,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004625660618980162,
      "loss": 5.3782,
      "mean_token_accuracy": 0.20062892138957977,
      "num_tokens": 40849590.0,
      "step": 21905
    },
    {
      "entropy": 5.906628751754761,
      "epoch": 1.934828682444366,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046254866642309067,
      "loss": 5.3512,
      "mean_token_accuracy": 0.2105919674038887,
      "num_tokens": 40859271.0,
      "step": 21910
    },
    {
      "entropy": 5.885888576507568,
      "epoch": 1.9352702225362064,
      "grad_norm": 1.0,
      "learning_rate": 0.0004625312672741935,
      "loss": 5.3613,
      "mean_token_accuracy": 0.2009398505091667,
      "num_tokens": 40869272.0,
      "step": 21915
    },
    {
      "entropy": 5.956877517700195,
      "epoch": 1.9357117626280467,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004625138644516655,
      "loss": 5.4494,
      "mean_token_accuracy": 0.19722941368818284,
      "num_tokens": 40880327.0,
      "step": 21920
    },
    {
      "entropy": 5.873924922943115,
      "epoch": 1.936153302719887,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004624964579558474,
      "loss": 5.3604,
      "mean_token_accuracy": 0.20672069489955902,
      "num_tokens": 40890407.0,
      "step": 21925
    },
    {
      "entropy": 5.893722009658814,
      "epoch": 1.9365948428117274,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004624790477870804,
      "loss": 5.3485,
      "mean_token_accuracy": 0.20273849219083787,
      "num_tokens": 40899306.0,
      "step": 21930
    },
    {
      "entropy": 5.911926555633545,
      "epoch": 1.9370363829035675,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004624616339457055,
      "loss": 5.4071,
      "mean_token_accuracy": 0.20775255858898162,
      "num_tokens": 40908908.0,
      "step": 21935
    },
    {
      "entropy": 5.931298780441284,
      "epoch": 1.937477922995408,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004624442164320638,
      "loss": 5.3654,
      "mean_token_accuracy": 0.20946729332208633,
      "num_tokens": 40917960.0,
      "step": 21940
    },
    {
      "entropy": 5.805299949645996,
      "epoch": 1.9379194630872483,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004624267952464965,
      "loss": 5.1944,
      "mean_token_accuracy": 0.2136712431907654,
      "num_tokens": 40926364.0,
      "step": 21945
    },
    {
      "entropy": 5.8090832233428955,
      "epoch": 1.9383610031790885,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046240937038934484,
      "loss": 5.3642,
      "mean_token_accuracy": 0.2044597715139389,
      "num_tokens": 40936558.0,
      "step": 21950
    },
    {
      "entropy": 5.841175937652588,
      "epoch": 1.938802543270929,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004623919418609502,
      "loss": 5.3497,
      "mean_token_accuracy": 0.21050939559936524,
      "num_tokens": 40945501.0,
      "step": 21955
    },
    {
      "entropy": 5.828894138336182,
      "epoch": 1.9392440833627693,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046237450966165407,
      "loss": 5.3282,
      "mean_token_accuracy": 0.20923472940921783,
      "num_tokens": 40954740.0,
      "step": 21960
    },
    {
      "entropy": 5.901995515823364,
      "epoch": 1.9396856234546096,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046235707379179775,
      "loss": 5.379,
      "mean_token_accuracy": 0.2061217874288559,
      "num_tokens": 40963748.0,
      "step": 21965
    },
    {
      "entropy": 5.956651735305786,
      "epoch": 1.94012716354645,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046233963425172306,
      "loss": 5.3655,
      "mean_token_accuracy": 0.20245062410831452,
      "num_tokens": 40973907.0,
      "step": 21970
    },
    {
      "entropy": 5.903942346572876,
      "epoch": 1.9405687036382904,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046232219104177136,
      "loss": 5.3956,
      "mean_token_accuracy": 0.19936734139919282,
      "num_tokens": 40983374.0,
      "step": 21975
    },
    {
      "entropy": 5.920079612731934,
      "epoch": 1.9410102437301306,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004623047441622845,
      "loss": 5.3908,
      "mean_token_accuracy": 0.1982327714562416,
      "num_tokens": 40993305.0,
      "step": 21980
    },
    {
      "entropy": 5.939617538452149,
      "epoch": 1.9414517838219711,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046228729361360427,
      "loss": 5.3369,
      "mean_token_accuracy": 0.19725982695817948,
      "num_tokens": 41002438.0,
      "step": 21985
    },
    {
      "entropy": 5.772862243652344,
      "epoch": 1.9418933239138114,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046226983939607236,
      "loss": 5.3019,
      "mean_token_accuracy": 0.20948296785354614,
      "num_tokens": 41011491.0,
      "step": 21990
    },
    {
      "entropy": 5.908985710144043,
      "epoch": 1.9423348640056517,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004622523815100309,
      "loss": 5.3901,
      "mean_token_accuracy": 0.1993059992790222,
      "num_tokens": 41021091.0,
      "step": 21995
    },
    {
      "entropy": 5.883903312683105,
      "epoch": 1.9427764040974922,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046223491995582187,
      "loss": 5.3328,
      "mean_token_accuracy": 0.20751825124025344,
      "num_tokens": 41030719.0,
      "step": 22000
    },
    {
      "entropy": 5.899147081375122,
      "epoch": 1.9432179441893322,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004622174547337871,
      "loss": 5.4061,
      "mean_token_accuracy": 0.2048639938235283,
      "num_tokens": 41039651.0,
      "step": 22005
    },
    {
      "entropy": 5.894042873382569,
      "epoch": 1.9436594842811727,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046219998584426885,
      "loss": 5.3031,
      "mean_token_accuracy": 0.20808445811271667,
      "num_tokens": 41047914.0,
      "step": 22010
    },
    {
      "entropy": 5.855243730545044,
      "epoch": 1.944101024373013,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004621825132876094,
      "loss": 5.4121,
      "mean_token_accuracy": 0.20523531883955,
      "num_tokens": 41056779.0,
      "step": 22015
    },
    {
      "entropy": 5.859929800033569,
      "epoch": 1.9445425644648533,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046216503706415096,
      "loss": 5.361,
      "mean_token_accuracy": 0.20483759939670562,
      "num_tokens": 41066356.0,
      "step": 22020
    },
    {
      "entropy": 5.94675235748291,
      "epoch": 1.9449841045566938,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004621475571742359,
      "loss": 5.3751,
      "mean_token_accuracy": 0.21237125992774963,
      "num_tokens": 41075766.0,
      "step": 22025
    },
    {
      "entropy": 5.8732929706573485,
      "epoch": 1.945425644648534,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046213007361820663,
      "loss": 5.3639,
      "mean_token_accuracy": 0.2066869780421257,
      "num_tokens": 41085503.0,
      "step": 22030
    },
    {
      "entropy": 6.006605625152588,
      "epoch": 1.9458671847403743,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004621125863964057,
      "loss": 5.5377,
      "mean_token_accuracy": 0.18834761828184127,
      "num_tokens": 41096581.0,
      "step": 22035
    },
    {
      "entropy": 5.945880317687989,
      "epoch": 1.9463087248322148,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046209509550917553,
      "loss": 5.3451,
      "mean_token_accuracy": 0.21192478984594346,
      "num_tokens": 41105226.0,
      "step": 22040
    },
    {
      "entropy": 5.95869255065918,
      "epoch": 1.9467502649240551,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004620776009568589,
      "loss": 5.3417,
      "mean_token_accuracy": 0.20510756820440293,
      "num_tokens": 41113998.0,
      "step": 22045
    },
    {
      "entropy": 5.841848468780517,
      "epoch": 1.9471918050158954,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004620601027397984,
      "loss": 5.3684,
      "mean_token_accuracy": 0.19611532390117645,
      "num_tokens": 41123147.0,
      "step": 22050
    },
    {
      "entropy": 5.871012496948242,
      "epoch": 1.947633345107736,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046204260085833695,
      "loss": 5.3515,
      "mean_token_accuracy": 0.20897799581289292,
      "num_tokens": 41132317.0,
      "step": 22055
    },
    {
      "entropy": 5.92749056816101,
      "epoch": 1.9480748851995762,
      "grad_norm": 1.125,
      "learning_rate": 0.0004620250953128173,
      "loss": 5.3976,
      "mean_token_accuracy": 0.197231525182724,
      "num_tokens": 41140795.0,
      "step": 22060
    },
    {
      "entropy": 5.919864130020142,
      "epoch": 1.9485164252914164,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004620075861035824,
      "loss": 5.3554,
      "mean_token_accuracy": 0.2035280168056488,
      "num_tokens": 41149959.0,
      "step": 22065
    },
    {
      "entropy": 5.929309368133545,
      "epoch": 1.948957965383257,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004619900732309753,
      "loss": 5.4304,
      "mean_token_accuracy": 0.19045210033655166,
      "num_tokens": 41159388.0,
      "step": 22070
    },
    {
      "entropy": 6.005502319335937,
      "epoch": 1.949399505475097,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046197255669533886,
      "loss": 5.6157,
      "mean_token_accuracy": 0.1881886512041092,
      "num_tokens": 41168875.0,
      "step": 22075
    },
    {
      "entropy": 5.972923135757446,
      "epoch": 1.9498410455669375,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046195503649701646,
      "loss": 5.3954,
      "mean_token_accuracy": 0.2040553331375122,
      "num_tokens": 41177621.0,
      "step": 22080
    },
    {
      "entropy": 5.989239120483399,
      "epoch": 1.9502825856587778,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004619375126363513,
      "loss": 5.4358,
      "mean_token_accuracy": 0.20210410058498382,
      "num_tokens": 41185911.0,
      "step": 22085
    },
    {
      "entropy": 5.921224975585938,
      "epoch": 1.950724125750618,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046191998511368646,
      "loss": 5.3268,
      "mean_token_accuracy": 0.20728100091218948,
      "num_tokens": 41194811.0,
      "step": 22090
    },
    {
      "entropy": 5.7517650604248045,
      "epoch": 1.9511656658424585,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046190245392936554,
      "loss": 5.2578,
      "mean_token_accuracy": 0.21629572361707688,
      "num_tokens": 41203955.0,
      "step": 22095
    },
    {
      "entropy": 5.836204576492309,
      "epoch": 1.9516072059342988,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004618849190837317,
      "loss": 5.2955,
      "mean_token_accuracy": 0.2150256708264351,
      "num_tokens": 41212635.0,
      "step": 22100
    },
    {
      "entropy": 5.896378040313721,
      "epoch": 1.952048746026139,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004618673805771287,
      "loss": 5.3621,
      "mean_token_accuracy": 0.20168523192405702,
      "num_tokens": 41221443.0,
      "step": 22105
    },
    {
      "entropy": 5.883521270751953,
      "epoch": 1.9524902861179796,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004618498384098999,
      "loss": 5.3303,
      "mean_token_accuracy": 0.2048863038420677,
      "num_tokens": 41230089.0,
      "step": 22110
    },
    {
      "entropy": 5.95154218673706,
      "epoch": 1.9529318262098199,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046183229258238905,
      "loss": 5.4687,
      "mean_token_accuracy": 0.19285901486873627,
      "num_tokens": 41239569.0,
      "step": 22115
    },
    {
      "entropy": 5.9417143821716305,
      "epoch": 1.9533733663016601,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004618147430949399,
      "loss": 5.3535,
      "mean_token_accuracy": 0.2068372204899788,
      "num_tokens": 41248394.0,
      "step": 22120
    },
    {
      "entropy": 5.878524303436279,
      "epoch": 1.9538149063935006,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046179718994789624,
      "loss": 5.3575,
      "mean_token_accuracy": 0.20982825756072998,
      "num_tokens": 41258148.0,
      "step": 22125
    },
    {
      "entropy": 5.876276159286499,
      "epoch": 1.954256446485341,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046177963314160175,
      "loss": 5.2879,
      "mean_token_accuracy": 0.21348503977060318,
      "num_tokens": 41266984.0,
      "step": 22130
    },
    {
      "entropy": 5.91729097366333,
      "epoch": 1.9546979865771812,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046176207267640054,
      "loss": 5.4063,
      "mean_token_accuracy": 0.19888876527547836,
      "num_tokens": 41276119.0,
      "step": 22135
    },
    {
      "entropy": 5.7885355949401855,
      "epoch": 1.9551395266690217,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004617445085526365,
      "loss": 5.3038,
      "mean_token_accuracy": 0.21451292783021927,
      "num_tokens": 41284945.0,
      "step": 22140
    },
    {
      "entropy": 5.871628189086914,
      "epoch": 1.9555810667608617,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004617269407706537,
      "loss": 5.291,
      "mean_token_accuracy": 0.21350942850112914,
      "num_tokens": 41294255.0,
      "step": 22145
    },
    {
      "entropy": 5.896035385131836,
      "epoch": 1.9560226068527022,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004617093693307964,
      "loss": 5.3977,
      "mean_token_accuracy": 0.20890733301639558,
      "num_tokens": 41303056.0,
      "step": 22150
    },
    {
      "entropy": 5.909364032745361,
      "epoch": 1.9564641469445425,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046169179423340866,
      "loss": 5.4023,
      "mean_token_accuracy": 0.2050107091665268,
      "num_tokens": 41312444.0,
      "step": 22155
    },
    {
      "entropy": 5.9884837627410885,
      "epoch": 1.9569056870363828,
      "grad_norm": 1.09375,
      "learning_rate": 0.000461674215478835,
      "loss": 5.4919,
      "mean_token_accuracy": 0.20010593831539153,
      "num_tokens": 41322507.0,
      "step": 22160
    },
    {
      "entropy": 5.947047901153565,
      "epoch": 1.9573472271282233,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004616566330674195,
      "loss": 5.3652,
      "mean_token_accuracy": 0.2078099951148033,
      "num_tokens": 41332507.0,
      "step": 22165
    },
    {
      "entropy": 5.927550983428955,
      "epoch": 1.9577887672200636,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004616390469995068,
      "loss": 5.3223,
      "mean_token_accuracy": 0.20718599557876588,
      "num_tokens": 41342085.0,
      "step": 22170
    },
    {
      "entropy": 5.835907220840454,
      "epoch": 1.9582303073119038,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046162145727544126,
      "loss": 5.3101,
      "mean_token_accuracy": 0.21252421885728837,
      "num_tokens": 41351422.0,
      "step": 22175
    },
    {
      "entropy": 5.84026517868042,
      "epoch": 1.9586718474037443,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004616038638955675,
      "loss": 5.345,
      "mean_token_accuracy": 0.21026050448417663,
      "num_tokens": 41361337.0,
      "step": 22180
    },
    {
      "entropy": 5.82983136177063,
      "epoch": 1.9591133874955846,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046158626686023014,
      "loss": 5.2765,
      "mean_token_accuracy": 0.21099374890327455,
      "num_tokens": 41370734.0,
      "step": 22185
    },
    {
      "entropy": 5.88320779800415,
      "epoch": 1.9595549275874249,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046156866616977406,
      "loss": 5.3056,
      "mean_token_accuracy": 0.20598259270191194,
      "num_tokens": 41380090.0,
      "step": 22190
    },
    {
      "entropy": 5.965227508544922,
      "epoch": 1.9599964676792654,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046155106182454384,
      "loss": 5.3792,
      "mean_token_accuracy": 0.2085856541991234,
      "num_tokens": 41388967.0,
      "step": 22195
    },
    {
      "entropy": 5.897452068328858,
      "epoch": 1.9604380077711057,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004615334538248844,
      "loss": 5.3517,
      "mean_token_accuracy": 0.2055166631937027,
      "num_tokens": 41397429.0,
      "step": 22200
    },
    {
      "entropy": 5.816228199005127,
      "epoch": 1.960879547862946,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046151584217114073,
      "loss": 5.3022,
      "mean_token_accuracy": 0.20593734234571456,
      "num_tokens": 41406420.0,
      "step": 22205
    },
    {
      "entropy": 5.8074297428131105,
      "epoch": 1.9613210879547864,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004614982268636577,
      "loss": 5.2603,
      "mean_token_accuracy": 0.22167766094207764,
      "num_tokens": 41415994.0,
      "step": 22210
    },
    {
      "entropy": 5.877642250061035,
      "epoch": 1.9617626280466265,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004614806079027806,
      "loss": 5.2613,
      "mean_token_accuracy": 0.215389646589756,
      "num_tokens": 41425744.0,
      "step": 22215
    },
    {
      "entropy": 5.904153633117676,
      "epoch": 1.962204168138467,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046146298528885435,
      "loss": 5.443,
      "mean_token_accuracy": 0.19678525477647782,
      "num_tokens": 41435452.0,
      "step": 22220
    },
    {
      "entropy": 5.941742277145385,
      "epoch": 1.9626457082303073,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004614453590222243,
      "loss": 5.4029,
      "mean_token_accuracy": 0.20584166049957275,
      "num_tokens": 41444853.0,
      "step": 22225
    },
    {
      "entropy": 5.794569444656372,
      "epoch": 1.9630872483221475,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004614277291032357,
      "loss": 5.2619,
      "mean_token_accuracy": 0.2254179000854492,
      "num_tokens": 41454569.0,
      "step": 22230
    },
    {
      "entropy": 5.952146291732788,
      "epoch": 1.963528788413988,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004614100955322339,
      "loss": 5.4187,
      "mean_token_accuracy": 0.2028772696852684,
      "num_tokens": 41464159.0,
      "step": 22235
    },
    {
      "entropy": 5.957342052459717,
      "epoch": 1.9639703285058283,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004613924583095644,
      "loss": 5.4372,
      "mean_token_accuracy": 0.1950291946530342,
      "num_tokens": 41474354.0,
      "step": 22240
    },
    {
      "entropy": 5.849534130096435,
      "epoch": 1.9644118685976686,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004613748174355726,
      "loss": 5.2956,
      "mean_token_accuracy": 0.20756881237030028,
      "num_tokens": 41484053.0,
      "step": 22245
    },
    {
      "entropy": 5.944089937210083,
      "epoch": 1.964853408689509,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046135717291060415,
      "loss": 5.3526,
      "mean_token_accuracy": 0.1971958577632904,
      "num_tokens": 41493377.0,
      "step": 22250
    },
    {
      "entropy": 5.902534770965576,
      "epoch": 1.9652949487813494,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046133952473500464,
      "loss": 5.3158,
      "mean_token_accuracy": 0.21343559622764588,
      "num_tokens": 41502361.0,
      "step": 22255
    },
    {
      "entropy": 5.886488914489746,
      "epoch": 1.9657364888731896,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004613218729091198,
      "loss": 5.3554,
      "mean_token_accuracy": 0.20211567282676696,
      "num_tokens": 41511083.0,
      "step": 22260
    },
    {
      "entropy": 5.896930932998657,
      "epoch": 1.9661780289650301,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004613042174332955,
      "loss": 5.4491,
      "mean_token_accuracy": 0.19491046518087388,
      "num_tokens": 41520264.0,
      "step": 22265
    },
    {
      "entropy": 5.846308040618896,
      "epoch": 1.9666195690568704,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004612865583078775,
      "loss": 5.2855,
      "mean_token_accuracy": 0.2126755714416504,
      "num_tokens": 41529805.0,
      "step": 22270
    },
    {
      "entropy": 5.827733278274536,
      "epoch": 1.9670611091487107,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004612688955332118,
      "loss": 5.2406,
      "mean_token_accuracy": 0.20872793793678285,
      "num_tokens": 41538348.0,
      "step": 22275
    },
    {
      "entropy": 5.87517786026001,
      "epoch": 1.9675026492405512,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046125122910964427,
      "loss": 5.3714,
      "mean_token_accuracy": 0.20601751357316972,
      "num_tokens": 41547433.0,
      "step": 22280
    },
    {
      "entropy": 5.931604194641113,
      "epoch": 1.9679441893323912,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004612335590375212,
      "loss": 5.3629,
      "mean_token_accuracy": 0.20532096326351165,
      "num_tokens": 41556162.0,
      "step": 22285
    },
    {
      "entropy": 5.912998962402344,
      "epoch": 1.9683857294242317,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004612158853171886,
      "loss": 5.3336,
      "mean_token_accuracy": 0.20705235749483109,
      "num_tokens": 41565415.0,
      "step": 22290
    },
    {
      "entropy": 5.832032537460327,
      "epoch": 1.968827269516072,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004611982079489928,
      "loss": 5.2932,
      "mean_token_accuracy": 0.21562063694000244,
      "num_tokens": 41574728.0,
      "step": 22295
    },
    {
      "entropy": 5.880598640441894,
      "epoch": 1.9692688096079123,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004611805269332798,
      "loss": 5.3814,
      "mean_token_accuracy": 0.19584305733442306,
      "num_tokens": 41584749.0,
      "step": 22300
    },
    {
      "entropy": 5.876179647445679,
      "epoch": 1.9697103496997528,
      "grad_norm": 1.0,
      "learning_rate": 0.00046116284227039626,
      "loss": 5.3288,
      "mean_token_accuracy": 0.2164471209049225,
      "num_tokens": 41594658.0,
      "step": 22305
    },
    {
      "entropy": 5.891310548782348,
      "epoch": 1.970151889791593,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004611451539606886,
      "loss": 5.4272,
      "mean_token_accuracy": 0.19682959616184234,
      "num_tokens": 41604548.0,
      "step": 22310
    },
    {
      "entropy": 5.8859165668487545,
      "epoch": 1.9705934298834333,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004611274620045031,
      "loss": 5.3778,
      "mean_token_accuracy": 0.20543351173400878,
      "num_tokens": 41614002.0,
      "step": 22315
    },
    {
      "entropy": 5.926422595977783,
      "epoch": 1.9710349699752738,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004611097664021866,
      "loss": 5.3785,
      "mean_token_accuracy": 0.19734903275966645,
      "num_tokens": 41623793.0,
      "step": 22320
    },
    {
      "entropy": 5.896677684783936,
      "epoch": 1.971476510067114,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046109206715408554,
      "loss": 5.3152,
      "mean_token_accuracy": 0.20300010740756988,
      "num_tokens": 41632608.0,
      "step": 22325
    },
    {
      "entropy": 5.921829748153686,
      "epoch": 1.9719180501589544,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046107436426054676,
      "loss": 5.4098,
      "mean_token_accuracy": 0.1996908262372017,
      "num_tokens": 41641814.0,
      "step": 22330
    },
    {
      "entropy": 5.9364342212677,
      "epoch": 1.9723595902507949,
      "grad_norm": 1.046875,
      "learning_rate": 0.000461056657721917,
      "loss": 5.3806,
      "mean_token_accuracy": 0.2009850949048996,
      "num_tokens": 41650643.0,
      "step": 22335
    },
    {
      "entropy": 5.912870216369629,
      "epoch": 1.9728011303426352,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046103894753854317,
      "loss": 5.3742,
      "mean_token_accuracy": 0.2038644790649414,
      "num_tokens": 41661308.0,
      "step": 22340
    },
    {
      "entropy": 5.947876596450806,
      "epoch": 1.9732426704344754,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004610212337107721,
      "loss": 5.42,
      "mean_token_accuracy": 0.196707184612751,
      "num_tokens": 41671090.0,
      "step": 22345
    },
    {
      "entropy": 5.856647539138794,
      "epoch": 1.973684210526316,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046100351623895087,
      "loss": 5.3623,
      "mean_token_accuracy": 0.2016085594892502,
      "num_tokens": 41679584.0,
      "step": 22350
    },
    {
      "entropy": 5.8358642101287845,
      "epoch": 1.974125750618156,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046098579512342654,
      "loss": 5.3486,
      "mean_token_accuracy": 0.19253745675086975,
      "num_tokens": 41689772.0,
      "step": 22355
    },
    {
      "entropy": 5.898290729522705,
      "epoch": 1.9745672907099965,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004609680703645463,
      "loss": 5.3724,
      "mean_token_accuracy": 0.20120088756084442,
      "num_tokens": 41699196.0,
      "step": 22360
    },
    {
      "entropy": 5.888697004318237,
      "epoch": 1.9750088308018368,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046095034196265726,
      "loss": 5.3224,
      "mean_token_accuracy": 0.2138876885175705,
      "num_tokens": 41707899.0,
      "step": 22365
    },
    {
      "entropy": 5.876837301254272,
      "epoch": 1.975450370893677,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046093260991810685,
      "loss": 5.381,
      "mean_token_accuracy": 0.20761237144470215,
      "num_tokens": 41716903.0,
      "step": 22370
    },
    {
      "entropy": 5.867739152908325,
      "epoch": 1.9758919109855175,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046091487423124234,
      "loss": 5.2375,
      "mean_token_accuracy": 0.21416876912117006,
      "num_tokens": 41725676.0,
      "step": 22375
    },
    {
      "entropy": 5.892343139648437,
      "epoch": 1.9763334510773578,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046089713490241117,
      "loss": 5.2567,
      "mean_token_accuracy": 0.21238938868045806,
      "num_tokens": 41734720.0,
      "step": 22380
    },
    {
      "entropy": 5.827096319198608,
      "epoch": 1.976774991169198,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004608793919319608,
      "loss": 5.3809,
      "mean_token_accuracy": 0.21058744490146636,
      "num_tokens": 41743866.0,
      "step": 22385
    },
    {
      "entropy": 5.839306926727295,
      "epoch": 1.9772165312610386,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046086164532023894,
      "loss": 5.3423,
      "mean_token_accuracy": 0.21011044085025787,
      "num_tokens": 41754157.0,
      "step": 22390
    },
    {
      "entropy": 5.941110277175904,
      "epoch": 1.9776580713528789,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046084389506759313,
      "loss": 5.357,
      "mean_token_accuracy": 0.20468229949474334,
      "num_tokens": 41763105.0,
      "step": 22395
    },
    {
      "entropy": 5.9145894050598145,
      "epoch": 1.9780996114447191,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046082614117437103,
      "loss": 5.2743,
      "mean_token_accuracy": 0.21250310987234117,
      "num_tokens": 41772623.0,
      "step": 22400
    },
    {
      "entropy": 5.852737665176392,
      "epoch": 1.9785411515365596,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046080838364092057,
      "loss": 5.3774,
      "mean_token_accuracy": 0.2026889517903328,
      "num_tokens": 41781331.0,
      "step": 22405
    },
    {
      "entropy": 5.911617851257324,
      "epoch": 1.9789826916284,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004607906224675896,
      "loss": 5.4981,
      "mean_token_accuracy": 0.19725169241428375,
      "num_tokens": 41789727.0,
      "step": 22410
    },
    {
      "entropy": 5.930763483047485,
      "epoch": 1.9794242317202402,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004607728576547259,
      "loss": 5.3589,
      "mean_token_accuracy": 0.2052120476961136,
      "num_tokens": 41798531.0,
      "step": 22415
    },
    {
      "entropy": 5.927769041061401,
      "epoch": 1.9798657718120807,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046075508920267766,
      "loss": 5.368,
      "mean_token_accuracy": 0.2038962408900261,
      "num_tokens": 41808304.0,
      "step": 22420
    },
    {
      "entropy": 5.828134202957154,
      "epoch": 1.9803073119039207,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004607373171117929,
      "loss": 5.2486,
      "mean_token_accuracy": 0.21249572932720184,
      "num_tokens": 41817427.0,
      "step": 22425
    },
    {
      "entropy": 5.837864875793457,
      "epoch": 1.9807488519957612,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004607195413824197,
      "loss": 5.3276,
      "mean_token_accuracy": 0.21120381057262422,
      "num_tokens": 41827202.0,
      "step": 22430
    },
    {
      "entropy": 5.897235250473022,
      "epoch": 1.9811903920876015,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004607017620149062,
      "loss": 5.3035,
      "mean_token_accuracy": 0.21286925226449965,
      "num_tokens": 41836213.0,
      "step": 22435
    },
    {
      "entropy": 5.875213384628296,
      "epoch": 1.9816319321794418,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004606839790096009,
      "loss": 5.4201,
      "mean_token_accuracy": 0.19661785364151002,
      "num_tokens": 41845358.0,
      "step": 22440
    },
    {
      "entropy": 5.918174362182617,
      "epoch": 1.9820734722712823,
      "grad_norm": 1.125,
      "learning_rate": 0.0004606661923668521,
      "loss": 5.3709,
      "mean_token_accuracy": 0.20897252708673478,
      "num_tokens": 41854572.0,
      "step": 22445
    },
    {
      "entropy": 5.858848142623901,
      "epoch": 1.9825150123631226,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004606484020870081,
      "loss": 5.3265,
      "mean_token_accuracy": 0.21139038801193238,
      "num_tokens": 41863643.0,
      "step": 22450
    },
    {
      "entropy": 5.858176612854004,
      "epoch": 1.9829565524549628,
      "grad_norm": 1.125,
      "learning_rate": 0.00046063060817041747,
      "loss": 5.2724,
      "mean_token_accuracy": 0.2067800909280777,
      "num_tokens": 41873004.0,
      "step": 22455
    },
    {
      "entropy": 5.946782112121582,
      "epoch": 1.9833980925468033,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004606128106174288,
      "loss": 5.339,
      "mean_token_accuracy": 0.21116142868995666,
      "num_tokens": 41881530.0,
      "step": 22460
    },
    {
      "entropy": 5.872727966308593,
      "epoch": 1.9838396326386436,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004605950094283908,
      "loss": 5.3938,
      "mean_token_accuracy": 0.1966492637991905,
      "num_tokens": 41890342.0,
      "step": 22465
    },
    {
      "entropy": 5.849273014068603,
      "epoch": 1.9842811727304839,
      "grad_norm": 1.0,
      "learning_rate": 0.00046057720460365205,
      "loss": 5.2872,
      "mean_token_accuracy": 0.21358196139335633,
      "num_tokens": 41899274.0,
      "step": 22470
    },
    {
      "entropy": 5.912584590911865,
      "epoch": 1.9847227128223244,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046055939614356146,
      "loss": 5.3354,
      "mean_token_accuracy": 0.20744515359401702,
      "num_tokens": 41908835.0,
      "step": 22475
    },
    {
      "entropy": 5.851005363464355,
      "epoch": 1.9851642529141647,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004605415840484678,
      "loss": 5.2682,
      "mean_token_accuracy": 0.21649398058652877,
      "num_tokens": 41917900.0,
      "step": 22480
    },
    {
      "entropy": 5.903795146942139,
      "epoch": 1.985605793006005,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046052376831872007,
      "loss": 5.3446,
      "mean_token_accuracy": 0.2121477320790291,
      "num_tokens": 41926964.0,
      "step": 22485
    },
    {
      "entropy": 5.944635820388794,
      "epoch": 1.9860473330978454,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004605059489546672,
      "loss": 5.4173,
      "mean_token_accuracy": 0.2030104786157608,
      "num_tokens": 41936401.0,
      "step": 22490
    },
    {
      "entropy": 5.830485725402832,
      "epoch": 1.9864888731896855,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046048812595665824,
      "loss": 5.3039,
      "mean_token_accuracy": 0.20996571332216263,
      "num_tokens": 41944665.0,
      "step": 22495
    },
    {
      "entropy": 5.844066381454468,
      "epoch": 1.986930413281526,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004604702993250425,
      "loss": 5.3662,
      "mean_token_accuracy": 0.1958061069250107,
      "num_tokens": 41953868.0,
      "step": 22500
    },
    {
      "entropy": 5.810266590118408,
      "epoch": 1.9873719533733663,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046045246906016904,
      "loss": 5.3681,
      "mean_token_accuracy": 0.2104953870177269,
      "num_tokens": 41963013.0,
      "step": 22505
    },
    {
      "entropy": 5.902640914916992,
      "epoch": 1.9878134934652065,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004604346351623872,
      "loss": 5.363,
      "mean_token_accuracy": 0.19927488267421722,
      "num_tokens": 41972926.0,
      "step": 22510
    },
    {
      "entropy": 5.942632150650025,
      "epoch": 1.988255033557047,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046041679763204625,
      "loss": 5.3581,
      "mean_token_accuracy": 0.21053942143917084,
      "num_tokens": 41981773.0,
      "step": 22515
    },
    {
      "entropy": 5.913764953613281,
      "epoch": 1.9886965736488873,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004603989564694957,
      "loss": 5.3457,
      "mean_token_accuracy": 0.20863389521837233,
      "num_tokens": 41990893.0,
      "step": 22520
    },
    {
      "entropy": 5.876496982574463,
      "epoch": 1.9891381137407276,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004603811116750851,
      "loss": 5.3682,
      "mean_token_accuracy": 0.2028349071741104,
      "num_tokens": 42000842.0,
      "step": 22525
    },
    {
      "entropy": 5.9543054580688475,
      "epoch": 1.989579653832568,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004603632632491639,
      "loss": 5.3356,
      "mean_token_accuracy": 0.21135483980178832,
      "num_tokens": 42009430.0,
      "step": 22530
    },
    {
      "entropy": 5.913153839111328,
      "epoch": 1.9900211939244083,
      "grad_norm": 1.0,
      "learning_rate": 0.00046034541119208177,
      "loss": 5.4722,
      "mean_token_accuracy": 0.2030464231967926,
      "num_tokens": 42018985.0,
      "step": 22535
    },
    {
      "entropy": 5.905248403549194,
      "epoch": 1.9904627340162486,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046032755550418845,
      "loss": 5.3344,
      "mean_token_accuracy": 0.20350456684827806,
      "num_tokens": 42028885.0,
      "step": 22540
    },
    {
      "entropy": 5.960259628295899,
      "epoch": 1.9909042741080891,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004603096961858338,
      "loss": 5.4422,
      "mean_token_accuracy": 0.19680802673101425,
      "num_tokens": 42037838.0,
      "step": 22545
    },
    {
      "entropy": 5.859803628921509,
      "epoch": 1.9913458141999294,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046029183323736755,
      "loss": 5.3297,
      "mean_token_accuracy": 0.2057755097746849,
      "num_tokens": 42047212.0,
      "step": 22550
    },
    {
      "entropy": 5.960439300537109,
      "epoch": 1.9917873542917697,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004602739666591397,
      "loss": 5.4123,
      "mean_token_accuracy": 0.20028246194124222,
      "num_tokens": 42056680.0,
      "step": 22555
    },
    {
      "entropy": 5.956166744232178,
      "epoch": 1.9922288943836102,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046025609645150014,
      "loss": 5.4116,
      "mean_token_accuracy": 0.19626805037260056,
      "num_tokens": 42065604.0,
      "step": 22560
    },
    {
      "entropy": 5.921209716796875,
      "epoch": 1.9926704344754502,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000460238222614799,
      "loss": 5.3433,
      "mean_token_accuracy": 0.20482654571533204,
      "num_tokens": 42074672.0,
      "step": 22565
    },
    {
      "entropy": 5.894555330276489,
      "epoch": 1.9931119745672907,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004602203451493865,
      "loss": 5.3036,
      "mean_token_accuracy": 0.20232925713062286,
      "num_tokens": 42083586.0,
      "step": 22570
    },
    {
      "entropy": 5.862616872787475,
      "epoch": 1.993553514659131,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046020246405561283,
      "loss": 5.326,
      "mean_token_accuracy": 0.20374557375907898,
      "num_tokens": 42092328.0,
      "step": 22575
    },
    {
      "entropy": 5.827028703689575,
      "epoch": 1.9939950547509713,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046018457933382807,
      "loss": 5.3187,
      "mean_token_accuracy": 0.21235123425722122,
      "num_tokens": 42102154.0,
      "step": 22580
    },
    {
      "entropy": 5.936002159118653,
      "epoch": 1.9944365948428118,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046016669098438286,
      "loss": 5.3829,
      "mean_token_accuracy": 0.20435209274291993,
      "num_tokens": 42111762.0,
      "step": 22585
    },
    {
      "entropy": 5.911132431030273,
      "epoch": 1.994878134934652,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004601487990076274,
      "loss": 5.3762,
      "mean_token_accuracy": 0.20171091109514236,
      "num_tokens": 42120631.0,
      "step": 22590
    },
    {
      "entropy": 5.822972202301026,
      "epoch": 1.9953196750264923,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004601309034039123,
      "loss": 5.2363,
      "mean_token_accuracy": 0.22829282730817796,
      "num_tokens": 42129311.0,
      "step": 22595
    },
    {
      "entropy": 5.937223196029663,
      "epoch": 1.9957612151183328,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004601130041735881,
      "loss": 5.4706,
      "mean_token_accuracy": 0.19585298597812653,
      "num_tokens": 42138735.0,
      "step": 22600
    },
    {
      "entropy": 5.997986698150635,
      "epoch": 1.996202755210173,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004600951013170054,
      "loss": 5.4305,
      "mean_token_accuracy": 0.20378115326166152,
      "num_tokens": 42148416.0,
      "step": 22605
    },
    {
      "entropy": 5.9393072605133055,
      "epoch": 1.9966442953020134,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046007719483451493,
      "loss": 5.3558,
      "mean_token_accuracy": 0.2071908250451088,
      "num_tokens": 42157191.0,
      "step": 22610
    },
    {
      "entropy": 5.775449609756469,
      "epoch": 1.9970858353938539,
      "grad_norm": 1.125,
      "learning_rate": 0.00046005928472646747,
      "loss": 5.2936,
      "mean_token_accuracy": 0.22011100500822067,
      "num_tokens": 42165798.0,
      "step": 22615
    },
    {
      "entropy": 5.871996688842773,
      "epoch": 1.9975273754856941,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046004137099321386,
      "loss": 5.4053,
      "mean_token_accuracy": 0.1983458638191223,
      "num_tokens": 42175318.0,
      "step": 22620
    },
    {
      "entropy": 5.954274415969849,
      "epoch": 1.9979689155775344,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046002345363510506,
      "loss": 5.4177,
      "mean_token_accuracy": 0.19837913960218428,
      "num_tokens": 42184800.0,
      "step": 22625
    },
    {
      "entropy": 5.900021648406982,
      "epoch": 1.998410455669375,
      "grad_norm": 1.0,
      "learning_rate": 0.0004600055326524921,
      "loss": 5.2444,
      "mean_token_accuracy": 0.21323984414339064,
      "num_tokens": 42194172.0,
      "step": 22630
    },
    {
      "entropy": 5.86443395614624,
      "epoch": 1.998851995761215,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045998760804572583,
      "loss": 5.2903,
      "mean_token_accuracy": 0.21045953035354614,
      "num_tokens": 42203102.0,
      "step": 22635
    },
    {
      "entropy": 5.787165307998658,
      "epoch": 1.9992935358530555,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045996967981515763,
      "loss": 5.2459,
      "mean_token_accuracy": 0.21351298987865447,
      "num_tokens": 42212421.0,
      "step": 22640
    },
    {
      "entropy": 5.909124040603638,
      "epoch": 1.9997350759448957,
      "grad_norm": 1.0,
      "learning_rate": 0.00045995174796113856,
      "loss": 5.3652,
      "mean_token_accuracy": 0.2063007727265358,
      "num_tokens": 42221528.0,
      "step": 22645
    },
    {
      "entropy": 5.87777943611145,
      "epoch": 2.000176616036736,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00045993381248402,
      "loss": 5.3305,
      "mean_token_accuracy": 0.2049049600958824,
      "num_tokens": 42230628.0,
      "step": 22650
    },
    {
      "entropy": 5.868379402160644,
      "epoch": 2.0006181561285765,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045991587338415306,
      "loss": 5.2927,
      "mean_token_accuracy": 0.20688465386629104,
      "num_tokens": 42239571.0,
      "step": 22655
    },
    {
      "entropy": 5.907155132293701,
      "epoch": 2.001059696220417,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045989793066188953,
      "loss": 5.2309,
      "mean_token_accuracy": 0.21225377470254897,
      "num_tokens": 42248010.0,
      "step": 22660
    },
    {
      "entropy": 5.803134965896606,
      "epoch": 2.001501236312257,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045987998431758063,
      "loss": 5.2375,
      "mean_token_accuracy": 0.21220990270376205,
      "num_tokens": 42257251.0,
      "step": 22665
    },
    {
      "entropy": 5.881656980514526,
      "epoch": 2.0019427764040976,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004598620343515779,
      "loss": 5.3549,
      "mean_token_accuracy": 0.20693680346012117,
      "num_tokens": 42266441.0,
      "step": 22670
    },
    {
      "entropy": 5.8739899635314945,
      "epoch": 2.0023843164959376,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00045984408076423313,
      "loss": 5.2753,
      "mean_token_accuracy": 0.2062024623155594,
      "num_tokens": 42275995.0,
      "step": 22675
    },
    {
      "entropy": 5.928394842147827,
      "epoch": 2.002825856587778,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004598261235558979,
      "loss": 5.2725,
      "mean_token_accuracy": 0.21312019377946853,
      "num_tokens": 42284065.0,
      "step": 22680
    },
    {
      "entropy": 5.872630262374878,
      "epoch": 2.0032673966796186,
      "grad_norm": 0.9375,
      "learning_rate": 0.00045980816272692403,
      "loss": 5.3725,
      "mean_token_accuracy": 0.20882306694984437,
      "num_tokens": 42295309.0,
      "step": 22685
    },
    {
      "entropy": 5.960452842712402,
      "epoch": 2.0037089367714587,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045979019827766345,
      "loss": 5.2739,
      "mean_token_accuracy": 0.21062087267637253,
      "num_tokens": 42304390.0,
      "step": 22690
    },
    {
      "entropy": 5.915776681900025,
      "epoch": 2.004150476863299,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045977223020846785,
      "loss": 5.3048,
      "mean_token_accuracy": 0.21240830272436143,
      "num_tokens": 42313830.0,
      "step": 22695
    },
    {
      "entropy": 5.913122510910034,
      "epoch": 2.0045920169551397,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004597542585196895,
      "loss": 5.2997,
      "mean_token_accuracy": 0.2076733887195587,
      "num_tokens": 42323827.0,
      "step": 22700
    },
    {
      "entropy": 5.964270734786988,
      "epoch": 2.0050335570469797,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045973628321168016,
      "loss": 5.3975,
      "mean_token_accuracy": 0.20324095636606215,
      "num_tokens": 42333221.0,
      "step": 22705
    },
    {
      "entropy": 5.819288206100464,
      "epoch": 2.00547509713882,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045971830428479215,
      "loss": 5.2696,
      "mean_token_accuracy": 0.21562483608722688,
      "num_tokens": 42343036.0,
      "step": 22710
    },
    {
      "entropy": 5.854319334030151,
      "epoch": 2.0059166372306607,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045970032173937766,
      "loss": 5.2503,
      "mean_token_accuracy": 0.21470864713191987,
      "num_tokens": 42351916.0,
      "step": 22715
    },
    {
      "entropy": 5.993334007263184,
      "epoch": 2.0063581773225008,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00045968233557578883,
      "loss": 5.3547,
      "mean_token_accuracy": 0.20781833529472352,
      "num_tokens": 42362412.0,
      "step": 22720
    },
    {
      "entropy": 5.867878150939942,
      "epoch": 2.0067997174143413,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004596643457943781,
      "loss": 5.2354,
      "mean_token_accuracy": 0.21803030967712403,
      "num_tokens": 42372773.0,
      "step": 22725
    },
    {
      "entropy": 5.886961364746094,
      "epoch": 2.0072412575061818,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004596463523954979,
      "loss": 5.2547,
      "mean_token_accuracy": 0.2088466003537178,
      "num_tokens": 42382009.0,
      "step": 22730
    },
    {
      "entropy": 5.882224369049072,
      "epoch": 2.007682797598022,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045962835537950063,
      "loss": 5.2758,
      "mean_token_accuracy": 0.21333387345075608,
      "num_tokens": 42390961.0,
      "step": 22735
    },
    {
      "entropy": 5.8336258888244625,
      "epoch": 2.0081243376898623,
      "grad_norm": 1.0,
      "learning_rate": 0.00045961035474673893,
      "loss": 5.2789,
      "mean_token_accuracy": 0.2099034771323204,
      "num_tokens": 42400483.0,
      "step": 22740
    },
    {
      "entropy": 5.759735345840454,
      "epoch": 2.0085658777817024,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045959235049756537,
      "loss": 5.1332,
      "mean_token_accuracy": 0.22602652311325072,
      "num_tokens": 42409285.0,
      "step": 22745
    },
    {
      "entropy": 5.9130964279174805,
      "epoch": 2.009007417873543,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045957434263233253,
      "loss": 5.3131,
      "mean_token_accuracy": 0.2079264223575592,
      "num_tokens": 42418950.0,
      "step": 22750
    },
    {
      "entropy": 5.894699907302856,
      "epoch": 2.0094489579653834,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045955633115139343,
      "loss": 5.3883,
      "mean_token_accuracy": 0.20179129242897034,
      "num_tokens": 42428285.0,
      "step": 22755
    },
    {
      "entropy": 5.87842812538147,
      "epoch": 2.0098904980572234,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004595383160551007,
      "loss": 5.2342,
      "mean_token_accuracy": 0.2104954943060875,
      "num_tokens": 42437832.0,
      "step": 22760
    },
    {
      "entropy": 5.872386884689331,
      "epoch": 2.010332038149064,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004595202973438074,
      "loss": 5.2547,
      "mean_token_accuracy": 0.2119743764400482,
      "num_tokens": 42447704.0,
      "step": 22765
    },
    {
      "entropy": 5.9236311435699465,
      "epoch": 2.0107735782409044,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004595022750178663,
      "loss": 5.3509,
      "mean_token_accuracy": 0.20961358696222304,
      "num_tokens": 42456571.0,
      "step": 22770
    },
    {
      "entropy": 5.90190806388855,
      "epoch": 2.0112151183327445,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004594842490776306,
      "loss": 5.2302,
      "mean_token_accuracy": 0.21186455190181733,
      "num_tokens": 42466522.0,
      "step": 22775
    },
    {
      "entropy": 5.840451097488403,
      "epoch": 2.011656658424585,
      "grad_norm": 1.0,
      "learning_rate": 0.00045946621952345335,
      "loss": 5.2938,
      "mean_token_accuracy": 0.20743368566036224,
      "num_tokens": 42475940.0,
      "step": 22780
    },
    {
      "entropy": 5.922690153121948,
      "epoch": 2.0120981985164255,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004594481863556879,
      "loss": 5.3625,
      "mean_token_accuracy": 0.20207197219133377,
      "num_tokens": 42486456.0,
      "step": 22785
    },
    {
      "entropy": 5.965627670288086,
      "epoch": 2.0125397386082655,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045943014957468726,
      "loss": 5.3418,
      "mean_token_accuracy": 0.21078796535730362,
      "num_tokens": 42495578.0,
      "step": 22790
    },
    {
      "entropy": 5.885344362258911,
      "epoch": 2.012981278700106,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004594121091808049,
      "loss": 5.1969,
      "mean_token_accuracy": 0.21920282989740372,
      "num_tokens": 42504532.0,
      "step": 22795
    },
    {
      "entropy": 5.872156238555908,
      "epoch": 2.0134228187919465,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045939406517439427,
      "loss": 5.2543,
      "mean_token_accuracy": 0.2080533429980278,
      "num_tokens": 42513683.0,
      "step": 22800
    },
    {
      "entropy": 5.92125301361084,
      "epoch": 2.0138643588837866,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045937601755580864,
      "loss": 5.3336,
      "mean_token_accuracy": 0.20605723708868026,
      "num_tokens": 42522717.0,
      "step": 22805
    },
    {
      "entropy": 5.932913970947266,
      "epoch": 2.014305898975627,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004593579663254018,
      "loss": 5.3516,
      "mean_token_accuracy": 0.20671084374189377,
      "num_tokens": 42531847.0,
      "step": 22810
    },
    {
      "entropy": 5.863706588745117,
      "epoch": 2.014747439067467,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004593399114835272,
      "loss": 5.2705,
      "mean_token_accuracy": 0.21442181318998338,
      "num_tokens": 42540869.0,
      "step": 22815
    },
    {
      "entropy": 5.8496442317962645,
      "epoch": 2.0151889791593076,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045932185303053857,
      "loss": 5.2563,
      "mean_token_accuracy": 0.21570290476083756,
      "num_tokens": 42550875.0,
      "step": 22820
    },
    {
      "entropy": 5.883614873886108,
      "epoch": 2.015630519251148,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004593037909667898,
      "loss": 5.2332,
      "mean_token_accuracy": 0.2173098623752594,
      "num_tokens": 42560066.0,
      "step": 22825
    },
    {
      "entropy": 5.836062431335449,
      "epoch": 2.016072059342988,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004592857252926344,
      "loss": 5.2487,
      "mean_token_accuracy": 0.2159641429781914,
      "num_tokens": 42568733.0,
      "step": 22830
    },
    {
      "entropy": 5.858053302764892,
      "epoch": 2.0165135994348287,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045926765600842653,
      "loss": 5.2856,
      "mean_token_accuracy": 0.20940887480974196,
      "num_tokens": 42577565.0,
      "step": 22835
    },
    {
      "entropy": 5.866206693649292,
      "epoch": 2.016955139526669,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045924958311452006,
      "loss": 5.3193,
      "mean_token_accuracy": 0.20851740688085557,
      "num_tokens": 42586489.0,
      "step": 22840
    },
    {
      "entropy": 5.892873382568359,
      "epoch": 2.017396679618509,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045923150661126904,
      "loss": 5.3567,
      "mean_token_accuracy": 0.20520659685134887,
      "num_tokens": 42596263.0,
      "step": 22845
    },
    {
      "entropy": 5.8912801265716555,
      "epoch": 2.0178382197103497,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004592134264990276,
      "loss": 5.232,
      "mean_token_accuracy": 0.21182719022035598,
      "num_tokens": 42605787.0,
      "step": 22850
    },
    {
      "entropy": 5.99868597984314,
      "epoch": 2.01827975980219,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045919534277814996,
      "loss": 5.4502,
      "mean_token_accuracy": 0.20145653635263444,
      "num_tokens": 42616014.0,
      "step": 22855
    },
    {
      "entropy": 5.913133955001831,
      "epoch": 2.0187212998940303,
      "grad_norm": 1.125,
      "learning_rate": 0.0004591772554489902,
      "loss": 5.3207,
      "mean_token_accuracy": 0.2100447028875351,
      "num_tokens": 42624839.0,
      "step": 22860
    },
    {
      "entropy": 5.905199146270752,
      "epoch": 2.0191628399858708,
      "grad_norm": 1.125,
      "learning_rate": 0.0004591591645119028,
      "loss": 5.3115,
      "mean_token_accuracy": 0.20897043645381927,
      "num_tokens": 42633811.0,
      "step": 22865
    },
    {
      "entropy": 5.8146604061126705,
      "epoch": 2.0196043800777113,
      "grad_norm": 1.125,
      "learning_rate": 0.00045914106996724217,
      "loss": 5.2131,
      "mean_token_accuracy": 0.21572656333446502,
      "num_tokens": 42643140.0,
      "step": 22870
    },
    {
      "entropy": 5.840064477920532,
      "epoch": 2.0200459201695513,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045912297181536257,
      "loss": 5.2172,
      "mean_token_accuracy": 0.2192785531282425,
      "num_tokens": 42651123.0,
      "step": 22875
    },
    {
      "entropy": 5.893158578872681,
      "epoch": 2.020487460261392,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045910487005661877,
      "loss": 5.2732,
      "mean_token_accuracy": 0.2179436832666397,
      "num_tokens": 42659880.0,
      "step": 22880
    },
    {
      "entropy": 5.878518772125244,
      "epoch": 2.020929000353232,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004590867646913653,
      "loss": 5.2157,
      "mean_token_accuracy": 0.22213126569986344,
      "num_tokens": 42668689.0,
      "step": 22885
    },
    {
      "entropy": 5.858295583724976,
      "epoch": 2.0213705404450724,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045906865571995666,
      "loss": 5.2611,
      "mean_token_accuracy": 0.21760706454515458,
      "num_tokens": 42678403.0,
      "step": 22890
    },
    {
      "entropy": 5.786852550506592,
      "epoch": 2.021812080536913,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004590505431427479,
      "loss": 5.2673,
      "mean_token_accuracy": 0.21441237777471542,
      "num_tokens": 42688203.0,
      "step": 22895
    },
    {
      "entropy": 5.888368701934814,
      "epoch": 2.022253620628753,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004590324269600936,
      "loss": 5.3581,
      "mean_token_accuracy": 0.20022109597921373,
      "num_tokens": 42698343.0,
      "step": 22900
    },
    {
      "entropy": 5.843066930770874,
      "epoch": 2.0226951607205934,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045901430717234875,
      "loss": 5.1911,
      "mean_token_accuracy": 0.22421941608190538,
      "num_tokens": 42707483.0,
      "step": 22905
    },
    {
      "entropy": 5.906141948699951,
      "epoch": 2.023136700812434,
      "grad_norm": 1.0,
      "learning_rate": 0.0004589961837798682,
      "loss": 5.3392,
      "mean_token_accuracy": 0.20942073464393615,
      "num_tokens": 42717820.0,
      "step": 22910
    },
    {
      "entropy": 5.895740938186646,
      "epoch": 2.023578240904274,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045897805678300714,
      "loss": 5.3087,
      "mean_token_accuracy": 0.21306860446929932,
      "num_tokens": 42726996.0,
      "step": 22915
    },
    {
      "entropy": 5.875725412368775,
      "epoch": 2.0240197809961145,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004589599261821205,
      "loss": 5.3059,
      "mean_token_accuracy": 0.2133346602320671,
      "num_tokens": 42735779.0,
      "step": 22920
    },
    {
      "entropy": 5.9298499584197994,
      "epoch": 2.024461321087955,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045894179197756364,
      "loss": 5.3848,
      "mean_token_accuracy": 0.2061198353767395,
      "num_tokens": 42745089.0,
      "step": 22925
    },
    {
      "entropy": 5.889540958404541,
      "epoch": 2.024902861179795,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00045892365416969164,
      "loss": 5.3199,
      "mean_token_accuracy": 0.2028687983751297,
      "num_tokens": 42754963.0,
      "step": 22930
    },
    {
      "entropy": 5.920861864089966,
      "epoch": 2.0253444012716355,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045890551275885985,
      "loss": 5.3594,
      "mean_token_accuracy": 0.20339042395353318,
      "num_tokens": 42764856.0,
      "step": 22935
    },
    {
      "entropy": 5.9101465225219725,
      "epoch": 2.025785941363476,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045888736774542363,
      "loss": 5.3182,
      "mean_token_accuracy": 0.2119409903883934,
      "num_tokens": 42775028.0,
      "step": 22940
    },
    {
      "entropy": 5.905062532424926,
      "epoch": 2.026227481455316,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004588692191297385,
      "loss": 5.3237,
      "mean_token_accuracy": 0.21250978857278824,
      "num_tokens": 42784288.0,
      "step": 22945
    },
    {
      "entropy": 5.912501239776612,
      "epoch": 2.0266690215471566,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004588510669121599,
      "loss": 5.3852,
      "mean_token_accuracy": 0.19859351217746735,
      "num_tokens": 42793356.0,
      "step": 22950
    },
    {
      "entropy": 5.805608034133911,
      "epoch": 2.0271105616389966,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045883291109304346,
      "loss": 5.2053,
      "mean_token_accuracy": 0.2177696481347084,
      "num_tokens": 42802287.0,
      "step": 22955
    },
    {
      "entropy": 5.923625135421753,
      "epoch": 2.027552101730837,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004588147516727449,
      "loss": 5.3307,
      "mean_token_accuracy": 0.21063922196626664,
      "num_tokens": 42812206.0,
      "step": 22960
    },
    {
      "entropy": 5.9731368064880375,
      "epoch": 2.0279936418226776,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004587965886516199,
      "loss": 5.3472,
      "mean_token_accuracy": 0.2017676427960396,
      "num_tokens": 42822046.0,
      "step": 22965
    },
    {
      "entropy": 5.87266206741333,
      "epoch": 2.0284351819145177,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045877842203002417,
      "loss": 5.2739,
      "mean_token_accuracy": 0.2126902461051941,
      "num_tokens": 42831228.0,
      "step": 22970
    },
    {
      "entropy": 5.8698663234710695,
      "epoch": 2.028876722006358,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045876025180831373,
      "loss": 5.3543,
      "mean_token_accuracy": 0.2101733535528183,
      "num_tokens": 42840683.0,
      "step": 22975
    },
    {
      "entropy": 5.90928955078125,
      "epoch": 2.0293182620981987,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045874207798684446,
      "loss": 5.273,
      "mean_token_accuracy": 0.21106504946947097,
      "num_tokens": 42849731.0,
      "step": 22980
    },
    {
      "entropy": 5.940430593490601,
      "epoch": 2.0297598021900387,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045872390056597235,
      "loss": 5.3328,
      "mean_token_accuracy": 0.20695223212242125,
      "num_tokens": 42858028.0,
      "step": 22985
    },
    {
      "entropy": 5.893157339096069,
      "epoch": 2.030201342281879,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004587057195460536,
      "loss": 5.3382,
      "mean_token_accuracy": 0.20177621990442277,
      "num_tokens": 42867340.0,
      "step": 22990
    },
    {
      "entropy": 5.964124965667724,
      "epoch": 2.0306428823737197,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045868753492744424,
      "loss": 5.3521,
      "mean_token_accuracy": 0.20110375732183455,
      "num_tokens": 42877006.0,
      "step": 22995
    },
    {
      "entropy": 5.850784778594971,
      "epoch": 2.0310844224655598,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004586693467105005,
      "loss": 5.2063,
      "mean_token_accuracy": 0.2156612440943718,
      "num_tokens": 42886324.0,
      "step": 23000
    },
    {
      "entropy": 5.92107253074646,
      "epoch": 2.0315259625574003,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004586511548955788,
      "loss": 5.3068,
      "mean_token_accuracy": 0.20806549489498138,
      "num_tokens": 42896411.0,
      "step": 23005
    },
    {
      "entropy": 5.804952096939087,
      "epoch": 2.0319675026492408,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004586329594830353,
      "loss": 5.2006,
      "mean_token_accuracy": 0.21858483254909516,
      "num_tokens": 42905606.0,
      "step": 23010
    },
    {
      "entropy": 5.783320331573487,
      "epoch": 2.032409042741081,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045861476047322664,
      "loss": 5.1788,
      "mean_token_accuracy": 0.2198178991675377,
      "num_tokens": 42914226.0,
      "step": 23015
    },
    {
      "entropy": 5.882395792007446,
      "epoch": 2.0328505828329213,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045859655786650925,
      "loss": 5.3622,
      "mean_token_accuracy": 0.2007863849401474,
      "num_tokens": 42923910.0,
      "step": 23020
    },
    {
      "entropy": 5.843041944503784,
      "epoch": 2.0332921229247614,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004585783516632397,
      "loss": 5.2073,
      "mean_token_accuracy": 0.23171699792146683,
      "num_tokens": 42933203.0,
      "step": 23025
    },
    {
      "entropy": 5.90801682472229,
      "epoch": 2.033733663016602,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004585601418637747,
      "loss": 5.2788,
      "mean_token_accuracy": 0.2077164724469185,
      "num_tokens": 42942143.0,
      "step": 23030
    },
    {
      "entropy": 5.861828327178955,
      "epoch": 2.0341752031084424,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004585419284684708,
      "loss": 5.3023,
      "mean_token_accuracy": 0.21032149642705916,
      "num_tokens": 42951432.0,
      "step": 23035
    },
    {
      "entropy": 5.8482099056243895,
      "epoch": 2.0346167432002824,
      "grad_norm": 1.078125,
      "learning_rate": 0.000458523711477685,
      "loss": 5.1956,
      "mean_token_accuracy": 0.21475542336702347,
      "num_tokens": 42961033.0,
      "step": 23040
    },
    {
      "entropy": 5.879715538024902,
      "epoch": 2.035058283292123,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045850549089177406,
      "loss": 5.1871,
      "mean_token_accuracy": 0.22608125060796738,
      "num_tokens": 42971371.0,
      "step": 23045
    },
    {
      "entropy": 5.969196462631226,
      "epoch": 2.0354998233839634,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045848726671109493,
      "loss": 5.3198,
      "mean_token_accuracy": 0.19820572286844254,
      "num_tokens": 42980445.0,
      "step": 23050
    },
    {
      "entropy": 5.8610227584838865,
      "epoch": 2.0359413634758035,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045846903893600457,
      "loss": 5.2904,
      "mean_token_accuracy": 0.20875540375709534,
      "num_tokens": 42989923.0,
      "step": 23055
    },
    {
      "entropy": 5.757601356506347,
      "epoch": 2.036382903567644,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045845080756686007,
      "loss": 5.1617,
      "mean_token_accuracy": 0.219550783932209,
      "num_tokens": 42998370.0,
      "step": 23060
    },
    {
      "entropy": 5.899701642990112,
      "epoch": 2.0368244436594845,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045843257260401855,
      "loss": 5.2656,
      "mean_token_accuracy": 0.21414129436016083,
      "num_tokens": 43007641.0,
      "step": 23065
    },
    {
      "entropy": 5.846440362930298,
      "epoch": 2.0372659837513245,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045841433404783733,
      "loss": 5.1923,
      "mean_token_accuracy": 0.21901296377182006,
      "num_tokens": 43016271.0,
      "step": 23070
    },
    {
      "entropy": 5.805743741989136,
      "epoch": 2.037707523843165,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004583960918986736,
      "loss": 5.1859,
      "mean_token_accuracy": 0.21661363542079926,
      "num_tokens": 43024525.0,
      "step": 23075
    },
    {
      "entropy": 5.8223387718200685,
      "epoch": 2.0381490639350055,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045837784615688473,
      "loss": 5.2871,
      "mean_token_accuracy": 0.21378793865442275,
      "num_tokens": 43033248.0,
      "step": 23080
    },
    {
      "entropy": 6.0096031665802006,
      "epoch": 2.0385906040268456,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004583595968228281,
      "loss": 5.4614,
      "mean_token_accuracy": 0.20177992582321166,
      "num_tokens": 43042171.0,
      "step": 23085
    },
    {
      "entropy": 5.879728603363037,
      "epoch": 2.039032144118686,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045834134389686137,
      "loss": 5.2216,
      "mean_token_accuracy": 0.22028322219848634,
      "num_tokens": 43051447.0,
      "step": 23090
    },
    {
      "entropy": 5.81450252532959,
      "epoch": 2.039473684210526,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004583230873793419,
      "loss": 5.1933,
      "mean_token_accuracy": 0.2075284019112587,
      "num_tokens": 43059869.0,
      "step": 23095
    },
    {
      "entropy": 5.7618450164794925,
      "epoch": 2.0399152243023666,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045830482727062736,
      "loss": 5.1734,
      "mean_token_accuracy": 0.21728577464818954,
      "num_tokens": 43068412.0,
      "step": 23100
    },
    {
      "entropy": 5.900337553024292,
      "epoch": 2.040356764394207,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045828656357107557,
      "loss": 5.3249,
      "mean_token_accuracy": 0.20258362889289855,
      "num_tokens": 43077130.0,
      "step": 23105
    },
    {
      "entropy": 5.941557168960571,
      "epoch": 2.040798304486047,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004582682962810442,
      "loss": 5.3504,
      "mean_token_accuracy": 0.20063544660806656,
      "num_tokens": 43086304.0,
      "step": 23110
    },
    {
      "entropy": 5.913349962234497,
      "epoch": 2.0412398445778877,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004582500254008912,
      "loss": 5.2715,
      "mean_token_accuracy": 0.21482429504394532,
      "num_tokens": 43095734.0,
      "step": 23115
    },
    {
      "entropy": 5.860885715484619,
      "epoch": 2.041681384669728,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045823175093097444,
      "loss": 5.255,
      "mean_token_accuracy": 0.2117638811469078,
      "num_tokens": 43104761.0,
      "step": 23120
    },
    {
      "entropy": 5.923091793060303,
      "epoch": 2.042122924761568,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004582134728716518,
      "loss": 5.3214,
      "mean_token_accuracy": 0.20944496244192123,
      "num_tokens": 43114182.0,
      "step": 23125
    },
    {
      "entropy": 5.832463979721069,
      "epoch": 2.0425644648534087,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004581951912232814,
      "loss": 5.2694,
      "mean_token_accuracy": 0.21437161713838576,
      "num_tokens": 43124893.0,
      "step": 23130
    },
    {
      "entropy": 5.894551181793213,
      "epoch": 2.043006004945249,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004581769059862215,
      "loss": 5.2739,
      "mean_token_accuracy": 0.21186769157648086,
      "num_tokens": 43133514.0,
      "step": 23135
    },
    {
      "entropy": 5.9472887992858885,
      "epoch": 2.0434475450370893,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004581586171608301,
      "loss": 5.366,
      "mean_token_accuracy": 0.2069222465157509,
      "num_tokens": 43142082.0,
      "step": 23140
    },
    {
      "entropy": 5.8757179260253904,
      "epoch": 2.0438890851289298,
      "grad_norm": 1.0,
      "learning_rate": 0.0004581403247474657,
      "loss": 5.3086,
      "mean_token_accuracy": 0.20761488974094391,
      "num_tokens": 43151466.0,
      "step": 23145
    },
    {
      "entropy": 5.928344774246216,
      "epoch": 2.0443306252207702,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004581220287464864,
      "loss": 5.3791,
      "mean_token_accuracy": 0.19838937073946,
      "num_tokens": 43160631.0,
      "step": 23150
    },
    {
      "entropy": 5.902462577819824,
      "epoch": 2.0447721653126103,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004581037291582507,
      "loss": 5.3461,
      "mean_token_accuracy": 0.21122662723064423,
      "num_tokens": 43170968.0,
      "step": 23155
    },
    {
      "entropy": 5.880038213729859,
      "epoch": 2.045213705404451,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004580854259831171,
      "loss": 5.2705,
      "mean_token_accuracy": 0.20816658437252045,
      "num_tokens": 43179661.0,
      "step": 23160
    },
    {
      "entropy": 5.806720542907715,
      "epoch": 2.045655245496291,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004580671192214441,
      "loss": 5.1291,
      "mean_token_accuracy": 0.22906865179538727,
      "num_tokens": 43188994.0,
      "step": 23165
    },
    {
      "entropy": 5.97022442817688,
      "epoch": 2.0460967855881314,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045804880887359036,
      "loss": 5.3974,
      "mean_token_accuracy": 0.20026474744081496,
      "num_tokens": 43199001.0,
      "step": 23170
    },
    {
      "entropy": 5.896676683425904,
      "epoch": 2.046538325679972,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004580304949399146,
      "loss": 5.2541,
      "mean_token_accuracy": 0.21216757595539093,
      "num_tokens": 43208113.0,
      "step": 23175
    },
    {
      "entropy": 5.838364839553833,
      "epoch": 2.046979865771812,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045801217742077544,
      "loss": 5.2629,
      "mean_token_accuracy": 0.2162158265709877,
      "num_tokens": 43216983.0,
      "step": 23180
    },
    {
      "entropy": 5.87812442779541,
      "epoch": 2.0474214058636524,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004579938563165319,
      "loss": 5.2582,
      "mean_token_accuracy": 0.21697108447551727,
      "num_tokens": 43225965.0,
      "step": 23185
    },
    {
      "entropy": 5.883720636367798,
      "epoch": 2.047862945955493,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004579755316275427,
      "loss": 5.1993,
      "mean_token_accuracy": 0.2154235154390335,
      "num_tokens": 43235078.0,
      "step": 23190
    },
    {
      "entropy": 5.854296731948852,
      "epoch": 2.048304486047333,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000457957203354167,
      "loss": 5.312,
      "mean_token_accuracy": 0.21079981178045273,
      "num_tokens": 43244398.0,
      "step": 23195
    },
    {
      "entropy": 5.898467397689819,
      "epoch": 2.0487460261391734,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045793887149676366,
      "loss": 5.3475,
      "mean_token_accuracy": 0.20842875689268112,
      "num_tokens": 43254638.0,
      "step": 23200
    },
    {
      "entropy": 5.853589296340942,
      "epoch": 2.049187566231014,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000457920536055692,
      "loss": 5.2556,
      "mean_token_accuracy": 0.20489853322505952,
      "num_tokens": 43263287.0,
      "step": 23205
    },
    {
      "entropy": 5.824107694625854,
      "epoch": 2.049629106322854,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045790219703131097,
      "loss": 5.1991,
      "mean_token_accuracy": 0.21740050315856935,
      "num_tokens": 43272921.0,
      "step": 23210
    },
    {
      "entropy": 5.834905529022217,
      "epoch": 2.0500706464146945,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045788385442397994,
      "loss": 5.2603,
      "mean_token_accuracy": 0.21566871106624602,
      "num_tokens": 43283423.0,
      "step": 23215
    },
    {
      "entropy": 5.955960464477539,
      "epoch": 2.050512186506535,
      "grad_norm": 1.0,
      "learning_rate": 0.0004578655082340582,
      "loss": 5.3722,
      "mean_token_accuracy": 0.2022867813706398,
      "num_tokens": 43293031.0,
      "step": 23220
    },
    {
      "entropy": 5.95057225227356,
      "epoch": 2.050953726598375,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004578471584619051,
      "loss": 5.3189,
      "mean_token_accuracy": 0.21219599545001983,
      "num_tokens": 43302775.0,
      "step": 23225
    },
    {
      "entropy": 5.794325494766236,
      "epoch": 2.0513952666902155,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004578288051078803,
      "loss": 5.2276,
      "mean_token_accuracy": 0.219016931951046,
      "num_tokens": 43312341.0,
      "step": 23230
    },
    {
      "entropy": 5.892101144790649,
      "epoch": 2.0518368067820556,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045781044817234305,
      "loss": 5.2703,
      "mean_token_accuracy": 0.20737400501966477,
      "num_tokens": 43321358.0,
      "step": 23235
    },
    {
      "entropy": 5.87251033782959,
      "epoch": 2.052278346873896,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045779208765565324,
      "loss": 5.2762,
      "mean_token_accuracy": 0.21060893982648848,
      "num_tokens": 43331839.0,
      "step": 23240
    },
    {
      "entropy": 5.8935997009277346,
      "epoch": 2.0527198869657366,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004577737235581702,
      "loss": 5.3401,
      "mean_token_accuracy": 0.20013512820005416,
      "num_tokens": 43340384.0,
      "step": 23245
    },
    {
      "entropy": 5.974007654190063,
      "epoch": 2.0531614270575766,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045775535588025404,
      "loss": 5.3282,
      "mean_token_accuracy": 0.20916105359792708,
      "num_tokens": 43349264.0,
      "step": 23250
    },
    {
      "entropy": 5.876816987991333,
      "epoch": 2.053602967149417,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004577369846222643,
      "loss": 5.3145,
      "mean_token_accuracy": 0.21036041229963304,
      "num_tokens": 43358598.0,
      "step": 23255
    },
    {
      "entropy": 5.8475799560546875,
      "epoch": 2.0540445072412576,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004577186097845609,
      "loss": 5.2732,
      "mean_token_accuracy": 0.20657162964344025,
      "num_tokens": 43368357.0,
      "step": 23260
    },
    {
      "entropy": 5.932975196838379,
      "epoch": 2.0544860473330977,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045770023136750405,
      "loss": 5.3351,
      "mean_token_accuracy": 0.21048603802919388,
      "num_tokens": 43378143.0,
      "step": 23265
    },
    {
      "entropy": 5.830256748199463,
      "epoch": 2.054927587424938,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004576818493714534,
      "loss": 5.2482,
      "mean_token_accuracy": 0.22011639922857285,
      "num_tokens": 43387010.0,
      "step": 23270
    },
    {
      "entropy": 5.770633840560913,
      "epoch": 2.0553691275167787,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004576634637967693,
      "loss": 5.1353,
      "mean_token_accuracy": 0.21366823017597197,
      "num_tokens": 43395864.0,
      "step": 23275
    },
    {
      "entropy": 5.874236440658569,
      "epoch": 2.0558106676086187,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004576450746438118,
      "loss": 5.238,
      "mean_token_accuracy": 0.21384260058403015,
      "num_tokens": 43404662.0,
      "step": 23280
    },
    {
      "entropy": 5.947582578659057,
      "epoch": 2.0562522077004592,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004576266819129412,
      "loss": 5.2634,
      "mean_token_accuracy": 0.21163098365068436,
      "num_tokens": 43413927.0,
      "step": 23285
    },
    {
      "entropy": 5.932430458068848,
      "epoch": 2.0566937477922997,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004576082856045177,
      "loss": 5.4263,
      "mean_token_accuracy": 0.1927229180932045,
      "num_tokens": 43424197.0,
      "step": 23290
    },
    {
      "entropy": 5.950639772415161,
      "epoch": 2.05713528788414,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045758988571890176,
      "loss": 5.3221,
      "mean_token_accuracy": 0.20873454362154006,
      "num_tokens": 43432630.0,
      "step": 23295
    },
    {
      "entropy": 5.867905569076538,
      "epoch": 2.0575768279759803,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045757148225645383,
      "loss": 5.3066,
      "mean_token_accuracy": 0.2055157884955406,
      "num_tokens": 43442186.0,
      "step": 23300
    },
    {
      "entropy": 5.831416416168213,
      "epoch": 2.0580183680678203,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045755307521753443,
      "loss": 5.2489,
      "mean_token_accuracy": 0.21740500181913375,
      "num_tokens": 43451927.0,
      "step": 23305
    },
    {
      "entropy": 5.938334369659424,
      "epoch": 2.058459908159661,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045753466460250405,
      "loss": 5.3021,
      "mean_token_accuracy": 0.2072618693113327,
      "num_tokens": 43460879.0,
      "step": 23310
    },
    {
      "entropy": 5.87259202003479,
      "epoch": 2.0589014482515013,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004575162504117234,
      "loss": 5.3632,
      "mean_token_accuracy": 0.20011452138423919,
      "num_tokens": 43469918.0,
      "step": 23315
    },
    {
      "entropy": 5.879549407958985,
      "epoch": 2.0593429883433414,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004574978326455532,
      "loss": 5.2864,
      "mean_token_accuracy": 0.20705562978982925,
      "num_tokens": 43478779.0,
      "step": 23320
    },
    {
      "entropy": 5.918912124633789,
      "epoch": 2.059784528435182,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004574794113043543,
      "loss": 5.3546,
      "mean_token_accuracy": 0.20848302245140077,
      "num_tokens": 43488128.0,
      "step": 23325
    },
    {
      "entropy": 5.86453948020935,
      "epoch": 2.0602260685270224,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004574609863884875,
      "loss": 5.2933,
      "mean_token_accuracy": 0.2075021594762802,
      "num_tokens": 43497567.0,
      "step": 23330
    },
    {
      "entropy": 5.893348693847656,
      "epoch": 2.0606676086188624,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045744255789831377,
      "loss": 5.3112,
      "mean_token_accuracy": 0.20413670539855958,
      "num_tokens": 43507360.0,
      "step": 23335
    },
    {
      "entropy": 5.879449033737183,
      "epoch": 2.061109148710703,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045742412583419403,
      "loss": 5.3127,
      "mean_token_accuracy": 0.20965750366449357,
      "num_tokens": 43517601.0,
      "step": 23340
    },
    {
      "entropy": 5.8456395149230955,
      "epoch": 2.0615506888025434,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045740569019648945,
      "loss": 5.2136,
      "mean_token_accuracy": 0.2215661346912384,
      "num_tokens": 43526348.0,
      "step": 23345
    },
    {
      "entropy": 5.860084772109985,
      "epoch": 2.0619922288943835,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004573872509855612,
      "loss": 5.2259,
      "mean_token_accuracy": 0.2159516081213951,
      "num_tokens": 43536106.0,
      "step": 23350
    },
    {
      "entropy": 5.958714532852173,
      "epoch": 2.062433768986224,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004573688082017703,
      "loss": 5.3625,
      "mean_token_accuracy": 0.20386123806238174,
      "num_tokens": 43545110.0,
      "step": 23355
    },
    {
      "entropy": 5.907396459579468,
      "epoch": 2.0628753090780645,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045735036184547825,
      "loss": 5.2947,
      "mean_token_accuracy": 0.203805835545063,
      "num_tokens": 43554226.0,
      "step": 23360
    },
    {
      "entropy": 5.851353502273559,
      "epoch": 2.0633168491699045,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045733191191704633,
      "loss": 5.2987,
      "mean_token_accuracy": 0.21612913608551027,
      "num_tokens": 43563184.0,
      "step": 23365
    },
    {
      "entropy": 5.859892988204956,
      "epoch": 2.063758389261745,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004573134584168359,
      "loss": 5.28,
      "mean_token_accuracy": 0.22314720451831818,
      "num_tokens": 43572595.0,
      "step": 23370
    },
    {
      "entropy": 5.91445026397705,
      "epoch": 2.064199929353585,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004572950013452086,
      "loss": 5.3716,
      "mean_token_accuracy": 0.19807493388652803,
      "num_tokens": 43582211.0,
      "step": 23375
    },
    {
      "entropy": 5.9036365985870365,
      "epoch": 2.0646414694454256,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004572765407025257,
      "loss": 5.2272,
      "mean_token_accuracy": 0.2168590322136879,
      "num_tokens": 43591060.0,
      "step": 23380
    },
    {
      "entropy": 5.975190877914429,
      "epoch": 2.065083009537266,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004572580764891493,
      "loss": 5.3986,
      "mean_token_accuracy": 0.20219420790672302,
      "num_tokens": 43600057.0,
      "step": 23385
    },
    {
      "entropy": 5.89881591796875,
      "epoch": 2.065524549629106,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004572396087054406,
      "loss": 5.1932,
      "mean_token_accuracy": 0.21935433149337769,
      "num_tokens": 43608609.0,
      "step": 23390
    },
    {
      "entropy": 5.852277374267578,
      "epoch": 2.0659660897209466,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004572211373517618,
      "loss": 5.3267,
      "mean_token_accuracy": 0.20932533144950866,
      "num_tokens": 43618751.0,
      "step": 23395
    },
    {
      "entropy": 5.926204824447632,
      "epoch": 2.066407629812787,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045720266242847446,
      "loss": 5.4134,
      "mean_token_accuracy": 0.20356207489967346,
      "num_tokens": 43628934.0,
      "step": 23400
    },
    {
      "entropy": 5.960275554656983,
      "epoch": 2.066849169904627,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045718418393594055,
      "loss": 5.2958,
      "mean_token_accuracy": 0.21271106153726577,
      "num_tokens": 43638512.0,
      "step": 23405
    },
    {
      "entropy": 5.8644452571868895,
      "epoch": 2.0672907099964677,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045716570187452214,
      "loss": 5.2471,
      "mean_token_accuracy": 0.20868025869131088,
      "num_tokens": 43647708.0,
      "step": 23410
    },
    {
      "entropy": 5.898849296569824,
      "epoch": 2.067732250088308,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004571472162445813,
      "loss": 5.3225,
      "mean_token_accuracy": 0.21653203368186952,
      "num_tokens": 43657096.0,
      "step": 23415
    },
    {
      "entropy": 5.916511583328247,
      "epoch": 2.0681737901801482,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045712872704647994,
      "loss": 5.3291,
      "mean_token_accuracy": 0.2047939881682396,
      "num_tokens": 43667410.0,
      "step": 23420
    },
    {
      "entropy": 5.86441388130188,
      "epoch": 2.0686153302719887,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004571102342805805,
      "loss": 5.2766,
      "mean_token_accuracy": 0.21467859596014022,
      "num_tokens": 43675800.0,
      "step": 23425
    },
    {
      "entropy": 5.853344249725342,
      "epoch": 2.0690568703638292,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004570917379472451,
      "loss": 5.2842,
      "mean_token_accuracy": 0.21802525073289872,
      "num_tokens": 43684802.0,
      "step": 23430
    },
    {
      "entropy": 5.852845191955566,
      "epoch": 2.0694984104556693,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045707323804683616,
      "loss": 5.2881,
      "mean_token_accuracy": 0.21087391823530197,
      "num_tokens": 43693968.0,
      "step": 23435
    },
    {
      "entropy": 5.971442413330078,
      "epoch": 2.06993995054751,
      "grad_norm": 1.15625,
      "learning_rate": 0.000457054734579716,
      "loss": 5.3038,
      "mean_token_accuracy": 0.20274436473846436,
      "num_tokens": 43703177.0,
      "step": 23440
    },
    {
      "entropy": 5.871698904037475,
      "epoch": 2.07038149063935,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00045703622754624716,
      "loss": 5.3242,
      "mean_token_accuracy": 0.20798488110303878,
      "num_tokens": 43712894.0,
      "step": 23445
    },
    {
      "entropy": 5.898809242248535,
      "epoch": 2.0708230307311903,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045701771694679213,
      "loss": 5.2313,
      "mean_token_accuracy": 0.2106014609336853,
      "num_tokens": 43722591.0,
      "step": 23450
    },
    {
      "entropy": 5.831428050994873,
      "epoch": 2.071264570823031,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045699920278171355,
      "loss": 5.2211,
      "mean_token_accuracy": 0.22353032380342483,
      "num_tokens": 43732251.0,
      "step": 23455
    },
    {
      "entropy": 5.955831289291382,
      "epoch": 2.071706110914871,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004569806850513741,
      "loss": 5.3866,
      "mean_token_accuracy": 0.20358723849058152,
      "num_tokens": 43741061.0,
      "step": 23460
    },
    {
      "entropy": 5.960063028335571,
      "epoch": 2.0721476510067114,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00045696216375613647,
      "loss": 5.344,
      "mean_token_accuracy": 0.20664846152067184,
      "num_tokens": 43750147.0,
      "step": 23465
    },
    {
      "entropy": 5.944277334213257,
      "epoch": 2.072589191098552,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004569436388963636,
      "loss": 5.3813,
      "mean_token_accuracy": 0.19914156794548035,
      "num_tokens": 43760960.0,
      "step": 23470
    },
    {
      "entropy": 5.876435232162476,
      "epoch": 2.073030731190392,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004569251104724184,
      "loss": 5.29,
      "mean_token_accuracy": 0.21085326373577118,
      "num_tokens": 43771306.0,
      "step": 23475
    },
    {
      "entropy": 5.827397775650025,
      "epoch": 2.0734722712822324,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004569065784846636,
      "loss": 5.2162,
      "mean_token_accuracy": 0.21310461759567262,
      "num_tokens": 43780520.0,
      "step": 23480
    },
    {
      "entropy": 5.924309492111206,
      "epoch": 2.073913811374073,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004568880429334624,
      "loss": 5.3474,
      "mean_token_accuracy": 0.20316143929958344,
      "num_tokens": 43789472.0,
      "step": 23485
    },
    {
      "entropy": 5.944648361206054,
      "epoch": 2.074355351465913,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045686950381917805,
      "loss": 5.3329,
      "mean_token_accuracy": 0.20691990852355957,
      "num_tokens": 43798728.0,
      "step": 23490
    },
    {
      "entropy": 5.900263738632202,
      "epoch": 2.0747968915577535,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045685096114217336,
      "loss": 5.2694,
      "mean_token_accuracy": 0.21408158540725708,
      "num_tokens": 43807959.0,
      "step": 23495
    },
    {
      "entropy": 5.9307873249053955,
      "epoch": 2.075238431649594,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004568324149028119,
      "loss": 5.2792,
      "mean_token_accuracy": 0.21293805688619613,
      "num_tokens": 43816922.0,
      "step": 23500
    },
    {
      "entropy": 5.865668439865113,
      "epoch": 2.075679971741434,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004568138651014567,
      "loss": 5.2798,
      "mean_token_accuracy": 0.21039648205041886,
      "num_tokens": 43827195.0,
      "step": 23505
    },
    {
      "entropy": 5.8485218524932865,
      "epoch": 2.0761215118332745,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004567953117384714,
      "loss": 5.2377,
      "mean_token_accuracy": 0.2087216168642044,
      "num_tokens": 43836289.0,
      "step": 23510
    },
    {
      "entropy": 5.925881385803223,
      "epoch": 2.0765630519251146,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004567767548142193,
      "loss": 5.3305,
      "mean_token_accuracy": 0.1997263640165329,
      "num_tokens": 43845918.0,
      "step": 23515
    },
    {
      "entropy": 5.949883651733399,
      "epoch": 2.077004592016955,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045675819432906394,
      "loss": 5.3831,
      "mean_token_accuracy": 0.20143934786319734,
      "num_tokens": 43853907.0,
      "step": 23520
    },
    {
      "entropy": 5.926950120925904,
      "epoch": 2.0774461321087956,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045673963028336896,
      "loss": 5.2985,
      "mean_token_accuracy": 0.2107247456908226,
      "num_tokens": 43862992.0,
      "step": 23525
    },
    {
      "entropy": 5.944992971420288,
      "epoch": 2.0778876722006356,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045672106267749795,
      "loss": 5.3076,
      "mean_token_accuracy": 0.20400300472974778,
      "num_tokens": 43871686.0,
      "step": 23530
    },
    {
      "entropy": 5.861617517471314,
      "epoch": 2.078329212292476,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004567024915118148,
      "loss": 5.2705,
      "mean_token_accuracy": 0.21350021064281463,
      "num_tokens": 43881302.0,
      "step": 23535
    },
    {
      "entropy": 5.98052225112915,
      "epoch": 2.0787707523843166,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045668391678668296,
      "loss": 5.3715,
      "mean_token_accuracy": 0.1990432098507881,
      "num_tokens": 43889982.0,
      "step": 23540
    },
    {
      "entropy": 5.944049739837647,
      "epoch": 2.0792122924761567,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045666533850246665,
      "loss": 5.3208,
      "mean_token_accuracy": 0.21239465922117234,
      "num_tokens": 43898385.0,
      "step": 23545
    },
    {
      "entropy": 5.903633546829224,
      "epoch": 2.079653832567997,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004566467566595297,
      "loss": 5.2903,
      "mean_token_accuracy": 0.20155944675207138,
      "num_tokens": 43907882.0,
      "step": 23550
    },
    {
      "entropy": 5.994525814056397,
      "epoch": 2.0800953726598377,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045662817125823605,
      "loss": 5.3112,
      "mean_token_accuracy": 0.20893317013978957,
      "num_tokens": 43916222.0,
      "step": 23555
    },
    {
      "entropy": 5.881528663635254,
      "epoch": 2.0805369127516777,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045660958229894985,
      "loss": 5.2574,
      "mean_token_accuracy": 0.2135061115026474,
      "num_tokens": 43925181.0,
      "step": 23560
    },
    {
      "entropy": 5.7951517581939695,
      "epoch": 2.0809784528435182,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045659098978203524,
      "loss": 5.3031,
      "mean_token_accuracy": 0.21031264066696168,
      "num_tokens": 43933973.0,
      "step": 23565
    },
    {
      "entropy": 5.808804893493653,
      "epoch": 2.0814199929353587,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004565723937078563,
      "loss": 5.291,
      "mean_token_accuracy": 0.2105001538991928,
      "num_tokens": 43943297.0,
      "step": 23570
    },
    {
      "entropy": 5.924958324432373,
      "epoch": 2.081861533027199,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045655379407677754,
      "loss": 5.325,
      "mean_token_accuracy": 0.21227133572101592,
      "num_tokens": 43952679.0,
      "step": 23575
    },
    {
      "entropy": 5.91604175567627,
      "epoch": 2.0823030731190393,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004565351908891632,
      "loss": 5.3078,
      "mean_token_accuracy": 0.20379597544670106,
      "num_tokens": 43961629.0,
      "step": 23580
    },
    {
      "entropy": 5.919429159164428,
      "epoch": 2.0827446132108793,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045651658414537765,
      "loss": 5.3176,
      "mean_token_accuracy": 0.20624782890081406,
      "num_tokens": 43970880.0,
      "step": 23585
    },
    {
      "entropy": 5.86760778427124,
      "epoch": 2.08318615330272,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004564979738457855,
      "loss": 5.243,
      "mean_token_accuracy": 0.21680566668510437,
      "num_tokens": 43980449.0,
      "step": 23590
    },
    {
      "entropy": 5.806300783157349,
      "epoch": 2.0836276933945603,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045647935999075127,
      "loss": 5.268,
      "mean_token_accuracy": 0.2086215451359749,
      "num_tokens": 43988716.0,
      "step": 23595
    },
    {
      "entropy": 5.899968242645263,
      "epoch": 2.0840692334864004,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045646074258063956,
      "loss": 5.3184,
      "mean_token_accuracy": 0.2169812336564064,
      "num_tokens": 43998464.0,
      "step": 23600
    },
    {
      "entropy": 5.960954332351685,
      "epoch": 2.084510773578241,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004564421216158152,
      "loss": 5.3207,
      "mean_token_accuracy": 0.21247612088918685,
      "num_tokens": 44006993.0,
      "step": 23605
    },
    {
      "entropy": 5.959644079208374,
      "epoch": 2.0849523136700814,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004564234970966428,
      "loss": 5.3605,
      "mean_token_accuracy": 0.20201978981494903,
      "num_tokens": 44015801.0,
      "step": 23610
    },
    {
      "entropy": 5.814644622802734,
      "epoch": 2.0853938537619214,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004564048690234872,
      "loss": 5.2859,
      "mean_token_accuracy": 0.20764555484056474,
      "num_tokens": 44025069.0,
      "step": 23615
    },
    {
      "entropy": 5.928499841690064,
      "epoch": 2.085835393853762,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045638623739671343,
      "loss": 5.4187,
      "mean_token_accuracy": 0.19506964981555938,
      "num_tokens": 44034700.0,
      "step": 23620
    },
    {
      "entropy": 5.920329284667969,
      "epoch": 2.0862769339456024,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045636760221668646,
      "loss": 5.2299,
      "mean_token_accuracy": 0.21019828021526338,
      "num_tokens": 44044417.0,
      "step": 23625
    },
    {
      "entropy": 5.955547571182251,
      "epoch": 2.0867184740374425,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004563489634837713,
      "loss": 5.2978,
      "mean_token_accuracy": 0.2106105163693428,
      "num_tokens": 44053617.0,
      "step": 23630
    },
    {
      "entropy": 5.844348382949829,
      "epoch": 2.087160014129283,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045633032119833314,
      "loss": 5.2449,
      "mean_token_accuracy": 0.20604305118322372,
      "num_tokens": 44062738.0,
      "step": 23635
    },
    {
      "entropy": 5.903376245498658,
      "epoch": 2.0876015542211235,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000456311675360737,
      "loss": 5.3416,
      "mean_token_accuracy": 0.19985350221395493,
      "num_tokens": 44072154.0,
      "step": 23640
    },
    {
      "entropy": 5.9602282524108885,
      "epoch": 2.0880430943129635,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004562930259713483,
      "loss": 5.4065,
      "mean_token_accuracy": 0.20159982442855834,
      "num_tokens": 44081239.0,
      "step": 23645
    },
    {
      "entropy": 5.894659423828125,
      "epoch": 2.088484634404804,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045627437303053236,
      "loss": 5.2513,
      "mean_token_accuracy": 0.21409705579280852,
      "num_tokens": 44090866.0,
      "step": 23650
    },
    {
      "entropy": 5.928786420822144,
      "epoch": 2.088926174496644,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004562557165386546,
      "loss": 5.4022,
      "mean_token_accuracy": 0.2024052545428276,
      "num_tokens": 44100225.0,
      "step": 23655
    },
    {
      "entropy": 5.900218963623047,
      "epoch": 2.0893677145884846,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004562370564960804,
      "loss": 5.3129,
      "mean_token_accuracy": 0.20917170196771623,
      "num_tokens": 44110139.0,
      "step": 23660
    },
    {
      "entropy": 5.883913373947143,
      "epoch": 2.089809254680325,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045621839290317526,
      "loss": 5.3128,
      "mean_token_accuracy": 0.2127288267016411,
      "num_tokens": 44118687.0,
      "step": 23665
    },
    {
      "entropy": 5.863697481155396,
      "epoch": 2.090250794772165,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045619972576030495,
      "loss": 5.295,
      "mean_token_accuracy": 0.20769568234682084,
      "num_tokens": 44127377.0,
      "step": 23670
    },
    {
      "entropy": 5.851103830337524,
      "epoch": 2.0906923348640056,
      "grad_norm": 1.015625,
      "learning_rate": 0.000456181055067835,
      "loss": 5.322,
      "mean_token_accuracy": 0.2059984490275383,
      "num_tokens": 44136583.0,
      "step": 23675
    },
    {
      "entropy": 5.866923856735229,
      "epoch": 2.091133874955846,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004561623808261313,
      "loss": 5.3066,
      "mean_token_accuracy": 0.20673833340406417,
      "num_tokens": 44145467.0,
      "step": 23680
    },
    {
      "entropy": 5.860962438583374,
      "epoch": 2.091575415047686,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004561437030355595,
      "loss": 5.2872,
      "mean_token_accuracy": 0.21174018681049347,
      "num_tokens": 44154932.0,
      "step": 23685
    },
    {
      "entropy": 5.835395765304566,
      "epoch": 2.0920169551395267,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004561250216964857,
      "loss": 5.2177,
      "mean_token_accuracy": 0.216437166929245,
      "num_tokens": 44162791.0,
      "step": 23690
    },
    {
      "entropy": 5.955163192749024,
      "epoch": 2.092458495231367,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004561063368092757,
      "loss": 5.3477,
      "mean_token_accuracy": 0.21063789576292039,
      "num_tokens": 44171993.0,
      "step": 23695
    },
    {
      "entropy": 5.9274516105651855,
      "epoch": 2.0929000353232072,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045608764837429543,
      "loss": 5.3483,
      "mean_token_accuracy": 0.207102532684803,
      "num_tokens": 44180807.0,
      "step": 23700
    },
    {
      "entropy": 5.9769903182983395,
      "epoch": 2.0933415754150477,
      "grad_norm": 1.0,
      "learning_rate": 0.00045606895639191126,
      "loss": 5.3712,
      "mean_token_accuracy": 0.20448126792907714,
      "num_tokens": 44191764.0,
      "step": 23705
    },
    {
      "entropy": 5.8805944442749025,
      "epoch": 2.0937831155068882,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045605026086248916,
      "loss": 5.3658,
      "mean_token_accuracy": 0.205495585501194,
      "num_tokens": 44201165.0,
      "step": 23710
    },
    {
      "entropy": 5.906459140777588,
      "epoch": 2.0942246555987283,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004560315617863954,
      "loss": 5.362,
      "mean_token_accuracy": 0.20707882344722747,
      "num_tokens": 44210709.0,
      "step": 23715
    },
    {
      "entropy": 5.90650110244751,
      "epoch": 2.0946661956905688,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004560128591639964,
      "loss": 5.322,
      "mean_token_accuracy": 0.20687424540519714,
      "num_tokens": 44219827.0,
      "step": 23720
    },
    {
      "entropy": 5.9476954460144045,
      "epoch": 2.095107735782409,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004559941529956584,
      "loss": 5.3852,
      "mean_token_accuracy": 0.20104854106903075,
      "num_tokens": 44229498.0,
      "step": 23725
    },
    {
      "entropy": 5.884517669677734,
      "epoch": 2.0955492758742493,
      "grad_norm": 1.0,
      "learning_rate": 0.0004559754432817479,
      "loss": 5.2882,
      "mean_token_accuracy": 0.20786643475294114,
      "num_tokens": 44238844.0,
      "step": 23730
    },
    {
      "entropy": 5.85141773223877,
      "epoch": 2.09599081596609,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045595673002263125,
      "loss": 5.2867,
      "mean_token_accuracy": 0.21021698862314225,
      "num_tokens": 44247512.0,
      "step": 23735
    },
    {
      "entropy": 5.870313405990601,
      "epoch": 2.09643235605793,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004559380132186753,
      "loss": 5.233,
      "mean_token_accuracy": 0.21559084057807923,
      "num_tokens": 44256485.0,
      "step": 23740
    },
    {
      "entropy": 5.785613679885865,
      "epoch": 2.0968738961497704,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045591929287024655,
      "loss": 5.1362,
      "mean_token_accuracy": 0.2268437996506691,
      "num_tokens": 44265095.0,
      "step": 23745
    },
    {
      "entropy": 5.952377986907959,
      "epoch": 2.097315436241611,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004559005689777117,
      "loss": 5.3197,
      "mean_token_accuracy": 0.20878552198410033,
      "num_tokens": 44274582.0,
      "step": 23750
    },
    {
      "entropy": 5.964988803863525,
      "epoch": 2.097756976333451,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004558818415414377,
      "loss": 5.3585,
      "mean_token_accuracy": 0.20715403109788894,
      "num_tokens": 44285752.0,
      "step": 23755
    },
    {
      "entropy": 5.932619190216064,
      "epoch": 2.0981985164252914,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004558631105617912,
      "loss": 5.3519,
      "mean_token_accuracy": 0.20505301356315614,
      "num_tokens": 44295315.0,
      "step": 23760
    },
    {
      "entropy": 5.882716035842895,
      "epoch": 2.098640056517132,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045584437603913923,
      "loss": 5.4127,
      "mean_token_accuracy": 0.20295526087284088,
      "num_tokens": 44305151.0,
      "step": 23765
    },
    {
      "entropy": 5.892424249649048,
      "epoch": 2.099081596608972,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045582563797384886,
      "loss": 5.3798,
      "mean_token_accuracy": 0.2029080092906952,
      "num_tokens": 44314462.0,
      "step": 23770
    },
    {
      "entropy": 5.89073224067688,
      "epoch": 2.0995231367008125,
      "grad_norm": 1.109375,
      "learning_rate": 0.000455806896366287,
      "loss": 5.1972,
      "mean_token_accuracy": 0.2278853550553322,
      "num_tokens": 44322782.0,
      "step": 23775
    },
    {
      "entropy": 5.886151266098023,
      "epoch": 2.099964676792653,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004557881512168209,
      "loss": 5.2521,
      "mean_token_accuracy": 0.21490204632282256,
      "num_tokens": 44331588.0,
      "step": 23780
    },
    {
      "entropy": 5.861545515060425,
      "epoch": 2.100406216884493,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004557694025258177,
      "loss": 5.297,
      "mean_token_accuracy": 0.218393275141716,
      "num_tokens": 44340997.0,
      "step": 23785
    },
    {
      "entropy": 5.890779113769531,
      "epoch": 2.1008477569763335,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004557506502936448,
      "loss": 5.3388,
      "mean_token_accuracy": 0.20636630952358245,
      "num_tokens": 44350679.0,
      "step": 23790
    },
    {
      "entropy": 5.870981740951538,
      "epoch": 2.1012892970681736,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045573189452066936,
      "loss": 5.2414,
      "mean_token_accuracy": 0.20990516096353531,
      "num_tokens": 44359934.0,
      "step": 23795
    },
    {
      "entropy": 5.852165365219117,
      "epoch": 2.101730837160014,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045571313520725897,
      "loss": 5.2844,
      "mean_token_accuracy": 0.2032035231590271,
      "num_tokens": 44369104.0,
      "step": 23800
    },
    {
      "entropy": 5.898313808441162,
      "epoch": 2.1021723772518546,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045569437235378103,
      "loss": 5.2945,
      "mean_token_accuracy": 0.211219222843647,
      "num_tokens": 44379645.0,
      "step": 23805
    },
    {
      "entropy": 5.959102010726928,
      "epoch": 2.1026139173436946,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004556756059606031,
      "loss": 5.3457,
      "mean_token_accuracy": 0.20944539457559586,
      "num_tokens": 44389276.0,
      "step": 23810
    },
    {
      "entropy": 5.868195581436157,
      "epoch": 2.103055457435535,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004556568360280928,
      "loss": 5.2925,
      "mean_token_accuracy": 0.20601432025432587,
      "num_tokens": 44399114.0,
      "step": 23815
    },
    {
      "entropy": 5.74198055267334,
      "epoch": 2.1034969975273756,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045563806255661784,
      "loss": 5.0541,
      "mean_token_accuracy": 0.236716628074646,
      "num_tokens": 44406917.0,
      "step": 23820
    },
    {
      "entropy": 5.911471319198609,
      "epoch": 2.1039385376192157,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004556192855465459,
      "loss": 5.3171,
      "mean_token_accuracy": 0.19750374257564546,
      "num_tokens": 44416678.0,
      "step": 23825
    },
    {
      "entropy": 5.739448165893554,
      "epoch": 2.104380077711056,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004556005049982449,
      "loss": 5.2176,
      "mean_token_accuracy": 0.21464794725179673,
      "num_tokens": 44425427.0,
      "step": 23830
    },
    {
      "entropy": 5.9034843921661375,
      "epoch": 2.1048216178028967,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045558172091208284,
      "loss": 5.2744,
      "mean_token_accuracy": 0.21139759868383406,
      "num_tokens": 44434267.0,
      "step": 23835
    },
    {
      "entropy": 5.930653715133667,
      "epoch": 2.1052631578947367,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045556293328842746,
      "loss": 5.2926,
      "mean_token_accuracy": 0.21125019043684007,
      "num_tokens": 44444037.0,
      "step": 23840
    },
    {
      "entropy": 5.889728116989136,
      "epoch": 2.1057046979865772,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004555441421276469,
      "loss": 5.2307,
      "mean_token_accuracy": 0.2160533383488655,
      "num_tokens": 44453707.0,
      "step": 23845
    },
    {
      "entropy": 5.87437105178833,
      "epoch": 2.1061462380784177,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045552534743010934,
      "loss": 5.2888,
      "mean_token_accuracy": 0.20803144723176956,
      "num_tokens": 44464357.0,
      "step": 23850
    },
    {
      "entropy": 5.936022567749023,
      "epoch": 2.1065877781702578,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045550654919618283,
      "loss": 5.4028,
      "mean_token_accuracy": 0.1988999828696251,
      "num_tokens": 44473803.0,
      "step": 23855
    },
    {
      "entropy": 5.94085168838501,
      "epoch": 2.1070293182620983,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004554877474262357,
      "loss": 5.3156,
      "mean_token_accuracy": 0.20876477062702178,
      "num_tokens": 44482830.0,
      "step": 23860
    },
    {
      "entropy": 5.863091659545899,
      "epoch": 2.1074708583539383,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004554689421206362,
      "loss": 5.2874,
      "mean_token_accuracy": 0.2077044203877449,
      "num_tokens": 44493045.0,
      "step": 23865
    },
    {
      "entropy": 5.886468124389649,
      "epoch": 2.107912398445779,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004554501332797529,
      "loss": 5.2631,
      "mean_token_accuracy": 0.21491543352603912,
      "num_tokens": 44503304.0,
      "step": 23870
    },
    {
      "entropy": 5.9362359046936035,
      "epoch": 2.1083539385376193,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045543132090395403,
      "loss": 5.3624,
      "mean_token_accuracy": 0.2076067328453064,
      "num_tokens": 44512337.0,
      "step": 23875
    },
    {
      "entropy": 5.8820459842681885,
      "epoch": 2.1087954786294594,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004554125049936082,
      "loss": 5.2974,
      "mean_token_accuracy": 0.2137009918689728,
      "num_tokens": 44521881.0,
      "step": 23880
    },
    {
      "entropy": 5.847499990463257,
      "epoch": 2.1092370187213,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000455393685549084,
      "loss": 5.3205,
      "mean_token_accuracy": 0.20926243662834168,
      "num_tokens": 44531351.0,
      "step": 23885
    },
    {
      "entropy": 5.9422526359558105,
      "epoch": 2.1096785588131404,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004553748625707501,
      "loss": 5.341,
      "mean_token_accuracy": 0.20805851221084595,
      "num_tokens": 44540592.0,
      "step": 23890
    },
    {
      "entropy": 5.85393443107605,
      "epoch": 2.1101200989049804,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045535603605897516,
      "loss": 5.253,
      "mean_token_accuracy": 0.21550757437944412,
      "num_tokens": 44550019.0,
      "step": 23895
    },
    {
      "entropy": 5.859655952453613,
      "epoch": 2.110561638996821,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004553372060141282,
      "loss": 5.2711,
      "mean_token_accuracy": 0.21352385729551315,
      "num_tokens": 44559423.0,
      "step": 23900
    },
    {
      "entropy": 5.809151220321655,
      "epoch": 2.1110031790886614,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004553183724365778,
      "loss": 5.2789,
      "mean_token_accuracy": 0.2176601141691208,
      "num_tokens": 44568792.0,
      "step": 23905
    },
    {
      "entropy": 5.8898368835449215,
      "epoch": 2.1114447191805015,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004552995353266931,
      "loss": 5.3005,
      "mean_token_accuracy": 0.2047635018825531,
      "num_tokens": 44578191.0,
      "step": 23910
    },
    {
      "entropy": 5.89700927734375,
      "epoch": 2.111886259272342,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004552806946848431,
      "loss": 5.2472,
      "mean_token_accuracy": 0.21541159152984618,
      "num_tokens": 44587402.0,
      "step": 23915
    },
    {
      "entropy": 5.892294073104859,
      "epoch": 2.1123277993641825,
      "grad_norm": 1.125,
      "learning_rate": 0.00045526185051139676,
      "loss": 5.4001,
      "mean_token_accuracy": 0.1966974601149559,
      "num_tokens": 44596241.0,
      "step": 23920
    },
    {
      "entropy": 5.860474920272827,
      "epoch": 2.1127693394560225,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045524300280672326,
      "loss": 5.1832,
      "mean_token_accuracy": 0.21655723303556443,
      "num_tokens": 44604796.0,
      "step": 23925
    },
    {
      "entropy": 5.904183626174927,
      "epoch": 2.113210879547863,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045522415157119185,
      "loss": 5.3388,
      "mean_token_accuracy": 0.20518684834241868,
      "num_tokens": 44615208.0,
      "step": 23930
    },
    {
      "entropy": 5.888106870651245,
      "epoch": 2.113652419639703,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004552052968051719,
      "loss": 5.3297,
      "mean_token_accuracy": 0.2135383367538452,
      "num_tokens": 44623937.0,
      "step": 23935
    },
    {
      "entropy": 5.874234342575074,
      "epoch": 2.1140939597315436,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004551864385090326,
      "loss": 5.2459,
      "mean_token_accuracy": 0.21329738050699235,
      "num_tokens": 44634116.0,
      "step": 23940
    },
    {
      "entropy": 5.931789445877075,
      "epoch": 2.114535499823384,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004551675766831434,
      "loss": 5.3283,
      "mean_token_accuracy": 0.20871264338493348,
      "num_tokens": 44643436.0,
      "step": 23945
    },
    {
      "entropy": 5.93265700340271,
      "epoch": 2.114977039915224,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045514871132787394,
      "loss": 5.3589,
      "mean_token_accuracy": 0.20905428230762482,
      "num_tokens": 44654077.0,
      "step": 23950
    },
    {
      "entropy": 5.929782104492188,
      "epoch": 2.1154185800070646,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045512984244359364,
      "loss": 5.3321,
      "mean_token_accuracy": 0.20821423083543777,
      "num_tokens": 44663785.0,
      "step": 23955
    },
    {
      "entropy": 5.7986588954925535,
      "epoch": 2.115860120098905,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045511097003067216,
      "loss": 5.1766,
      "mean_token_accuracy": 0.2192887246608734,
      "num_tokens": 44672750.0,
      "step": 23960
    },
    {
      "entropy": 5.83315076828003,
      "epoch": 2.116301660190745,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045509209408947914,
      "loss": 5.2325,
      "mean_token_accuracy": 0.22007079273462296,
      "num_tokens": 44681814.0,
      "step": 23965
    },
    {
      "entropy": 5.837835454940796,
      "epoch": 2.1167432002825857,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004550732146203845,
      "loss": 5.2804,
      "mean_token_accuracy": 0.21167240887880326,
      "num_tokens": 44690926.0,
      "step": 23970
    },
    {
      "entropy": 5.964428472518921,
      "epoch": 2.117184740374426,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045505433162375806,
      "loss": 5.3964,
      "mean_token_accuracy": 0.2062529668211937,
      "num_tokens": 44700625.0,
      "step": 23975
    },
    {
      "entropy": 5.9191807270050045,
      "epoch": 2.117626280466266,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004550354450999696,
      "loss": 5.3378,
      "mean_token_accuracy": 0.21039126217365264,
      "num_tokens": 44710703.0,
      "step": 23980
    },
    {
      "entropy": 5.89006028175354,
      "epoch": 2.1180678205581067,
      "grad_norm": 1.125,
      "learning_rate": 0.0004550165550493891,
      "loss": 5.2458,
      "mean_token_accuracy": 0.21144918948411942,
      "num_tokens": 44719412.0,
      "step": 23985
    },
    {
      "entropy": 5.969325304031372,
      "epoch": 2.118509360649947,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045499766147238675,
      "loss": 5.3802,
      "mean_token_accuracy": 0.2084307700395584,
      "num_tokens": 44729161.0,
      "step": 23990
    },
    {
      "entropy": 6.003976058959961,
      "epoch": 2.1189509007417873,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045497876436933257,
      "loss": 5.3664,
      "mean_token_accuracy": 0.2052161544561386,
      "num_tokens": 44738985.0,
      "step": 23995
    },
    {
      "entropy": 5.8369025707244875,
      "epoch": 2.1193924408336278,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045495986374059674,
      "loss": 5.2828,
      "mean_token_accuracy": 0.21516386717557906,
      "num_tokens": 44748070.0,
      "step": 24000
    },
    {
      "epoch": 2.1193924408336278,
      "eval_entropy": 5.6000492876275345,
      "eval_loss": 5.37293815612793,
      "eval_mean_token_accuracy": 0.21494417624761133,
      "eval_num_tokens": 44748070.0,
      "eval_runtime": 20.1103,
      "eval_samples_per_second": 1756.066,
      "eval_steps_per_second": 219.539,
      "step": 24000
    },
    {
      "entropy": 5.840093088150025,
      "epoch": 2.119833980925468,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004549409595865496,
      "loss": 5.245,
      "mean_token_accuracy": 0.21080335974693298,
      "num_tokens": 44756607.0,
      "step": 24005
    },
    {
      "entropy": 5.908659410476685,
      "epoch": 2.1202755210173083,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045492205190756133,
      "loss": 5.3401,
      "mean_token_accuracy": 0.21164225786924362,
      "num_tokens": 44765435.0,
      "step": 24010
    },
    {
      "entropy": 5.837518882751465,
      "epoch": 2.120717061109149,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004549031407040024,
      "loss": 5.2953,
      "mean_token_accuracy": 0.2108205586671829,
      "num_tokens": 44775589.0,
      "step": 24015
    },
    {
      "entropy": 5.82951250076294,
      "epoch": 2.121158601200989,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004548842259762432,
      "loss": 5.3095,
      "mean_token_accuracy": 0.20778565406799315,
      "num_tokens": 44785692.0,
      "step": 24020
    },
    {
      "entropy": 5.839303016662598,
      "epoch": 2.1216001412928294,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004548653077246544,
      "loss": 5.2111,
      "mean_token_accuracy": 0.2144887626171112,
      "num_tokens": 44794074.0,
      "step": 24025
    },
    {
      "entropy": 5.919186305999756,
      "epoch": 2.12204168138467,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004548463859496065,
      "loss": 5.3064,
      "mean_token_accuracy": 0.2060460314154625,
      "num_tokens": 44803959.0,
      "step": 24030
    },
    {
      "entropy": 5.827262878417969,
      "epoch": 2.12248322147651,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045482746065147014,
      "loss": 5.2547,
      "mean_token_accuracy": 0.21491232961416246,
      "num_tokens": 44813294.0,
      "step": 24035
    },
    {
      "entropy": 5.909788274765015,
      "epoch": 2.1229247615683504,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004548085318306162,
      "loss": 5.2704,
      "mean_token_accuracy": 0.2111326724290848,
      "num_tokens": 44822788.0,
      "step": 24040
    },
    {
      "entropy": 5.879832172393799,
      "epoch": 2.123366301660191,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045478959948741526,
      "loss": 5.2203,
      "mean_token_accuracy": 0.21731253564357758,
      "num_tokens": 44832236.0,
      "step": 24045
    },
    {
      "entropy": 5.890810585021972,
      "epoch": 2.123807841752031,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004547706636222384,
      "loss": 5.3052,
      "mean_token_accuracy": 0.21453931629657746,
      "num_tokens": 44841258.0,
      "step": 24050
    },
    {
      "entropy": 5.879990100860596,
      "epoch": 2.1242493818438715,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004547517242354565,
      "loss": 5.2852,
      "mean_token_accuracy": 0.21117878407239915,
      "num_tokens": 44850365.0,
      "step": 24055
    },
    {
      "entropy": 5.933914661407471,
      "epoch": 2.124690921935712,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004547327813274405,
      "loss": 5.3128,
      "mean_token_accuracy": 0.20933755487203598,
      "num_tokens": 44860451.0,
      "step": 24060
    },
    {
      "entropy": 5.914949178695679,
      "epoch": 2.125132462027552,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004547138348985615,
      "loss": 5.229,
      "mean_token_accuracy": 0.21384336054325104,
      "num_tokens": 44868815.0,
      "step": 24065
    },
    {
      "entropy": 5.891973161697388,
      "epoch": 2.1255740021193925,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045469488494919076,
      "loss": 5.3474,
      "mean_token_accuracy": 0.20765818804502487,
      "num_tokens": 44878046.0,
      "step": 24070
    },
    {
      "entropy": 5.852796030044556,
      "epoch": 2.126015542211233,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004546759314796994,
      "loss": 5.2491,
      "mean_token_accuracy": 0.21183056086301805,
      "num_tokens": 44886915.0,
      "step": 24075
    },
    {
      "entropy": 5.947250699996948,
      "epoch": 2.126457082303073,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004546569744904586,
      "loss": 5.2978,
      "mean_token_accuracy": 0.21096168011426925,
      "num_tokens": 44895629.0,
      "step": 24080
    },
    {
      "entropy": 5.915944147109985,
      "epoch": 2.1268986223949136,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045463801398184004,
      "loss": 5.4305,
      "mean_token_accuracy": 0.19379859417676926,
      "num_tokens": 44905913.0,
      "step": 24085
    },
    {
      "entropy": 5.990835237503052,
      "epoch": 2.1273401624867536,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004546190499542149,
      "loss": 5.3184,
      "mean_token_accuracy": 0.20539964288473128,
      "num_tokens": 44915329.0,
      "step": 24090
    },
    {
      "entropy": 5.945349216461182,
      "epoch": 2.127781702578594,
      "grad_norm": 1.125,
      "learning_rate": 0.0004546000824079547,
      "loss": 5.2814,
      "mean_token_accuracy": 0.21771723330020903,
      "num_tokens": 44923810.0,
      "step": 24095
    },
    {
      "entropy": 5.89575457572937,
      "epoch": 2.1282232426704346,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000454581111343431,
      "loss": 5.3159,
      "mean_token_accuracy": 0.20633020401000976,
      "num_tokens": 44932375.0,
      "step": 24100
    },
    {
      "entropy": 5.936436319351197,
      "epoch": 2.1286647827622747,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004545621367610155,
      "loss": 5.2982,
      "mean_token_accuracy": 0.20895180851221085,
      "num_tokens": 44941341.0,
      "step": 24105
    },
    {
      "entropy": 5.955826473236084,
      "epoch": 2.129106322854115,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004545431586610799,
      "loss": 5.3346,
      "mean_token_accuracy": 0.21046103835105895,
      "num_tokens": 44951221.0,
      "step": 24110
    },
    {
      "entropy": 5.829993438720703,
      "epoch": 2.1295478629459557,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004545241770439958,
      "loss": 5.2231,
      "mean_token_accuracy": 0.22138845771551133,
      "num_tokens": 44960254.0,
      "step": 24115
    },
    {
      "entropy": 5.931124496459961,
      "epoch": 2.1299894030377957,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004545051919101353,
      "loss": 5.4155,
      "mean_token_accuracy": 0.20018528252840043,
      "num_tokens": 44969023.0,
      "step": 24120
    },
    {
      "entropy": 5.86586709022522,
      "epoch": 2.130430943129636,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045448620325987013,
      "loss": 5.2894,
      "mean_token_accuracy": 0.20908062011003495,
      "num_tokens": 44978942.0,
      "step": 24125
    },
    {
      "entropy": 5.921107578277588,
      "epoch": 2.1308724832214767,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00045446721109357243,
      "loss": 5.2214,
      "mean_token_accuracy": 0.2165845513343811,
      "num_tokens": 44987794.0,
      "step": 24130
    },
    {
      "entropy": 5.948254728317261,
      "epoch": 2.1313140233133168,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045444821541161404,
      "loss": 5.3669,
      "mean_token_accuracy": 0.20377086102962494,
      "num_tokens": 44995781.0,
      "step": 24135
    },
    {
      "entropy": 5.889995145797729,
      "epoch": 2.1317555634051573,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004544292162143671,
      "loss": 5.3078,
      "mean_token_accuracy": 0.20938256084918977,
      "num_tokens": 45004731.0,
      "step": 24140
    },
    {
      "entropy": 5.9272013187408445,
      "epoch": 2.1321971034969973,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004544102135022039,
      "loss": 5.3361,
      "mean_token_accuracy": 0.20272944867610931,
      "num_tokens": 45014312.0,
      "step": 24145
    },
    {
      "entropy": 5.949671792984009,
      "epoch": 2.132638643588838,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004543912072754967,
      "loss": 5.3983,
      "mean_token_accuracy": 0.19977204501628876,
      "num_tokens": 45023903.0,
      "step": 24150
    },
    {
      "entropy": 5.915448331832886,
      "epoch": 2.1330801836806783,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004543721975346177,
      "loss": 5.3078,
      "mean_token_accuracy": 0.20177028626203536,
      "num_tokens": 45033330.0,
      "step": 24155
    },
    {
      "entropy": 5.796429586410523,
      "epoch": 2.1335217237725184,
      "grad_norm": 1.0,
      "learning_rate": 0.00045435318427993946,
      "loss": 5.1934,
      "mean_token_accuracy": 0.21853149682283401,
      "num_tokens": 45042862.0,
      "step": 24160
    },
    {
      "entropy": 5.894321537017822,
      "epoch": 2.133963263864359,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004543341675118343,
      "loss": 5.2621,
      "mean_token_accuracy": 0.2157619684934616,
      "num_tokens": 45052198.0,
      "step": 24165
    },
    {
      "entropy": 5.954981088638306,
      "epoch": 2.1344048039561994,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045431514723067487,
      "loss": 5.3183,
      "mean_token_accuracy": 0.21114950031042098,
      "num_tokens": 45060625.0,
      "step": 24170
    },
    {
      "entropy": 5.871349906921386,
      "epoch": 2.1348463440480394,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045429612343683354,
      "loss": 5.2435,
      "mean_token_accuracy": 0.2122351974248886,
      "num_tokens": 45069557.0,
      "step": 24175
    },
    {
      "entropy": 5.893993139266968,
      "epoch": 2.13528788413988,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045427709613068317,
      "loss": 5.3464,
      "mean_token_accuracy": 0.20821600705385207,
      "num_tokens": 45079281.0,
      "step": 24180
    },
    {
      "entropy": 5.951321220397949,
      "epoch": 2.1357294242317204,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004542580653125965,
      "loss": 5.3411,
      "mean_token_accuracy": 0.20824411809444426,
      "num_tokens": 45089363.0,
      "step": 24185
    },
    {
      "entropy": 5.839923429489136,
      "epoch": 2.1361709643235605,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045423903098294624,
      "loss": 5.2353,
      "mean_token_accuracy": 0.2098870813846588,
      "num_tokens": 45098735.0,
      "step": 24190
    },
    {
      "entropy": 5.921649742126465,
      "epoch": 2.136612504415401,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045421999314210535,
      "loss": 5.4091,
      "mean_token_accuracy": 0.20201925039291382,
      "num_tokens": 45109173.0,
      "step": 24195
    },
    {
      "entropy": 5.909635829925537,
      "epoch": 2.1370540445072415,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004542009517904466,
      "loss": 5.4144,
      "mean_token_accuracy": 0.1985637828707695,
      "num_tokens": 45119153.0,
      "step": 24200
    },
    {
      "entropy": 5.8969038963317875,
      "epoch": 2.1374955845990815,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004541819069283432,
      "loss": 5.2721,
      "mean_token_accuracy": 0.21434079706668854,
      "num_tokens": 45128219.0,
      "step": 24205
    },
    {
      "entropy": 5.911312007904053,
      "epoch": 2.137937124690922,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004541628585561682,
      "loss": 5.3395,
      "mean_token_accuracy": 0.206658037006855,
      "num_tokens": 45137777.0,
      "step": 24210
    },
    {
      "entropy": 5.834524297714234,
      "epoch": 2.1383786647827625,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004541438066742946,
      "loss": 5.2711,
      "mean_token_accuracy": 0.214461150765419,
      "num_tokens": 45147553.0,
      "step": 24215
    },
    {
      "entropy": 5.907449769973755,
      "epoch": 2.1388202048746026,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004541247512830958,
      "loss": 5.3231,
      "mean_token_accuracy": 0.20177627354860306,
      "num_tokens": 45157443.0,
      "step": 24220
    },
    {
      "entropy": 5.983698511123658,
      "epoch": 2.139261744966443,
      "grad_norm": 1.0,
      "learning_rate": 0.000454105692382945,
      "loss": 5.3731,
      "mean_token_accuracy": 0.2040179491043091,
      "num_tokens": 45167279.0,
      "step": 24225
    },
    {
      "entropy": 5.987958002090454,
      "epoch": 2.139703285058283,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045408662997421546,
      "loss": 5.3972,
      "mean_token_accuracy": 0.20381687134504317,
      "num_tokens": 45176515.0,
      "step": 24230
    },
    {
      "entropy": 5.981200933456421,
      "epoch": 2.1401448251501236,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004540675640572807,
      "loss": 5.3407,
      "mean_token_accuracy": 0.20446030497550965,
      "num_tokens": 45185501.0,
      "step": 24235
    },
    {
      "entropy": 5.940980339050293,
      "epoch": 2.140586365241964,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004540484946325143,
      "loss": 5.3981,
      "mean_token_accuracy": 0.2035575717687607,
      "num_tokens": 45194881.0,
      "step": 24240
    },
    {
      "entropy": 5.922618722915649,
      "epoch": 2.141027905333804,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045402942170028965,
      "loss": 5.3016,
      "mean_token_accuracy": 0.21147134751081467,
      "num_tokens": 45204463.0,
      "step": 24245
    },
    {
      "entropy": 5.915569210052491,
      "epoch": 2.1414694454256447,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045401034526098053,
      "loss": 5.3208,
      "mean_token_accuracy": 0.20379938781261445,
      "num_tokens": 45214302.0,
      "step": 24250
    },
    {
      "entropy": 5.920975065231323,
      "epoch": 2.141910985517485,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004539912653149604,
      "loss": 5.3472,
      "mean_token_accuracy": 0.20329059958457946,
      "num_tokens": 45224565.0,
      "step": 24255
    },
    {
      "entropy": 5.831735324859619,
      "epoch": 2.142352525609325,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004539721818626033,
      "loss": 5.2301,
      "mean_token_accuracy": 0.22341244518756867,
      "num_tokens": 45234574.0,
      "step": 24260
    },
    {
      "entropy": 5.797258806228638,
      "epoch": 2.1427940657011657,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004539530949042829,
      "loss": 5.213,
      "mean_token_accuracy": 0.21867974549531938,
      "num_tokens": 45242707.0,
      "step": 24265
    },
    {
      "entropy": 5.9111264705657955,
      "epoch": 2.143235605793006,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004539340044403731,
      "loss": 5.4577,
      "mean_token_accuracy": 0.19683025628328324,
      "num_tokens": 45252873.0,
      "step": 24270
    },
    {
      "entropy": 5.904419612884522,
      "epoch": 2.1436771458848463,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000453914910471248,
      "loss": 5.3181,
      "mean_token_accuracy": 0.20716434717178345,
      "num_tokens": 45262833.0,
      "step": 24275
    },
    {
      "entropy": 5.88218412399292,
      "epoch": 2.1441186859766868,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004538958129972815,
      "loss": 5.2941,
      "mean_token_accuracy": 0.21218712031841278,
      "num_tokens": 45271798.0,
      "step": 24280
    },
    {
      "entropy": 5.882066345214843,
      "epoch": 2.144560226068527,
      "grad_norm": 1.0,
      "learning_rate": 0.0004538767120188479,
      "loss": 5.2506,
      "mean_token_accuracy": 0.21435187011957169,
      "num_tokens": 45281470.0,
      "step": 24285
    },
    {
      "entropy": 5.934221267700195,
      "epoch": 2.1450017661603673,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004538576075363211,
      "loss": 5.2922,
      "mean_token_accuracy": 0.2095536097884178,
      "num_tokens": 45290685.0,
      "step": 24290
    },
    {
      "entropy": 5.903287601470947,
      "epoch": 2.145443306252208,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045383849955007554,
      "loss": 5.2919,
      "mean_token_accuracy": 0.20858779102563857,
      "num_tokens": 45299631.0,
      "step": 24295
    },
    {
      "entropy": 5.934343481063843,
      "epoch": 2.145884846344048,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045381938806048544,
      "loss": 5.3976,
      "mean_token_accuracy": 0.20383385866880416,
      "num_tokens": 45309871.0,
      "step": 24300
    },
    {
      "entropy": 5.84241509437561,
      "epoch": 2.1463263864358884,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004538002730679253,
      "loss": 5.1652,
      "mean_token_accuracy": 0.22025881409645082,
      "num_tokens": 45319233.0,
      "step": 24305
    },
    {
      "entropy": 5.813080883026123,
      "epoch": 2.146767926527729,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00045378115457276946,
      "loss": 5.2671,
      "mean_token_accuracy": 0.21412792056798935,
      "num_tokens": 45329467.0,
      "step": 24310
    },
    {
      "entropy": 5.916565704345703,
      "epoch": 2.147209466619569,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045376203257539244,
      "loss": 5.2989,
      "mean_token_accuracy": 0.2064487800002098,
      "num_tokens": 45339124.0,
      "step": 24315
    },
    {
      "entropy": 5.94030179977417,
      "epoch": 2.1476510067114094,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045374290707616893,
      "loss": 5.4022,
      "mean_token_accuracy": 0.19933412969112396,
      "num_tokens": 45348973.0,
      "step": 24320
    },
    {
      "entropy": 5.962058353424072,
      "epoch": 2.14809254680325,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004537237780754735,
      "loss": 5.3328,
      "mean_token_accuracy": 0.20377595573663712,
      "num_tokens": 45359161.0,
      "step": 24325
    },
    {
      "entropy": 5.935665798187256,
      "epoch": 2.14853408689509,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004537046455736809,
      "loss": 5.329,
      "mean_token_accuracy": 0.21389723867177962,
      "num_tokens": 45369019.0,
      "step": 24330
    },
    {
      "entropy": 5.875147485733033,
      "epoch": 2.1489756269869305,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004536855095711659,
      "loss": 5.2599,
      "mean_token_accuracy": 0.21821779012680054,
      "num_tokens": 45378491.0,
      "step": 24335
    },
    {
      "entropy": 5.8914107322692875,
      "epoch": 2.149417167078771,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004536663700683035,
      "loss": 5.3364,
      "mean_token_accuracy": 0.20960796773433685,
      "num_tokens": 45387651.0,
      "step": 24340
    },
    {
      "entropy": 5.941328382492065,
      "epoch": 2.149858707170611,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004536472270654685,
      "loss": 5.2897,
      "mean_token_accuracy": 0.21224784553050996,
      "num_tokens": 45396524.0,
      "step": 24345
    },
    {
      "entropy": 5.9586982250213625,
      "epoch": 2.1503002472624515,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004536280805630358,
      "loss": 5.3598,
      "mean_token_accuracy": 0.209202179312706,
      "num_tokens": 45404849.0,
      "step": 24350
    },
    {
      "entropy": 5.910099649429322,
      "epoch": 2.150741787354292,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045360893056138076,
      "loss": 5.2743,
      "mean_token_accuracy": 0.21451979726552964,
      "num_tokens": 45413628.0,
      "step": 24355
    },
    {
      "entropy": 5.88285117149353,
      "epoch": 2.151183327446132,
      "grad_norm": 1.125,
      "learning_rate": 0.00045358977706087827,
      "loss": 5.2805,
      "mean_token_accuracy": 0.2138541266322136,
      "num_tokens": 45423190.0,
      "step": 24360
    },
    {
      "entropy": 5.816536855697632,
      "epoch": 2.1516248675379726,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004535706200619036,
      "loss": 5.2784,
      "mean_token_accuracy": 0.213318532705307,
      "num_tokens": 45432075.0,
      "step": 24365
    },
    {
      "entropy": 5.905173301696777,
      "epoch": 2.1520664076298126,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045355145956483205,
      "loss": 5.3989,
      "mean_token_accuracy": 0.2014447495341301,
      "num_tokens": 45442627.0,
      "step": 24370
    },
    {
      "entropy": 5.8832025051116945,
      "epoch": 2.152507947721653,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045353229557003896,
      "loss": 5.2673,
      "mean_token_accuracy": 0.21728233247995377,
      "num_tokens": 45452818.0,
      "step": 24375
    },
    {
      "entropy": 5.8580810546875,
      "epoch": 2.1529494878134936,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004535131280778997,
      "loss": 5.2678,
      "mean_token_accuracy": 0.21723175048828125,
      "num_tokens": 45461964.0,
      "step": 24380
    },
    {
      "entropy": 5.853044319152832,
      "epoch": 2.1533910279053337,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004534939570887898,
      "loss": 5.2763,
      "mean_token_accuracy": 0.21086964011192322,
      "num_tokens": 45471700.0,
      "step": 24385
    },
    {
      "entropy": 5.911209630966186,
      "epoch": 2.153832567997174,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004534747826030848,
      "loss": 5.2594,
      "mean_token_accuracy": 0.20904617607593537,
      "num_tokens": 45481291.0,
      "step": 24390
    },
    {
      "entropy": 5.8785828113555905,
      "epoch": 2.1542741080890146,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004534556046211604,
      "loss": 5.3133,
      "mean_token_accuracy": 0.21149694919586182,
      "num_tokens": 45490317.0,
      "step": 24395
    },
    {
      "entropy": 5.9012034893035885,
      "epoch": 2.1547156481808547,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045343642314339206,
      "loss": 5.3569,
      "mean_token_accuracy": 0.20782990753650665,
      "num_tokens": 45499688.0,
      "step": 24400
    },
    {
      "entropy": 5.9482227802276615,
      "epoch": 2.155157188272695,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004534172381701558,
      "loss": 5.3079,
      "mean_token_accuracy": 0.2127251073718071,
      "num_tokens": 45508951.0,
      "step": 24405
    },
    {
      "entropy": 5.969006061553955,
      "epoch": 2.1555987283645357,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004533980497018272,
      "loss": 5.3491,
      "mean_token_accuracy": 0.21004294008016586,
      "num_tokens": 45518955.0,
      "step": 24410
    },
    {
      "entropy": 5.871236658096313,
      "epoch": 2.1560402684563758,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004533788577387824,
      "loss": 5.3287,
      "mean_token_accuracy": 0.2022874802350998,
      "num_tokens": 45527869.0,
      "step": 24415
    },
    {
      "entropy": 5.859797620773316,
      "epoch": 2.1564818085482163,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004533596622813971,
      "loss": 5.2316,
      "mean_token_accuracy": 0.22090501338243484,
      "num_tokens": 45537250.0,
      "step": 24420
    },
    {
      "entropy": 5.885863399505615,
      "epoch": 2.1569233486400563,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004533404633300475,
      "loss": 5.2608,
      "mean_token_accuracy": 0.2091406211256981,
      "num_tokens": 45547204.0,
      "step": 24425
    },
    {
      "entropy": 5.851839160919189,
      "epoch": 2.157364888731897,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045332126088510965,
      "loss": 5.27,
      "mean_token_accuracy": 0.20369155555963517,
      "num_tokens": 45555464.0,
      "step": 24430
    },
    {
      "entropy": 5.9216224193573,
      "epoch": 2.1578064288237373,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004533020549469597,
      "loss": 5.2419,
      "mean_token_accuracy": 0.21041091084480285,
      "num_tokens": 45564797.0,
      "step": 24435
    },
    {
      "entropy": 5.9219653606414795,
      "epoch": 2.1582479689155774,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004532828455159739,
      "loss": 5.3354,
      "mean_token_accuracy": 0.20181020647287368,
      "num_tokens": 45573317.0,
      "step": 24440
    },
    {
      "entropy": 5.792671012878418,
      "epoch": 2.158689509007418,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004532636325925286,
      "loss": 5.2378,
      "mean_token_accuracy": 0.21440215408802032,
      "num_tokens": 45582838.0,
      "step": 24445
    },
    {
      "entropy": 5.901033687591553,
      "epoch": 2.1591310490992583,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045324441617700014,
      "loss": 5.1783,
      "mean_token_accuracy": 0.2169872060418129,
      "num_tokens": 45592031.0,
      "step": 24450
    },
    {
      "entropy": 5.918382692337036,
      "epoch": 2.1595725891910984,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004532251962697649,
      "loss": 5.3899,
      "mean_token_accuracy": 0.20761216729879378,
      "num_tokens": 45601353.0,
      "step": 24455
    },
    {
      "entropy": 5.858606624603271,
      "epoch": 2.160014129282939,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045320597287119943,
      "loss": 5.2656,
      "mean_token_accuracy": 0.20914039611816407,
      "num_tokens": 45611064.0,
      "step": 24460
    },
    {
      "entropy": 5.893822145462036,
      "epoch": 2.1604556693747794,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004531867459816803,
      "loss": 5.2934,
      "mean_token_accuracy": 0.20908606946468353,
      "num_tokens": 45620704.0,
      "step": 24465
    },
    {
      "entropy": 5.9575080394744875,
      "epoch": 2.1608972094666195,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045316751560158414,
      "loss": 5.3544,
      "mean_token_accuracy": 0.20919648110866546,
      "num_tokens": 45629967.0,
      "step": 24470
    },
    {
      "entropy": 5.922767400741577,
      "epoch": 2.16133874955846,
      "grad_norm": 1.25,
      "learning_rate": 0.0004531482817312877,
      "loss": 5.3619,
      "mean_token_accuracy": 0.19891661554574966,
      "num_tokens": 45639104.0,
      "step": 24475
    },
    {
      "entropy": 5.93939094543457,
      "epoch": 2.1617802896503004,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045312904437116774,
      "loss": 5.4344,
      "mean_token_accuracy": 0.19522229135036467,
      "num_tokens": 45649217.0,
      "step": 24480
    },
    {
      "entropy": 5.978772830963135,
      "epoch": 2.1622218297421405,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004531098035216011,
      "loss": 5.3878,
      "mean_token_accuracy": 0.20174179822206498,
      "num_tokens": 45658969.0,
      "step": 24485
    },
    {
      "entropy": 5.801855945587159,
      "epoch": 2.162663369833981,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045309055918296465,
      "loss": 5.2146,
      "mean_token_accuracy": 0.21612147092819214,
      "num_tokens": 45668427.0,
      "step": 24490
    },
    {
      "entropy": 5.809905576705932,
      "epoch": 2.1631049099258215,
      "grad_norm": 1.0,
      "learning_rate": 0.0004530713113556355,
      "loss": 5.1604,
      "mean_token_accuracy": 0.21866879016160964,
      "num_tokens": 45676675.0,
      "step": 24495
    },
    {
      "entropy": 5.827397584915161,
      "epoch": 2.1635464500176615,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045305206003999057,
      "loss": 5.1987,
      "mean_token_accuracy": 0.21538830250501634,
      "num_tokens": 45685756.0,
      "step": 24500
    },
    {
      "entropy": 5.885875606536866,
      "epoch": 2.163987990109502,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004530328052364071,
      "loss": 5.3073,
      "mean_token_accuracy": 0.21500690877437592,
      "num_tokens": 45694770.0,
      "step": 24505
    },
    {
      "entropy": 5.865076971054077,
      "epoch": 2.164429530201342,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004530135469452622,
      "loss": 5.2033,
      "mean_token_accuracy": 0.21871818006038665,
      "num_tokens": 45703940.0,
      "step": 24510
    },
    {
      "entropy": 5.887880182266235,
      "epoch": 2.1648710702931826,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004529942851669333,
      "loss": 5.3443,
      "mean_token_accuracy": 0.20593252927064895,
      "num_tokens": 45713875.0,
      "step": 24515
    },
    {
      "entropy": 5.930170011520386,
      "epoch": 2.165312610385023,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004529750199017974,
      "loss": 5.3522,
      "mean_token_accuracy": 0.1997353181242943,
      "num_tokens": 45723870.0,
      "step": 24520
    },
    {
      "entropy": 5.891964721679687,
      "epoch": 2.165754150476863,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004529557511502322,
      "loss": 5.2898,
      "mean_token_accuracy": 0.2156054347753525,
      "num_tokens": 45733269.0,
      "step": 24525
    },
    {
      "entropy": 5.912575960159302,
      "epoch": 2.1661956905687036,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045293647891261494,
      "loss": 5.3112,
      "mean_token_accuracy": 0.20392105281352996,
      "num_tokens": 45743208.0,
      "step": 24530
    },
    {
      "entropy": 5.927594232559204,
      "epoch": 2.166637230660544,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004529172031893233,
      "loss": 5.2794,
      "mean_token_accuracy": 0.21576096266508102,
      "num_tokens": 45752411.0,
      "step": 24535
    },
    {
      "entropy": 5.822519731521607,
      "epoch": 2.167078770752384,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045289792398073487,
      "loss": 5.2951,
      "mean_token_accuracy": 0.21168990880250932,
      "num_tokens": 45761770.0,
      "step": 24540
    },
    {
      "entropy": 5.892732286453247,
      "epoch": 2.1675203108442247,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045287864128722724,
      "loss": 5.266,
      "mean_token_accuracy": 0.2090234711766243,
      "num_tokens": 45771227.0,
      "step": 24545
    },
    {
      "entropy": 5.861612939834595,
      "epoch": 2.167961850936065,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004528593551091782,
      "loss": 5.2215,
      "mean_token_accuracy": 0.21584538072347642,
      "num_tokens": 45780558.0,
      "step": 24550
    },
    {
      "entropy": 5.8548243045806885,
      "epoch": 2.1684033910279052,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045284006544696554,
      "loss": 5.3329,
      "mean_token_accuracy": 0.20803416520357132,
      "num_tokens": 45790078.0,
      "step": 24555
    },
    {
      "entropy": 5.882325220108032,
      "epoch": 2.1688449311197457,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045282077230096715,
      "loss": 5.2546,
      "mean_token_accuracy": 0.21524593383073806,
      "num_tokens": 45799962.0,
      "step": 24560
    },
    {
      "entropy": 5.817895078659058,
      "epoch": 2.169286471211586,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045280147567156097,
      "loss": 5.1613,
      "mean_token_accuracy": 0.2302367866039276,
      "num_tokens": 45809599.0,
      "step": 24565
    },
    {
      "entropy": 5.8085283756256105,
      "epoch": 2.1697280113034263,
      "grad_norm": 1.109375,
      "learning_rate": 0.000452782175559125,
      "loss": 5.2447,
      "mean_token_accuracy": 0.22017981112003326,
      "num_tokens": 45818318.0,
      "step": 24570
    },
    {
      "entropy": 5.892239046096802,
      "epoch": 2.170169551395267,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004527628719640374,
      "loss": 5.3492,
      "mean_token_accuracy": 0.20504026412963866,
      "num_tokens": 45828083.0,
      "step": 24575
    },
    {
      "entropy": 5.935414838790893,
      "epoch": 2.170611091487107,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045274356488667614,
      "loss": 5.3017,
      "mean_token_accuracy": 0.21164270639419555,
      "num_tokens": 45838441.0,
      "step": 24580
    },
    {
      "entropy": 5.934444761276245,
      "epoch": 2.1710526315789473,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045272425432741953,
      "loss": 5.3014,
      "mean_token_accuracy": 0.20749692767858505,
      "num_tokens": 45848047.0,
      "step": 24585
    },
    {
      "entropy": 5.8710541248321535,
      "epoch": 2.171494171670788,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045270494028664586,
      "loss": 5.3091,
      "mean_token_accuracy": 0.20810611099004744,
      "num_tokens": 45857675.0,
      "step": 24590
    },
    {
      "entropy": 5.8301526546478275,
      "epoch": 2.171935711762628,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004526856227647335,
      "loss": 5.2426,
      "mean_token_accuracy": 0.21092915683984756,
      "num_tokens": 45866506.0,
      "step": 24595
    },
    {
      "entropy": 5.9587842464447025,
      "epoch": 2.1723772518544684,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004526663017620609,
      "loss": 5.3155,
      "mean_token_accuracy": 0.2150665834546089,
      "num_tokens": 45875955.0,
      "step": 24600
    },
    {
      "entropy": 5.936282682418823,
      "epoch": 2.172818791946309,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004526469772790063,
      "loss": 5.2992,
      "mean_token_accuracy": 0.2054099589586258,
      "num_tokens": 45884518.0,
      "step": 24605
    },
    {
      "entropy": 5.855802059173584,
      "epoch": 2.173260332038149,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004526276493159487,
      "loss": 5.2583,
      "mean_token_accuracy": 0.2131984516978264,
      "num_tokens": 45894006.0,
      "step": 24610
    },
    {
      "entropy": 5.805609273910522,
      "epoch": 2.1737018721299894,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004526083178732663,
      "loss": 5.2151,
      "mean_token_accuracy": 0.21467824131250382,
      "num_tokens": 45902739.0,
      "step": 24615
    },
    {
      "entropy": 5.7868249893188475,
      "epoch": 2.17414341222183,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045258898295133797,
      "loss": 5.1477,
      "mean_token_accuracy": 0.22033517062664032,
      "num_tokens": 45911418.0,
      "step": 24620
    },
    {
      "entropy": 5.86443076133728,
      "epoch": 2.17458495231367,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045256964455054255,
      "loss": 5.3076,
      "mean_token_accuracy": 0.20846038460731506,
      "num_tokens": 45921135.0,
      "step": 24625
    },
    {
      "entropy": 5.882087326049804,
      "epoch": 2.1750264924055105,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045255030267125874,
      "loss": 5.2392,
      "mean_token_accuracy": 0.21579915434122085,
      "num_tokens": 45930276.0,
      "step": 24630
    },
    {
      "entropy": 5.908164930343628,
      "epoch": 2.175468032497351,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045253095731386553,
      "loss": 5.2635,
      "mean_token_accuracy": 0.20852722823619843,
      "num_tokens": 45939466.0,
      "step": 24635
    },
    {
      "entropy": 5.874197101593017,
      "epoch": 2.175909572589191,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045251160847874174,
      "loss": 5.2183,
      "mean_token_accuracy": 0.21040224730968476,
      "num_tokens": 45948158.0,
      "step": 24640
    },
    {
      "entropy": 5.911040735244751,
      "epoch": 2.1763511126810315,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045249225616626656,
      "loss": 5.4035,
      "mean_token_accuracy": 0.19876639544963837,
      "num_tokens": 45957759.0,
      "step": 24645
    },
    {
      "entropy": 5.907107543945313,
      "epoch": 2.1767926527728716,
      "grad_norm": 1.09375,
      "learning_rate": 0.000452472900376819,
      "loss": 5.2749,
      "mean_token_accuracy": 0.20810548663139344,
      "num_tokens": 45966689.0,
      "step": 24650
    },
    {
      "entropy": 5.90785174369812,
      "epoch": 2.177234192864712,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045245354111077827,
      "loss": 5.3534,
      "mean_token_accuracy": 0.20499788969755173,
      "num_tokens": 45975618.0,
      "step": 24655
    },
    {
      "entropy": 5.877581405639648,
      "epoch": 2.1776757329565526,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045243417836852355,
      "loss": 5.2893,
      "mean_token_accuracy": 0.21044217497110368,
      "num_tokens": 45984500.0,
      "step": 24660
    },
    {
      "entropy": 5.919665718078614,
      "epoch": 2.1781172730483926,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004524148121504343,
      "loss": 5.2821,
      "mean_token_accuracy": 0.20708625465631486,
      "num_tokens": 45994471.0,
      "step": 24665
    },
    {
      "entropy": 6.013688278198242,
      "epoch": 2.178558813140233,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045239544245688967,
      "loss": 5.3607,
      "mean_token_accuracy": 0.20263609439134597,
      "num_tokens": 46003757.0,
      "step": 24670
    },
    {
      "entropy": 5.843350505828857,
      "epoch": 2.1790003532320736,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004523760692882693,
      "loss": 5.207,
      "mean_token_accuracy": 0.22185277789831162,
      "num_tokens": 46013512.0,
      "step": 24675
    },
    {
      "entropy": 5.8821147918701175,
      "epoch": 2.1794418933239137,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004523566926449526,
      "loss": 5.2902,
      "mean_token_accuracy": 0.21643492728471755,
      "num_tokens": 46022867.0,
      "step": 24680
    },
    {
      "entropy": 5.897031164169311,
      "epoch": 2.179883433415754,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045233731252731905,
      "loss": 5.3253,
      "mean_token_accuracy": 0.2071937397122383,
      "num_tokens": 46031799.0,
      "step": 24685
    },
    {
      "entropy": 5.886275148391723,
      "epoch": 2.1803249735075947,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045231792893574846,
      "loss": 5.3355,
      "mean_token_accuracy": 0.2045131057500839,
      "num_tokens": 46041917.0,
      "step": 24690
    },
    {
      "entropy": 5.919997787475586,
      "epoch": 2.1807665135994347,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045229854187062054,
      "loss": 5.3231,
      "mean_token_accuracy": 0.21723242104053497,
      "num_tokens": 46051629.0,
      "step": 24695
    },
    {
      "entropy": 5.8439795017242435,
      "epoch": 2.1812080536912752,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004522791513323149,
      "loss": 5.217,
      "mean_token_accuracy": 0.2257873997092247,
      "num_tokens": 46060840.0,
      "step": 24700
    },
    {
      "entropy": 5.8255692481994625,
      "epoch": 2.1816495937831153,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045225975732121165,
      "loss": 5.2464,
      "mean_token_accuracy": 0.21378356367349624,
      "num_tokens": 46069774.0,
      "step": 24705
    },
    {
      "entropy": 5.941590452194214,
      "epoch": 2.182091133874956,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045224035983769054,
      "loss": 5.3464,
      "mean_token_accuracy": 0.20414550006389617,
      "num_tokens": 46079377.0,
      "step": 24710
    },
    {
      "entropy": 5.8948118686676025,
      "epoch": 2.1825326739667963,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045222095888213155,
      "loss": 5.2737,
      "mean_token_accuracy": 0.21850760281085968,
      "num_tokens": 46088190.0,
      "step": 24715
    },
    {
      "entropy": 5.88330717086792,
      "epoch": 2.1829742140586363,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045220155445491465,
      "loss": 5.3242,
      "mean_token_accuracy": 0.2182814046740532,
      "num_tokens": 46097487.0,
      "step": 24720
    },
    {
      "entropy": 5.988511943817139,
      "epoch": 2.183415754150477,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004521821465564203,
      "loss": 5.3989,
      "mean_token_accuracy": 0.20368327498435973,
      "num_tokens": 46106554.0,
      "step": 24725
    },
    {
      "entropy": 5.910604333877563,
      "epoch": 2.1838572942423173,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004521627351870283,
      "loss": 5.3665,
      "mean_token_accuracy": 0.19943333566188812,
      "num_tokens": 46115665.0,
      "step": 24730
    },
    {
      "entropy": 5.875160169601441,
      "epoch": 2.1842988343341574,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004521433203471192,
      "loss": 5.2427,
      "mean_token_accuracy": 0.21412675380706786,
      "num_tokens": 46124863.0,
      "step": 24735
    },
    {
      "entropy": 5.914878892898559,
      "epoch": 2.184740374425998,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045212390203707315,
      "loss": 5.2673,
      "mean_token_accuracy": 0.20714907199144364,
      "num_tokens": 46133986.0,
      "step": 24740
    },
    {
      "entropy": 5.867431211471557,
      "epoch": 2.1851819145178384,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004521044802572705,
      "loss": 5.2326,
      "mean_token_accuracy": 0.2173826053738594,
      "num_tokens": 46142234.0,
      "step": 24745
    },
    {
      "entropy": 5.871627235412598,
      "epoch": 2.1856234546096784,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045208505500809196,
      "loss": 5.3164,
      "mean_token_accuracy": 0.21288973242044448,
      "num_tokens": 46150976.0,
      "step": 24750
    },
    {
      "entropy": 5.9165575981140135,
      "epoch": 2.186064994701519,
      "grad_norm": 1.125,
      "learning_rate": 0.00045206562628991785,
      "loss": 5.3212,
      "mean_token_accuracy": 0.20821150541305541,
      "num_tokens": 46160541.0,
      "step": 24755
    },
    {
      "entropy": 5.983079338073731,
      "epoch": 2.1865065347933594,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045204619410312875,
      "loss": 5.3598,
      "mean_token_accuracy": 0.2055536165833473,
      "num_tokens": 46170958.0,
      "step": 24760
    },
    {
      "entropy": 5.911781120300293,
      "epoch": 2.1869480748851995,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004520267584481055,
      "loss": 5.3002,
      "mean_token_accuracy": 0.21306820958852768,
      "num_tokens": 46180293.0,
      "step": 24765
    },
    {
      "entropy": 5.854779386520386,
      "epoch": 2.18738961497704,
      "grad_norm": 1.125,
      "learning_rate": 0.00045200731932522873,
      "loss": 5.2644,
      "mean_token_accuracy": 0.21739444881677628,
      "num_tokens": 46189264.0,
      "step": 24770
    },
    {
      "entropy": 5.958702898025512,
      "epoch": 2.1878311550688805,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004519878767348792,
      "loss": 5.3999,
      "mean_token_accuracy": 0.20089254528284073,
      "num_tokens": 46199417.0,
      "step": 24775
    },
    {
      "entropy": 5.87711820602417,
      "epoch": 2.1882726951607205,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004519684306774378,
      "loss": 5.2395,
      "mean_token_accuracy": 0.21633054614067077,
      "num_tokens": 46209360.0,
      "step": 24780
    },
    {
      "entropy": 5.8455109119415285,
      "epoch": 2.188714235252561,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045194898115328555,
      "loss": 5.2822,
      "mean_token_accuracy": 0.20362259894609452,
      "num_tokens": 46218414.0,
      "step": 24785
    },
    {
      "entropy": 5.867178297042846,
      "epoch": 2.189155775344401,
      "grad_norm": 1.125,
      "learning_rate": 0.0004519295281628033,
      "loss": 5.2841,
      "mean_token_accuracy": 0.20989346206188203,
      "num_tokens": 46227526.0,
      "step": 24790
    },
    {
      "entropy": 5.910853910446167,
      "epoch": 2.1895973154362416,
      "grad_norm": 1.125,
      "learning_rate": 0.0004519100717063723,
      "loss": 5.2326,
      "mean_token_accuracy": 0.2142106845974922,
      "num_tokens": 46236041.0,
      "step": 24795
    },
    {
      "entropy": 5.8417548656463625,
      "epoch": 2.190038855528082,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045189061178437363,
      "loss": 5.2359,
      "mean_token_accuracy": 0.2107262760400772,
      "num_tokens": 46244557.0,
      "step": 24800
    },
    {
      "entropy": 5.863319206237793,
      "epoch": 2.190480395619922,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004518711483971885,
      "loss": 5.2374,
      "mean_token_accuracy": 0.21817210614681243,
      "num_tokens": 46253957.0,
      "step": 24805
    },
    {
      "entropy": 5.948736715316772,
      "epoch": 2.1909219357117626,
      "grad_norm": 1.203125,
      "learning_rate": 0.000451851681545198,
      "loss": 5.3347,
      "mean_token_accuracy": 0.20921835005283357,
      "num_tokens": 46263110.0,
      "step": 24810
    },
    {
      "entropy": 5.872492361068725,
      "epoch": 2.191363475803603,
      "grad_norm": 1.125,
      "learning_rate": 0.00045183221122878375,
      "loss": 5.3469,
      "mean_token_accuracy": 0.20458832383155823,
      "num_tokens": 46272163.0,
      "step": 24815
    },
    {
      "entropy": 5.835750770568848,
      "epoch": 2.191805015895443,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045181273744832707,
      "loss": 5.3153,
      "mean_token_accuracy": 0.2052254855632782,
      "num_tokens": 46281097.0,
      "step": 24820
    },
    {
      "entropy": 5.933538150787354,
      "epoch": 2.1922465559872837,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004517932602042094,
      "loss": 5.2769,
      "mean_token_accuracy": 0.20738780945539476,
      "num_tokens": 46291552.0,
      "step": 24825
    },
    {
      "entropy": 5.8951469421386715,
      "epoch": 2.192688096079124,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045177377949681235,
      "loss": 5.3115,
      "mean_token_accuracy": 0.21112328469753266,
      "num_tokens": 46300661.0,
      "step": 24830
    },
    {
      "entropy": 5.8536769390106205,
      "epoch": 2.1931296361709642,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045175429532651743,
      "loss": 5.2531,
      "mean_token_accuracy": 0.21318843215703964,
      "num_tokens": 46309705.0,
      "step": 24835
    },
    {
      "entropy": 5.842858219146729,
      "epoch": 2.1935711762628047,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004517348076937064,
      "loss": 5.2745,
      "mean_token_accuracy": 0.21584783345460892,
      "num_tokens": 46319701.0,
      "step": 24840
    },
    {
      "entropy": 5.804854965209961,
      "epoch": 2.194012716354645,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045171531659876104,
      "loss": 5.2619,
      "mean_token_accuracy": 0.21641617715358735,
      "num_tokens": 46328195.0,
      "step": 24845
    },
    {
      "entropy": 5.920786809921265,
      "epoch": 2.1944542564464853,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004516958220420631,
      "loss": 5.3088,
      "mean_token_accuracy": 0.21326993256807328,
      "num_tokens": 46337518.0,
      "step": 24850
    },
    {
      "entropy": 5.860372734069824,
      "epoch": 2.194895796538326,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004516763240239945,
      "loss": 5.235,
      "mean_token_accuracy": 0.2159891277551651,
      "num_tokens": 46346406.0,
      "step": 24855
    },
    {
      "entropy": 5.843536233901977,
      "epoch": 2.195337336630166,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004516568225449372,
      "loss": 5.2314,
      "mean_token_accuracy": 0.21313668191432952,
      "num_tokens": 46355284.0,
      "step": 24860
    },
    {
      "entropy": 5.902785110473633,
      "epoch": 2.1957788767220063,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045163731760527326,
      "loss": 5.3466,
      "mean_token_accuracy": 0.2091919496655464,
      "num_tokens": 46364982.0,
      "step": 24865
    },
    {
      "entropy": 5.814175987243653,
      "epoch": 2.196220416813847,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004516178092053847,
      "loss": 5.1796,
      "mean_token_accuracy": 0.22120079398155212,
      "num_tokens": 46373923.0,
      "step": 24870
    },
    {
      "entropy": 5.854439735412598,
      "epoch": 2.196661956905687,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004515982973456537,
      "loss": 5.2101,
      "mean_token_accuracy": 0.21815921664237975,
      "num_tokens": 46382194.0,
      "step": 24875
    },
    {
      "entropy": 5.829131126403809,
      "epoch": 2.1971034969975274,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004515787820264624,
      "loss": 5.2868,
      "mean_token_accuracy": 0.20817655622959136,
      "num_tokens": 46391463.0,
      "step": 24880
    },
    {
      "entropy": 5.891905784606934,
      "epoch": 2.197545037089368,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045155926324819335,
      "loss": 5.3097,
      "mean_token_accuracy": 0.21281156688928604,
      "num_tokens": 46401235.0,
      "step": 24885
    },
    {
      "entropy": 5.895998191833496,
      "epoch": 2.197986577181208,
      "grad_norm": 1.125,
      "learning_rate": 0.00045153974101122866,
      "loss": 5.3426,
      "mean_token_accuracy": 0.20992677211761473,
      "num_tokens": 46410631.0,
      "step": 24890
    },
    {
      "entropy": 5.933129453659058,
      "epoch": 2.1984281172730484,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045152021531595085,
      "loss": 5.3483,
      "mean_token_accuracy": 0.20004705488681793,
      "num_tokens": 46420467.0,
      "step": 24895
    },
    {
      "entropy": 5.877530336380005,
      "epoch": 2.198869657364889,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004515006861627424,
      "loss": 5.2475,
      "mean_token_accuracy": 0.21700648814439774,
      "num_tokens": 46430276.0,
      "step": 24900
    },
    {
      "entropy": 5.972809648513794,
      "epoch": 2.199311197456729,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045148115355198596,
      "loss": 5.3574,
      "mean_token_accuracy": 0.21151867508888245,
      "num_tokens": 46439227.0,
      "step": 24905
    },
    {
      "entropy": 5.8662396430969235,
      "epoch": 2.1997527375485695,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000451461617484064,
      "loss": 5.3092,
      "mean_token_accuracy": 0.20609249621629716,
      "num_tokens": 46449043.0,
      "step": 24910
    },
    {
      "entropy": 5.8734776973724365,
      "epoch": 2.20019427764041,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004514420779593594,
      "loss": 5.2975,
      "mean_token_accuracy": 0.20904383957386016,
      "num_tokens": 46458332.0,
      "step": 24915
    },
    {
      "entropy": 5.971198177337646,
      "epoch": 2.20063581773225,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045142253497825477,
      "loss": 5.2397,
      "mean_token_accuracy": 0.21337376832962035,
      "num_tokens": 46467208.0,
      "step": 24920
    },
    {
      "entropy": 5.8712622165679935,
      "epoch": 2.2010773578240905,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045140298854113306,
      "loss": 5.2608,
      "mean_token_accuracy": 0.21790449917316437,
      "num_tokens": 46476715.0,
      "step": 24925
    },
    {
      "entropy": 5.7847902297973635,
      "epoch": 2.2015188979159306,
      "grad_norm": 1.125,
      "learning_rate": 0.00045138343864837703,
      "loss": 5.208,
      "mean_token_accuracy": 0.22131199687719344,
      "num_tokens": 46484783.0,
      "step": 24930
    },
    {
      "entropy": 5.83673677444458,
      "epoch": 2.201960438007771,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004513638853003699,
      "loss": 5.1931,
      "mean_token_accuracy": 0.21337351500988005,
      "num_tokens": 46493616.0,
      "step": 24935
    },
    {
      "entropy": 5.920994091033935,
      "epoch": 2.2024019780996116,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045134432849749446,
      "loss": 5.2948,
      "mean_token_accuracy": 0.21216038912534713,
      "num_tokens": 46502779.0,
      "step": 24940
    },
    {
      "entropy": 5.855118370056152,
      "epoch": 2.2028435181914516,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045132476824013396,
      "loss": 5.2946,
      "mean_token_accuracy": 0.21107901632785797,
      "num_tokens": 46511506.0,
      "step": 24945
    },
    {
      "entropy": 5.926555442810058,
      "epoch": 2.203285058283292,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004513052045286715,
      "loss": 5.3223,
      "mean_token_accuracy": 0.20706935822963715,
      "num_tokens": 46520219.0,
      "step": 24950
    },
    {
      "entropy": 5.947846698760986,
      "epoch": 2.2037265983751326,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004512856373634904,
      "loss": 5.2831,
      "mean_token_accuracy": 0.21450922787189483,
      "num_tokens": 46528737.0,
      "step": 24955
    },
    {
      "entropy": 5.925307989120483,
      "epoch": 2.2041681384669727,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045126606674497394,
      "loss": 5.374,
      "mean_token_accuracy": 0.20536865592002868,
      "num_tokens": 46538802.0,
      "step": 24960
    },
    {
      "entropy": 5.87228307723999,
      "epoch": 2.204609678558813,
      "grad_norm": 1.125,
      "learning_rate": 0.0004512464926735054,
      "loss": 5.2969,
      "mean_token_accuracy": 0.21601083278656005,
      "num_tokens": 46548369.0,
      "step": 24965
    },
    {
      "entropy": 5.873328018188476,
      "epoch": 2.2050512186506537,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045122691514946835,
      "loss": 5.3408,
      "mean_token_accuracy": 0.207840633392334,
      "num_tokens": 46557876.0,
      "step": 24970
    },
    {
      "entropy": 5.839500808715821,
      "epoch": 2.2054927587424937,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004512073341732463,
      "loss": 5.2873,
      "mean_token_accuracy": 0.22232381552457808,
      "num_tokens": 46566747.0,
      "step": 24975
    },
    {
      "entropy": 5.923096609115601,
      "epoch": 2.2059342988343342,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004511877497452227,
      "loss": 5.2837,
      "mean_token_accuracy": 0.21339815557003022,
      "num_tokens": 46576196.0,
      "step": 24980
    },
    {
      "entropy": 5.970819520950317,
      "epoch": 2.2063758389261743,
      "grad_norm": 1.0,
      "learning_rate": 0.0004511681618657813,
      "loss": 5.3767,
      "mean_token_accuracy": 0.20547797977924348,
      "num_tokens": 46585734.0,
      "step": 24985
    },
    {
      "entropy": 5.813763380050659,
      "epoch": 2.206817379018015,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045114857053530586,
      "loss": 5.2274,
      "mean_token_accuracy": 0.21588525772094727,
      "num_tokens": 46595931.0,
      "step": 24990
    },
    {
      "entropy": 5.910294151306152,
      "epoch": 2.2072589191098553,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00045112897575418006,
      "loss": 5.2965,
      "mean_token_accuracy": 0.20620531141757964,
      "num_tokens": 46604688.0,
      "step": 24995
    },
    {
      "entropy": 5.848907613754273,
      "epoch": 2.2077004592016953,
      "grad_norm": 1.125,
      "learning_rate": 0.0004511093775227878,
      "loss": 5.2798,
      "mean_token_accuracy": 0.21882594972848893,
      "num_tokens": 46613860.0,
      "step": 25000
    },
    {
      "entropy": 5.929471445083618,
      "epoch": 2.208141999293536,
      "grad_norm": 1.0625,
      "learning_rate": 0.000451089775841513,
      "loss": 5.3365,
      "mean_token_accuracy": 0.20733083933591842,
      "num_tokens": 46623041.0,
      "step": 25005
    },
    {
      "entropy": 5.898074197769165,
      "epoch": 2.2085835393853763,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045107017071073963,
      "loss": 5.3134,
      "mean_token_accuracy": 0.20037408620119096,
      "num_tokens": 46633025.0,
      "step": 25010
    },
    {
      "entropy": 5.954243850708008,
      "epoch": 2.2090250794772164,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004510505621308517,
      "loss": 5.3108,
      "mean_token_accuracy": 0.20995305925607682,
      "num_tokens": 46641826.0,
      "step": 25015
    },
    {
      "entropy": 5.968498039245605,
      "epoch": 2.209466619569057,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045103095010223354,
      "loss": 5.395,
      "mean_token_accuracy": 0.20329425781965255,
      "num_tokens": 46652028.0,
      "step": 25020
    },
    {
      "entropy": 5.86537013053894,
      "epoch": 2.2099081596608974,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045101133462526903,
      "loss": 5.3911,
      "mean_token_accuracy": 0.20344689041376113,
      "num_tokens": 46660763.0,
      "step": 25025
    },
    {
      "entropy": 5.909679746627807,
      "epoch": 2.2103496997527374,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004509917157003427,
      "loss": 5.3289,
      "mean_token_accuracy": 0.20585323572158815,
      "num_tokens": 46670213.0,
      "step": 25030
    },
    {
      "entropy": 5.984507894515991,
      "epoch": 2.210791239844578,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004509720933278386,
      "loss": 5.3113,
      "mean_token_accuracy": 0.21056326180696489,
      "num_tokens": 46679658.0,
      "step": 25035
    },
    {
      "entropy": 5.881831407546997,
      "epoch": 2.2112327799364184,
      "grad_norm": 1.125,
      "learning_rate": 0.0004509524675081414,
      "loss": 5.2593,
      "mean_token_accuracy": 0.21351644545793533,
      "num_tokens": 46687920.0,
      "step": 25040
    },
    {
      "entropy": 5.897279882431031,
      "epoch": 2.2116743200282585,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004509328382416354,
      "loss": 5.3571,
      "mean_token_accuracy": 0.20236642360687257,
      "num_tokens": 46696779.0,
      "step": 25045
    },
    {
      "entropy": 5.918566751480102,
      "epoch": 2.212115860120099,
      "grad_norm": 1.125,
      "learning_rate": 0.0004509132055287051,
      "loss": 5.3122,
      "mean_token_accuracy": 0.19762982279062272,
      "num_tokens": 46706568.0,
      "step": 25050
    },
    {
      "entropy": 5.9651261329650875,
      "epoch": 2.2125574002119395,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004508935693697352,
      "loss": 5.3682,
      "mean_token_accuracy": 0.21093694418668746,
      "num_tokens": 46715369.0,
      "step": 25055
    },
    {
      "entropy": 5.8661106586456295,
      "epoch": 2.2129989403037795,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004508739297651103,
      "loss": 5.2658,
      "mean_token_accuracy": 0.21148535013198852,
      "num_tokens": 46724867.0,
      "step": 25060
    },
    {
      "entropy": 5.866259765625,
      "epoch": 2.21344048039562,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045085428671521514,
      "loss": 5.3114,
      "mean_token_accuracy": 0.21071349531412126,
      "num_tokens": 46734611.0,
      "step": 25065
    },
    {
      "entropy": 5.935286378860473,
      "epoch": 2.21388202048746,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045083464022043456,
      "loss": 5.3203,
      "mean_token_accuracy": 0.20520623475313188,
      "num_tokens": 46744031.0,
      "step": 25070
    },
    {
      "entropy": 5.873317766189575,
      "epoch": 2.2143235605793006,
      "grad_norm": 1.125,
      "learning_rate": 0.00045081499028115337,
      "loss": 5.2467,
      "mean_token_accuracy": 0.22226097583770751,
      "num_tokens": 46753379.0,
      "step": 25075
    },
    {
      "entropy": 5.963942050933838,
      "epoch": 2.214765100671141,
      "grad_norm": 1.125,
      "learning_rate": 0.0004507953368977565,
      "loss": 5.3827,
      "mean_token_accuracy": 0.19638515263795853,
      "num_tokens": 46762625.0,
      "step": 25080
    },
    {
      "entropy": 5.924911308288574,
      "epoch": 2.215206640762981,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045077568007062893,
      "loss": 5.3529,
      "mean_token_accuracy": 0.21058709174394608,
      "num_tokens": 46772577.0,
      "step": 25085
    },
    {
      "entropy": 5.929365015029907,
      "epoch": 2.2156481808548216,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045075601980015573,
      "loss": 5.3221,
      "mean_token_accuracy": 0.21401261538267136,
      "num_tokens": 46782226.0,
      "step": 25090
    },
    {
      "entropy": 5.872507476806641,
      "epoch": 2.216089720946662,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045073635608672215,
      "loss": 5.2176,
      "mean_token_accuracy": 0.21572758108377457,
      "num_tokens": 46789988.0,
      "step": 25095
    },
    {
      "entropy": 5.910791254043579,
      "epoch": 2.216531261038502,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045071668893071324,
      "loss": 5.3543,
      "mean_token_accuracy": 0.19882809668779372,
      "num_tokens": 46799448.0,
      "step": 25100
    },
    {
      "entropy": 5.93634033203125,
      "epoch": 2.2169728011303427,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045069701833251434,
      "loss": 5.3458,
      "mean_token_accuracy": 0.20672561824321747,
      "num_tokens": 46809829.0,
      "step": 25105
    },
    {
      "entropy": 5.8733679294586185,
      "epoch": 2.217414341222183,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045067734429251086,
      "loss": 5.2877,
      "mean_token_accuracy": 0.21472548097372054,
      "num_tokens": 46817975.0,
      "step": 25110
    },
    {
      "entropy": 5.829145097732544,
      "epoch": 2.2178558813140232,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045065766681108793,
      "loss": 5.243,
      "mean_token_accuracy": 0.2141059309244156,
      "num_tokens": 46827350.0,
      "step": 25115
    },
    {
      "entropy": 5.910316228866577,
      "epoch": 2.2182974214058637,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004506379858886314,
      "loss": 5.303,
      "mean_token_accuracy": 0.21683756709098817,
      "num_tokens": 46835944.0,
      "step": 25120
    },
    {
      "entropy": 5.8803784370422365,
      "epoch": 2.2187389614977038,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045061830152552646,
      "loss": 5.2655,
      "mean_token_accuracy": 0.21124572157859803,
      "num_tokens": 46844843.0,
      "step": 25125
    },
    {
      "entropy": 5.902810001373291,
      "epoch": 2.2191805015895443,
      "grad_norm": 1.1875,
      "learning_rate": 0.000450598613722159,
      "loss": 5.3852,
      "mean_token_accuracy": 0.19811780899763107,
      "num_tokens": 46854188.0,
      "step": 25130
    },
    {
      "entropy": 5.913742113113403,
      "epoch": 2.2196220416813848,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004505789224789145,
      "loss": 5.3385,
      "mean_token_accuracy": 0.20688226521015168,
      "num_tokens": 46863279.0,
      "step": 25135
    },
    {
      "entropy": 5.931808853149414,
      "epoch": 2.220063581773225,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045055922779617874,
      "loss": 5.2956,
      "mean_token_accuracy": 0.20689380913972855,
      "num_tokens": 46873336.0,
      "step": 25140
    },
    {
      "entropy": 5.912406778335571,
      "epoch": 2.2205051218650653,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004505395296743376,
      "loss": 5.2988,
      "mean_token_accuracy": 0.2146769717335701,
      "num_tokens": 46882497.0,
      "step": 25145
    },
    {
      "entropy": 5.846728801727295,
      "epoch": 2.220946661956906,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00045051982811377696,
      "loss": 5.2445,
      "mean_token_accuracy": 0.2113554820418358,
      "num_tokens": 46891260.0,
      "step": 25150
    },
    {
      "entropy": 5.896357488632202,
      "epoch": 2.221388202048746,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045050012311488265,
      "loss": 5.2155,
      "mean_token_accuracy": 0.21709609925746917,
      "num_tokens": 46899969.0,
      "step": 25155
    },
    {
      "entropy": 5.842054843902588,
      "epoch": 2.2218297421405864,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045048041467804067,
      "loss": 5.2331,
      "mean_token_accuracy": 0.21707316786050795,
      "num_tokens": 46909206.0,
      "step": 25160
    },
    {
      "entropy": 5.862126064300537,
      "epoch": 2.222271282232427,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045046070280363725,
      "loss": 5.2447,
      "mean_token_accuracy": 0.21472883969545364,
      "num_tokens": 46918423.0,
      "step": 25165
    },
    {
      "entropy": 5.918490266799926,
      "epoch": 2.222712822324267,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004504409874920585,
      "loss": 5.2886,
      "mean_token_accuracy": 0.20576518774032593,
      "num_tokens": 46927817.0,
      "step": 25170
    },
    {
      "entropy": 5.988033962249756,
      "epoch": 2.2231543624161074,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004504212687436905,
      "loss": 5.3587,
      "mean_token_accuracy": 0.20495160669088364,
      "num_tokens": 46936880.0,
      "step": 25175
    },
    {
      "entropy": 5.996985864639282,
      "epoch": 2.223595902507948,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004504015465589196,
      "loss": 5.2994,
      "mean_token_accuracy": 0.1994699567556381,
      "num_tokens": 46946722.0,
      "step": 25180
    },
    {
      "entropy": 5.881836557388306,
      "epoch": 2.224037442599788,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004503818209381322,
      "loss": 5.2676,
      "mean_token_accuracy": 0.21085475087165834,
      "num_tokens": 46956660.0,
      "step": 25185
    },
    {
      "entropy": 5.929487466812134,
      "epoch": 2.2244789826916285,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004503620918817147,
      "loss": 5.3206,
      "mean_token_accuracy": 0.19917973428964614,
      "num_tokens": 46965823.0,
      "step": 25190
    },
    {
      "entropy": 5.926318836212158,
      "epoch": 2.224920522783469,
      "grad_norm": 1.125,
      "learning_rate": 0.00045034235939005353,
      "loss": 5.2955,
      "mean_token_accuracy": 0.2085857257246971,
      "num_tokens": 46974933.0,
      "step": 25195
    },
    {
      "entropy": 5.936947011947632,
      "epoch": 2.225362062875309,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045032262346353526,
      "loss": 5.3779,
      "mean_token_accuracy": 0.20409031063318253,
      "num_tokens": 46983506.0,
      "step": 25200
    },
    {
      "entropy": 5.879003047943115,
      "epoch": 2.2258036029671495,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004503028841025465,
      "loss": 5.276,
      "mean_token_accuracy": 0.20933781564235687,
      "num_tokens": 46992234.0,
      "step": 25205
    },
    {
      "entropy": 5.8367249965667725,
      "epoch": 2.2262451430589896,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045028314130747397,
      "loss": 5.2185,
      "mean_token_accuracy": 0.21828076839447022,
      "num_tokens": 47002249.0,
      "step": 25210
    },
    {
      "entropy": 5.859875059127807,
      "epoch": 2.22668668315083,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045026339507870435,
      "loss": 5.3495,
      "mean_token_accuracy": 0.21007241904735566,
      "num_tokens": 47010721.0,
      "step": 25215
    },
    {
      "entropy": 5.956480169296265,
      "epoch": 2.2271282232426706,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045024364541662457,
      "loss": 5.3518,
      "mean_token_accuracy": 0.204175928235054,
      "num_tokens": 47019675.0,
      "step": 25220
    },
    {
      "entropy": 5.9527867317199705,
      "epoch": 2.2275697633345106,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004502238923216213,
      "loss": 5.2812,
      "mean_token_accuracy": 0.21272652447223664,
      "num_tokens": 47028293.0,
      "step": 25225
    },
    {
      "entropy": 5.965755987167358,
      "epoch": 2.228011303426351,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004502041357940817,
      "loss": 5.3419,
      "mean_token_accuracy": 0.2071531221270561,
      "num_tokens": 47037605.0,
      "step": 25230
    },
    {
      "entropy": 5.868108034133911,
      "epoch": 2.2284528435181916,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045018437583439275,
      "loss": 5.2762,
      "mean_token_accuracy": 0.21650639474391936,
      "num_tokens": 47045889.0,
      "step": 25235
    },
    {
      "entropy": 5.867903137207032,
      "epoch": 2.2288943836100317,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045016461244294154,
      "loss": 5.322,
      "mean_token_accuracy": 0.2104552283883095,
      "num_tokens": 47055896.0,
      "step": 25240
    },
    {
      "entropy": 5.935128831863404,
      "epoch": 2.229335923701872,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045014484562011514,
      "loss": 5.3473,
      "mean_token_accuracy": 0.20300930440425874,
      "num_tokens": 47064980.0,
      "step": 25245
    },
    {
      "entropy": 5.853549432754517,
      "epoch": 2.2297774637937127,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004501250753663008,
      "loss": 5.2202,
      "mean_token_accuracy": 0.21265652030706406,
      "num_tokens": 47074045.0,
      "step": 25250
    },
    {
      "entropy": 5.849943590164185,
      "epoch": 2.2302190038855527,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004501053016818859,
      "loss": 5.2379,
      "mean_token_accuracy": 0.21710557788610457,
      "num_tokens": 47083439.0,
      "step": 25255
    },
    {
      "entropy": 5.870268821716309,
      "epoch": 2.230660543977393,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045008552456725766,
      "loss": 5.276,
      "mean_token_accuracy": 0.21208076030015946,
      "num_tokens": 47091977.0,
      "step": 25260
    },
    {
      "entropy": 5.877220201492309,
      "epoch": 2.2311020840692333,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004500657440228037,
      "loss": 5.3058,
      "mean_token_accuracy": 0.20941203236579894,
      "num_tokens": 47102197.0,
      "step": 25265
    },
    {
      "entropy": 5.966434955596924,
      "epoch": 2.2315436241610738,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045004596004891124,
      "loss": 5.3353,
      "mean_token_accuracy": 0.2138953238725662,
      "num_tokens": 47111563.0,
      "step": 25270
    },
    {
      "entropy": 5.888266372680664,
      "epoch": 2.2319851642529143,
      "grad_norm": 1.125,
      "learning_rate": 0.00045002617264596807,
      "loss": 5.3041,
      "mean_token_accuracy": 0.21606009900569917,
      "num_tokens": 47120272.0,
      "step": 25275
    },
    {
      "entropy": 5.863013458251953,
      "epoch": 2.2324267043447543,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004500063818143617,
      "loss": 5.2776,
      "mean_token_accuracy": 0.21193020939826965,
      "num_tokens": 47131023.0,
      "step": 25280
    },
    {
      "entropy": 5.828634023666382,
      "epoch": 2.232868244436595,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004499865875544797,
      "loss": 5.2548,
      "mean_token_accuracy": 0.21766403317451477,
      "num_tokens": 47139968.0,
      "step": 25285
    },
    {
      "entropy": 5.859275817871094,
      "epoch": 2.2333097845284353,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004499667898667101,
      "loss": 5.258,
      "mean_token_accuracy": 0.21532260477542878,
      "num_tokens": 47149843.0,
      "step": 25290
    },
    {
      "entropy": 5.895445394515991,
      "epoch": 2.2337513246202754,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044994698875144064,
      "loss": 5.3087,
      "mean_token_accuracy": 0.2108255445957184,
      "num_tokens": 47159025.0,
      "step": 25295
    },
    {
      "entropy": 5.904364681243896,
      "epoch": 2.234192864712116,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004499271842090591,
      "loss": 5.3293,
      "mean_token_accuracy": 0.2063928499817848,
      "num_tokens": 47168644.0,
      "step": 25300
    },
    {
      "entropy": 5.87594690322876,
      "epoch": 2.2346344048039564,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044990737623995357,
      "loss": 5.2394,
      "mean_token_accuracy": 0.2168074816465378,
      "num_tokens": 47178051.0,
      "step": 25305
    },
    {
      "entropy": 5.957010126113891,
      "epoch": 2.2350759448957964,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004498875648445119,
      "loss": 5.3934,
      "mean_token_accuracy": 0.20169862508773803,
      "num_tokens": 47187855.0,
      "step": 25310
    },
    {
      "entropy": 5.852364349365234,
      "epoch": 2.235517484987637,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004498677500231224,
      "loss": 5.3207,
      "mean_token_accuracy": 0.2070738285779953,
      "num_tokens": 47196700.0,
      "step": 25315
    },
    {
      "entropy": 5.932506465911866,
      "epoch": 2.2359590250794774,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004498479317761731,
      "loss": 5.3812,
      "mean_token_accuracy": 0.2044564366340637,
      "num_tokens": 47205202.0,
      "step": 25320
    },
    {
      "entropy": 5.848980379104614,
      "epoch": 2.2364005651713175,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004498281101040522,
      "loss": 5.2853,
      "mean_token_accuracy": 0.2187192052602768,
      "num_tokens": 47214803.0,
      "step": 25325
    },
    {
      "entropy": 5.972981214523315,
      "epoch": 2.236842105263158,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044980828500714814,
      "loss": 5.3826,
      "mean_token_accuracy": 0.20191094726324083,
      "num_tokens": 47224641.0,
      "step": 25330
    },
    {
      "entropy": 5.915935611724853,
      "epoch": 2.2372836453549985,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004497884564858492,
      "loss": 5.339,
      "mean_token_accuracy": 0.1996872067451477,
      "num_tokens": 47233905.0,
      "step": 25335
    },
    {
      "entropy": 5.903089284896851,
      "epoch": 2.2377251854468385,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044976862454054376,
      "loss": 5.2588,
      "mean_token_accuracy": 0.21022669225931168,
      "num_tokens": 47242209.0,
      "step": 25340
    },
    {
      "entropy": 5.9644183158874515,
      "epoch": 2.238166725538679,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004497487891716204,
      "loss": 5.3399,
      "mean_token_accuracy": 0.21070777326822282,
      "num_tokens": 47252609.0,
      "step": 25345
    },
    {
      "entropy": 5.962730026245117,
      "epoch": 2.238608265630519,
      "grad_norm": 1.125,
      "learning_rate": 0.00044972895037946755,
      "loss": 5.3551,
      "mean_token_accuracy": 0.20537231415510177,
      "num_tokens": 47261868.0,
      "step": 25350
    },
    {
      "entropy": 5.935928535461426,
      "epoch": 2.2390498057223596,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044970910816447403,
      "loss": 5.2872,
      "mean_token_accuracy": 0.21497126966714858,
      "num_tokens": 47271404.0,
      "step": 25355
    },
    {
      "entropy": 5.922522735595703,
      "epoch": 2.2394913458142,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044968926252702835,
      "loss": 5.3435,
      "mean_token_accuracy": 0.2059139981865883,
      "num_tokens": 47280669.0,
      "step": 25360
    },
    {
      "entropy": 5.938412523269653,
      "epoch": 2.23993288590604,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004496694134675195,
      "loss": 5.3536,
      "mean_token_accuracy": 0.2070578634738922,
      "num_tokens": 47290096.0,
      "step": 25365
    },
    {
      "entropy": 5.964613246917724,
      "epoch": 2.2403744259978806,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004496495609863361,
      "loss": 5.3442,
      "mean_token_accuracy": 0.20849495381116867,
      "num_tokens": 47298767.0,
      "step": 25370
    },
    {
      "entropy": 5.834093189239502,
      "epoch": 2.240815966089721,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044962970508386716,
      "loss": 5.2142,
      "mean_token_accuracy": 0.2193813592195511,
      "num_tokens": 47307950.0,
      "step": 25375
    },
    {
      "entropy": 5.842846775054932,
      "epoch": 2.241257506181561,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004496098457605016,
      "loss": 5.3171,
      "mean_token_accuracy": 0.21000723540782928,
      "num_tokens": 47317963.0,
      "step": 25380
    },
    {
      "entropy": 5.89036283493042,
      "epoch": 2.2416990462734017,
      "grad_norm": 1.125,
      "learning_rate": 0.0004495899830166284,
      "loss": 5.3154,
      "mean_token_accuracy": 0.2095501020550728,
      "num_tokens": 47326728.0,
      "step": 25385
    },
    {
      "entropy": 6.011849403381348,
      "epoch": 2.242140586365242,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004495701168526368,
      "loss": 5.3265,
      "mean_token_accuracy": 0.20927484929561616,
      "num_tokens": 47336386.0,
      "step": 25390
    },
    {
      "entropy": 5.95103268623352,
      "epoch": 2.242582126457082,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044955024726891585,
      "loss": 5.3366,
      "mean_token_accuracy": 0.2021625816822052,
      "num_tokens": 47345958.0,
      "step": 25395
    },
    {
      "entropy": 5.902404022216797,
      "epoch": 2.2430236665489227,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004495303742658549,
      "loss": 5.3458,
      "mean_token_accuracy": 0.20588893741369246,
      "num_tokens": 47355758.0,
      "step": 25400
    },
    {
      "entropy": 5.965249490737915,
      "epoch": 2.2434652066407628,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004495104978438431,
      "loss": 5.3252,
      "mean_token_accuracy": 0.21320498287677764,
      "num_tokens": 47365186.0,
      "step": 25405
    },
    {
      "entropy": 5.91492018699646,
      "epoch": 2.2439067467326033,
      "grad_norm": 1.40625,
      "learning_rate": 0.00044949061800327,
      "loss": 5.3093,
      "mean_token_accuracy": 0.21295351833105086,
      "num_tokens": 47374443.0,
      "step": 25410
    },
    {
      "entropy": 5.8478227138519285,
      "epoch": 2.2443482868244438,
      "grad_norm": 1.125,
      "learning_rate": 0.0004494707347445248,
      "loss": 5.2356,
      "mean_token_accuracy": 0.21066135615110398,
      "num_tokens": 47383369.0,
      "step": 25415
    },
    {
      "entropy": 5.853169679641724,
      "epoch": 2.244789826916284,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004494508480679972,
      "loss": 5.2627,
      "mean_token_accuracy": 0.2113923028111458,
      "num_tokens": 47393104.0,
      "step": 25420
    },
    {
      "entropy": 5.840248966217041,
      "epoch": 2.2452313670081243,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044943095797407676,
      "loss": 5.2599,
      "mean_token_accuracy": 0.2124545991420746,
      "num_tokens": 47401954.0,
      "step": 25425
    },
    {
      "entropy": 5.949639463424683,
      "epoch": 2.245672907099965,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000449411064463153,
      "loss": 5.2631,
      "mean_token_accuracy": 0.2147961214184761,
      "num_tokens": 47411463.0,
      "step": 25430
    },
    {
      "entropy": 5.9110359191894535,
      "epoch": 2.246114447191805,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044939116753561564,
      "loss": 5.2016,
      "mean_token_accuracy": 0.2140165835618973,
      "num_tokens": 47421405.0,
      "step": 25435
    },
    {
      "entropy": 5.8378636837005615,
      "epoch": 2.2465559872836454,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004493712671918546,
      "loss": 5.3247,
      "mean_token_accuracy": 0.2112680584192276,
      "num_tokens": 47430268.0,
      "step": 25440
    },
    {
      "entropy": 5.9168859958648685,
      "epoch": 2.246997527375486,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004493513634322595,
      "loss": 5.3532,
      "mean_token_accuracy": 0.20273826718330384,
      "num_tokens": 47439571.0,
      "step": 25445
    },
    {
      "entropy": 5.908704090118408,
      "epoch": 2.247439067467326,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004493314562572204,
      "loss": 5.3002,
      "mean_token_accuracy": 0.20778064131736756,
      "num_tokens": 47449085.0,
      "step": 25450
    },
    {
      "entropy": 5.846994256973266,
      "epoch": 2.2478806075591664,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004493115456671272,
      "loss": 5.2136,
      "mean_token_accuracy": 0.21585794240236283,
      "num_tokens": 47458740.0,
      "step": 25455
    },
    {
      "entropy": 5.9260420322418215,
      "epoch": 2.248322147651007,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044929163166236997,
      "loss": 5.3499,
      "mean_token_accuracy": 0.20734633058309554,
      "num_tokens": 47467458.0,
      "step": 25460
    },
    {
      "entropy": 5.906539440155029,
      "epoch": 2.248763687742847,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004492717142433388,
      "loss": 5.3123,
      "mean_token_accuracy": 0.20516202747821807,
      "num_tokens": 47476845.0,
      "step": 25465
    },
    {
      "entropy": 5.882449293136597,
      "epoch": 2.2492052278346875,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044925179341042385,
      "loss": 5.3097,
      "mean_token_accuracy": 0.20874747037887573,
      "num_tokens": 47485632.0,
      "step": 25470
    },
    {
      "entropy": 5.944317388534546,
      "epoch": 2.249646767926528,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044923186916401536,
      "loss": 5.3056,
      "mean_token_accuracy": 0.21193400025367737,
      "num_tokens": 47494681.0,
      "step": 25475
    },
    {
      "entropy": 5.867374753952026,
      "epoch": 2.250088308018368,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004492119415045036,
      "loss": 5.3371,
      "mean_token_accuracy": 0.21191453486680983,
      "num_tokens": 47504482.0,
      "step": 25480
    },
    {
      "entropy": 5.869369745254517,
      "epoch": 2.2505298481102085,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004491920104322791,
      "loss": 5.3113,
      "mean_token_accuracy": 0.21246003955602646,
      "num_tokens": 47513601.0,
      "step": 25485
    },
    {
      "entropy": 5.910813283920288,
      "epoch": 2.2509713882020486,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004491720759477321,
      "loss": 5.2678,
      "mean_token_accuracy": 0.21182847023010254,
      "num_tokens": 47522811.0,
      "step": 25490
    },
    {
      "entropy": 5.912567615509033,
      "epoch": 2.251412928293889,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004491521380512533,
      "loss": 5.2812,
      "mean_token_accuracy": 0.2175702065229416,
      "num_tokens": 47532471.0,
      "step": 25495
    },
    {
      "entropy": 5.828784799575805,
      "epoch": 2.2518544683857296,
      "grad_norm": 1.203125,
      "learning_rate": 0.000449132196743233,
      "loss": 5.2998,
      "mean_token_accuracy": 0.21837954074144364,
      "num_tokens": 47541336.0,
      "step": 25500
    },
    {
      "entropy": 5.864141893386841,
      "epoch": 2.2522960084775696,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004491122520240621,
      "loss": 5.2223,
      "mean_token_accuracy": 0.21477721333503724,
      "num_tokens": 47551446.0,
      "step": 25505
    },
    {
      "entropy": 5.889760589599609,
      "epoch": 2.25273754856941,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004490923038941312,
      "loss": 5.287,
      "mean_token_accuracy": 0.21306386291980745,
      "num_tokens": 47560596.0,
      "step": 25510
    },
    {
      "entropy": 5.958892393112182,
      "epoch": 2.2531790886612506,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000449072352353831,
      "loss": 5.3854,
      "mean_token_accuracy": 0.20812500715255738,
      "num_tokens": 47570612.0,
      "step": 25515
    },
    {
      "entropy": 5.92105450630188,
      "epoch": 2.2536206287530907,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004490523974035525,
      "loss": 5.3114,
      "mean_token_accuracy": 0.2044833332300186,
      "num_tokens": 47580234.0,
      "step": 25520
    },
    {
      "entropy": 5.953958892822266,
      "epoch": 2.254062168844931,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004490324390436866,
      "loss": 5.3402,
      "mean_token_accuracy": 0.2086621806025505,
      "num_tokens": 47589777.0,
      "step": 25525
    },
    {
      "entropy": 5.9193359375,
      "epoch": 2.254503708936771,
      "grad_norm": 1.21875,
      "learning_rate": 0.000449012477274624,
      "loss": 5.2469,
      "mean_token_accuracy": 0.21461936533451081,
      "num_tokens": 47597951.0,
      "step": 25530
    },
    {
      "entropy": 5.904752635955811,
      "epoch": 2.2549452490286117,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044899251209675606,
      "loss": 5.352,
      "mean_token_accuracy": 0.20868321657180786,
      "num_tokens": 47606054.0,
      "step": 25535
    },
    {
      "entropy": 5.992991876602173,
      "epoch": 2.255386789120452,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004489725435104737,
      "loss": 5.3342,
      "mean_token_accuracy": 0.21043829768896102,
      "num_tokens": 47614704.0,
      "step": 25540
    },
    {
      "entropy": 5.888143825531006,
      "epoch": 2.2558283292122923,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004489525715161682,
      "loss": 5.3159,
      "mean_token_accuracy": 0.2094903200864792,
      "num_tokens": 47623189.0,
      "step": 25545
    },
    {
      "entropy": 5.8683079242706295,
      "epoch": 2.2562698693041328,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044893259611423076,
      "loss": 5.281,
      "mean_token_accuracy": 0.20954559892416,
      "num_tokens": 47633831.0,
      "step": 25550
    },
    {
      "entropy": 5.979770135879517,
      "epoch": 2.2567114093959733,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004489126173050527,
      "loss": 5.2176,
      "mean_token_accuracy": 0.21380435079336166,
      "num_tokens": 47642155.0,
      "step": 25555
    },
    {
      "entropy": 6.064649438858032,
      "epoch": 2.2571529494878133,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004488926350890253,
      "loss": 5.4504,
      "mean_token_accuracy": 0.19683613181114196,
      "num_tokens": 47651042.0,
      "step": 25560
    },
    {
      "entropy": 5.95261516571045,
      "epoch": 2.257594489579654,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004488726494665402,
      "loss": 5.3637,
      "mean_token_accuracy": 0.20659175515174866,
      "num_tokens": 47661424.0,
      "step": 25565
    },
    {
      "entropy": 5.918775033950806,
      "epoch": 2.2580360296714943,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044885266043798866,
      "loss": 5.3001,
      "mean_token_accuracy": 0.20935991108417512,
      "num_tokens": 47671603.0,
      "step": 25570
    },
    {
      "entropy": 5.903744792938232,
      "epoch": 2.2584775697633344,
      "grad_norm": 1.25,
      "learning_rate": 0.0004488326680037624,
      "loss": 5.301,
      "mean_token_accuracy": 0.20374856442213057,
      "num_tokens": 47679861.0,
      "step": 25575
    },
    {
      "entropy": 5.78634729385376,
      "epoch": 2.258919109855175,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004488126721642531,
      "loss": 5.195,
      "mean_token_accuracy": 0.21474214494228364,
      "num_tokens": 47689702.0,
      "step": 25580
    },
    {
      "entropy": 5.812030076980591,
      "epoch": 2.2593606499470154,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044879267291985243,
      "loss": 5.2884,
      "mean_token_accuracy": 0.2032901331782341,
      "num_tokens": 47699261.0,
      "step": 25585
    },
    {
      "entropy": 5.962048864364624,
      "epoch": 2.2598021900388554,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044877267027095203,
      "loss": 5.3635,
      "mean_token_accuracy": 0.20693973153829576,
      "num_tokens": 47709079.0,
      "step": 25590
    },
    {
      "entropy": 5.875109338760376,
      "epoch": 2.260243730130696,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004487526642179439,
      "loss": 5.2452,
      "mean_token_accuracy": 0.21441958844661713,
      "num_tokens": 47718641.0,
      "step": 25595
    },
    {
      "entropy": 5.996947765350342,
      "epoch": 2.2606852702225364,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044873265476121993,
      "loss": 5.3722,
      "mean_token_accuracy": 0.20505607426166533,
      "num_tokens": 47729156.0,
      "step": 25600
    },
    {
      "entropy": 5.91913628578186,
      "epoch": 2.2611268103143765,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044871264190117196,
      "loss": 5.3121,
      "mean_token_accuracy": 0.20826969146728516,
      "num_tokens": 47738424.0,
      "step": 25605
    },
    {
      "entropy": 5.9685125827789305,
      "epoch": 2.261568350406217,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004486926256381922,
      "loss": 5.4466,
      "mean_token_accuracy": 0.19461868405342103,
      "num_tokens": 47749032.0,
      "step": 25610
    },
    {
      "entropy": 5.977568626403809,
      "epoch": 2.2620098904980575,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004486726059726727,
      "loss": 5.3256,
      "mean_token_accuracy": 0.2057436317205429,
      "num_tokens": 47758979.0,
      "step": 25615
    },
    {
      "entropy": 5.892540216445923,
      "epoch": 2.2624514305898975,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004486525829050056,
      "loss": 5.3005,
      "mean_token_accuracy": 0.20875338912010194,
      "num_tokens": 47768752.0,
      "step": 25620
    },
    {
      "entropy": 5.929640245437622,
      "epoch": 2.262892970681738,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004486325564355831,
      "loss": 5.3552,
      "mean_token_accuracy": 0.20305759459733963,
      "num_tokens": 47778175.0,
      "step": 25625
    },
    {
      "entropy": 5.866891765594483,
      "epoch": 2.263334510773578,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004486125265647977,
      "loss": 5.2947,
      "mean_token_accuracy": 0.21630572229623796,
      "num_tokens": 47787864.0,
      "step": 25630
    },
    {
      "entropy": 5.955856084823608,
      "epoch": 2.2637760508654186,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004485924932930416,
      "loss": 5.3574,
      "mean_token_accuracy": 0.19902995526790618,
      "num_tokens": 47796912.0,
      "step": 25635
    },
    {
      "entropy": 5.969979858398437,
      "epoch": 2.264217590957259,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004485724566207073,
      "loss": 5.3896,
      "mean_token_accuracy": 0.2024100422859192,
      "num_tokens": 47807452.0,
      "step": 25640
    },
    {
      "entropy": 6.00482211112976,
      "epoch": 2.264659131049099,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044855241654818723,
      "loss": 5.4097,
      "mean_token_accuracy": 0.20293576568365096,
      "num_tokens": 47817575.0,
      "step": 25645
    },
    {
      "entropy": 5.956986618041992,
      "epoch": 2.2651006711409396,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004485323730758741,
      "loss": 5.3141,
      "mean_token_accuracy": 0.21868909746408463,
      "num_tokens": 47826177.0,
      "step": 25650
    },
    {
      "entropy": 5.873044681549072,
      "epoch": 2.26554221123278,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004485123262041604,
      "loss": 5.2547,
      "mean_token_accuracy": 0.20949017405509948,
      "num_tokens": 47835622.0,
      "step": 25655
    },
    {
      "entropy": 5.871547794342041,
      "epoch": 2.26598375132462,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004484922759334389,
      "loss": 5.2971,
      "mean_token_accuracy": 0.21253005564212799,
      "num_tokens": 47845552.0,
      "step": 25660
    },
    {
      "entropy": 5.9136628150939945,
      "epoch": 2.2664252914164607,
      "grad_norm": 1.25,
      "learning_rate": 0.0004484722222641024,
      "loss": 5.2477,
      "mean_token_accuracy": 0.20884334594011306,
      "num_tokens": 47853820.0,
      "step": 25665
    },
    {
      "entropy": 5.895097112655639,
      "epoch": 2.2668668315083007,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004484521651965438,
      "loss": 5.2528,
      "mean_token_accuracy": 0.21776512116193772,
      "num_tokens": 47863416.0,
      "step": 25670
    },
    {
      "entropy": 5.919981908798218,
      "epoch": 2.267308371600141,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044843210473115584,
      "loss": 5.3251,
      "mean_token_accuracy": 0.21240510940551757,
      "num_tokens": 47871407.0,
      "step": 25675
    },
    {
      "entropy": 5.927334308624268,
      "epoch": 2.2677499116919817,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004484120408683316,
      "loss": 5.276,
      "mean_token_accuracy": 0.20964734852313996,
      "num_tokens": 47880742.0,
      "step": 25680
    },
    {
      "entropy": 5.8885444641113285,
      "epoch": 2.2681914517838218,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044839197360846416,
      "loss": 5.3543,
      "mean_token_accuracy": 0.20122861117124557,
      "num_tokens": 47889573.0,
      "step": 25685
    },
    {
      "entropy": 5.901750230789185,
      "epoch": 2.2686329918756623,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044837190295194654,
      "loss": 5.2868,
      "mean_token_accuracy": 0.21235938370227814,
      "num_tokens": 47898855.0,
      "step": 25690
    },
    {
      "entropy": 5.921929597854614,
      "epoch": 2.2690745319675028,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044835182889917195,
      "loss": 5.3224,
      "mean_token_accuracy": 0.20815297216176987,
      "num_tokens": 47908981.0,
      "step": 25695
    },
    {
      "entropy": 5.908852291107178,
      "epoch": 2.269516072059343,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004483317514505336,
      "loss": 5.3144,
      "mean_token_accuracy": 0.2115495041012764,
      "num_tokens": 47918313.0,
      "step": 25700
    },
    {
      "entropy": 5.861187314987182,
      "epoch": 2.2699576121511833,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004483116706064248,
      "loss": 5.2471,
      "mean_token_accuracy": 0.21142636239528656,
      "num_tokens": 47927838.0,
      "step": 25705
    },
    {
      "entropy": 5.905887460708618,
      "epoch": 2.270399152243024,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004482915863672389,
      "loss": 5.2902,
      "mean_token_accuracy": 0.20914349853992462,
      "num_tokens": 47937774.0,
      "step": 25710
    },
    {
      "entropy": 5.943370914459228,
      "epoch": 2.270840692334864,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004482714987333695,
      "loss": 5.2514,
      "mean_token_accuracy": 0.2146176055073738,
      "num_tokens": 47947091.0,
      "step": 25715
    },
    {
      "entropy": 5.8821768283844,
      "epoch": 2.2712822324267044,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044825140770520997,
      "loss": 5.2402,
      "mean_token_accuracy": 0.21849192529916764,
      "num_tokens": 47955715.0,
      "step": 25720
    },
    {
      "entropy": 5.741145372390747,
      "epoch": 2.271723772518545,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044823131328315387,
      "loss": 5.1576,
      "mean_token_accuracy": 0.23379732221364974,
      "num_tokens": 47965991.0,
      "step": 25725
    },
    {
      "entropy": 5.924877166748047,
      "epoch": 2.272165312610385,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004482112154675948,
      "loss": 5.3498,
      "mean_token_accuracy": 0.20598774254322053,
      "num_tokens": 47975610.0,
      "step": 25730
    },
    {
      "entropy": 5.876081752777099,
      "epoch": 2.2726068527022254,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044819111425892667,
      "loss": 5.2691,
      "mean_token_accuracy": 0.21684676706790923,
      "num_tokens": 47984977.0,
      "step": 25735
    },
    {
      "entropy": 5.863063144683838,
      "epoch": 2.273048392794066,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044817100965754297,
      "loss": 5.2424,
      "mean_token_accuracy": 0.21109579503536224,
      "num_tokens": 47994205.0,
      "step": 25740
    },
    {
      "entropy": 5.877716207504273,
      "epoch": 2.273489932885906,
      "grad_norm": 1.125,
      "learning_rate": 0.00044815090166383775,
      "loss": 5.3187,
      "mean_token_accuracy": 0.21038877815008164,
      "num_tokens": 48003715.0,
      "step": 25745
    },
    {
      "entropy": 5.9276769161224365,
      "epoch": 2.2739314729777464,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004481307902782048,
      "loss": 5.4015,
      "mean_token_accuracy": 0.20905213952064514,
      "num_tokens": 48013200.0,
      "step": 25750
    },
    {
      "entropy": 5.955729198455811,
      "epoch": 2.274373013069587,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004481106755010382,
      "loss": 5.3283,
      "mean_token_accuracy": 0.20823585689067842,
      "num_tokens": 48022366.0,
      "step": 25755
    },
    {
      "entropy": 5.931550931930542,
      "epoch": 2.274814553161427,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044809055733273183,
      "loss": 5.347,
      "mean_token_accuracy": 0.20925108939409257,
      "num_tokens": 48031968.0,
      "step": 25760
    },
    {
      "entropy": 5.894931507110596,
      "epoch": 2.2752560932532675,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044807043577367996,
      "loss": 5.2863,
      "mean_token_accuracy": 0.2153531491756439,
      "num_tokens": 48041069.0,
      "step": 25765
    },
    {
      "entropy": 5.795672035217285,
      "epoch": 2.2756976333451076,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044805031082427666,
      "loss": 5.1953,
      "mean_token_accuracy": 0.21762861013412477,
      "num_tokens": 48050550.0,
      "step": 25770
    },
    {
      "entropy": 5.922796010971069,
      "epoch": 2.276139173436948,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044803018248491615,
      "loss": 5.2726,
      "mean_token_accuracy": 0.20829552114009858,
      "num_tokens": 48059971.0,
      "step": 25775
    },
    {
      "entropy": 5.890300512313843,
      "epoch": 2.2765807135287885,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004480100507559928,
      "loss": 5.1586,
      "mean_token_accuracy": 0.21602099090814592,
      "num_tokens": 48068566.0,
      "step": 25780
    },
    {
      "entropy": 5.7922278881073,
      "epoch": 2.2770222536206286,
      "grad_norm": 1.0625,
      "learning_rate": 0.000447989915637901,
      "loss": 5.2006,
      "mean_token_accuracy": 0.21922920793294906,
      "num_tokens": 48078352.0,
      "step": 25785
    },
    {
      "entropy": 5.849106931686402,
      "epoch": 2.277463793712469,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004479697771310351,
      "loss": 5.3018,
      "mean_token_accuracy": 0.2083287462592125,
      "num_tokens": 48087089.0,
      "step": 25790
    },
    {
      "entropy": 5.9186639308929445,
      "epoch": 2.2779053338043096,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044794963523578965,
      "loss": 5.3621,
      "mean_token_accuracy": 0.20507449954748153,
      "num_tokens": 48096762.0,
      "step": 25795
    },
    {
      "entropy": 5.919962739944458,
      "epoch": 2.2783468738961496,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044792948995255917,
      "loss": 5.2812,
      "mean_token_accuracy": 0.21261240243911744,
      "num_tokens": 48105994.0,
      "step": 25800
    },
    {
      "entropy": 5.888382148742676,
      "epoch": 2.27878841398799,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004479093412817383,
      "loss": 5.2587,
      "mean_token_accuracy": 0.21734432876110077,
      "num_tokens": 48115148.0,
      "step": 25805
    },
    {
      "entropy": 5.841091680526733,
      "epoch": 2.27922995407983,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044788918922372196,
      "loss": 5.3449,
      "mean_token_accuracy": 0.20284795761108398,
      "num_tokens": 48124696.0,
      "step": 25810
    },
    {
      "entropy": 5.868496656417847,
      "epoch": 2.2796714941716707,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004478690337789046,
      "loss": 5.2343,
      "mean_token_accuracy": 0.21560890972614288,
      "num_tokens": 48134165.0,
      "step": 25815
    },
    {
      "entropy": 5.968175220489502,
      "epoch": 2.280113034263511,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044784887494768113,
      "loss": 5.2773,
      "mean_token_accuracy": 0.20464987605810164,
      "num_tokens": 48144080.0,
      "step": 25820
    },
    {
      "entropy": 5.8811687469482425,
      "epoch": 2.2805545743553512,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044782871273044654,
      "loss": 5.3186,
      "mean_token_accuracy": 0.21887168139219285,
      "num_tokens": 48156161.0,
      "step": 25825
    },
    {
      "entropy": 5.89983115196228,
      "epoch": 2.2809961144471917,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004478085471275958,
      "loss": 5.3069,
      "mean_token_accuracy": 0.20691936910152436,
      "num_tokens": 48165317.0,
      "step": 25830
    },
    {
      "entropy": 5.8984099388122555,
      "epoch": 2.2814376545390322,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004477883781395239,
      "loss": 5.2647,
      "mean_token_accuracy": 0.21246760487556457,
      "num_tokens": 48173639.0,
      "step": 25835
    },
    {
      "entropy": 5.8816565990448,
      "epoch": 2.2818791946308723,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044776820576662593,
      "loss": 5.283,
      "mean_token_accuracy": 0.21220593750476838,
      "num_tokens": 48183596.0,
      "step": 25840
    },
    {
      "entropy": 5.985248136520386,
      "epoch": 2.282320734722713,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004477480300092971,
      "loss": 5.3315,
      "mean_token_accuracy": 0.21500713378190994,
      "num_tokens": 48194269.0,
      "step": 25845
    },
    {
      "entropy": 5.915961074829101,
      "epoch": 2.2827622748145533,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044772785086793257,
      "loss": 5.3138,
      "mean_token_accuracy": 0.20815593749284744,
      "num_tokens": 48203180.0,
      "step": 25850
    },
    {
      "entropy": 5.898174190521241,
      "epoch": 2.2832038149063933,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044770766834292765,
      "loss": 5.3275,
      "mean_token_accuracy": 0.21373473405838012,
      "num_tokens": 48212768.0,
      "step": 25855
    },
    {
      "entropy": 5.915253353118897,
      "epoch": 2.283645354998234,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004476874824346777,
      "loss": 5.3488,
      "mean_token_accuracy": 0.20545073449611664,
      "num_tokens": 48221620.0,
      "step": 25860
    },
    {
      "entropy": 5.917410564422608,
      "epoch": 2.2840868950900743,
      "grad_norm": 1.125,
      "learning_rate": 0.00044766729314357827,
      "loss": 5.3288,
      "mean_token_accuracy": 0.20083833932876588,
      "num_tokens": 48231379.0,
      "step": 25865
    },
    {
      "entropy": 5.9638396263122555,
      "epoch": 2.2845284351819144,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004476471004700248,
      "loss": 5.3471,
      "mean_token_accuracy": 0.20801860839128494,
      "num_tokens": 48241827.0,
      "step": 25870
    },
    {
      "entropy": 5.978330278396607,
      "epoch": 2.284969975273755,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004476269044144127,
      "loss": 5.3073,
      "mean_token_accuracy": 0.206272092461586,
      "num_tokens": 48250747.0,
      "step": 25875
    },
    {
      "entropy": 5.931081247329712,
      "epoch": 2.2854115153655954,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004476067049771378,
      "loss": 5.2985,
      "mean_token_accuracy": 0.20079461336135865,
      "num_tokens": 48260909.0,
      "step": 25880
    },
    {
      "entropy": 5.9064287662506105,
      "epoch": 2.2858530554574354,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004475865021585957,
      "loss": 5.3647,
      "mean_token_accuracy": 0.19917707443237304,
      "num_tokens": 48270974.0,
      "step": 25885
    },
    {
      "entropy": 5.884914541244507,
      "epoch": 2.286294595549276,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004475662959591822,
      "loss": 5.2302,
      "mean_token_accuracy": 0.21621289253234863,
      "num_tokens": 48279466.0,
      "step": 25890
    },
    {
      "entropy": 5.915259218215942,
      "epoch": 2.2867361356411164,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000447546086379293,
      "loss": 5.3452,
      "mean_token_accuracy": 0.20903617590665818,
      "num_tokens": 48289830.0,
      "step": 25895
    },
    {
      "entropy": 5.91605019569397,
      "epoch": 2.2871776757329565,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004475258734193242,
      "loss": 5.3006,
      "mean_token_accuracy": 0.21248300671577453,
      "num_tokens": 48299493.0,
      "step": 25900
    },
    {
      "entropy": 5.925859689712524,
      "epoch": 2.287619215824797,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004475056570796717,
      "loss": 5.3196,
      "mean_token_accuracy": 0.21409946382045747,
      "num_tokens": 48308366.0,
      "step": 25905
    },
    {
      "entropy": 5.910337591171265,
      "epoch": 2.288060755916637,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004474854373607314,
      "loss": 5.2834,
      "mean_token_accuracy": 0.20431629717350006,
      "num_tokens": 48317760.0,
      "step": 25910
    },
    {
      "entropy": 5.890132856369019,
      "epoch": 2.2885022960084775,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004474652142628995,
      "loss": 5.3346,
      "mean_token_accuracy": 0.20410245954990386,
      "num_tokens": 48325847.0,
      "step": 25915
    },
    {
      "entropy": 5.9430078029632565,
      "epoch": 2.288943836100318,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004474449877865722,
      "loss": 5.3304,
      "mean_token_accuracy": 0.2001043140888214,
      "num_tokens": 48334938.0,
      "step": 25920
    },
    {
      "entropy": 5.903149366378784,
      "epoch": 2.289385376192158,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004474247579321456,
      "loss": 5.3239,
      "mean_token_accuracy": 0.2027680292725563,
      "num_tokens": 48344414.0,
      "step": 25925
    },
    {
      "entropy": 5.902489852905274,
      "epoch": 2.2898269162839986,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000447404524700016,
      "loss": 5.2523,
      "mean_token_accuracy": 0.20890749841928483,
      "num_tokens": 48353711.0,
      "step": 25930
    },
    {
      "entropy": 5.819829511642456,
      "epoch": 2.290268456375839,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004473842880905799,
      "loss": 5.2066,
      "mean_token_accuracy": 0.22657289505004882,
      "num_tokens": 48362907.0,
      "step": 25935
    },
    {
      "entropy": 5.915345096588135,
      "epoch": 2.290709996467679,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044736404810423366,
      "loss": 5.2713,
      "mean_token_accuracy": 0.21852019280195237,
      "num_tokens": 48372450.0,
      "step": 25940
    },
    {
      "entropy": 5.842714166641235,
      "epoch": 2.2911515365595196,
      "grad_norm": 1.125,
      "learning_rate": 0.0004473438047413737,
      "loss": 5.2266,
      "mean_token_accuracy": 0.2167292207479477,
      "num_tokens": 48381521.0,
      "step": 25945
    },
    {
      "entropy": 5.887005186080932,
      "epoch": 2.2915930766513597,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004473235580023966,
      "loss": 5.2517,
      "mean_token_accuracy": 0.21939159333705902,
      "num_tokens": 48390635.0,
      "step": 25950
    },
    {
      "entropy": 5.887283658981323,
      "epoch": 2.2920346167432,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044730330788769903,
      "loss": 5.3246,
      "mean_token_accuracy": 0.20345648676156997,
      "num_tokens": 48400199.0,
      "step": 25955
    },
    {
      "entropy": 5.862120914459228,
      "epoch": 2.2924761568350407,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044728305439767764,
      "loss": 5.289,
      "mean_token_accuracy": 0.21052179336547852,
      "num_tokens": 48409266.0,
      "step": 25960
    },
    {
      "entropy": 5.927556467056275,
      "epoch": 2.2929176969268807,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004472627975327293,
      "loss": 5.3299,
      "mean_token_accuracy": 0.20798514932394027,
      "num_tokens": 48419195.0,
      "step": 25965
    },
    {
      "entropy": 5.97014627456665,
      "epoch": 2.2933592370187212,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004472425372932506,
      "loss": 5.3744,
      "mean_token_accuracy": 0.205880506336689,
      "num_tokens": 48430637.0,
      "step": 25970
    },
    {
      "entropy": 5.794405126571656,
      "epoch": 2.2938007771105617,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004472222736796386,
      "loss": 5.1628,
      "mean_token_accuracy": 0.21999103128910064,
      "num_tokens": 48439929.0,
      "step": 25975
    },
    {
      "entropy": 5.959406042098999,
      "epoch": 2.294242317202402,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004472020066922902,
      "loss": 5.4216,
      "mean_token_accuracy": 0.20152697712183,
      "num_tokens": 48449398.0,
      "step": 25980
    },
    {
      "entropy": 5.979047536849976,
      "epoch": 2.2946838572942423,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004471817363316024,
      "loss": 5.2966,
      "mean_token_accuracy": 0.21314818859100343,
      "num_tokens": 48458843.0,
      "step": 25985
    },
    {
      "entropy": 5.9526755809783936,
      "epoch": 2.295125397386083,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004471614625979723,
      "loss": 5.2409,
      "mean_token_accuracy": 0.21272154301404952,
      "num_tokens": 48468111.0,
      "step": 25990
    },
    {
      "entropy": 5.882488059997558,
      "epoch": 2.295566937477923,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004471411854917972,
      "loss": 5.3028,
      "mean_token_accuracy": 0.21438165307044982,
      "num_tokens": 48477501.0,
      "step": 25995
    },
    {
      "entropy": 6.024596929550171,
      "epoch": 2.2960084775697633,
      "grad_norm": 1.09375,
      "learning_rate": 0.000447120905013474,
      "loss": 5.44,
      "mean_token_accuracy": 0.19502519816160202,
      "num_tokens": 48487576.0,
      "step": 26000
    },
    {
      "entropy": 5.89276614189148,
      "epoch": 2.296450017661604,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004471006211634002,
      "loss": 5.2236,
      "mean_token_accuracy": 0.22616962045431138,
      "num_tokens": 48496230.0,
      "step": 26005
    },
    {
      "entropy": 5.959626388549805,
      "epoch": 2.296891557753444,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044708033394197323,
      "loss": 5.3099,
      "mean_token_accuracy": 0.20844869464635848,
      "num_tokens": 48505871.0,
      "step": 26010
    },
    {
      "entropy": 5.901160430908203,
      "epoch": 2.2973330978452844,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044706004334959024,
      "loss": 5.3114,
      "mean_token_accuracy": 0.2045974016189575,
      "num_tokens": 48515298.0,
      "step": 26015
    },
    {
      "entropy": 5.828893661499023,
      "epoch": 2.297774637937125,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004470397493866489,
      "loss": 5.3616,
      "mean_token_accuracy": 0.20081913471221924,
      "num_tokens": 48524573.0,
      "step": 26020
    },
    {
      "entropy": 5.954742431640625,
      "epoch": 2.298216178028965,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004470194520535467,
      "loss": 5.3858,
      "mean_token_accuracy": 0.20398117303848268,
      "num_tokens": 48533244.0,
      "step": 26025
    },
    {
      "entropy": 5.976409244537353,
      "epoch": 2.2986577181208054,
      "grad_norm": 1.0,
      "learning_rate": 0.0004469991513506814,
      "loss": 5.2495,
      "mean_token_accuracy": 0.2126106098294258,
      "num_tokens": 48542464.0,
      "step": 26030
    },
    {
      "entropy": 5.865573120117188,
      "epoch": 2.299099258212646,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004469788472784504,
      "loss": 5.3244,
      "mean_token_accuracy": 0.21156174093484878,
      "num_tokens": 48551947.0,
      "step": 26035
    },
    {
      "entropy": 5.892144346237183,
      "epoch": 2.299540798304486,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044695853983725163,
      "loss": 5.2394,
      "mean_token_accuracy": 0.2179814860224724,
      "num_tokens": 48561287.0,
      "step": 26040
    },
    {
      "entropy": 5.940769100189209,
      "epoch": 2.2999823383963265,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004469382290274829,
      "loss": 5.3803,
      "mean_token_accuracy": 0.20299497991800308,
      "num_tokens": 48570800.0,
      "step": 26045
    },
    {
      "entropy": 5.924418830871582,
      "epoch": 2.3004238784881665,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044691791484954195,
      "loss": 5.2126,
      "mean_token_accuracy": 0.2202403575181961,
      "num_tokens": 48579705.0,
      "step": 26050
    },
    {
      "entropy": 5.889727973937989,
      "epoch": 2.300865418580007,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000446897597303827,
      "loss": 5.3412,
      "mean_token_accuracy": 0.2094974100589752,
      "num_tokens": 48588793.0,
      "step": 26055
    },
    {
      "entropy": 5.894078111648559,
      "epoch": 2.3013069586718475,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004468772763907357,
      "loss": 5.188,
      "mean_token_accuracy": 0.21566384434700012,
      "num_tokens": 48598361.0,
      "step": 26060
    },
    {
      "entropy": 5.9139727592468265,
      "epoch": 2.3017484987636876,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004468569521106664,
      "loss": 5.3145,
      "mean_token_accuracy": 0.20633782893419267,
      "num_tokens": 48607703.0,
      "step": 26065
    },
    {
      "entropy": 5.889852142333984,
      "epoch": 2.302190038855528,
      "grad_norm": 1.125,
      "learning_rate": 0.0004468366244640171,
      "loss": 5.2644,
      "mean_token_accuracy": 0.21064090579748154,
      "num_tokens": 48616345.0,
      "step": 26070
    },
    {
      "entropy": 5.980006980895996,
      "epoch": 2.3026315789473686,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044681629345118615,
      "loss": 5.3295,
      "mean_token_accuracy": 0.2057981863617897,
      "num_tokens": 48624907.0,
      "step": 26075
    },
    {
      "entropy": 5.897442245483399,
      "epoch": 2.3030731190392086,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004467959590725716,
      "loss": 5.262,
      "mean_token_accuracy": 0.21115820109844208,
      "num_tokens": 48634766.0,
      "step": 26080
    },
    {
      "entropy": 5.91874737739563,
      "epoch": 2.303514659131049,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044677562132857204,
      "loss": 5.3042,
      "mean_token_accuracy": 0.20883141010999678,
      "num_tokens": 48644321.0,
      "step": 26085
    },
    {
      "entropy": 5.8903004169464115,
      "epoch": 2.303956199222889,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044675528021958567,
      "loss": 5.312,
      "mean_token_accuracy": 0.21108170002698898,
      "num_tokens": 48653507.0,
      "step": 26090
    },
    {
      "entropy": 5.924942922592163,
      "epoch": 2.3043977393147297,
      "grad_norm": 1.078125,
      "learning_rate": 0.000446734935746011,
      "loss": 5.3224,
      "mean_token_accuracy": 0.2184051901102066,
      "num_tokens": 48662186.0,
      "step": 26095
    },
    {
      "entropy": 5.869546031951904,
      "epoch": 2.30483927940657,
      "grad_norm": 1.25,
      "learning_rate": 0.0004467145879082467,
      "loss": 5.2877,
      "mean_token_accuracy": 0.21554356813430786,
      "num_tokens": 48671605.0,
      "step": 26100
    },
    {
      "entropy": 5.82264199256897,
      "epoch": 2.3052808194984102,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004466942367066912,
      "loss": 5.2211,
      "mean_token_accuracy": 0.21485772579908372,
      "num_tokens": 48679873.0,
      "step": 26105
    },
    {
      "entropy": 5.875986337661743,
      "epoch": 2.3057223595902507,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044667388214174336,
      "loss": 5.2074,
      "mean_token_accuracy": 0.2163578301668167,
      "num_tokens": 48688678.0,
      "step": 26110
    },
    {
      "entropy": 5.903117370605469,
      "epoch": 2.3061638996820912,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004466535242138018,
      "loss": 5.3039,
      "mean_token_accuracy": 0.2125939905643463,
      "num_tokens": 48697794.0,
      "step": 26115
    },
    {
      "entropy": 5.9491369247436525,
      "epoch": 2.3066054397739313,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044663316292326526,
      "loss": 5.4288,
      "mean_token_accuracy": 0.2034354031085968,
      "num_tokens": 48706859.0,
      "step": 26120
    },
    {
      "entropy": 5.871827459335327,
      "epoch": 2.307046979865772,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044661279827053267,
      "loss": 5.2511,
      "mean_token_accuracy": 0.22039943486452102,
      "num_tokens": 48716627.0,
      "step": 26125
    },
    {
      "entropy": 5.94290862083435,
      "epoch": 2.3074885199576123,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044659243025600297,
      "loss": 5.3204,
      "mean_token_accuracy": 0.2023373529314995,
      "num_tokens": 48726514.0,
      "step": 26130
    },
    {
      "entropy": 5.884207439422608,
      "epoch": 2.3079300600494523,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044657205888007525,
      "loss": 5.2866,
      "mean_token_accuracy": 0.21925854682922363,
      "num_tokens": 48735546.0,
      "step": 26135
    },
    {
      "entropy": 5.79543981552124,
      "epoch": 2.308371600141293,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004465516841431483,
      "loss": 5.2103,
      "mean_token_accuracy": 0.21801828145980834,
      "num_tokens": 48744719.0,
      "step": 26140
    },
    {
      "entropy": 5.971226692199707,
      "epoch": 2.3088131402331333,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004465313060456216,
      "loss": 5.2802,
      "mean_token_accuracy": 0.21436866670846938,
      "num_tokens": 48754201.0,
      "step": 26145
    },
    {
      "entropy": 5.951769828796387,
      "epoch": 2.3092546803249734,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044651092458789405,
      "loss": 5.3033,
      "mean_token_accuracy": 0.2043192520737648,
      "num_tokens": 48763552.0,
      "step": 26150
    },
    {
      "entropy": 5.84910364151001,
      "epoch": 2.309696220416814,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004464905397703651,
      "loss": 5.2791,
      "mean_token_accuracy": 0.2093621864914894,
      "num_tokens": 48772494.0,
      "step": 26155
    },
    {
      "entropy": 5.858048295974731,
      "epoch": 2.3101377605086544,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044647015159343396,
      "loss": 5.2995,
      "mean_token_accuracy": 0.20694191455841066,
      "num_tokens": 48782121.0,
      "step": 26160
    },
    {
      "entropy": 5.913721656799316,
      "epoch": 2.3105793006004944,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044644976005750003,
      "loss": 5.2368,
      "mean_token_accuracy": 0.2198876455426216,
      "num_tokens": 48792040.0,
      "step": 26165
    },
    {
      "entropy": 5.891841840744019,
      "epoch": 2.311020840692335,
      "grad_norm": 1.125,
      "learning_rate": 0.0004464293651629629,
      "loss": 5.261,
      "mean_token_accuracy": 0.2119569733738899,
      "num_tokens": 48801429.0,
      "step": 26170
    },
    {
      "entropy": 5.940471649169922,
      "epoch": 2.3114623807841754,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044640896691022195,
      "loss": 5.3195,
      "mean_token_accuracy": 0.2093986988067627,
      "num_tokens": 48811185.0,
      "step": 26175
    },
    {
      "entropy": 5.980248498916626,
      "epoch": 2.3119039208760155,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004463885652996768,
      "loss": 5.3611,
      "mean_token_accuracy": 0.2037942513823509,
      "num_tokens": 48820691.0,
      "step": 26180
    },
    {
      "entropy": 5.964035177230835,
      "epoch": 2.312345460967856,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044636816033172716,
      "loss": 5.3715,
      "mean_token_accuracy": 0.20366516411304475,
      "num_tokens": 48829960.0,
      "step": 26185
    },
    {
      "entropy": 5.895146179199219,
      "epoch": 2.312787001059696,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004463477520067728,
      "loss": 5.2499,
      "mean_token_accuracy": 0.21204532384872438,
      "num_tokens": 48839107.0,
      "step": 26190
    },
    {
      "entropy": 5.9575910568237305,
      "epoch": 2.3132285411515365,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004463273403252133,
      "loss": 5.3508,
      "mean_token_accuracy": 0.2099679559469223,
      "num_tokens": 48848430.0,
      "step": 26195
    },
    {
      "entropy": 5.867958211898804,
      "epoch": 2.313670081243377,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004463069252874487,
      "loss": 5.2424,
      "mean_token_accuracy": 0.21630232334136962,
      "num_tokens": 48857869.0,
      "step": 26200
    },
    {
      "entropy": 5.959478378295898,
      "epoch": 2.314111621335217,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044628650689387887,
      "loss": 5.3443,
      "mean_token_accuracy": 0.20582959800958633,
      "num_tokens": 48867730.0,
      "step": 26205
    },
    {
      "entropy": 5.962307596206665,
      "epoch": 2.3145531614270576,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00044626608514490384,
      "loss": 5.3177,
      "mean_token_accuracy": 0.20978791266679764,
      "num_tokens": 48876615.0,
      "step": 26210
    },
    {
      "entropy": 5.934621667861938,
      "epoch": 2.314994701518898,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004462456600409235,
      "loss": 5.3276,
      "mean_token_accuracy": 0.21311743706464767,
      "num_tokens": 48885732.0,
      "step": 26215
    },
    {
      "entropy": 5.945902109146118,
      "epoch": 2.315436241610738,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004462252315823381,
      "loss": 5.2594,
      "mean_token_accuracy": 0.2124856159090996,
      "num_tokens": 48894372.0,
      "step": 26220
    },
    {
      "entropy": 5.9715066909790036,
      "epoch": 2.3158777817025786,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004462047997695478,
      "loss": 5.3978,
      "mean_token_accuracy": 0.2053362786769867,
      "num_tokens": 48904321.0,
      "step": 26225
    },
    {
      "entropy": 5.874537229537964,
      "epoch": 2.3163193217944187,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044618436460295294,
      "loss": 5.328,
      "mean_token_accuracy": 0.20437328964471818,
      "num_tokens": 48914215.0,
      "step": 26230
    },
    {
      "entropy": 5.889763832092285,
      "epoch": 2.316760861886259,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004461639260829535,
      "loss": 5.281,
      "mean_token_accuracy": 0.2028007388114929,
      "num_tokens": 48923383.0,
      "step": 26235
    },
    {
      "entropy": 5.960892152786255,
      "epoch": 2.3172024019780997,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004461434842099503,
      "loss": 5.3207,
      "mean_token_accuracy": 0.20639059990644454,
      "num_tokens": 48933560.0,
      "step": 26240
    },
    {
      "entropy": 5.837919044494629,
      "epoch": 2.3176439420699397,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004461230389843435,
      "loss": 5.2407,
      "mean_token_accuracy": 0.21555689424276353,
      "num_tokens": 48943111.0,
      "step": 26245
    },
    {
      "entropy": 5.88907470703125,
      "epoch": 2.3180854821617802,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004461025904065338,
      "loss": 5.2866,
      "mean_token_accuracy": 0.21421474665403367,
      "num_tokens": 48952470.0,
      "step": 26250
    },
    {
      "entropy": 5.950889253616333,
      "epoch": 2.3185270222536207,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044608213847692154,
      "loss": 5.3453,
      "mean_token_accuracy": 0.20455030649900435,
      "num_tokens": 48961273.0,
      "step": 26255
    },
    {
      "entropy": 5.9744789600372314,
      "epoch": 2.318968562345461,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004460616831959075,
      "loss": 5.3756,
      "mean_token_accuracy": 0.20767430067062378,
      "num_tokens": 48971078.0,
      "step": 26260
    },
    {
      "entropy": 5.938797855377198,
      "epoch": 2.3194101024373013,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004460412245638924,
      "loss": 5.3388,
      "mean_token_accuracy": 0.21122367531061173,
      "num_tokens": 48980711.0,
      "step": 26265
    },
    {
      "entropy": 5.935947322845459,
      "epoch": 2.3198516425291418,
      "grad_norm": 1.09375,
      "learning_rate": 0.000446020762581277,
      "loss": 5.2563,
      "mean_token_accuracy": 0.21150058656930923,
      "num_tokens": 48990195.0,
      "step": 26270
    },
    {
      "entropy": 5.8040759563446045,
      "epoch": 2.320293182620982,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00044600029724846206,
      "loss": 5.15,
      "mean_token_accuracy": 0.2249244526028633,
      "num_tokens": 49000271.0,
      "step": 26275
    },
    {
      "entropy": 5.876610994338989,
      "epoch": 2.3207347227128223,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004459798285658487,
      "loss": 5.2848,
      "mean_token_accuracy": 0.2026217296719551,
      "num_tokens": 49009624.0,
      "step": 26280
    },
    {
      "entropy": 5.8926310539245605,
      "epoch": 2.321176262804663,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044595935653383756,
      "loss": 5.2677,
      "mean_token_accuracy": 0.21139924526214598,
      "num_tokens": 49018845.0,
      "step": 26285
    },
    {
      "entropy": 5.843112373352051,
      "epoch": 2.321617802896503,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044593888115282995,
      "loss": 5.229,
      "mean_token_accuracy": 0.2147248864173889,
      "num_tokens": 49028352.0,
      "step": 26290
    },
    {
      "entropy": 5.889179420471192,
      "epoch": 2.3220593429883434,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044591840242322686,
      "loss": 5.3083,
      "mean_token_accuracy": 0.21021538376808166,
      "num_tokens": 49038180.0,
      "step": 26295
    },
    {
      "entropy": 5.894275569915772,
      "epoch": 2.322500883080184,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004458979203454295,
      "loss": 5.2238,
      "mean_token_accuracy": 0.2269373804330826,
      "num_tokens": 49048038.0,
      "step": 26300
    },
    {
      "entropy": 5.911482334136963,
      "epoch": 2.322942423172024,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044587743491983903,
      "loss": 5.2979,
      "mean_token_accuracy": 0.21373720616102218,
      "num_tokens": 49057304.0,
      "step": 26305
    },
    {
      "entropy": 5.915894842147827,
      "epoch": 2.3233839632638644,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004458569461468568,
      "loss": 5.3971,
      "mean_token_accuracy": 0.20674056857824324,
      "num_tokens": 49067071.0,
      "step": 26310
    },
    {
      "entropy": 5.865229368209839,
      "epoch": 2.323825503355705,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044583645402688414,
      "loss": 5.3305,
      "mean_token_accuracy": 0.2091332569718361,
      "num_tokens": 49075814.0,
      "step": 26315
    },
    {
      "entropy": 5.980800628662109,
      "epoch": 2.324267043447545,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004458159585603226,
      "loss": 5.402,
      "mean_token_accuracy": 0.20223182886838914,
      "num_tokens": 49085831.0,
      "step": 26320
    },
    {
      "entropy": 5.985586261749267,
      "epoch": 2.3247085835393855,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044579545974757353,
      "loss": 5.3144,
      "mean_token_accuracy": 0.20553084909915925,
      "num_tokens": 49095893.0,
      "step": 26325
    },
    {
      "entropy": 5.854241371154785,
      "epoch": 2.3251501236312255,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004457749575890385,
      "loss": 5.2765,
      "mean_token_accuracy": 0.2071152910590172,
      "num_tokens": 49104810.0,
      "step": 26330
    },
    {
      "entropy": 5.874606037139893,
      "epoch": 2.325591663723066,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004457544520851192,
      "loss": 5.3381,
      "mean_token_accuracy": 0.20311879068613053,
      "num_tokens": 49114108.0,
      "step": 26335
    },
    {
      "entropy": 5.94635796546936,
      "epoch": 2.3260332038149065,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044573394323621735,
      "loss": 5.2883,
      "mean_token_accuracy": 0.2100869044661522,
      "num_tokens": 49123371.0,
      "step": 26340
    },
    {
      "entropy": 5.985779094696045,
      "epoch": 2.3264747439067466,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004457134310427346,
      "loss": 5.3304,
      "mean_token_accuracy": 0.2075203150510788,
      "num_tokens": 49132056.0,
      "step": 26345
    },
    {
      "entropy": 5.931562328338623,
      "epoch": 2.326916283998587,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004456929155050729,
      "loss": 5.3258,
      "mean_token_accuracy": 0.206938473880291,
      "num_tokens": 49141316.0,
      "step": 26350
    },
    {
      "entropy": 5.8523143291473385,
      "epoch": 2.3273578240904276,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00044567239662363393,
      "loss": 5.2719,
      "mean_token_accuracy": 0.2166082426905632,
      "num_tokens": 49151337.0,
      "step": 26355
    },
    {
      "entropy": 5.958336544036865,
      "epoch": 2.3277993641822676,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004456518743988199,
      "loss": 5.3066,
      "mean_token_accuracy": 0.20598851144313812,
      "num_tokens": 49160970.0,
      "step": 26360
    },
    {
      "entropy": 5.935576772689819,
      "epoch": 2.328240904274108,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004456313488310326,
      "loss": 5.2792,
      "mean_token_accuracy": 0.2192042887210846,
      "num_tokens": 49170727.0,
      "step": 26365
    },
    {
      "entropy": 5.898619794845581,
      "epoch": 2.328682444365948,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044561081992067425,
      "loss": 5.2957,
      "mean_token_accuracy": 0.21167370080947875,
      "num_tokens": 49179485.0,
      "step": 26370
    },
    {
      "entropy": 5.903181552886963,
      "epoch": 2.3291239844577887,
      "grad_norm": 1.078125,
      "learning_rate": 0.000445590287668147,
      "loss": 5.3931,
      "mean_token_accuracy": 0.2014477387070656,
      "num_tokens": 49189339.0,
      "step": 26375
    },
    {
      "entropy": 5.890514945983886,
      "epoch": 2.329565524549629,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044556975207385305,
      "loss": 5.2362,
      "mean_token_accuracy": 0.21826662868261337,
      "num_tokens": 49198387.0,
      "step": 26380
    },
    {
      "entropy": 5.932420825958252,
      "epoch": 2.3300070646414692,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044554921313819465,
      "loss": 5.3376,
      "mean_token_accuracy": 0.2069243609905243,
      "num_tokens": 49207518.0,
      "step": 26385
    },
    {
      "entropy": 5.832262420654297,
      "epoch": 2.3304486047333097,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004455286708615742,
      "loss": 5.2288,
      "mean_token_accuracy": 0.21882445514202117,
      "num_tokens": 49217031.0,
      "step": 26390
    },
    {
      "entropy": 5.973717975616455,
      "epoch": 2.3308901448251502,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000445508125244394,
      "loss": 5.3063,
      "mean_token_accuracy": 0.21102532297372817,
      "num_tokens": 49225500.0,
      "step": 26395
    },
    {
      "entropy": 5.912530755996704,
      "epoch": 2.3313316849169903,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044548757628705654,
      "loss": 5.2322,
      "mean_token_accuracy": 0.2207881659269333,
      "num_tokens": 49235088.0,
      "step": 26400
    },
    {
      "entropy": 5.84417462348938,
      "epoch": 2.3317732250088308,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004454670239899646,
      "loss": 5.3429,
      "mean_token_accuracy": 0.19991041272878646,
      "num_tokens": 49245597.0,
      "step": 26405
    },
    {
      "entropy": 5.827473831176758,
      "epoch": 2.3322147651006713,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044544646835352047,
      "loss": 5.1475,
      "mean_token_accuracy": 0.22434647381305695,
      "num_tokens": 49254284.0,
      "step": 26410
    },
    {
      "entropy": 5.904017591476441,
      "epoch": 2.3326563051925113,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000445425909378127,
      "loss": 5.3749,
      "mean_token_accuracy": 0.20371006727218627,
      "num_tokens": 49264226.0,
      "step": 26415
    },
    {
      "entropy": 5.859605073928833,
      "epoch": 2.333097845284352,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044540534706418697,
      "loss": 5.3265,
      "mean_token_accuracy": 0.20770861357450485,
      "num_tokens": 49274022.0,
      "step": 26420
    },
    {
      "entropy": 5.953948211669922,
      "epoch": 2.3335393853761923,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000445384781412103,
      "loss": 5.3061,
      "mean_token_accuracy": 0.20780505686998368,
      "num_tokens": 49283376.0,
      "step": 26425
    },
    {
      "entropy": 5.95822057723999,
      "epoch": 2.3339809254680324,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000445364212422278,
      "loss": 5.3445,
      "mean_token_accuracy": 0.20842400193214417,
      "num_tokens": 49292994.0,
      "step": 26430
    },
    {
      "entropy": 5.990125274658203,
      "epoch": 2.334422465559873,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004453436400951152,
      "loss": 5.414,
      "mean_token_accuracy": 0.21070096492767335,
      "num_tokens": 49303599.0,
      "step": 26435
    },
    {
      "entropy": 6.002164125442505,
      "epoch": 2.3348640056517134,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004453230644310172,
      "loss": 5.3829,
      "mean_token_accuracy": 0.20825206637382507,
      "num_tokens": 49313397.0,
      "step": 26440
    },
    {
      "entropy": 5.943878889083862,
      "epoch": 2.3353055457435534,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044530248543038725,
      "loss": 5.3043,
      "mean_token_accuracy": 0.21043455004692077,
      "num_tokens": 49323288.0,
      "step": 26445
    },
    {
      "entropy": 5.71579065322876,
      "epoch": 2.335747085835394,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044528190309362853,
      "loss": 5.1433,
      "mean_token_accuracy": 0.2238108515739441,
      "num_tokens": 49332153.0,
      "step": 26450
    },
    {
      "entropy": 5.880008697509766,
      "epoch": 2.3361886259272344,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004452613174211441,
      "loss": 5.3063,
      "mean_token_accuracy": 0.20676835775375366,
      "num_tokens": 49340773.0,
      "step": 26455
    },
    {
      "entropy": 5.866214990615845,
      "epoch": 2.3366301660190745,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044524072841333736,
      "loss": 5.3217,
      "mean_token_accuracy": 0.20998685210943221,
      "num_tokens": 49349863.0,
      "step": 26460
    },
    {
      "entropy": 5.8846845626831055,
      "epoch": 2.337071706110915,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004452201360706115,
      "loss": 5.1621,
      "mean_token_accuracy": 0.22079674452543258,
      "num_tokens": 49358518.0,
      "step": 26465
    },
    {
      "entropy": 5.891619682312012,
      "epoch": 2.337513246202755,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004451995403933701,
      "loss": 5.2496,
      "mean_token_accuracy": 0.2173308476805687,
      "num_tokens": 49367509.0,
      "step": 26470
    },
    {
      "entropy": 5.885852527618408,
      "epoch": 2.3379547862945955,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004451789413820164,
      "loss": 5.3115,
      "mean_token_accuracy": 0.21425034403800963,
      "num_tokens": 49377157.0,
      "step": 26475
    },
    {
      "entropy": 5.943954420089722,
      "epoch": 2.338396326386436,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004451583390369541,
      "loss": 5.3486,
      "mean_token_accuracy": 0.20654962211847305,
      "num_tokens": 49386580.0,
      "step": 26480
    },
    {
      "entropy": 5.85311861038208,
      "epoch": 2.338837866478276,
      "grad_norm": 1.125,
      "learning_rate": 0.0004451377333585866,
      "loss": 5.2346,
      "mean_token_accuracy": 0.21684196591377258,
      "num_tokens": 49395894.0,
      "step": 26485
    },
    {
      "entropy": 5.826545286178589,
      "epoch": 2.3392794065701166,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044511712434731774,
      "loss": 5.2746,
      "mean_token_accuracy": 0.2159070134162903,
      "num_tokens": 49405475.0,
      "step": 26490
    },
    {
      "entropy": 5.9277873039245605,
      "epoch": 2.339720946661957,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004450965120035511,
      "loss": 5.297,
      "mean_token_accuracy": 0.2140595495700836,
      "num_tokens": 49414982.0,
      "step": 26495
    },
    {
      "entropy": 5.861628770828247,
      "epoch": 2.340162486753797,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004450758963276907,
      "loss": 5.2433,
      "mean_token_accuracy": 0.21881699711084365,
      "num_tokens": 49425502.0,
      "step": 26500
    },
    {
      "entropy": 5.864910364151001,
      "epoch": 2.3406040268456376,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044505527732014,
      "loss": 5.2322,
      "mean_token_accuracy": 0.211099337041378,
      "num_tokens": 49435278.0,
      "step": 26505
    },
    {
      "entropy": 5.856877279281616,
      "epoch": 2.3410455669374777,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004450346549813033,
      "loss": 5.2796,
      "mean_token_accuracy": 0.21108997315168382,
      "num_tokens": 49444505.0,
      "step": 26510
    },
    {
      "entropy": 5.871217346191406,
      "epoch": 2.341487107029318,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044501402931158433,
      "loss": 5.2587,
      "mean_token_accuracy": 0.21675732880830764,
      "num_tokens": 49453485.0,
      "step": 26515
    },
    {
      "entropy": 5.868362808227539,
      "epoch": 2.3419286471211587,
      "grad_norm": 1.125,
      "learning_rate": 0.0004449934003113872,
      "loss": 5.223,
      "mean_token_accuracy": 0.21355945467948914,
      "num_tokens": 49462319.0,
      "step": 26520
    },
    {
      "entropy": 5.894816637039185,
      "epoch": 2.3423701872129987,
      "grad_norm": 1.125,
      "learning_rate": 0.0004449727679811161,
      "loss": 5.2638,
      "mean_token_accuracy": 0.21920399963855744,
      "num_tokens": 49470949.0,
      "step": 26525
    },
    {
      "entropy": 5.891136217117309,
      "epoch": 2.342811727304839,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004449521323211751,
      "loss": 5.2676,
      "mean_token_accuracy": 0.21321358233690263,
      "num_tokens": 49480269.0,
      "step": 26530
    },
    {
      "entropy": 5.891949939727783,
      "epoch": 2.3432532673966797,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004449314933319685,
      "loss": 5.276,
      "mean_token_accuracy": 0.21341665536165239,
      "num_tokens": 49489364.0,
      "step": 26535
    },
    {
      "entropy": 5.95717420578003,
      "epoch": 2.3436948074885198,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044491085101390065,
      "loss": 5.2907,
      "mean_token_accuracy": 0.21524744033813475,
      "num_tokens": 49499360.0,
      "step": 26540
    },
    {
      "entropy": 5.87869930267334,
      "epoch": 2.3441363475803603,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004448902053673758,
      "loss": 5.2151,
      "mean_token_accuracy": 0.21302966475486756,
      "num_tokens": 49507770.0,
      "step": 26545
    },
    {
      "entropy": 5.7908093452453615,
      "epoch": 2.3445778876722008,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044486955639279845,
      "loss": 5.3046,
      "mean_token_accuracy": 0.2130023419857025,
      "num_tokens": 49516824.0,
      "step": 26550
    },
    {
      "entropy": 5.848095989227295,
      "epoch": 2.345019427764041,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004448489040905731,
      "loss": 5.288,
      "mean_token_accuracy": 0.2065812900662422,
      "num_tokens": 49525176.0,
      "step": 26555
    },
    {
      "entropy": 5.945406532287597,
      "epoch": 2.3454609678558813,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004448282484611043,
      "loss": 5.3914,
      "mean_token_accuracy": 0.20405150800943375,
      "num_tokens": 49535701.0,
      "step": 26560
    },
    {
      "entropy": 5.916968822479248,
      "epoch": 2.345902507947722,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044480758950479673,
      "loss": 5.3,
      "mean_token_accuracy": 0.2253708213567734,
      "num_tokens": 49545473.0,
      "step": 26565
    },
    {
      "entropy": 5.832232427597046,
      "epoch": 2.346344048039562,
      "grad_norm": 1.125,
      "learning_rate": 0.00044478692722205507,
      "loss": 5.2189,
      "mean_token_accuracy": 0.21153826713562013,
      "num_tokens": 49553955.0,
      "step": 26570
    },
    {
      "entropy": 5.882525444030762,
      "epoch": 2.3467855881314024,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000444766261613284,
      "loss": 5.3385,
      "mean_token_accuracy": 0.20648249983787537,
      "num_tokens": 49563305.0,
      "step": 26575
    },
    {
      "entropy": 5.914910078048706,
      "epoch": 2.347227128223243,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044474559267888847,
      "loss": 5.3204,
      "mean_token_accuracy": 0.2100345715880394,
      "num_tokens": 49571841.0,
      "step": 26580
    },
    {
      "entropy": 5.906492614746094,
      "epoch": 2.347668668315083,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004447249204192733,
      "loss": 5.2841,
      "mean_token_accuracy": 0.2114686354994774,
      "num_tokens": 49582862.0,
      "step": 26585
    },
    {
      "entropy": 5.954437923431397,
      "epoch": 2.3481102084069234,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004447042448348435,
      "loss": 5.344,
      "mean_token_accuracy": 0.20796673893928527,
      "num_tokens": 49592410.0,
      "step": 26590
    },
    {
      "entropy": 5.903286790847778,
      "epoch": 2.348551748498764,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044468356592600405,
      "loss": 5.2978,
      "mean_token_accuracy": 0.217125041782856,
      "num_tokens": 49600582.0,
      "step": 26595
    },
    {
      "entropy": 5.8548342227935795,
      "epoch": 2.348993288590604,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044466288369316006,
      "loss": 5.2796,
      "mean_token_accuracy": 0.21384060382843018,
      "num_tokens": 49610165.0,
      "step": 26600
    },
    {
      "entropy": 5.838822698593139,
      "epoch": 2.3494348286824445,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044464219813671664,
      "loss": 5.2901,
      "mean_token_accuracy": 0.22041625827550887,
      "num_tokens": 49619231.0,
      "step": 26605
    },
    {
      "entropy": 5.875089073181153,
      "epoch": 2.3498763687742845,
      "grad_norm": 1.15625,
      "learning_rate": 0.000444621509257079,
      "loss": 5.1968,
      "mean_token_accuracy": 0.21730110347270964,
      "num_tokens": 49627899.0,
      "step": 26610
    },
    {
      "entropy": 5.93037896156311,
      "epoch": 2.350317908866125,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044460081705465247,
      "loss": 5.3638,
      "mean_token_accuracy": 0.1980777531862259,
      "num_tokens": 49637866.0,
      "step": 26615
    },
    {
      "entropy": 5.964282608032226,
      "epoch": 2.3507594489579655,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044458012152984253,
      "loss": 5.3392,
      "mean_token_accuracy": 0.2110477700829506,
      "num_tokens": 49648273.0,
      "step": 26620
    },
    {
      "entropy": 5.955967378616333,
      "epoch": 2.3512009890498056,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004445594226830544,
      "loss": 5.3139,
      "mean_token_accuracy": 0.20588536113500594,
      "num_tokens": 49657944.0,
      "step": 26625
    },
    {
      "entropy": 5.939783000946045,
      "epoch": 2.351642529141646,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044453872051469345,
      "loss": 5.3178,
      "mean_token_accuracy": 0.21336936950683594,
      "num_tokens": 49667158.0,
      "step": 26630
    },
    {
      "entropy": 5.8636696338653564,
      "epoch": 2.3520840692334866,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044451801502516555,
      "loss": 5.2274,
      "mean_token_accuracy": 0.21531042009592055,
      "num_tokens": 49675535.0,
      "step": 26635
    },
    {
      "entropy": 5.888688945770264,
      "epoch": 2.3525256093253266,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004444973062148761,
      "loss": 5.3121,
      "mean_token_accuracy": 0.20918356031179428,
      "num_tokens": 49684922.0,
      "step": 26640
    },
    {
      "entropy": 5.8577906608581545,
      "epoch": 2.352967149417167,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004444765940842308,
      "loss": 5.2419,
      "mean_token_accuracy": 0.21934637576341628,
      "num_tokens": 49695302.0,
      "step": 26645
    },
    {
      "entropy": 5.891083431243897,
      "epoch": 2.3534086895090076,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004444558786336355,
      "loss": 5.3084,
      "mean_token_accuracy": 0.21281784772872925,
      "num_tokens": 49704668.0,
      "step": 26650
    },
    {
      "entropy": 5.853655529022217,
      "epoch": 2.3538502296008477,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004444351598634958,
      "loss": 5.2457,
      "mean_token_accuracy": 0.21667134761810303,
      "num_tokens": 49713704.0,
      "step": 26655
    },
    {
      "entropy": 5.905678462982178,
      "epoch": 2.354291769692688,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004444144377742178,
      "loss": 5.3267,
      "mean_token_accuracy": 0.2111634835600853,
      "num_tokens": 49723743.0,
      "step": 26660
    },
    {
      "entropy": 5.967500114440918,
      "epoch": 2.354733309784528,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004443937123662071,
      "loss": 5.3906,
      "mean_token_accuracy": 0.20422385782003402,
      "num_tokens": 49734760.0,
      "step": 26665
    },
    {
      "entropy": 5.878413343429566,
      "epoch": 2.3551748498763687,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004443729836398702,
      "loss": 5.2663,
      "mean_token_accuracy": 0.2070953667163849,
      "num_tokens": 49744836.0,
      "step": 26670
    },
    {
      "entropy": 5.8912896633148195,
      "epoch": 2.355616389968209,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044435225159561263,
      "loss": 5.2528,
      "mean_token_accuracy": 0.2180324152112007,
      "num_tokens": 49753646.0,
      "step": 26675
    },
    {
      "entropy": 5.88151707649231,
      "epoch": 2.3560579300600493,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004443315162338409,
      "loss": 5.2941,
      "mean_token_accuracy": 0.21318313479423523,
      "num_tokens": 49762576.0,
      "step": 26680
    },
    {
      "entropy": 5.943893098831177,
      "epoch": 2.3564994701518898,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000444310777554961,
      "loss": 5.2926,
      "mean_token_accuracy": 0.20822067260742189,
      "num_tokens": 49772614.0,
      "step": 26685
    },
    {
      "entropy": 5.859329748153686,
      "epoch": 2.3569410102437303,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004442900355593793,
      "loss": 5.2595,
      "mean_token_accuracy": 0.21090115755796432,
      "num_tokens": 49781005.0,
      "step": 26690
    },
    {
      "entropy": 5.828287792205811,
      "epoch": 2.3573825503355703,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044426929024750204,
      "loss": 5.2705,
      "mean_token_accuracy": 0.21130183190107346,
      "num_tokens": 49790639.0,
      "step": 26695
    },
    {
      "entropy": 5.938632488250732,
      "epoch": 2.357824090427411,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004442485416197357,
      "loss": 5.3764,
      "mean_token_accuracy": 0.20766025334596633,
      "num_tokens": 49800010.0,
      "step": 26700
    },
    {
      "entropy": 5.926017236709595,
      "epoch": 2.3582656305192513,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044422778967648675,
      "loss": 5.2738,
      "mean_token_accuracy": 0.20750439614057542,
      "num_tokens": 49808510.0,
      "step": 26705
    },
    {
      "entropy": 5.874882411956787,
      "epoch": 2.3587071706110914,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044420703441816153,
      "loss": 5.3053,
      "mean_token_accuracy": 0.2100327119231224,
      "num_tokens": 49818017.0,
      "step": 26710
    },
    {
      "entropy": 5.845489645004273,
      "epoch": 2.359148710702932,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004441862758451668,
      "loss": 5.279,
      "mean_token_accuracy": 0.21044786274433136,
      "num_tokens": 49826797.0,
      "step": 26715
    },
    {
      "entropy": 5.910222911834717,
      "epoch": 2.3595902507947724,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044416551395790916,
      "loss": 5.2785,
      "mean_token_accuracy": 0.20705964863300325,
      "num_tokens": 49835632.0,
      "step": 26720
    },
    {
      "entropy": 5.903456354141236,
      "epoch": 2.3600317908866124,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004441447487567953,
      "loss": 5.3059,
      "mean_token_accuracy": 0.21210643649101257,
      "num_tokens": 49844770.0,
      "step": 26725
    },
    {
      "entropy": 5.877901554107666,
      "epoch": 2.360473330978453,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044412398024223194,
      "loss": 5.3167,
      "mean_token_accuracy": 0.21286253929138182,
      "num_tokens": 49854170.0,
      "step": 26730
    },
    {
      "entropy": 5.969999217987061,
      "epoch": 2.3609148710702934,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004441032084146261,
      "loss": 5.338,
      "mean_token_accuracy": 0.20706886053085327,
      "num_tokens": 49863541.0,
      "step": 26735
    },
    {
      "entropy": 5.947088289260864,
      "epoch": 2.3613564111621335,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044408243327438446,
      "loss": 5.2531,
      "mean_token_accuracy": 0.2140100210905075,
      "num_tokens": 49872556.0,
      "step": 26740
    },
    {
      "entropy": 5.900307989120483,
      "epoch": 2.361797951253974,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044406165482191424,
      "loss": 5.3271,
      "mean_token_accuracy": 0.21021255254745483,
      "num_tokens": 49882188.0,
      "step": 26745
    },
    {
      "entropy": 5.9232886791229244,
      "epoch": 2.362239491345814,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004440408730576223,
      "loss": 5.3914,
      "mean_token_accuracy": 0.20298184603452682,
      "num_tokens": 49891980.0,
      "step": 26750
    },
    {
      "entropy": 5.940526247024536,
      "epoch": 2.3626810314376545,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044402008798191575,
      "loss": 5.2782,
      "mean_token_accuracy": 0.2152009680867195,
      "num_tokens": 49900783.0,
      "step": 26755
    },
    {
      "entropy": 5.854488849639893,
      "epoch": 2.363122571529495,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004439992995952019,
      "loss": 5.317,
      "mean_token_accuracy": 0.21335611790418624,
      "num_tokens": 49910525.0,
      "step": 26760
    },
    {
      "entropy": 5.903363609313965,
      "epoch": 2.363564111621335,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044397850789788774,
      "loss": 5.3084,
      "mean_token_accuracy": 0.20832739770412445,
      "num_tokens": 49919286.0,
      "step": 26765
    },
    {
      "entropy": 5.925664663314819,
      "epoch": 2.3640056517131756,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004439577128903808,
      "loss": 5.2568,
      "mean_token_accuracy": 0.21320044696331025,
      "num_tokens": 49927882.0,
      "step": 26770
    },
    {
      "entropy": 5.95701413154602,
      "epoch": 2.364447191805016,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004439369145730884,
      "loss": 5.307,
      "mean_token_accuracy": 0.20887205749750137,
      "num_tokens": 49936289.0,
      "step": 26775
    },
    {
      "entropy": 5.903356981277466,
      "epoch": 2.364888731896856,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004439161129464179,
      "loss": 5.3161,
      "mean_token_accuracy": 0.19926695972681047,
      "num_tokens": 49945356.0,
      "step": 26780
    },
    {
      "entropy": 5.908850574493409,
      "epoch": 2.3653302719886966,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044389530801077675,
      "loss": 5.2252,
      "mean_token_accuracy": 0.2146916702389717,
      "num_tokens": 49954216.0,
      "step": 26785
    },
    {
      "entropy": 5.960436105728149,
      "epoch": 2.365771812080537,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004438744997665726,
      "loss": 5.3547,
      "mean_token_accuracy": 0.20617856830358505,
      "num_tokens": 49963179.0,
      "step": 26790
    },
    {
      "entropy": 5.909926271438598,
      "epoch": 2.366213352172377,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004438536882142131,
      "loss": 5.3587,
      "mean_token_accuracy": 0.20382177382707595,
      "num_tokens": 49973108.0,
      "step": 26795
    },
    {
      "entropy": 5.912693452835083,
      "epoch": 2.3666548922642177,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004438328733541059,
      "loss": 5.2506,
      "mean_token_accuracy": 0.21181951016187667,
      "num_tokens": 49982003.0,
      "step": 26800
    },
    {
      "entropy": 5.870848941802978,
      "epoch": 2.3670964323560577,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004438120551866586,
      "loss": 5.2925,
      "mean_token_accuracy": 0.20937933623790742,
      "num_tokens": 49992034.0,
      "step": 26805
    },
    {
      "entropy": 5.828803634643554,
      "epoch": 2.367537972447898,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004437912337122792,
      "loss": 5.291,
      "mean_token_accuracy": 0.21592638045549392,
      "num_tokens": 50001379.0,
      "step": 26810
    },
    {
      "entropy": 5.945402050018311,
      "epoch": 2.3679795125397387,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004437704089313756,
      "loss": 5.3812,
      "mean_token_accuracy": 0.20421904921531678,
      "num_tokens": 50010721.0,
      "step": 26815
    },
    {
      "entropy": 5.942852115631103,
      "epoch": 2.3684210526315788,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044374958084435564,
      "loss": 5.363,
      "mean_token_accuracy": 0.210134793817997,
      "num_tokens": 50019571.0,
      "step": 26820
    },
    {
      "entropy": 5.925140047073365,
      "epoch": 2.3688625927234193,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044372874945162734,
      "loss": 5.3012,
      "mean_token_accuracy": 0.21074575036764145,
      "num_tokens": 50028045.0,
      "step": 26825
    },
    {
      "entropy": 5.942230081558227,
      "epoch": 2.3693041328152598,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044370791475359877,
      "loss": 5.2767,
      "mean_token_accuracy": 0.2164221227169037,
      "num_tokens": 50036769.0,
      "step": 26830
    },
    {
      "entropy": 5.896571350097656,
      "epoch": 2.3697456729071,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044368707675067814,
      "loss": 5.3065,
      "mean_token_accuracy": 0.2126169517636299,
      "num_tokens": 50045973.0,
      "step": 26835
    },
    {
      "entropy": 5.834485340118408,
      "epoch": 2.3701872129989403,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004436662354432736,
      "loss": 5.2297,
      "mean_token_accuracy": 0.2074983149766922,
      "num_tokens": 50055243.0,
      "step": 26840
    },
    {
      "entropy": 5.943623208999634,
      "epoch": 2.370628753090781,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044364539083179347,
      "loss": 5.3003,
      "mean_token_accuracy": 0.21723183244466782,
      "num_tokens": 50064573.0,
      "step": 26845
    },
    {
      "entropy": 5.822544050216675,
      "epoch": 2.371070293182621,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000443624542916646,
      "loss": 5.1537,
      "mean_token_accuracy": 0.2214773789048195,
      "num_tokens": 50073578.0,
      "step": 26850
    },
    {
      "entropy": 5.8752302646636965,
      "epoch": 2.3715118332744614,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004436036916982397,
      "loss": 5.2996,
      "mean_token_accuracy": 0.2183082014322281,
      "num_tokens": 50083620.0,
      "step": 26855
    },
    {
      "entropy": 5.857642459869385,
      "epoch": 2.371953373366302,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004435828371769829,
      "loss": 5.2327,
      "mean_token_accuracy": 0.2220786392688751,
      "num_tokens": 50092050.0,
      "step": 26860
    },
    {
      "entropy": 5.918768882751465,
      "epoch": 2.372394913458142,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044356197935328426,
      "loss": 5.3242,
      "mean_token_accuracy": 0.20807345062494279,
      "num_tokens": 50102425.0,
      "step": 26865
    },
    {
      "entropy": 5.88598861694336,
      "epoch": 2.3728364535499824,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044354111822755225,
      "loss": 5.226,
      "mean_token_accuracy": 0.21028743386268617,
      "num_tokens": 50112338.0,
      "step": 26870
    },
    {
      "entropy": 5.932241296768188,
      "epoch": 2.373277993641823,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044352025380019563,
      "loss": 5.2958,
      "mean_token_accuracy": 0.21172044426202774,
      "num_tokens": 50121302.0,
      "step": 26875
    },
    {
      "entropy": 5.817448234558105,
      "epoch": 2.373719533733663,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044349938607162305,
      "loss": 5.2381,
      "mean_token_accuracy": 0.21841894388198851,
      "num_tokens": 50129820.0,
      "step": 26880
    },
    {
      "entropy": 5.915347385406494,
      "epoch": 2.3741610738255035,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044347851504224336,
      "loss": 5.3644,
      "mean_token_accuracy": 0.20760912746191024,
      "num_tokens": 50138860.0,
      "step": 26885
    },
    {
      "entropy": 5.894362831115723,
      "epoch": 2.3746026139173435,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044345764071246536,
      "loss": 5.2826,
      "mean_token_accuracy": 0.2128836765885353,
      "num_tokens": 50148257.0,
      "step": 26890
    },
    {
      "entropy": 5.845790719985962,
      "epoch": 2.375044154009184,
      "grad_norm": 1.125,
      "learning_rate": 0.00044343676308269796,
      "loss": 5.2711,
      "mean_token_accuracy": 0.2171759709715843,
      "num_tokens": 50157380.0,
      "step": 26895
    },
    {
      "entropy": 5.970784759521484,
      "epoch": 2.3754856941010245,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044341588215335016,
      "loss": 5.3054,
      "mean_token_accuracy": 0.20860047340393068,
      "num_tokens": 50167806.0,
      "step": 26900
    },
    {
      "entropy": 5.986636066436768,
      "epoch": 2.3759272341928646,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044339499792483104,
      "loss": 5.3353,
      "mean_token_accuracy": 0.20293398052453995,
      "num_tokens": 50176763.0,
      "step": 26905
    },
    {
      "entropy": 5.816183471679688,
      "epoch": 2.376368774284705,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004433741103975498,
      "loss": 5.3014,
      "mean_token_accuracy": 0.21069140285253524,
      "num_tokens": 50186735.0,
      "step": 26910
    },
    {
      "entropy": 5.8311268329620365,
      "epoch": 2.3768103143765456,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044335321957191535,
      "loss": 5.2535,
      "mean_token_accuracy": 0.21647219210863114,
      "num_tokens": 50195206.0,
      "step": 26915
    },
    {
      "entropy": 5.972170972824097,
      "epoch": 2.3772518544683856,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044333232544833707,
      "loss": 5.3435,
      "mean_token_accuracy": 0.20145729631185533,
      "num_tokens": 50203706.0,
      "step": 26920
    },
    {
      "entropy": 5.909353971481323,
      "epoch": 2.377693394560226,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004433114280272243,
      "loss": 5.3159,
      "mean_token_accuracy": 0.20624017119407653,
      "num_tokens": 50213007.0,
      "step": 26925
    },
    {
      "entropy": 5.921986865997314,
      "epoch": 2.3781349346520666,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004432905273089864,
      "loss": 5.2271,
      "mean_token_accuracy": 0.21782964169979097,
      "num_tokens": 50223155.0,
      "step": 26930
    },
    {
      "entropy": 5.795108938217163,
      "epoch": 2.3785764747439067,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044326962329403275,
      "loss": 5.161,
      "mean_token_accuracy": 0.22062039226293564,
      "num_tokens": 50231788.0,
      "step": 26935
    },
    {
      "entropy": 5.8474769115448,
      "epoch": 2.379018014835747,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044324871598277294,
      "loss": 5.2554,
      "mean_token_accuracy": 0.21597664207220077,
      "num_tokens": 50241557.0,
      "step": 26940
    },
    {
      "entropy": 5.905368709564209,
      "epoch": 2.379459554927587,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044322780537561643,
      "loss": 5.3213,
      "mean_token_accuracy": 0.21200947612524032,
      "num_tokens": 50252141.0,
      "step": 26945
    },
    {
      "entropy": 5.933778619766235,
      "epoch": 2.3799010950194277,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004432068914729729,
      "loss": 5.358,
      "mean_token_accuracy": 0.2091843694448471,
      "num_tokens": 50261361.0,
      "step": 26950
    },
    {
      "entropy": 5.907635068893432,
      "epoch": 2.380342635111268,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044318597427525197,
      "loss": 5.3747,
      "mean_token_accuracy": 0.21069760769605636,
      "num_tokens": 50270193.0,
      "step": 26955
    },
    {
      "entropy": 5.977096796035767,
      "epoch": 2.3807841752031083,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044316505378286356,
      "loss": 5.3233,
      "mean_token_accuracy": 0.20597683787345886,
      "num_tokens": 50280217.0,
      "step": 26960
    },
    {
      "entropy": 5.882652139663696,
      "epoch": 2.3812257152949488,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004431441299962173,
      "loss": 5.2369,
      "mean_token_accuracy": 0.21186361312866211,
      "num_tokens": 50289610.0,
      "step": 26965
    },
    {
      "entropy": 5.848551607131958,
      "epoch": 2.3816672553867893,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004431232029157232,
      "loss": 5.2106,
      "mean_token_accuracy": 0.21853208541870117,
      "num_tokens": 50298734.0,
      "step": 26970
    },
    {
      "entropy": 5.852073717117309,
      "epoch": 2.3821087954786293,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004431022725417912,
      "loss": 5.2288,
      "mean_token_accuracy": 0.21615064442157744,
      "num_tokens": 50307679.0,
      "step": 26975
    },
    {
      "entropy": 5.928052520751953,
      "epoch": 2.38255033557047,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004430813388748312,
      "loss": 5.2182,
      "mean_token_accuracy": 0.21318549811840057,
      "num_tokens": 50315983.0,
      "step": 26980
    },
    {
      "entropy": 5.951604557037354,
      "epoch": 2.3829918756623103,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044306040191525347,
      "loss": 5.3343,
      "mean_token_accuracy": 0.2079811379313469,
      "num_tokens": 50325222.0,
      "step": 26985
    },
    {
      "entropy": 5.83290319442749,
      "epoch": 2.3834334157541504,
      "grad_norm": 1.1875,
      "learning_rate": 0.000443039461663468,
      "loss": 5.2398,
      "mean_token_accuracy": 0.21699571907520293,
      "num_tokens": 50333417.0,
      "step": 26990
    },
    {
      "entropy": 5.878643608093261,
      "epoch": 2.383874955845991,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044301851811988504,
      "loss": 5.2488,
      "mean_token_accuracy": 0.22254746854305268,
      "num_tokens": 50342836.0,
      "step": 26995
    },
    {
      "entropy": 5.94852123260498,
      "epoch": 2.3843164959378313,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044299757128491487,
      "loss": 5.3788,
      "mean_token_accuracy": 0.20585856586694717,
      "num_tokens": 50351988.0,
      "step": 27000
    },
    {
      "epoch": 2.3843164959378313,
      "eval_entropy": 5.618579518889517,
      "eval_loss": 5.360966682434082,
      "eval_mean_token_accuracy": 0.2160217850480655,
      "eval_num_tokens": 50351988.0,
      "eval_runtime": 20.2811,
      "eval_samples_per_second": 1741.276,
      "eval_steps_per_second": 217.69,
      "step": 27000
    },
    {
      "entropy": 5.869129753112793,
      "epoch": 2.3847580360296714,
      "grad_norm": 1.125,
      "learning_rate": 0.00044297662115896783,
      "loss": 5.3286,
      "mean_token_accuracy": 0.20879713594913482,
      "num_tokens": 50361087.0,
      "step": 27005
    },
    {
      "entropy": 5.878187227249145,
      "epoch": 2.385199576121512,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00044295566774245435,
      "loss": 5.314,
      "mean_token_accuracy": 0.21083126962184906,
      "num_tokens": 50369842.0,
      "step": 27010
    },
    {
      "entropy": 6.013025617599487,
      "epoch": 2.3856411162133524,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044293471103578475,
      "loss": 5.3935,
      "mean_token_accuracy": 0.20158904045820236,
      "num_tokens": 50379125.0,
      "step": 27015
    },
    {
      "entropy": 5.881494045257568,
      "epoch": 2.3860826563051925,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044291375103936977,
      "loss": 5.2025,
      "mean_token_accuracy": 0.2193581134080887,
      "num_tokens": 50388031.0,
      "step": 27020
    },
    {
      "entropy": 5.895949935913086,
      "epoch": 2.386524196397033,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044289278775361984,
      "loss": 5.3242,
      "mean_token_accuracy": 0.20894320011138917,
      "num_tokens": 50397348.0,
      "step": 27025
    },
    {
      "entropy": 5.887780618667603,
      "epoch": 2.386965736488873,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044287182117894576,
      "loss": 5.3093,
      "mean_token_accuracy": 0.21733276695013046,
      "num_tokens": 50406782.0,
      "step": 27030
    },
    {
      "entropy": 5.918292140960693,
      "epoch": 2.3874072765807135,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00044285085131575807,
      "loss": 5.2489,
      "mean_token_accuracy": 0.2138858899474144,
      "num_tokens": 50415527.0,
      "step": 27035
    },
    {
      "entropy": 5.993356561660766,
      "epoch": 2.387848816672554,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044282987816446774,
      "loss": 5.3522,
      "mean_token_accuracy": 0.21257390677928925,
      "num_tokens": 50424852.0,
      "step": 27040
    },
    {
      "entropy": 5.886292028427124,
      "epoch": 2.388290356764394,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004428089017254856,
      "loss": 5.3026,
      "mean_token_accuracy": 0.20935603231191635,
      "num_tokens": 50434371.0,
      "step": 27045
    },
    {
      "entropy": 5.977764463424682,
      "epoch": 2.3887318968562345,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044278792199922245,
      "loss": 5.3685,
      "mean_token_accuracy": 0.20573052316904067,
      "num_tokens": 50443358.0,
      "step": 27050
    },
    {
      "entropy": 5.961162757873535,
      "epoch": 2.389173436948075,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044276693898608933,
      "loss": 5.2949,
      "mean_token_accuracy": 0.2088039606809616,
      "num_tokens": 50453132.0,
      "step": 27055
    },
    {
      "entropy": 5.90708646774292,
      "epoch": 2.389614977039915,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044274595268649727,
      "loss": 5.2541,
      "mean_token_accuracy": 0.21330638527870177,
      "num_tokens": 50461837.0,
      "step": 27060
    },
    {
      "entropy": 5.8635557174682615,
      "epoch": 2.3900565171317556,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044272496310085745,
      "loss": 5.2614,
      "mean_token_accuracy": 0.2185961902141571,
      "num_tokens": 50470482.0,
      "step": 27065
    },
    {
      "entropy": 5.869186115264893,
      "epoch": 2.390498057223596,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044270397022958094,
      "loss": 5.2909,
      "mean_token_accuracy": 0.21192137002944947,
      "num_tokens": 50480168.0,
      "step": 27070
    },
    {
      "entropy": 5.8967235565185545,
      "epoch": 2.390939597315436,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044268297407307905,
      "loss": 5.2932,
      "mean_token_accuracy": 0.20982505679130553,
      "num_tokens": 50489937.0,
      "step": 27075
    },
    {
      "entropy": 5.840878438949585,
      "epoch": 2.3913811374072766,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044266197463176305,
      "loss": 5.2886,
      "mean_token_accuracy": 0.2081238806247711,
      "num_tokens": 50499151.0,
      "step": 27080
    },
    {
      "entropy": 5.962720727920532,
      "epoch": 2.3918226774991167,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044264097190604436,
      "loss": 5.353,
      "mean_token_accuracy": 0.21375546902418135,
      "num_tokens": 50508413.0,
      "step": 27085
    },
    {
      "entropy": 5.971755504608154,
      "epoch": 2.392264217590957,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004426199658963344,
      "loss": 5.3911,
      "mean_token_accuracy": 0.2080995336174965,
      "num_tokens": 50517513.0,
      "step": 27090
    },
    {
      "entropy": 5.952742052078247,
      "epoch": 2.3927057576827977,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044259895660304455,
      "loss": 5.2618,
      "mean_token_accuracy": 0.21463502645492555,
      "num_tokens": 50526057.0,
      "step": 27095
    },
    {
      "entropy": 5.858213901519775,
      "epoch": 2.3931472977746377,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044257794402658655,
      "loss": 5.2432,
      "mean_token_accuracy": 0.20845317095518112,
      "num_tokens": 50535388.0,
      "step": 27100
    },
    {
      "entropy": 5.915638113021851,
      "epoch": 2.3935888378664782,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004425569281673719,
      "loss": 5.3102,
      "mean_token_accuracy": 0.21171536892652512,
      "num_tokens": 50544606.0,
      "step": 27105
    },
    {
      "entropy": 5.932204723358154,
      "epoch": 2.3940303779583187,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044253590902581223,
      "loss": 5.2896,
      "mean_token_accuracy": 0.21097355484962463,
      "num_tokens": 50553822.0,
      "step": 27110
    },
    {
      "entropy": 5.895391035079956,
      "epoch": 2.394471918050159,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004425148866023195,
      "loss": 5.2678,
      "mean_token_accuracy": 0.21966320723295213,
      "num_tokens": 50562879.0,
      "step": 27115
    },
    {
      "entropy": 5.912628173828125,
      "epoch": 2.3949134581419993,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044249386089730534,
      "loss": 5.2059,
      "mean_token_accuracy": 0.2158907324075699,
      "num_tokens": 50571536.0,
      "step": 27120
    },
    {
      "entropy": 5.8657153129577635,
      "epoch": 2.39535499823384,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004424728319111818,
      "loss": 5.2341,
      "mean_token_accuracy": 0.22003188580274582,
      "num_tokens": 50580692.0,
      "step": 27125
    },
    {
      "entropy": 5.996710157394409,
      "epoch": 2.39579653832568,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004424517996443606,
      "loss": 5.464,
      "mean_token_accuracy": 0.20087900310754775,
      "num_tokens": 50589456.0,
      "step": 27130
    },
    {
      "entropy": 5.879535913467407,
      "epoch": 2.3962380784175203,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044243076409725395,
      "loss": 5.2568,
      "mean_token_accuracy": 0.21488114595413207,
      "num_tokens": 50598947.0,
      "step": 27135
    },
    {
      "entropy": 5.89729528427124,
      "epoch": 2.396679618509361,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044240972527027385,
      "loss": 5.3349,
      "mean_token_accuracy": 0.2167168065905571,
      "num_tokens": 50607373.0,
      "step": 27140
    },
    {
      "entropy": 5.974016571044922,
      "epoch": 2.397121158601201,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044238868316383246,
      "loss": 5.4165,
      "mean_token_accuracy": 0.20460365414619447,
      "num_tokens": 50616930.0,
      "step": 27145
    },
    {
      "entropy": 5.884567928314209,
      "epoch": 2.3975626986930414,
      "grad_norm": 1.078125,
      "learning_rate": 0.000442367637778342,
      "loss": 5.3029,
      "mean_token_accuracy": 0.2128095731139183,
      "num_tokens": 50626658.0,
      "step": 27150
    },
    {
      "entropy": 5.887493848800659,
      "epoch": 2.398004238784882,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004423465891142146,
      "loss": 5.3607,
      "mean_token_accuracy": 0.20511319041252135,
      "num_tokens": 50635692.0,
      "step": 27155
    },
    {
      "entropy": 5.916338872909546,
      "epoch": 2.398445778876722,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044232553717186274,
      "loss": 5.2112,
      "mean_token_accuracy": 0.21953658610582352,
      "num_tokens": 50644036.0,
      "step": 27160
    },
    {
      "entropy": 5.957617139816284,
      "epoch": 2.3988873189685624,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004423044819516988,
      "loss": 5.3772,
      "mean_token_accuracy": 0.2041273072361946,
      "num_tokens": 50652363.0,
      "step": 27165
    },
    {
      "entropy": 5.9244499683380125,
      "epoch": 2.3993288590604025,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004422834234541352,
      "loss": 5.2736,
      "mean_token_accuracy": 0.2100076898932457,
      "num_tokens": 50661985.0,
      "step": 27170
    },
    {
      "entropy": 5.879900407791138,
      "epoch": 2.399770399152243,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004422623616795845,
      "loss": 5.2857,
      "mean_token_accuracy": 0.21499712765216827,
      "num_tokens": 50671171.0,
      "step": 27175
    },
    {
      "entropy": 5.896852445602417,
      "epoch": 2.4002119392440835,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044224129662845916,
      "loss": 5.2316,
      "mean_token_accuracy": 0.21163083463907242,
      "num_tokens": 50679792.0,
      "step": 27180
    },
    {
      "entropy": 5.91713399887085,
      "epoch": 2.4006534793359235,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000442220228301172,
      "loss": 5.3739,
      "mean_token_accuracy": 0.20006223022937775,
      "num_tokens": 50690082.0,
      "step": 27185
    },
    {
      "entropy": 5.887785863876343,
      "epoch": 2.401095019427764,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004421991566981357,
      "loss": 5.2434,
      "mean_token_accuracy": 0.22010890543460845,
      "num_tokens": 50698351.0,
      "step": 27190
    },
    {
      "entropy": 5.8420814037322994,
      "epoch": 2.4015365595196045,
      "grad_norm": 1.234375,
      "learning_rate": 0.000442178081819763,
      "loss": 5.2396,
      "mean_token_accuracy": 0.21667411923408508,
      "num_tokens": 50706977.0,
      "step": 27195
    },
    {
      "entropy": 5.848702430725098,
      "epoch": 2.4019780996114446,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004421570036664667,
      "loss": 5.3129,
      "mean_token_accuracy": 0.21142877787351608,
      "num_tokens": 50716168.0,
      "step": 27200
    },
    {
      "entropy": 5.919884204864502,
      "epoch": 2.402419639703285,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044213592223865976,
      "loss": 5.328,
      "mean_token_accuracy": 0.20722629725933076,
      "num_tokens": 50725301.0,
      "step": 27205
    },
    {
      "entropy": 5.909435701370239,
      "epoch": 2.4028611797951256,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044211483753675524,
      "loss": 5.2861,
      "mean_token_accuracy": 0.21097408086061478,
      "num_tokens": 50734494.0,
      "step": 27210
    },
    {
      "entropy": 5.89926290512085,
      "epoch": 2.4033027198869656,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004420937495611661,
      "loss": 5.3264,
      "mean_token_accuracy": 0.20398035049438476,
      "num_tokens": 50743650.0,
      "step": 27215
    },
    {
      "entropy": 5.873824787139893,
      "epoch": 2.403744259978806,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004420726583123054,
      "loss": 5.2375,
      "mean_token_accuracy": 0.21353846192359924,
      "num_tokens": 50753538.0,
      "step": 27220
    },
    {
      "entropy": 5.9598548412323,
      "epoch": 2.404185800070646,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004420515637905863,
      "loss": 5.355,
      "mean_token_accuracy": 0.208790223300457,
      "num_tokens": 50763349.0,
      "step": 27225
    },
    {
      "entropy": 5.960746812820434,
      "epoch": 2.4046273401624867,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004420304659964221,
      "loss": 5.3474,
      "mean_token_accuracy": 0.20182476490736007,
      "num_tokens": 50772348.0,
      "step": 27230
    },
    {
      "entropy": 5.984484434127808,
      "epoch": 2.405068880254327,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044200936493022616,
      "loss": 5.4158,
      "mean_token_accuracy": 0.20486751794815064,
      "num_tokens": 50782054.0,
      "step": 27235
    },
    {
      "entropy": 5.951259517669678,
      "epoch": 2.4055104203461672,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004419882605924116,
      "loss": 5.3067,
      "mean_token_accuracy": 0.2121528849005699,
      "num_tokens": 50791955.0,
      "step": 27240
    },
    {
      "entropy": 5.88605523109436,
      "epoch": 2.4059519604380077,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004419671529833921,
      "loss": 5.3436,
      "mean_token_accuracy": 0.20783793032169343,
      "num_tokens": 50801907.0,
      "step": 27245
    },
    {
      "entropy": 5.939907121658325,
      "epoch": 2.4063935005298482,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004419460421035809,
      "loss": 5.3946,
      "mean_token_accuracy": 0.20269893556833268,
      "num_tokens": 50811413.0,
      "step": 27250
    },
    {
      "entropy": 5.917592573165893,
      "epoch": 2.4068350406216883,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004419249279533918,
      "loss": 5.2812,
      "mean_token_accuracy": 0.21199343204498292,
      "num_tokens": 50820080.0,
      "step": 27255
    },
    {
      "entropy": 5.941693067550659,
      "epoch": 2.407276580713529,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044190381053323824,
      "loss": 5.3906,
      "mean_token_accuracy": 0.20564500093460084,
      "num_tokens": 50829622.0,
      "step": 27260
    },
    {
      "entropy": 5.905082941055298,
      "epoch": 2.4077181208053693,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044188268984353396,
      "loss": 5.227,
      "mean_token_accuracy": 0.22203828990459443,
      "num_tokens": 50838112.0,
      "step": 27265
    },
    {
      "entropy": 5.866824197769165,
      "epoch": 2.4081596608972093,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004418615658846927,
      "loss": 5.3369,
      "mean_token_accuracy": 0.20572246760129928,
      "num_tokens": 50848014.0,
      "step": 27270
    },
    {
      "entropy": 6.020290327072144,
      "epoch": 2.40860120098905,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004418404386571283,
      "loss": 5.4019,
      "mean_token_accuracy": 0.19499764889478682,
      "num_tokens": 50857103.0,
      "step": 27275
    },
    {
      "entropy": 5.912293529510498,
      "epoch": 2.4090427410808903,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004418193081612545,
      "loss": 5.267,
      "mean_token_accuracy": 0.21246202290058136,
      "num_tokens": 50865866.0,
      "step": 27280
    },
    {
      "entropy": 5.942117309570312,
      "epoch": 2.4094842811727304,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044179817439748547,
      "loss": 5.343,
      "mean_token_accuracy": 0.20735968500375748,
      "num_tokens": 50874694.0,
      "step": 27285
    },
    {
      "entropy": 5.868934965133667,
      "epoch": 2.409925821264571,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044177703736623497,
      "loss": 5.2473,
      "mean_token_accuracy": 0.21567634344100953,
      "num_tokens": 50884965.0,
      "step": 27290
    },
    {
      "entropy": 5.957904958724976,
      "epoch": 2.4103673613564114,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044175589706791724,
      "loss": 5.3604,
      "mean_token_accuracy": 0.20950464010238648,
      "num_tokens": 50894192.0,
      "step": 27295
    },
    {
      "entropy": 5.908656597137451,
      "epoch": 2.4108089014482514,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044173475350294623,
      "loss": 5.2774,
      "mean_token_accuracy": 0.20949270129203795,
      "num_tokens": 50902458.0,
      "step": 27300
    },
    {
      "entropy": 5.90416316986084,
      "epoch": 2.411250441540092,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044171360667173633,
      "loss": 5.2317,
      "mean_token_accuracy": 0.2146620810031891,
      "num_tokens": 50912038.0,
      "step": 27305
    },
    {
      "entropy": 5.958814716339111,
      "epoch": 2.411691981631932,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004416924565747016,
      "loss": 5.4353,
      "mean_token_accuracy": 0.19050152003765106,
      "num_tokens": 50921164.0,
      "step": 27310
    },
    {
      "entropy": 5.916319465637207,
      "epoch": 2.4121335217237725,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044167130321225644,
      "loss": 5.289,
      "mean_token_accuracy": 0.20970964580774307,
      "num_tokens": 50930359.0,
      "step": 27315
    },
    {
      "entropy": 5.860442209243774,
      "epoch": 2.412575061815613,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044165014658481533,
      "loss": 5.3172,
      "mean_token_accuracy": 0.20729777812957764,
      "num_tokens": 50940675.0,
      "step": 27320
    },
    {
      "entropy": 5.934039211273193,
      "epoch": 2.413016601907453,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004416289866927926,
      "loss": 5.414,
      "mean_token_accuracy": 0.19985995441675186,
      "num_tokens": 50949402.0,
      "step": 27325
    },
    {
      "entropy": 5.893720293045044,
      "epoch": 2.4134581419992935,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004416078235366027,
      "loss": 5.2522,
      "mean_token_accuracy": 0.21008396595716478,
      "num_tokens": 50958348.0,
      "step": 27330
    },
    {
      "entropy": 5.8870007514953615,
      "epoch": 2.413899682091134,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044158665711666034,
      "loss": 5.2759,
      "mean_token_accuracy": 0.21114024370908738,
      "num_tokens": 50967348.0,
      "step": 27335
    },
    {
      "entropy": 5.946973991394043,
      "epoch": 2.414341222182974,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044156548743338016,
      "loss": 5.3263,
      "mean_token_accuracy": 0.21287312358617783,
      "num_tokens": 50975882.0,
      "step": 27340
    },
    {
      "entropy": 5.970800065994263,
      "epoch": 2.4147827622748146,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044154431448717674,
      "loss": 5.2849,
      "mean_token_accuracy": 0.2098815768957138,
      "num_tokens": 50985241.0,
      "step": 27345
    },
    {
      "entropy": 5.941105031967163,
      "epoch": 2.415224302366655,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044152313827846494,
      "loss": 5.2917,
      "mean_token_accuracy": 0.21210358589887618,
      "num_tokens": 50994834.0,
      "step": 27350
    },
    {
      "entropy": 5.873815727233887,
      "epoch": 2.415665842458495,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044150195880765964,
      "loss": 5.3036,
      "mean_token_accuracy": 0.21397734433412552,
      "num_tokens": 51003919.0,
      "step": 27355
    },
    {
      "entropy": 5.972154951095581,
      "epoch": 2.4161073825503356,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044148077607517554,
      "loss": 5.2581,
      "mean_token_accuracy": 0.211700876057148,
      "num_tokens": 51012868.0,
      "step": 27360
    },
    {
      "entropy": 5.942103290557862,
      "epoch": 2.4165489226421757,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044145959008142784,
      "loss": 5.327,
      "mean_token_accuracy": 0.20591636300086974,
      "num_tokens": 51021693.0,
      "step": 27365
    },
    {
      "entropy": 5.908239459991455,
      "epoch": 2.416990462734016,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004414384008268314,
      "loss": 5.32,
      "mean_token_accuracy": 0.2089933604001999,
      "num_tokens": 51031100.0,
      "step": 27370
    },
    {
      "entropy": 5.918517827987671,
      "epoch": 2.4174320028258567,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044141720831180134,
      "loss": 5.3441,
      "mean_token_accuracy": 0.19942449629306794,
      "num_tokens": 51041077.0,
      "step": 27375
    },
    {
      "entropy": 5.917243003845215,
      "epoch": 2.4178735429176967,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004413960125367527,
      "loss": 5.2809,
      "mean_token_accuracy": 0.20920685678720474,
      "num_tokens": 51050403.0,
      "step": 27380
    },
    {
      "entropy": 5.890940189361572,
      "epoch": 2.4183150830095372,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044137481350210093,
      "loss": 5.3071,
      "mean_token_accuracy": 0.20343202650547026,
      "num_tokens": 51060119.0,
      "step": 27385
    },
    {
      "entropy": 5.951772975921631,
      "epoch": 2.4187566231013777,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044135361120826116,
      "loss": 5.3786,
      "mean_token_accuracy": 0.20055670142173768,
      "num_tokens": 51070355.0,
      "step": 27390
    },
    {
      "entropy": 5.898914623260498,
      "epoch": 2.419198163193218,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004413324056556488,
      "loss": 5.2273,
      "mean_token_accuracy": 0.21486866176128389,
      "num_tokens": 51079386.0,
      "step": 27395
    },
    {
      "entropy": 5.90981159210205,
      "epoch": 2.4196397032850583,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004413111968446792,
      "loss": 5.3881,
      "mean_token_accuracy": 0.19863603860139847,
      "num_tokens": 51089234.0,
      "step": 27400
    },
    {
      "entropy": 5.98210768699646,
      "epoch": 2.420081243376899,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004412899847757678,
      "loss": 5.2945,
      "mean_token_accuracy": 0.21305158287286757,
      "num_tokens": 51099062.0,
      "step": 27405
    },
    {
      "entropy": 5.957066297531128,
      "epoch": 2.420522783468739,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004412687694493302,
      "loss": 5.2644,
      "mean_token_accuracy": 0.2079029783606529,
      "num_tokens": 51108278.0,
      "step": 27410
    },
    {
      "entropy": 5.967505741119385,
      "epoch": 2.4209643235605793,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000441247550865782,
      "loss": 5.369,
      "mean_token_accuracy": 0.21181105822324753,
      "num_tokens": 51117378.0,
      "step": 27415
    },
    {
      "entropy": 5.871611833572388,
      "epoch": 2.42140586365242,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044122632902553877,
      "loss": 5.2887,
      "mean_token_accuracy": 0.22086175233125688,
      "num_tokens": 51126323.0,
      "step": 27420
    },
    {
      "entropy": 5.8886247158050535,
      "epoch": 2.42184740374426,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004412051039290163,
      "loss": 5.2959,
      "mean_token_accuracy": 0.2203296795487404,
      "num_tokens": 51135734.0,
      "step": 27425
    },
    {
      "entropy": 5.786067485809326,
      "epoch": 2.4222889438361004,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044118387557663034,
      "loss": 5.1848,
      "mean_token_accuracy": 0.22002641558647157,
      "num_tokens": 51145020.0,
      "step": 27430
    },
    {
      "entropy": 5.959189510345459,
      "epoch": 2.422730483927941,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044116264396879684,
      "loss": 5.3212,
      "mean_token_accuracy": 0.20634805411100388,
      "num_tokens": 51153908.0,
      "step": 27435
    },
    {
      "entropy": 5.87118706703186,
      "epoch": 2.423172024019781,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004411414091059316,
      "loss": 5.2402,
      "mean_token_accuracy": 0.21175404489040375,
      "num_tokens": 51163547.0,
      "step": 27440
    },
    {
      "entropy": 5.965648365020752,
      "epoch": 2.4236135641116214,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044112017098845064,
      "loss": 5.3876,
      "mean_token_accuracy": 0.2016305848956108,
      "num_tokens": 51172559.0,
      "step": 27445
    },
    {
      "entropy": 5.909197282791138,
      "epoch": 2.4240551042034615,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004410989296167701,
      "loss": 5.2429,
      "mean_token_accuracy": 0.21567276120185852,
      "num_tokens": 51181455.0,
      "step": 27450
    },
    {
      "entropy": 5.9022523880004885,
      "epoch": 2.424496644295302,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044107768499130587,
      "loss": 5.3465,
      "mean_token_accuracy": 0.206267312169075,
      "num_tokens": 51191149.0,
      "step": 27455
    },
    {
      "entropy": 5.863588762283325,
      "epoch": 2.4249381843871425,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004410564371124743,
      "loss": 5.2655,
      "mean_token_accuracy": 0.22086622565984726,
      "num_tokens": 51200702.0,
      "step": 27460
    },
    {
      "entropy": 5.953502035140991,
      "epoch": 2.4253797244789825,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044103518598069156,
      "loss": 5.3367,
      "mean_token_accuracy": 0.2079252779483795,
      "num_tokens": 51209058.0,
      "step": 27465
    },
    {
      "entropy": 5.946819400787353,
      "epoch": 2.425821264570823,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044101393159637396,
      "loss": 5.3144,
      "mean_token_accuracy": 0.21250441074371337,
      "num_tokens": 51219295.0,
      "step": 27470
    },
    {
      "entropy": 5.870768880844116,
      "epoch": 2.4262628046626635,
      "grad_norm": 1.125,
      "learning_rate": 0.00044099267395993785,
      "loss": 5.241,
      "mean_token_accuracy": 0.207552233338356,
      "num_tokens": 51228430.0,
      "step": 27475
    },
    {
      "entropy": 5.846934080123901,
      "epoch": 2.4267043447545036,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004409714130717997,
      "loss": 5.1932,
      "mean_token_accuracy": 0.21826155930757524,
      "num_tokens": 51237804.0,
      "step": 27480
    },
    {
      "entropy": 5.880660676956177,
      "epoch": 2.427145884846344,
      "grad_norm": 1.0,
      "learning_rate": 0.000440950148932376,
      "loss": 5.2558,
      "mean_token_accuracy": 0.22079368382692338,
      "num_tokens": 51247313.0,
      "step": 27485
    },
    {
      "entropy": 5.892316293716431,
      "epoch": 2.4275874249381846,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044092888154208314,
      "loss": 5.3025,
      "mean_token_accuracy": 0.21051884442567825,
      "num_tokens": 51257310.0,
      "step": 27490
    },
    {
      "entropy": 5.92861967086792,
      "epoch": 2.4280289650300246,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004409076109013379,
      "loss": 5.2727,
      "mean_token_accuracy": 0.20940722525119781,
      "num_tokens": 51267028.0,
      "step": 27495
    },
    {
      "entropy": 5.940636157989502,
      "epoch": 2.428470505121865,
      "grad_norm": 1.078125,
      "learning_rate": 0.000440886337010557,
      "loss": 5.2566,
      "mean_token_accuracy": 0.2121024399995804,
      "num_tokens": 51276756.0,
      "step": 27500
    },
    {
      "entropy": 5.920091915130615,
      "epoch": 2.428912045213705,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000440865059870157,
      "loss": 5.3041,
      "mean_token_accuracy": 0.20553562194108962,
      "num_tokens": 51286491.0,
      "step": 27505
    },
    {
      "entropy": 5.985690355300903,
      "epoch": 2.4293535853055457,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044084377948055487,
      "loss": 5.3647,
      "mean_token_accuracy": 0.20088395923376084,
      "num_tokens": 51295221.0,
      "step": 27510
    },
    {
      "entropy": 5.9276206493377686,
      "epoch": 2.429795125397386,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044082249584216735,
      "loss": 5.3294,
      "mean_token_accuracy": 0.21328363567590714,
      "num_tokens": 51304637.0,
      "step": 27515
    },
    {
      "entropy": 5.834335803985596,
      "epoch": 2.4302366654892262,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004408012089554115,
      "loss": 5.2229,
      "mean_token_accuracy": 0.21640851199626923,
      "num_tokens": 51314107.0,
      "step": 27520
    },
    {
      "entropy": 5.918609714508056,
      "epoch": 2.4306782055810667,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044077991882070423,
      "loss": 5.2564,
      "mean_token_accuracy": 0.21038911044597625,
      "num_tokens": 51323353.0,
      "step": 27525
    },
    {
      "entropy": 5.891938495635986,
      "epoch": 2.4311197456729072,
      "grad_norm": 1.125,
      "learning_rate": 0.0004407586254384627,
      "loss": 5.268,
      "mean_token_accuracy": 0.2134750634431839,
      "num_tokens": 51333771.0,
      "step": 27530
    },
    {
      "entropy": 5.879325675964355,
      "epoch": 2.4315612857647473,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004407373288091039,
      "loss": 5.285,
      "mean_token_accuracy": 0.20853112190961837,
      "num_tokens": 51343476.0,
      "step": 27535
    },
    {
      "entropy": 5.936458730697632,
      "epoch": 2.432002825856588,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004407160289330452,
      "loss": 5.2637,
      "mean_token_accuracy": 0.20930998772382736,
      "num_tokens": 51352903.0,
      "step": 27540
    },
    {
      "entropy": 5.948486995697022,
      "epoch": 2.4324443659484283,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00044069472581070363,
      "loss": 5.2981,
      "mean_token_accuracy": 0.2118260756134987,
      "num_tokens": 51362922.0,
      "step": 27545
    },
    {
      "entropy": 5.967051839828491,
      "epoch": 2.4328859060402683,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044067341944249667,
      "loss": 5.3201,
      "mean_token_accuracy": 0.20072970241308213,
      "num_tokens": 51371789.0,
      "step": 27550
    },
    {
      "entropy": 5.977655076980591,
      "epoch": 2.433327446132109,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044065210982884164,
      "loss": 5.3681,
      "mean_token_accuracy": 0.20817289054393767,
      "num_tokens": 51380989.0,
      "step": 27555
    },
    {
      "entropy": 5.90216817855835,
      "epoch": 2.4337689862239493,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004406307969701559,
      "loss": 5.3576,
      "mean_token_accuracy": 0.20791536271572114,
      "num_tokens": 51390191.0,
      "step": 27560
    },
    {
      "entropy": 5.949291372299195,
      "epoch": 2.4342105263157894,
      "grad_norm": 1.125,
      "learning_rate": 0.00044060948086685714,
      "loss": 5.3225,
      "mean_token_accuracy": 0.20892780721187593,
      "num_tokens": 51399133.0,
      "step": 27565
    },
    {
      "entropy": 5.87764573097229,
      "epoch": 2.43465206640763,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044058816151936283,
      "loss": 5.199,
      "mean_token_accuracy": 0.21638662070035936,
      "num_tokens": 51408473.0,
      "step": 27570
    },
    {
      "entropy": 5.870560598373413,
      "epoch": 2.4350936064994704,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044056683892809055,
      "loss": 5.2115,
      "mean_token_accuracy": 0.21610395312309266,
      "num_tokens": 51416944.0,
      "step": 27575
    },
    {
      "entropy": 5.84603180885315,
      "epoch": 2.4355351465913104,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044054551309345816,
      "loss": 5.4273,
      "mean_token_accuracy": 0.20335961431264876,
      "num_tokens": 51426193.0,
      "step": 27580
    },
    {
      "entropy": 5.912672328948974,
      "epoch": 2.435976686683151,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004405241840158833,
      "loss": 5.3539,
      "mean_token_accuracy": 0.21305938810110092,
      "num_tokens": 51435718.0,
      "step": 27585
    },
    {
      "entropy": 5.931331014633178,
      "epoch": 2.436418226774991,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004405028516957837,
      "loss": 5.2748,
      "mean_token_accuracy": 0.20944211781024932,
      "num_tokens": 51445127.0,
      "step": 27590
    },
    {
      "entropy": 6.023457336425781,
      "epoch": 2.4368597668668315,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044048151613357745,
      "loss": 5.3724,
      "mean_token_accuracy": 0.20370213240385054,
      "num_tokens": 51454813.0,
      "step": 27595
    },
    {
      "entropy": 5.903233575820923,
      "epoch": 2.437301306958672,
      "grad_norm": 1.125,
      "learning_rate": 0.0004404601773296824,
      "loss": 5.2979,
      "mean_token_accuracy": 0.21144006103277208,
      "num_tokens": 51464454.0,
      "step": 27600
    },
    {
      "entropy": 5.875844049453735,
      "epoch": 2.437742847050512,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004404388352845166,
      "loss": 5.3056,
      "mean_token_accuracy": 0.20363294035196305,
      "num_tokens": 51473535.0,
      "step": 27605
    },
    {
      "entropy": 5.913086223602295,
      "epoch": 2.4381843871423525,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044041748999849807,
      "loss": 5.2822,
      "mean_token_accuracy": 0.2077423483133316,
      "num_tokens": 51482490.0,
      "step": 27610
    },
    {
      "entropy": 5.941003227233887,
      "epoch": 2.438625927234193,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000440396141472045,
      "loss": 5.3042,
      "mean_token_accuracy": 0.21714147478342055,
      "num_tokens": 51493604.0,
      "step": 27615
    },
    {
      "entropy": 5.899606037139892,
      "epoch": 2.439067467326033,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004403747897055756,
      "loss": 5.3199,
      "mean_token_accuracy": 0.21122273206710815,
      "num_tokens": 51502980.0,
      "step": 27620
    },
    {
      "entropy": 5.957281589508057,
      "epoch": 2.4395090074178736,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004403534346995081,
      "loss": 5.2963,
      "mean_token_accuracy": 0.2119792267680168,
      "num_tokens": 51512131.0,
      "step": 27625
    },
    {
      "entropy": 5.988756275177002,
      "epoch": 2.439950547509714,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044033207645426093,
      "loss": 5.3752,
      "mean_token_accuracy": 0.1983141392469406,
      "num_tokens": 51521600.0,
      "step": 27630
    },
    {
      "entropy": 5.870926284790039,
      "epoch": 2.440392087601554,
      "grad_norm": 1.125,
      "learning_rate": 0.0004403107149702524,
      "loss": 5.1817,
      "mean_token_accuracy": 0.2303866758942604,
      "num_tokens": 51531123.0,
      "step": 27635
    },
    {
      "entropy": 5.888161849975586,
      "epoch": 2.4408336276933946,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004402893502479009,
      "loss": 5.2877,
      "mean_token_accuracy": 0.2155813455581665,
      "num_tokens": 51540395.0,
      "step": 27640
    },
    {
      "entropy": 5.892059469223023,
      "epoch": 2.4412751677852347,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000440267982287625,
      "loss": 5.2559,
      "mean_token_accuracy": 0.2157188445329666,
      "num_tokens": 51549683.0,
      "step": 27645
    },
    {
      "entropy": 5.992727136611938,
      "epoch": 2.441716707877075,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004402466110898435,
      "loss": 5.4445,
      "mean_token_accuracy": 0.19387487322092056,
      "num_tokens": 51559800.0,
      "step": 27650
    },
    {
      "entropy": 5.906308507919311,
      "epoch": 2.4421582479689157,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044022523665497473,
      "loss": 5.2697,
      "mean_token_accuracy": 0.21126964390277864,
      "num_tokens": 51569788.0,
      "step": 27655
    },
    {
      "entropy": 5.908762693405151,
      "epoch": 2.4425997880607557,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004402038589834375,
      "loss": 5.3264,
      "mean_token_accuracy": 0.2027886062860489,
      "num_tokens": 51578598.0,
      "step": 27660
    },
    {
      "entropy": 5.939718627929688,
      "epoch": 2.4430413281525962,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004401824780756507,
      "loss": 5.388,
      "mean_token_accuracy": 0.20246291905641556,
      "num_tokens": 51589203.0,
      "step": 27665
    },
    {
      "entropy": 5.969191360473633,
      "epoch": 2.4434828682444367,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004401610939320331,
      "loss": 5.2881,
      "mean_token_accuracy": 0.21210476160049438,
      "num_tokens": 51597813.0,
      "step": 27670
    },
    {
      "entropy": 5.959642028808593,
      "epoch": 2.4439244083362768,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044013970655300364,
      "loss": 5.3312,
      "mean_token_accuracy": 0.2086146056652069,
      "num_tokens": 51607386.0,
      "step": 27675
    },
    {
      "entropy": 5.942275667190552,
      "epoch": 2.4443659484281173,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044011831593898117,
      "loss": 5.3778,
      "mean_token_accuracy": 0.20626229494810105,
      "num_tokens": 51617091.0,
      "step": 27680
    },
    {
      "entropy": 5.947418832778931,
      "epoch": 2.4448074885199578,
      "grad_norm": 1.125,
      "learning_rate": 0.00044009692209038485,
      "loss": 5.3817,
      "mean_token_accuracy": 0.20702431946992875,
      "num_tokens": 51625616.0,
      "step": 27685
    },
    {
      "entropy": 5.815789222717285,
      "epoch": 2.445249028611798,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004400755250076337,
      "loss": 5.2074,
      "mean_token_accuracy": 0.22401528805494308,
      "num_tokens": 51634693.0,
      "step": 27690
    },
    {
      "entropy": 5.856943273544312,
      "epoch": 2.4456905687036383,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044005412469114684,
      "loss": 5.205,
      "mean_token_accuracy": 0.22315629422664643,
      "num_tokens": 51642549.0,
      "step": 27695
    },
    {
      "entropy": 5.942341279983521,
      "epoch": 2.446132108795479,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044003272114134367,
      "loss": 5.281,
      "mean_token_accuracy": 0.21207432299852372,
      "num_tokens": 51651742.0,
      "step": 27700
    },
    {
      "entropy": 5.945241069793701,
      "epoch": 2.446573648887319,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004400113143586433,
      "loss": 5.3421,
      "mean_token_accuracy": 0.2087707817554474,
      "num_tokens": 51660929.0,
      "step": 27705
    },
    {
      "entropy": 5.937140226364136,
      "epoch": 2.4470151889791594,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043998990434346514,
      "loss": 5.3312,
      "mean_token_accuracy": 0.20369307249784468,
      "num_tokens": 51670580.0,
      "step": 27710
    },
    {
      "entropy": 5.895580720901489,
      "epoch": 2.447456729071,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004399684910962286,
      "loss": 5.325,
      "mean_token_accuracy": 0.2112492635846138,
      "num_tokens": 51680184.0,
      "step": 27715
    },
    {
      "entropy": 5.88524227142334,
      "epoch": 2.44789826916284,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004399470746173531,
      "loss": 5.2777,
      "mean_token_accuracy": 0.20961735695600509,
      "num_tokens": 51690277.0,
      "step": 27720
    },
    {
      "entropy": 5.886123371124268,
      "epoch": 2.4483398092546804,
      "grad_norm": 1.125,
      "learning_rate": 0.0004399256549072583,
      "loss": 5.2619,
      "mean_token_accuracy": 0.2162192165851593,
      "num_tokens": 51699316.0,
      "step": 27725
    },
    {
      "entropy": 5.905518960952759,
      "epoch": 2.4487813493465205,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043990423196636357,
      "loss": 5.3017,
      "mean_token_accuracy": 0.21099428683519364,
      "num_tokens": 51708552.0,
      "step": 27730
    },
    {
      "entropy": 5.997363567352295,
      "epoch": 2.449222889438361,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004398828057950888,
      "loss": 5.2676,
      "mean_token_accuracy": 0.20974707156419753,
      "num_tokens": 51717082.0,
      "step": 27735
    },
    {
      "entropy": 5.882771730422974,
      "epoch": 2.4496644295302015,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004398613763938537,
      "loss": 5.3424,
      "mean_token_accuracy": 0.21198190599679947,
      "num_tokens": 51727059.0,
      "step": 27740
    },
    {
      "entropy": 5.818633890151977,
      "epoch": 2.4501059696220415,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043983994376307797,
      "loss": 5.2232,
      "mean_token_accuracy": 0.22189587950706482,
      "num_tokens": 51735889.0,
      "step": 27745
    },
    {
      "entropy": 5.961389827728271,
      "epoch": 2.450547509713882,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043981850790318145,
      "loss": 5.3701,
      "mean_token_accuracy": 0.20809839963912963,
      "num_tokens": 51744856.0,
      "step": 27750
    },
    {
      "entropy": 5.881601572036743,
      "epoch": 2.4509890498057225,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043979706881458415,
      "loss": 5.2656,
      "mean_token_accuracy": 0.22031942009925842,
      "num_tokens": 51753552.0,
      "step": 27755
    },
    {
      "entropy": 5.9547035694122314,
      "epoch": 2.4514305898975626,
      "grad_norm": 1.125,
      "learning_rate": 0.00043977562649770587,
      "loss": 5.3936,
      "mean_token_accuracy": 0.20053804963827132,
      "num_tokens": 51762929.0,
      "step": 27760
    },
    {
      "entropy": 6.016264915466309,
      "epoch": 2.451872129989403,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043975418095296696,
      "loss": 5.3862,
      "mean_token_accuracy": 0.2072538897395134,
      "num_tokens": 51772732.0,
      "step": 27765
    },
    {
      "entropy": 5.9943207740783695,
      "epoch": 2.4523136700812436,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043973273218078716,
      "loss": 5.3875,
      "mean_token_accuracy": 0.20567516535520552,
      "num_tokens": 51782304.0,
      "step": 27770
    },
    {
      "entropy": 5.899684715270996,
      "epoch": 2.4527552101730836,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043971128018158693,
      "loss": 5.2428,
      "mean_token_accuracy": 0.2212805226445198,
      "num_tokens": 51791492.0,
      "step": 27775
    },
    {
      "entropy": 5.88948163986206,
      "epoch": 2.453196750264924,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004396898249557864,
      "loss": 5.3291,
      "mean_token_accuracy": 0.21388096064329148,
      "num_tokens": 51800906.0,
      "step": 27780
    },
    {
      "entropy": 5.835018491744995,
      "epoch": 2.453638290356764,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043966836650380587,
      "loss": 5.2089,
      "mean_token_accuracy": 0.22278151363134385,
      "num_tokens": 51809071.0,
      "step": 27785
    },
    {
      "entropy": 5.911023187637329,
      "epoch": 2.4540798304486047,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043964690482606566,
      "loss": 5.2794,
      "mean_token_accuracy": 0.21338130682706832,
      "num_tokens": 51817450.0,
      "step": 27790
    },
    {
      "entropy": 5.974888896942138,
      "epoch": 2.454521370540445,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043962543992298624,
      "loss": 5.3218,
      "mean_token_accuracy": 0.21255274266004562,
      "num_tokens": 51826400.0,
      "step": 27795
    },
    {
      "entropy": 5.9243310451507565,
      "epoch": 2.4549629106322852,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004396039717949881,
      "loss": 5.3112,
      "mean_token_accuracy": 0.2160378649830818,
      "num_tokens": 51835415.0,
      "step": 27800
    },
    {
      "entropy": 5.900667905807495,
      "epoch": 2.4554044507241257,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004395825004424917,
      "loss": 5.3317,
      "mean_token_accuracy": 0.2162179708480835,
      "num_tokens": 51844914.0,
      "step": 27805
    },
    {
      "entropy": 5.876917743682862,
      "epoch": 2.455845990815966,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004395610258659178,
      "loss": 5.3123,
      "mean_token_accuracy": 0.21062172055244446,
      "num_tokens": 51854178.0,
      "step": 27810
    },
    {
      "entropy": 5.932152795791626,
      "epoch": 2.4562875309078063,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043953954806568693,
      "loss": 5.416,
      "mean_token_accuracy": 0.2007034808397293,
      "num_tokens": 51863184.0,
      "step": 27815
    },
    {
      "entropy": 5.88009786605835,
      "epoch": 2.4567290709996468,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043951806704221994,
      "loss": 5.3402,
      "mean_token_accuracy": 0.21763793230056763,
      "num_tokens": 51872651.0,
      "step": 27820
    },
    {
      "entropy": 5.879913663864135,
      "epoch": 2.4571706110914873,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004394965827959376,
      "loss": 5.2639,
      "mean_token_accuracy": 0.2185867816209793,
      "num_tokens": 51881527.0,
      "step": 27825
    },
    {
      "entropy": 6.017736864089966,
      "epoch": 2.4576121511833273,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043947509532726067,
      "loss": 5.3998,
      "mean_token_accuracy": 0.20268522650003434,
      "num_tokens": 51892083.0,
      "step": 27830
    },
    {
      "entropy": 5.900433206558228,
      "epoch": 2.458053691275168,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004394536046366102,
      "loss": 5.2271,
      "mean_token_accuracy": 0.21191155463457106,
      "num_tokens": 51900758.0,
      "step": 27835
    },
    {
      "entropy": 5.927733039855957,
      "epoch": 2.4584952313670083,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004394321107244072,
      "loss": 5.3477,
      "mean_token_accuracy": 0.2092591032385826,
      "num_tokens": 51910624.0,
      "step": 27840
    },
    {
      "entropy": 5.738432741165161,
      "epoch": 2.4589367714588484,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004394106135910726,
      "loss": 5.1105,
      "mean_token_accuracy": 0.220411516726017,
      "num_tokens": 51919306.0,
      "step": 27845
    },
    {
      "entropy": 5.9758576393127445,
      "epoch": 2.459378311550689,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043938911323702766,
      "loss": 5.3207,
      "mean_token_accuracy": 0.20858184844255448,
      "num_tokens": 51928650.0,
      "step": 27850
    },
    {
      "entropy": 5.97742190361023,
      "epoch": 2.4598198516425294,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004393676096626934,
      "loss": 5.3638,
      "mean_token_accuracy": 0.204018497467041,
      "num_tokens": 51939198.0,
      "step": 27855
    },
    {
      "entropy": 5.770755195617676,
      "epoch": 2.4602613917343694,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043934610286849114,
      "loss": 5.0553,
      "mean_token_accuracy": 0.24026506394147873,
      "num_tokens": 51948641.0,
      "step": 27860
    },
    {
      "entropy": 5.914965724945068,
      "epoch": 2.46070293182621,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004393245928548423,
      "loss": 5.2688,
      "mean_token_accuracy": 0.21708640605211257,
      "num_tokens": 51958258.0,
      "step": 27865
    },
    {
      "entropy": 5.860828876495361,
      "epoch": 2.46114447191805,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043930307962216806,
      "loss": 5.2348,
      "mean_token_accuracy": 0.21189690977334977,
      "num_tokens": 51967295.0,
      "step": 27870
    },
    {
      "entropy": 5.7797869682312015,
      "epoch": 2.4615860120098905,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043928156317088994,
      "loss": 5.1903,
      "mean_token_accuracy": 0.22431680262088777,
      "num_tokens": 51976450.0,
      "step": 27875
    },
    {
      "entropy": 5.927054786682129,
      "epoch": 2.462027552101731,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004392600435014295,
      "loss": 5.2914,
      "mean_token_accuracy": 0.20668855905532837,
      "num_tokens": 51985782.0,
      "step": 27880
    },
    {
      "entropy": 6.017324447631836,
      "epoch": 2.462469092193571,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043923852061420815,
      "loss": 5.3556,
      "mean_token_accuracy": 0.21441836953163146,
      "num_tokens": 51994745.0,
      "step": 27885
    },
    {
      "entropy": 5.9987016201019285,
      "epoch": 2.4629106322854115,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004392169945096476,
      "loss": 5.3539,
      "mean_token_accuracy": 0.20949921607971192,
      "num_tokens": 52004369.0,
      "step": 27890
    },
    {
      "entropy": 5.920764970779419,
      "epoch": 2.463352172377252,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043919546518816953,
      "loss": 5.3409,
      "mean_token_accuracy": 0.21419924795627593,
      "num_tokens": 52012556.0,
      "step": 27895
    },
    {
      "entropy": 5.899825096130371,
      "epoch": 2.463793712469092,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004391739326501957,
      "loss": 5.389,
      "mean_token_accuracy": 0.2033422186970711,
      "num_tokens": 52022854.0,
      "step": 27900
    },
    {
      "entropy": 5.994083070755005,
      "epoch": 2.4642352525609326,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004391523968961479,
      "loss": 5.3232,
      "mean_token_accuracy": 0.2052223190665245,
      "num_tokens": 52032924.0,
      "step": 27905
    },
    {
      "entropy": 5.938289213180542,
      "epoch": 2.464676792652773,
      "grad_norm": 1.09375,
      "learning_rate": 0.000439130857926448,
      "loss": 5.2887,
      "mean_token_accuracy": 0.21431934237480163,
      "num_tokens": 52042625.0,
      "step": 27910
    },
    {
      "entropy": 5.894624185562134,
      "epoch": 2.465118332744613,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004391093157415179,
      "loss": 5.3076,
      "mean_token_accuracy": 0.21275169402360916,
      "num_tokens": 52051964.0,
      "step": 27915
    },
    {
      "entropy": 5.8985813617706295,
      "epoch": 2.4655598728364536,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004390877703417797,
      "loss": 5.3094,
      "mean_token_accuracy": 0.2128138393163681,
      "num_tokens": 52060877.0,
      "step": 27920
    },
    {
      "entropy": 5.951144361495972,
      "epoch": 2.4660014129282937,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043906622172765533,
      "loss": 5.3994,
      "mean_token_accuracy": 0.20297456234693528,
      "num_tokens": 52069585.0,
      "step": 27925
    },
    {
      "entropy": 5.947353982925415,
      "epoch": 2.466442953020134,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043904466989956706,
      "loss": 5.3513,
      "mean_token_accuracy": 0.208660988509655,
      "num_tokens": 52078435.0,
      "step": 27930
    },
    {
      "entropy": 5.970871353149414,
      "epoch": 2.4668844931119747,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043902311485793694,
      "loss": 5.3306,
      "mean_token_accuracy": 0.20841691344976426,
      "num_tokens": 52088227.0,
      "step": 27935
    },
    {
      "entropy": 5.928849411010742,
      "epoch": 2.4673260332038147,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004390015566031873,
      "loss": 5.3313,
      "mean_token_accuracy": 0.20298113077878951,
      "num_tokens": 52098101.0,
      "step": 27940
    },
    {
      "entropy": 5.939661073684692,
      "epoch": 2.467767573295655,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043897999513574045,
      "loss": 5.2789,
      "mean_token_accuracy": 0.21402155011892318,
      "num_tokens": 52107370.0,
      "step": 27945
    },
    {
      "entropy": 5.828591346740723,
      "epoch": 2.4682091133874957,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043895843045601874,
      "loss": 5.297,
      "mean_token_accuracy": 0.21102840900421144,
      "num_tokens": 52116808.0,
      "step": 27950
    },
    {
      "entropy": 5.906517267227173,
      "epoch": 2.4686506534793358,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043893686256444463,
      "loss": 5.2412,
      "mean_token_accuracy": 0.2173311248421669,
      "num_tokens": 52125724.0,
      "step": 27955
    },
    {
      "entropy": 6.007397270202636,
      "epoch": 2.4690921935711763,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004389152914614407,
      "loss": 5.3862,
      "mean_token_accuracy": 0.20011391788721083,
      "num_tokens": 52135530.0,
      "step": 27960
    },
    {
      "entropy": 5.953218030929565,
      "epoch": 2.4695337336630168,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004388937171474293,
      "loss": 5.3403,
      "mean_token_accuracy": 0.21143974959850312,
      "num_tokens": 52144195.0,
      "step": 27965
    },
    {
      "entropy": 5.968507289886475,
      "epoch": 2.469975273754857,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043887213962283335,
      "loss": 5.3653,
      "mean_token_accuracy": 0.2077018067240715,
      "num_tokens": 52153675.0,
      "step": 27970
    },
    {
      "entropy": 5.894257259368897,
      "epoch": 2.4704168138466973,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043885055888807516,
      "loss": 5.3076,
      "mean_token_accuracy": 0.20812157839536666,
      "num_tokens": 52163455.0,
      "step": 27975
    },
    {
      "entropy": 5.905441617965698,
      "epoch": 2.470858353938538,
      "grad_norm": 1.125,
      "learning_rate": 0.0004388289749435779,
      "loss": 5.3033,
      "mean_token_accuracy": 0.21255824565887452,
      "num_tokens": 52172054.0,
      "step": 27980
    },
    {
      "entropy": 5.966552257537842,
      "epoch": 2.471299894030378,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004388073877897641,
      "loss": 5.3911,
      "mean_token_accuracy": 0.20351026505231856,
      "num_tokens": 52181667.0,
      "step": 27985
    },
    {
      "entropy": 5.95760293006897,
      "epoch": 2.4717414341222184,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043878579742705675,
      "loss": 5.2903,
      "mean_token_accuracy": 0.21335915923118592,
      "num_tokens": 52190472.0,
      "step": 27990
    },
    {
      "entropy": 5.890383625030518,
      "epoch": 2.472182974214059,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043876420385587886,
      "loss": 5.2232,
      "mean_token_accuracy": 0.22119523584842682,
      "num_tokens": 52199905.0,
      "step": 27995
    },
    {
      "entropy": 5.888234949111938,
      "epoch": 2.472624514305899,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043874260707665327,
      "loss": 5.3321,
      "mean_token_accuracy": 0.20483122915029525,
      "num_tokens": 52209270.0,
      "step": 28000
    },
    {
      "entropy": 5.873456001281738,
      "epoch": 2.4730660543977394,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004387210070898032,
      "loss": 5.3169,
      "mean_token_accuracy": 0.20711778104305267,
      "num_tokens": 52218144.0,
      "step": 28005
    },
    {
      "entropy": 5.969706869125366,
      "epoch": 2.4735075944895795,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004386994038957516,
      "loss": 5.2459,
      "mean_token_accuracy": 0.2161187544465065,
      "num_tokens": 52227128.0,
      "step": 28010
    },
    {
      "entropy": 5.918254756927491,
      "epoch": 2.47394913458142,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004386777974949219,
      "loss": 5.2983,
      "mean_token_accuracy": 0.20576366037130356,
      "num_tokens": 52235823.0,
      "step": 28015
    },
    {
      "entropy": 5.914078283309936,
      "epoch": 2.4743906746732605,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004386561878877372,
      "loss": 5.424,
      "mean_token_accuracy": 0.20312600284814836,
      "num_tokens": 52245510.0,
      "step": 28020
    },
    {
      "entropy": 5.965360116958618,
      "epoch": 2.4748322147651005,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043863457507462076,
      "loss": 5.2644,
      "mean_token_accuracy": 0.2169720321893692,
      "num_tokens": 52254583.0,
      "step": 28025
    },
    {
      "entropy": 5.960397434234619,
      "epoch": 2.475273754856941,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004386129590559962,
      "loss": 5.3061,
      "mean_token_accuracy": 0.2091424733400345,
      "num_tokens": 52264778.0,
      "step": 28030
    },
    {
      "entropy": 5.8918462753295895,
      "epoch": 2.4757152949487815,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004385913398322867,
      "loss": 5.3405,
      "mean_token_accuracy": 0.21004281789064408,
      "num_tokens": 52274959.0,
      "step": 28035
    },
    {
      "entropy": 5.9388001441955565,
      "epoch": 2.4761568350406216,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000438569717403916,
      "loss": 5.2941,
      "mean_token_accuracy": 0.20815666913986205,
      "num_tokens": 52284606.0,
      "step": 28040
    },
    {
      "entropy": 5.915408945083618,
      "epoch": 2.476598375132462,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004385480917713075,
      "loss": 5.3302,
      "mean_token_accuracy": 0.20920236259698868,
      "num_tokens": 52294571.0,
      "step": 28045
    },
    {
      "entropy": 5.925726222991943,
      "epoch": 2.4770399152243026,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004385264629348849,
      "loss": 5.2735,
      "mean_token_accuracy": 0.2120954692363739,
      "num_tokens": 52303575.0,
      "step": 28050
    },
    {
      "entropy": 5.862115573883057,
      "epoch": 2.4774814553161426,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004385048308950719,
      "loss": 5.2579,
      "mean_token_accuracy": 0.21174015998840331,
      "num_tokens": 52312863.0,
      "step": 28055
    },
    {
      "entropy": 5.927686357498169,
      "epoch": 2.477922995407983,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004384831956522922,
      "loss": 5.3248,
      "mean_token_accuracy": 0.21127449721097946,
      "num_tokens": 52322493.0,
      "step": 28060
    },
    {
      "entropy": 5.946255111694336,
      "epoch": 2.478364535499823,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043846155720696976,
      "loss": 5.2369,
      "mean_token_accuracy": 0.21889870166778563,
      "num_tokens": 52331159.0,
      "step": 28065
    },
    {
      "entropy": 5.935913515090943,
      "epoch": 2.4788060755916637,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00043843991555952847,
      "loss": 5.2885,
      "mean_token_accuracy": 0.20991544872522355,
      "num_tokens": 52340276.0,
      "step": 28070
    },
    {
      "entropy": 5.907207345962524,
      "epoch": 2.479247615683504,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043841827071039197,
      "loss": 5.3101,
      "mean_token_accuracy": 0.21458655297756196,
      "num_tokens": 52349666.0,
      "step": 28075
    },
    {
      "entropy": 5.835780000686645,
      "epoch": 2.479689155775344,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043839662265998463,
      "loss": 5.2466,
      "mean_token_accuracy": 0.2235080972313881,
      "num_tokens": 52358238.0,
      "step": 28080
    },
    {
      "entropy": 5.9305213451385494,
      "epoch": 2.4801306958671847,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043837497140873037,
      "loss": 5.2706,
      "mean_token_accuracy": 0.222248674929142,
      "num_tokens": 52367663.0,
      "step": 28085
    },
    {
      "entropy": 5.919798517227173,
      "epoch": 2.480572235959025,
      "grad_norm": 1.25,
      "learning_rate": 0.00043835331695705326,
      "loss": 5.2842,
      "mean_token_accuracy": 0.21688373684883117,
      "num_tokens": 52377320.0,
      "step": 28090
    },
    {
      "entropy": 5.924546957015991,
      "epoch": 2.4810137760508653,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004383316593053777,
      "loss": 5.3431,
      "mean_token_accuracy": 0.20754344314336776,
      "num_tokens": 52386231.0,
      "step": 28095
    },
    {
      "entropy": 5.962829113006592,
      "epoch": 2.4814553161427058,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004383099984541277,
      "loss": 5.3831,
      "mean_token_accuracy": 0.19705422073602677,
      "num_tokens": 52395487.0,
      "step": 28100
    },
    {
      "entropy": 5.95886344909668,
      "epoch": 2.4818968562345463,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043828833440372775,
      "loss": 5.304,
      "mean_token_accuracy": 0.21045807749032974,
      "num_tokens": 52404372.0,
      "step": 28105
    },
    {
      "entropy": 5.840684652328491,
      "epoch": 2.4823383963263863,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004382666671546022,
      "loss": 5.1579,
      "mean_token_accuracy": 0.22269117087125778,
      "num_tokens": 52413156.0,
      "step": 28110
    },
    {
      "entropy": 5.9180396556854244,
      "epoch": 2.482779936418227,
      "grad_norm": 1.125,
      "learning_rate": 0.00043824499670717547,
      "loss": 5.3014,
      "mean_token_accuracy": 0.21077678650617598,
      "num_tokens": 52422782.0,
      "step": 28115
    },
    {
      "entropy": 5.983758974075317,
      "epoch": 2.4832214765100673,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043822332306187213,
      "loss": 5.3206,
      "mean_token_accuracy": 0.20122966170310974,
      "num_tokens": 52432363.0,
      "step": 28120
    },
    {
      "entropy": 5.925285959243775,
      "epoch": 2.4836630166019074,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004382016462191167,
      "loss": 5.2458,
      "mean_token_accuracy": 0.21362931728363038,
      "num_tokens": 52440995.0,
      "step": 28125
    },
    {
      "entropy": 5.886008310317993,
      "epoch": 2.484104556693748,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043817996617933393,
      "loss": 5.2762,
      "mean_token_accuracy": 0.21155107468366624,
      "num_tokens": 52450135.0,
      "step": 28130
    },
    {
      "entropy": 5.948771047592163,
      "epoch": 2.4845460967855884,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004381582829429483,
      "loss": 5.2726,
      "mean_token_accuracy": 0.21573687195777894,
      "num_tokens": 52459065.0,
      "step": 28135
    },
    {
      "entropy": 5.934777545928955,
      "epoch": 2.4849876368774284,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004381365965103848,
      "loss": 5.2701,
      "mean_token_accuracy": 0.21677020639181138,
      "num_tokens": 52468382.0,
      "step": 28140
    },
    {
      "entropy": 5.898145914077759,
      "epoch": 2.485429176969269,
      "grad_norm": 1.125,
      "learning_rate": 0.0004381149068820681,
      "loss": 5.2793,
      "mean_token_accuracy": 0.2134518012404442,
      "num_tokens": 52476940.0,
      "step": 28145
    },
    {
      "entropy": 5.952098321914673,
      "epoch": 2.485870717061109,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043809321405842314,
      "loss": 5.3421,
      "mean_token_accuracy": 0.20668397694826127,
      "num_tokens": 52486496.0,
      "step": 28150
    },
    {
      "entropy": 6.014197492599488,
      "epoch": 2.4863122571529495,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004380715180398749,
      "loss": 5.3543,
      "mean_token_accuracy": 0.20514219701290132,
      "num_tokens": 52495778.0,
      "step": 28155
    },
    {
      "entropy": 5.936219549179077,
      "epoch": 2.48675379724479,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004380498188268483,
      "loss": 5.2062,
      "mean_token_accuracy": 0.22205806225538255,
      "num_tokens": 52504855.0,
      "step": 28160
    },
    {
      "entropy": 5.897814607620239,
      "epoch": 2.48719533733663,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043802811641976855,
      "loss": 5.2568,
      "mean_token_accuracy": 0.20964664816856385,
      "num_tokens": 52513159.0,
      "step": 28165
    },
    {
      "entropy": 5.969333219528198,
      "epoch": 2.4876368774284705,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004380064108190607,
      "loss": 5.3113,
      "mean_token_accuracy": 0.2087924987077713,
      "num_tokens": 52521869.0,
      "step": 28170
    },
    {
      "entropy": 5.947543096542359,
      "epoch": 2.488078417520311,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004379847020251499,
      "loss": 5.3841,
      "mean_token_accuracy": 0.20693899244070052,
      "num_tokens": 52531886.0,
      "step": 28175
    },
    {
      "entropy": 5.914919424057007,
      "epoch": 2.488519957612151,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043796299003846155,
      "loss": 5.252,
      "mean_token_accuracy": 0.21773796826601027,
      "num_tokens": 52540331.0,
      "step": 28180
    },
    {
      "entropy": 5.955459356307983,
      "epoch": 2.4889614977039916,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043794127485942094,
      "loss": 5.3946,
      "mean_token_accuracy": 0.20597021132707596,
      "num_tokens": 52549686.0,
      "step": 28185
    },
    {
      "entropy": 5.870624828338623,
      "epoch": 2.489403037795832,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004379195564884534,
      "loss": 5.2484,
      "mean_token_accuracy": 0.21358037292957305,
      "num_tokens": 52558722.0,
      "step": 28190
    },
    {
      "entropy": 5.973939657211304,
      "epoch": 2.489844577887672,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004378978349259844,
      "loss": 5.359,
      "mean_token_accuracy": 0.2102392941713333,
      "num_tokens": 52568884.0,
      "step": 28195
    },
    {
      "entropy": 5.939207220077515,
      "epoch": 2.4902861179795126,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043787611017243944,
      "loss": 5.3072,
      "mean_token_accuracy": 0.21111804097890854,
      "num_tokens": 52578519.0,
      "step": 28200
    },
    {
      "entropy": 5.887236547470093,
      "epoch": 2.4907276580713527,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043785438222824406,
      "loss": 5.2261,
      "mean_token_accuracy": 0.21920061558485032,
      "num_tokens": 52588496.0,
      "step": 28205
    },
    {
      "entropy": 5.957219839096069,
      "epoch": 2.491169198163193,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043783265109382407,
      "loss": 5.339,
      "mean_token_accuracy": 0.20118987262248994,
      "num_tokens": 52598463.0,
      "step": 28210
    },
    {
      "entropy": 5.871705818176269,
      "epoch": 2.4916107382550337,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043781091676960507,
      "loss": 5.2325,
      "mean_token_accuracy": 0.21095334887504577,
      "num_tokens": 52607257.0,
      "step": 28215
    },
    {
      "entropy": 5.972517299652099,
      "epoch": 2.4920522783468737,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004377891792560127,
      "loss": 5.342,
      "mean_token_accuracy": 0.2064761996269226,
      "num_tokens": 52616434.0,
      "step": 28220
    },
    {
      "entropy": 5.966226959228516,
      "epoch": 2.492493818438714,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043776743855347287,
      "loss": 5.4207,
      "mean_token_accuracy": 0.20471108853816986,
      "num_tokens": 52625564.0,
      "step": 28225
    },
    {
      "entropy": 5.975445938110352,
      "epoch": 2.4929353585305547,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043774569466241165,
      "loss": 5.4278,
      "mean_token_accuracy": 0.20721568912267685,
      "num_tokens": 52634081.0,
      "step": 28230
    },
    {
      "entropy": 5.872003602981567,
      "epoch": 2.4933768986223948,
      "grad_norm": 1.0,
      "learning_rate": 0.0004377239475832547,
      "loss": 5.2795,
      "mean_token_accuracy": 0.21807872056961058,
      "num_tokens": 52643722.0,
      "step": 28235
    },
    {
      "entropy": 5.927830839157105,
      "epoch": 2.4938184387142353,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004377021973164281,
      "loss": 5.2912,
      "mean_token_accuracy": 0.2126550629734993,
      "num_tokens": 52652831.0,
      "step": 28240
    },
    {
      "entropy": 6.061373567581176,
      "epoch": 2.4942599788060758,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004376804438623582,
      "loss": 5.4512,
      "mean_token_accuracy": 0.20207064896821975,
      "num_tokens": 52663139.0,
      "step": 28245
    },
    {
      "entropy": 5.96367335319519,
      "epoch": 2.494701518897916,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043765868722147076,
      "loss": 5.3089,
      "mean_token_accuracy": 0.21384908109903336,
      "num_tokens": 52672281.0,
      "step": 28250
    },
    {
      "entropy": 5.892288303375244,
      "epoch": 2.4951430589897563,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043763692739419224,
      "loss": 5.2602,
      "mean_token_accuracy": 0.2156843960285187,
      "num_tokens": 52681644.0,
      "step": 28255
    },
    {
      "entropy": 5.989990329742431,
      "epoch": 2.495584599081597,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004376151643809487,
      "loss": 5.3551,
      "mean_token_accuracy": 0.20546325594186782,
      "num_tokens": 52690706.0,
      "step": 28260
    },
    {
      "entropy": 5.994533634185791,
      "epoch": 2.496026139173437,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004375933981821667,
      "loss": 5.3155,
      "mean_token_accuracy": 0.2104247733950615,
      "num_tokens": 52700391.0,
      "step": 28265
    },
    {
      "entropy": 5.9365886688232425,
      "epoch": 2.4964676792652774,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043757162879827245,
      "loss": 5.3283,
      "mean_token_accuracy": 0.20595275312662126,
      "num_tokens": 52709586.0,
      "step": 28270
    },
    {
      "entropy": 5.837874126434326,
      "epoch": 2.496909219357118,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004375498562296924,
      "loss": 5.285,
      "mean_token_accuracy": 0.21209441721439362,
      "num_tokens": 52718926.0,
      "step": 28275
    },
    {
      "entropy": 5.9122637748718265,
      "epoch": 2.497350759448958,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004375280804768532,
      "loss": 5.2553,
      "mean_token_accuracy": 0.2144363984465599,
      "num_tokens": 52728311.0,
      "step": 28280
    },
    {
      "entropy": 6.03780951499939,
      "epoch": 2.4977922995407984,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004375063015401813,
      "loss": 5.358,
      "mean_token_accuracy": 0.20636511445045472,
      "num_tokens": 52737758.0,
      "step": 28285
    },
    {
      "entropy": 5.976330089569092,
      "epoch": 2.4982338396326385,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043748451942010337,
      "loss": 5.3591,
      "mean_token_accuracy": 0.21436747908592224,
      "num_tokens": 52746959.0,
      "step": 28290
    },
    {
      "entropy": 5.931586694717407,
      "epoch": 2.498675379724479,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004374627341170461,
      "loss": 5.3491,
      "mean_token_accuracy": 0.20356991440057753,
      "num_tokens": 52757315.0,
      "step": 28295
    },
    {
      "entropy": 5.974539136886596,
      "epoch": 2.4991169198163194,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004374409456314362,
      "loss": 5.3505,
      "mean_token_accuracy": 0.21006207764148713,
      "num_tokens": 52766927.0,
      "step": 28300
    },
    {
      "entropy": 5.955719518661499,
      "epoch": 2.4995584599081595,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004374191539637007,
      "loss": 5.3414,
      "mean_token_accuracy": 0.20606182813644408,
      "num_tokens": 52775103.0,
      "step": 28305
    },
    {
      "entropy": 5.94163498878479,
      "epoch": 2.5,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004373973591142663,
      "loss": 5.3757,
      "mean_token_accuracy": 0.20909145772457122,
      "num_tokens": 52785281.0,
      "step": 28310
    },
    {
      "entropy": 5.8582042217254635,
      "epoch": 2.5004415400918405,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004373755610835599,
      "loss": 5.2023,
      "mean_token_accuracy": 0.21804287731647493,
      "num_tokens": 52793593.0,
      "step": 28315
    },
    {
      "entropy": 5.958576202392578,
      "epoch": 2.5008830801836806,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043735375987200877,
      "loss": 5.2939,
      "mean_token_accuracy": 0.2117796942591667,
      "num_tokens": 52803103.0,
      "step": 28320
    },
    {
      "entropy": 6.00177583694458,
      "epoch": 2.501324620275521,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004373319554800397,
      "loss": 5.3447,
      "mean_token_accuracy": 0.2099045068025589,
      "num_tokens": 52811928.0,
      "step": 28325
    },
    {
      "entropy": 5.875462436676026,
      "epoch": 2.501766160367361,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043731014790808,
      "loss": 5.178,
      "mean_token_accuracy": 0.22477563768625258,
      "num_tokens": 52821352.0,
      "step": 28330
    },
    {
      "entropy": 5.863584709167481,
      "epoch": 2.5022077004592016,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004372883371565568,
      "loss": 5.2274,
      "mean_token_accuracy": 0.22447099387645722,
      "num_tokens": 52830688.0,
      "step": 28335
    },
    {
      "entropy": 5.913503551483155,
      "epoch": 2.502649240551042,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043726652322589735,
      "loss": 5.2553,
      "mean_token_accuracy": 0.2177971363067627,
      "num_tokens": 52840804.0,
      "step": 28340
    },
    {
      "entropy": 5.9881562232971195,
      "epoch": 2.503090780642882,
      "grad_norm": 1.140625,
      "learning_rate": 0.000437244706116529,
      "loss": 5.2936,
      "mean_token_accuracy": 0.20659193843603135,
      "num_tokens": 52850476.0,
      "step": 28345
    },
    {
      "entropy": 5.964049434661865,
      "epoch": 2.5035323207347226,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043722288582887913,
      "loss": 5.3741,
      "mean_token_accuracy": 0.20472928285598754,
      "num_tokens": 52859254.0,
      "step": 28350
    },
    {
      "entropy": 5.870206308364868,
      "epoch": 2.503973860826563,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004372010623633753,
      "loss": 5.3016,
      "mean_token_accuracy": 0.20638857036828995,
      "num_tokens": 52867814.0,
      "step": 28355
    },
    {
      "entropy": 5.911526393890381,
      "epoch": 2.504415400918403,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043717923572044473,
      "loss": 5.1868,
      "mean_token_accuracy": 0.215077805519104,
      "num_tokens": 52876958.0,
      "step": 28360
    },
    {
      "entropy": 5.901006603240967,
      "epoch": 2.5048569410102437,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043715740590051527,
      "loss": 5.2295,
      "mean_token_accuracy": 0.2152526095509529,
      "num_tokens": 52885978.0,
      "step": 28365
    },
    {
      "entropy": 5.85199556350708,
      "epoch": 2.505298481102084,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004371355729040144,
      "loss": 5.2133,
      "mean_token_accuracy": 0.21765127032995224,
      "num_tokens": 52894081.0,
      "step": 28370
    },
    {
      "entropy": 5.884133148193359,
      "epoch": 2.5057400211939242,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004371137367313699,
      "loss": 5.3272,
      "mean_token_accuracy": 0.21255928426980972,
      "num_tokens": 52904012.0,
      "step": 28375
    },
    {
      "entropy": 5.951147937774659,
      "epoch": 2.5061815612857647,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043709189738300956,
      "loss": 5.2962,
      "mean_token_accuracy": 0.20573777407407762,
      "num_tokens": 52913319.0,
      "step": 28380
    },
    {
      "entropy": 5.939668941497803,
      "epoch": 2.5066231013776052,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004370700548593611,
      "loss": 5.2872,
      "mean_token_accuracy": 0.20664036124944687,
      "num_tokens": 52923539.0,
      "step": 28385
    },
    {
      "entropy": 5.872171306610108,
      "epoch": 2.5070646414694453,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004370482091608523,
      "loss": 5.228,
      "mean_token_accuracy": 0.21582048386335373,
      "num_tokens": 52931856.0,
      "step": 28390
    },
    {
      "entropy": 5.9336615085601805,
      "epoch": 2.507506181561286,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043702636028791135,
      "loss": 5.3444,
      "mean_token_accuracy": 0.2015314444899559,
      "num_tokens": 52942270.0,
      "step": 28395
    },
    {
      "entropy": 5.877850198745728,
      "epoch": 2.5079477216531263,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004370045082409662,
      "loss": 5.2366,
      "mean_token_accuracy": 0.21975299715995789,
      "num_tokens": 52951503.0,
      "step": 28400
    },
    {
      "entropy": 5.932393312454224,
      "epoch": 2.5083892617449663,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004369826530204449,
      "loss": 5.3192,
      "mean_token_accuracy": 0.20674848258495332,
      "num_tokens": 52960288.0,
      "step": 28405
    },
    {
      "entropy": 5.940851354598999,
      "epoch": 2.508830801836807,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043696079462677537,
      "loss": 5.3447,
      "mean_token_accuracy": 0.2133128121495247,
      "num_tokens": 52969313.0,
      "step": 28410
    },
    {
      "entropy": 5.906246185302734,
      "epoch": 2.5092723419286473,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004369389330603861,
      "loss": 5.3399,
      "mean_token_accuracy": 0.20142117142677307,
      "num_tokens": 52978924.0,
      "step": 28415
    },
    {
      "entropy": 5.940451812744141,
      "epoch": 2.5097138820204874,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043691706832170527,
      "loss": 5.3077,
      "mean_token_accuracy": 0.20725252777338027,
      "num_tokens": 52989027.0,
      "step": 28420
    },
    {
      "entropy": 5.981883096694946,
      "epoch": 2.510155422112328,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043689520041116116,
      "loss": 5.3229,
      "mean_token_accuracy": 0.21033522635698318,
      "num_tokens": 52999436.0,
      "step": 28425
    },
    {
      "entropy": 5.922626399993897,
      "epoch": 2.5105969622041684,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043687332932918204,
      "loss": 5.32,
      "mean_token_accuracy": 0.2131922259926796,
      "num_tokens": 53008153.0,
      "step": 28430
    },
    {
      "entropy": 5.871616983413697,
      "epoch": 2.5110385022960084,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004368514550761965,
      "loss": 5.3147,
      "mean_token_accuracy": 0.21697081476449967,
      "num_tokens": 53017344.0,
      "step": 28435
    },
    {
      "entropy": 5.8953453540802006,
      "epoch": 2.511480042387849,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043682957765263304,
      "loss": 5.2412,
      "mean_token_accuracy": 0.2200936958193779,
      "num_tokens": 53026966.0,
      "step": 28440
    },
    {
      "entropy": 5.907599830627442,
      "epoch": 2.511921582479689,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043680769705892013,
      "loss": 5.2675,
      "mean_token_accuracy": 0.21449509263038635,
      "num_tokens": 53035879.0,
      "step": 28445
    },
    {
      "entropy": 5.901587915420532,
      "epoch": 2.5123631225715295,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004367858132954866,
      "loss": 5.3102,
      "mean_token_accuracy": 0.20541498363018035,
      "num_tokens": 53044727.0,
      "step": 28450
    },
    {
      "entropy": 5.801606702804565,
      "epoch": 2.51280466266337,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004367639263627609,
      "loss": 5.1271,
      "mean_token_accuracy": 0.2262067198753357,
      "num_tokens": 53054152.0,
      "step": 28455
    },
    {
      "entropy": 5.932039642333985,
      "epoch": 2.51324620275521,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004367420362611718,
      "loss": 5.3266,
      "mean_token_accuracy": 0.20783986896276474,
      "num_tokens": 53063259.0,
      "step": 28460
    },
    {
      "entropy": 5.959280490875244,
      "epoch": 2.5136877428470505,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004367201429911483,
      "loss": 5.259,
      "mean_token_accuracy": 0.21542590111494064,
      "num_tokens": 53072139.0,
      "step": 28465
    },
    {
      "entropy": 5.940527820587159,
      "epoch": 2.5141292829388906,
      "grad_norm": 1.25,
      "learning_rate": 0.0004366982465531192,
      "loss": 5.31,
      "mean_token_accuracy": 0.20827726274728775,
      "num_tokens": 53081833.0,
      "step": 28470
    },
    {
      "entropy": 5.8209006786346436,
      "epoch": 2.514570823030731,
      "grad_norm": 1.125,
      "learning_rate": 0.0004366763469475134,
      "loss": 5.2574,
      "mean_token_accuracy": 0.22331633120775224,
      "num_tokens": 53091952.0,
      "step": 28475
    },
    {
      "entropy": 5.877666807174682,
      "epoch": 2.5150123631225716,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004366544441747599,
      "loss": 5.3191,
      "mean_token_accuracy": 0.214091719686985,
      "num_tokens": 53101619.0,
      "step": 28480
    },
    {
      "entropy": 5.9487916946411135,
      "epoch": 2.5154539032144116,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004366325382352878,
      "loss": 5.2719,
      "mean_token_accuracy": 0.2041495516896248,
      "num_tokens": 53110931.0,
      "step": 28485
    },
    {
      "entropy": 5.9291887283325195,
      "epoch": 2.515895443306252,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043661062912952607,
      "loss": 5.2954,
      "mean_token_accuracy": 0.2095704957842827,
      "num_tokens": 53119949.0,
      "step": 28490
    },
    {
      "entropy": 5.883489418029785,
      "epoch": 2.5163369833980926,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004365887168579042,
      "loss": 5.2944,
      "mean_token_accuracy": 0.21369004398584365,
      "num_tokens": 53129048.0,
      "step": 28495
    },
    {
      "entropy": 5.96072154045105,
      "epoch": 2.5167785234899327,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004365668014208512,
      "loss": 5.3466,
      "mean_token_accuracy": 0.20351312160491944,
      "num_tokens": 53139121.0,
      "step": 28500
    },
    {
      "entropy": 5.970879602432251,
      "epoch": 2.517220063581773,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043654488281879645,
      "loss": 5.2517,
      "mean_token_accuracy": 0.2110189288854599,
      "num_tokens": 53148098.0,
      "step": 28505
    },
    {
      "entropy": 5.903145742416382,
      "epoch": 2.5176616036736137,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004365229610521693,
      "loss": 5.2957,
      "mean_token_accuracy": 0.20802702605724335,
      "num_tokens": 53156992.0,
      "step": 28510
    },
    {
      "entropy": 5.907380819320679,
      "epoch": 2.5181031437654537,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043650103612139917,
      "loss": 5.3135,
      "mean_token_accuracy": 0.2089930847287178,
      "num_tokens": 53166985.0,
      "step": 28515
    },
    {
      "entropy": 5.994011354446411,
      "epoch": 2.5185446838572942,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004364791080269156,
      "loss": 5.3148,
      "mean_token_accuracy": 0.2114214077591896,
      "num_tokens": 53176924.0,
      "step": 28520
    },
    {
      "entropy": 5.955840253829956,
      "epoch": 2.5189862239491347,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004364571767691482,
      "loss": 5.3078,
      "mean_token_accuracy": 0.21268728822469712,
      "num_tokens": 53186966.0,
      "step": 28525
    },
    {
      "entropy": 5.951052904129028,
      "epoch": 2.519427764040975,
      "grad_norm": 1.25,
      "learning_rate": 0.00043643524234852655,
      "loss": 5.3562,
      "mean_token_accuracy": 0.20697386413812638,
      "num_tokens": 53196147.0,
      "step": 28530
    },
    {
      "entropy": 5.863919115066528,
      "epoch": 2.5198693041328153,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043641330476548024,
      "loss": 5.2232,
      "mean_token_accuracy": 0.22046734690666198,
      "num_tokens": 53204446.0,
      "step": 28535
    },
    {
      "entropy": 5.906647777557373,
      "epoch": 2.520310844224656,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004363913640204392,
      "loss": 5.3324,
      "mean_token_accuracy": 0.2043924868106842,
      "num_tokens": 53213528.0,
      "step": 28540
    },
    {
      "entropy": 5.950580596923828,
      "epoch": 2.520752384316496,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000436369420113833,
      "loss": 5.3415,
      "mean_token_accuracy": 0.20853210836648942,
      "num_tokens": 53223574.0,
      "step": 28545
    },
    {
      "entropy": 5.828311920166016,
      "epoch": 2.5211939244083363,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043634747304609173,
      "loss": 5.1916,
      "mean_token_accuracy": 0.21760771572589874,
      "num_tokens": 53233011.0,
      "step": 28550
    },
    {
      "entropy": 5.939109802246094,
      "epoch": 2.521635464500177,
      "grad_norm": 1.125,
      "learning_rate": 0.0004363255228176452,
      "loss": 5.3707,
      "mean_token_accuracy": 0.2100078657269478,
      "num_tokens": 53241835.0,
      "step": 28555
    },
    {
      "entropy": 6.01583423614502,
      "epoch": 2.522077004592017,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004363035694289234,
      "loss": 5.307,
      "mean_token_accuracy": 0.2139008477330208,
      "num_tokens": 53250666.0,
      "step": 28560
    },
    {
      "entropy": 5.914329099655151,
      "epoch": 2.5225185446838574,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043628161288035644,
      "loss": 5.2558,
      "mean_token_accuracy": 0.2156945899128914,
      "num_tokens": 53259386.0,
      "step": 28565
    },
    {
      "entropy": 5.886383390426635,
      "epoch": 2.522960084775698,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043625965317237435,
      "loss": 5.3179,
      "mean_token_accuracy": 0.20526764839887618,
      "num_tokens": 53268760.0,
      "step": 28570
    },
    {
      "entropy": 5.928327035903931,
      "epoch": 2.523401624867538,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043623769030540747,
      "loss": 5.3016,
      "mean_token_accuracy": 0.21419572532176973,
      "num_tokens": 53277817.0,
      "step": 28575
    },
    {
      "entropy": 5.916297101974488,
      "epoch": 2.5238431649593784,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043621572427988585,
      "loss": 5.2124,
      "mean_token_accuracy": 0.22644717395305633,
      "num_tokens": 53286699.0,
      "step": 28580
    },
    {
      "entropy": 5.903418254852295,
      "epoch": 2.5242847050512185,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004361937550962399,
      "loss": 5.2493,
      "mean_token_accuracy": 0.2131398618221283,
      "num_tokens": 53295431.0,
      "step": 28585
    },
    {
      "entropy": 5.858805131912232,
      "epoch": 2.524726245143059,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043617178275489997,
      "loss": 5.237,
      "mean_token_accuracy": 0.21085935980081558,
      "num_tokens": 53304986.0,
      "step": 28590
    },
    {
      "entropy": 5.9547985076904295,
      "epoch": 2.5251677852348995,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043614980725629663,
      "loss": 5.3755,
      "mean_token_accuracy": 0.20055561810731887,
      "num_tokens": 53313908.0,
      "step": 28595
    },
    {
      "entropy": 5.934130907058716,
      "epoch": 2.5256093253267395,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043612782860086,
      "loss": 5.2517,
      "mean_token_accuracy": 0.21327352225780488,
      "num_tokens": 53323553.0,
      "step": 28600
    },
    {
      "entropy": 6.000094747543335,
      "epoch": 2.52605086541858,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043610584678902096,
      "loss": 5.362,
      "mean_token_accuracy": 0.20739827752113343,
      "num_tokens": 53333236.0,
      "step": 28605
    },
    {
      "entropy": 5.934924554824829,
      "epoch": 2.52649240551042,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00043608386182120997,
      "loss": 5.3292,
      "mean_token_accuracy": 0.206508831679821,
      "num_tokens": 53342344.0,
      "step": 28610
    },
    {
      "entropy": 5.867638683319091,
      "epoch": 2.5269339456022606,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004360618736978577,
      "loss": 5.2999,
      "mean_token_accuracy": 0.21623359322547914,
      "num_tokens": 53351764.0,
      "step": 28615
    },
    {
      "entropy": 5.792739009857177,
      "epoch": 2.527375485694101,
      "grad_norm": 1.171875,
      "learning_rate": 0.000436039882419395,
      "loss": 5.1472,
      "mean_token_accuracy": 0.22465425431728364,
      "num_tokens": 53360667.0,
      "step": 28620
    },
    {
      "entropy": 5.947846174240112,
      "epoch": 2.527817025785941,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043601788798625253,
      "loss": 5.3269,
      "mean_token_accuracy": 0.2047562852501869,
      "num_tokens": 53370011.0,
      "step": 28625
    },
    {
      "entropy": 5.951759147644043,
      "epoch": 2.5282585658777816,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043599589039886125,
      "loss": 5.3922,
      "mean_token_accuracy": 0.20921133309602738,
      "num_tokens": 53379035.0,
      "step": 28630
    },
    {
      "entropy": 5.93406491279602,
      "epoch": 2.528700105969622,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000435973889657652,
      "loss": 5.2805,
      "mean_token_accuracy": 0.21744174510240555,
      "num_tokens": 53387739.0,
      "step": 28635
    },
    {
      "entropy": 5.985143232345581,
      "epoch": 2.529141646061462,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043595188576305585,
      "loss": 5.3622,
      "mean_token_accuracy": 0.20534148514270784,
      "num_tokens": 53398219.0,
      "step": 28640
    },
    {
      "entropy": 5.998261022567749,
      "epoch": 2.5295831861533027,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043592987871550373,
      "loss": 5.3368,
      "mean_token_accuracy": 0.2070773273706436,
      "num_tokens": 53407337.0,
      "step": 28645
    },
    {
      "entropy": 6.0023926258087155,
      "epoch": 2.530024726245143,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043590786851542687,
      "loss": 5.3537,
      "mean_token_accuracy": 0.20647776424884795,
      "num_tokens": 53416223.0,
      "step": 28650
    },
    {
      "entropy": 5.897646093368531,
      "epoch": 2.5304662663369832,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043588585516325635,
      "loss": 5.2594,
      "mean_token_accuracy": 0.2101878583431244,
      "num_tokens": 53424719.0,
      "step": 28655
    },
    {
      "entropy": 5.942400312423706,
      "epoch": 2.5309078064288237,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043586383865942347,
      "loss": 5.281,
      "mean_token_accuracy": 0.21358522176742553,
      "num_tokens": 53433763.0,
      "step": 28660
    },
    {
      "entropy": 5.9134784698486325,
      "epoch": 2.5313493465206642,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043584181900435947,
      "loss": 5.314,
      "mean_token_accuracy": 0.21268401443958282,
      "num_tokens": 53443270.0,
      "step": 28665
    },
    {
      "entropy": 6.006677055358887,
      "epoch": 2.5317908866125043,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004358197961984957,
      "loss": 5.2848,
      "mean_token_accuracy": 0.20642824321985245,
      "num_tokens": 53451555.0,
      "step": 28670
    },
    {
      "entropy": 5.952741575241089,
      "epoch": 2.532232426704345,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004357977702422635,
      "loss": 5.3277,
      "mean_token_accuracy": 0.2049025610089302,
      "num_tokens": 53460845.0,
      "step": 28675
    },
    {
      "entropy": 5.836100292205811,
      "epoch": 2.5326739667961853,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043577574113609456,
      "loss": 5.2348,
      "mean_token_accuracy": 0.22356937527656556,
      "num_tokens": 53470169.0,
      "step": 28680
    },
    {
      "entropy": 5.901159143447876,
      "epoch": 2.5331155068880253,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004357537088804203,
      "loss": 5.3004,
      "mean_token_accuracy": 0.2164873421192169,
      "num_tokens": 53478484.0,
      "step": 28685
    },
    {
      "entropy": 5.891032028198242,
      "epoch": 2.533557046979866,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004357316734756722,
      "loss": 5.2905,
      "mean_token_accuracy": 0.20862107574939728,
      "num_tokens": 53488791.0,
      "step": 28690
    },
    {
      "entropy": 5.9253740310668945,
      "epoch": 2.5339985870717063,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004357096349222821,
      "loss": 5.2739,
      "mean_token_accuracy": 0.2140319287776947,
      "num_tokens": 53498310.0,
      "step": 28695
    },
    {
      "entropy": 5.928901815414429,
      "epoch": 2.5344401271635464,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043568759322068165,
      "loss": 5.3179,
      "mean_token_accuracy": 0.2158198431134224,
      "num_tokens": 53508145.0,
      "step": 28700
    },
    {
      "entropy": 5.9225671768188475,
      "epoch": 2.534881667255387,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004356655483713026,
      "loss": 5.3383,
      "mean_token_accuracy": 0.21078818738460542,
      "num_tokens": 53517473.0,
      "step": 28705
    },
    {
      "entropy": 5.903563117980957,
      "epoch": 2.5353232073472274,
      "grad_norm": 1.125,
      "learning_rate": 0.00043564350037457683,
      "loss": 5.3363,
      "mean_token_accuracy": 0.20895638167858124,
      "num_tokens": 53526625.0,
      "step": 28710
    },
    {
      "entropy": 5.889246606826783,
      "epoch": 2.5357647474390674,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043562144923093627,
      "loss": 5.1876,
      "mean_token_accuracy": 0.21991955488920212,
      "num_tokens": 53535480.0,
      "step": 28715
    },
    {
      "entropy": 5.881613826751709,
      "epoch": 2.536206287530908,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043559939494081286,
      "loss": 5.3501,
      "mean_token_accuracy": 0.2082623705267906,
      "num_tokens": 53544368.0,
      "step": 28720
    },
    {
      "entropy": 5.880730104446411,
      "epoch": 2.536647827622748,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004355773375046387,
      "loss": 5.2991,
      "mean_token_accuracy": 0.21242556273937224,
      "num_tokens": 53553235.0,
      "step": 28725
    },
    {
      "entropy": 5.952645301818848,
      "epoch": 2.5370893677145885,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004355552769228457,
      "loss": 5.3365,
      "mean_token_accuracy": 0.1999655619263649,
      "num_tokens": 53562635.0,
      "step": 28730
    },
    {
      "entropy": 5.930995035171509,
      "epoch": 2.537530907806429,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004355332131958663,
      "loss": 5.2549,
      "mean_token_accuracy": 0.20702547281980516,
      "num_tokens": 53571394.0,
      "step": 28735
    },
    {
      "entropy": 5.987088251113891,
      "epoch": 2.537972447898269,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004355111463241324,
      "loss": 5.3422,
      "mean_token_accuracy": 0.20331409871578215,
      "num_tokens": 53580718.0,
      "step": 28740
    },
    {
      "entropy": 5.956633710861206,
      "epoch": 2.5384139879901095,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004354890763080765,
      "loss": 5.376,
      "mean_token_accuracy": 0.20374861508607864,
      "num_tokens": 53590120.0,
      "step": 28745
    },
    {
      "entropy": 5.950815629959107,
      "epoch": 2.5388555280819496,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004354670031481309,
      "loss": 5.3111,
      "mean_token_accuracy": 0.2089494436979294,
      "num_tokens": 53599783.0,
      "step": 28750
    },
    {
      "entropy": 5.984275579452515,
      "epoch": 2.53929706817379,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004354449268447279,
      "loss": 5.2806,
      "mean_token_accuracy": 0.20911658704280853,
      "num_tokens": 53608663.0,
      "step": 28755
    },
    {
      "entropy": 5.967377519607544,
      "epoch": 2.5397386082656306,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043542284739830006,
      "loss": 5.386,
      "mean_token_accuracy": 0.20675139129161835,
      "num_tokens": 53618239.0,
      "step": 28760
    },
    {
      "entropy": 5.972046661376953,
      "epoch": 2.5401801483574706,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043540076480927994,
      "loss": 5.331,
      "mean_token_accuracy": 0.21449458003044128,
      "num_tokens": 53627683.0,
      "step": 28765
    },
    {
      "entropy": 5.892813396453858,
      "epoch": 2.540621688449311,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043537867907809995,
      "loss": 5.2186,
      "mean_token_accuracy": 0.21554412841796874,
      "num_tokens": 53636980.0,
      "step": 28770
    },
    {
      "entropy": 5.895318555831909,
      "epoch": 2.5410632285411516,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004353565902051929,
      "loss": 5.2931,
      "mean_token_accuracy": 0.20791402459144592,
      "num_tokens": 53647310.0,
      "step": 28775
    },
    {
      "entropy": 5.915997171401978,
      "epoch": 2.5415047686329917,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043533449819099145,
      "loss": 5.3608,
      "mean_token_accuracy": 0.209612612426281,
      "num_tokens": 53657433.0,
      "step": 28780
    },
    {
      "entropy": 5.926881885528564,
      "epoch": 2.541946308724832,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004353124030359284,
      "loss": 5.2815,
      "mean_token_accuracy": 0.20911362767219543,
      "num_tokens": 53666969.0,
      "step": 28785
    },
    {
      "entropy": 5.990406799316406,
      "epoch": 2.5423878488166727,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043529030474043657,
      "loss": 5.3167,
      "mean_token_accuracy": 0.2074495419859886,
      "num_tokens": 53676332.0,
      "step": 28790
    },
    {
      "entropy": 5.931350040435791,
      "epoch": 2.5428293889085127,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043526820330494877,
      "loss": 5.3546,
      "mean_token_accuracy": 0.20901401937007905,
      "num_tokens": 53685766.0,
      "step": 28795
    },
    {
      "entropy": 5.955107259750366,
      "epoch": 2.5432709290003532,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004352460987298981,
      "loss": 5.2834,
      "mean_token_accuracy": 0.20815054774284364,
      "num_tokens": 53694744.0,
      "step": 28800
    },
    {
      "entropy": 5.85746431350708,
      "epoch": 2.5437124690921937,
      "grad_norm": 1.125,
      "learning_rate": 0.0004352239910157174,
      "loss": 5.1696,
      "mean_token_accuracy": 0.22617327123880387,
      "num_tokens": 53704266.0,
      "step": 28805
    },
    {
      "entropy": 5.886471748352051,
      "epoch": 2.544154009184034,
      "grad_norm": 1.25,
      "learning_rate": 0.00043520188016283977,
      "loss": 5.2477,
      "mean_token_accuracy": 0.2141741171479225,
      "num_tokens": 53712070.0,
      "step": 28810
    },
    {
      "entropy": 5.905902481079101,
      "epoch": 2.5445955492758743,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043517976617169857,
      "loss": 5.3034,
      "mean_token_accuracy": 0.205298313498497,
      "num_tokens": 53720152.0,
      "step": 28815
    },
    {
      "entropy": 5.816996335983276,
      "epoch": 2.5450370893677148,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004351576490427268,
      "loss": 5.1404,
      "mean_token_accuracy": 0.2271268680691719,
      "num_tokens": 53729451.0,
      "step": 28820
    },
    {
      "entropy": 5.883787345886231,
      "epoch": 2.545478629459555,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004351355287763578,
      "loss": 5.3037,
      "mean_token_accuracy": 0.21037020087242125,
      "num_tokens": 53738866.0,
      "step": 28825
    },
    {
      "entropy": 5.876320266723633,
      "epoch": 2.5459201695513953,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043511340537302485,
      "loss": 5.1717,
      "mean_token_accuracy": 0.2207638218998909,
      "num_tokens": 53748006.0,
      "step": 28830
    },
    {
      "entropy": 5.9471558094024655,
      "epoch": 2.546361709643236,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004350912788331613,
      "loss": 5.2175,
      "mean_token_accuracy": 0.22418534904718398,
      "num_tokens": 53756412.0,
      "step": 28835
    },
    {
      "entropy": 5.945276689529419,
      "epoch": 2.546803249735076,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004350691491572007,
      "loss": 5.328,
      "mean_token_accuracy": 0.2092160016298294,
      "num_tokens": 53765509.0,
      "step": 28840
    },
    {
      "entropy": 5.898266363143921,
      "epoch": 2.5472447898269164,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004350470163455764,
      "loss": 5.3403,
      "mean_token_accuracy": 0.20662661343812944,
      "num_tokens": 53774883.0,
      "step": 28845
    },
    {
      "entropy": 5.946175479888916,
      "epoch": 2.547686329918757,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043502488039872224,
      "loss": 5.3134,
      "mean_token_accuracy": 0.20263454914093018,
      "num_tokens": 53783650.0,
      "step": 28850
    },
    {
      "entropy": 5.969151163101197,
      "epoch": 2.548127870010597,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043500274131707156,
      "loss": 5.3685,
      "mean_token_accuracy": 0.20084885209798814,
      "num_tokens": 53793031.0,
      "step": 28855
    },
    {
      "entropy": 6.040827083587646,
      "epoch": 2.5485694101024374,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004349805991010582,
      "loss": 5.3812,
      "mean_token_accuracy": 0.19600951671600342,
      "num_tokens": 53803967.0,
      "step": 28860
    },
    {
      "entropy": 5.89522008895874,
      "epoch": 2.5490109501942775,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043495845375111594,
      "loss": 5.2967,
      "mean_token_accuracy": 0.20914786905050278,
      "num_tokens": 53813575.0,
      "step": 28865
    },
    {
      "entropy": 5.906645679473877,
      "epoch": 2.549452490286118,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043493630526767846,
      "loss": 5.2746,
      "mean_token_accuracy": 0.21194277703762054,
      "num_tokens": 53823250.0,
      "step": 28870
    },
    {
      "entropy": 5.971550989151001,
      "epoch": 2.5498940303779585,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043491415365117984,
      "loss": 5.2808,
      "mean_token_accuracy": 0.2145892858505249,
      "num_tokens": 53832300.0,
      "step": 28875
    },
    {
      "entropy": 5.903762626647949,
      "epoch": 2.5503355704697985,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043489199890205376,
      "loss": 5.2324,
      "mean_token_accuracy": 0.21277001053094863,
      "num_tokens": 53841385.0,
      "step": 28880
    },
    {
      "entropy": 5.977278232574463,
      "epoch": 2.550777110561639,
      "grad_norm": 1.125,
      "learning_rate": 0.00043486984102073445,
      "loss": 5.3547,
      "mean_token_accuracy": 0.20658586472272872,
      "num_tokens": 53850349.0,
      "step": 28885
    },
    {
      "entropy": 5.847180891036987,
      "epoch": 2.551218650653479,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043484768000765585,
      "loss": 5.1936,
      "mean_token_accuracy": 0.2215432345867157,
      "num_tokens": 53859547.0,
      "step": 28890
    },
    {
      "entropy": 5.905997467041016,
      "epoch": 2.5516601907453196,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004348255158632521,
      "loss": 5.3066,
      "mean_token_accuracy": 0.21306932866573333,
      "num_tokens": 53868497.0,
      "step": 28895
    },
    {
      "entropy": 5.853203582763672,
      "epoch": 2.55210173083716,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043480334858795737,
      "loss": 5.2463,
      "mean_token_accuracy": 0.2165161669254303,
      "num_tokens": 53877200.0,
      "step": 28900
    },
    {
      "entropy": 5.873076486587524,
      "epoch": 2.552543270929,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000434781178182206,
      "loss": 5.2025,
      "mean_token_accuracy": 0.2198852077126503,
      "num_tokens": 53887081.0,
      "step": 28905
    },
    {
      "entropy": 5.872828149795533,
      "epoch": 2.5529848110208406,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004347590046464322,
      "loss": 5.244,
      "mean_token_accuracy": 0.21027422845363616,
      "num_tokens": 53896334.0,
      "step": 28910
    },
    {
      "entropy": 5.885881853103638,
      "epoch": 2.553426351112681,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043473682798107034,
      "loss": 5.2733,
      "mean_token_accuracy": 0.21409823894500732,
      "num_tokens": 53905658.0,
      "step": 28915
    },
    {
      "entropy": 5.998106384277344,
      "epoch": 2.553867891204521,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004347146481865549,
      "loss": 5.3866,
      "mean_token_accuracy": 0.20790007561445237,
      "num_tokens": 53915747.0,
      "step": 28920
    },
    {
      "entropy": 5.984354305267334,
      "epoch": 2.5543094312963617,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004346924652633203,
      "loss": 5.3314,
      "mean_token_accuracy": 0.20827525705099106,
      "num_tokens": 53924636.0,
      "step": 28925
    },
    {
      "entropy": 5.857732534408569,
      "epoch": 2.554750971388202,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004346702792118011,
      "loss": 5.2379,
      "mean_token_accuracy": 0.2204352781176567,
      "num_tokens": 53934104.0,
      "step": 28930
    },
    {
      "entropy": 5.92335844039917,
      "epoch": 2.5551925114800422,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043464809003243197,
      "loss": 5.3359,
      "mean_token_accuracy": 0.21232685595750808,
      "num_tokens": 53943055.0,
      "step": 28935
    },
    {
      "entropy": 5.926897144317627,
      "epoch": 2.5556340515718827,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004346258977256475,
      "loss": 5.2987,
      "mean_token_accuracy": 0.21971941739320755,
      "num_tokens": 53952489.0,
      "step": 28940
    },
    {
      "entropy": 5.949771070480347,
      "epoch": 2.5560755916637232,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043460370229188257,
      "loss": 5.2291,
      "mean_token_accuracy": 0.21267286390066148,
      "num_tokens": 53962274.0,
      "step": 28945
    },
    {
      "entropy": 5.932826185226441,
      "epoch": 2.5565171317555633,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043458150373157183,
      "loss": 5.3258,
      "mean_token_accuracy": 0.20649837851524352,
      "num_tokens": 53971152.0,
      "step": 28950
    },
    {
      "entropy": 5.922298097610474,
      "epoch": 2.5569586718474038,
      "grad_norm": 1.25,
      "learning_rate": 0.0004345593020451503,
      "loss": 5.2889,
      "mean_token_accuracy": 0.2098290890455246,
      "num_tokens": 53979510.0,
      "step": 28955
    },
    {
      "entropy": 5.829650068283081,
      "epoch": 2.5574002119392443,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043453709723305264,
      "loss": 5.2313,
      "mean_token_accuracy": 0.22153203785419465,
      "num_tokens": 53987917.0,
      "step": 28960
    },
    {
      "entropy": 5.919157886505127,
      "epoch": 2.5578417520310843,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043451488929571405,
      "loss": 5.325,
      "mean_token_accuracy": 0.2069593071937561,
      "num_tokens": 53997924.0,
      "step": 28965
    },
    {
      "entropy": 5.9032261848449705,
      "epoch": 2.558283292122925,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004344926782335695,
      "loss": 5.2953,
      "mean_token_accuracy": 0.20828854590654372,
      "num_tokens": 54006505.0,
      "step": 28970
    },
    {
      "entropy": 5.98134298324585,
      "epoch": 2.5587248322147653,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043447046404705404,
      "loss": 5.4096,
      "mean_token_accuracy": 0.2034696653485298,
      "num_tokens": 54016794.0,
      "step": 28975
    },
    {
      "entropy": 5.850054454803467,
      "epoch": 2.5591663723066054,
      "grad_norm": 1.125,
      "learning_rate": 0.0004344482467366029,
      "loss": 5.1935,
      "mean_token_accuracy": 0.22895340472459794,
      "num_tokens": 54025995.0,
      "step": 28980
    },
    {
      "entropy": 5.925017309188843,
      "epoch": 2.559607912398446,
      "grad_norm": 1.25,
      "learning_rate": 0.0004344260263026513,
      "loss": 5.2688,
      "mean_token_accuracy": 0.21536701917648315,
      "num_tokens": 54035501.0,
      "step": 28985
    },
    {
      "entropy": 5.928806114196777,
      "epoch": 2.5600494524902864,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004344038027456346,
      "loss": 5.3559,
      "mean_token_accuracy": 0.20691070407629014,
      "num_tokens": 54045283.0,
      "step": 28990
    },
    {
      "entropy": 5.947187900543213,
      "epoch": 2.5604909925821264,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004343815760659879,
      "loss": 5.3706,
      "mean_token_accuracy": 0.20990268439054488,
      "num_tokens": 54056250.0,
      "step": 28995
    },
    {
      "entropy": 5.945185279846191,
      "epoch": 2.560932532673967,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004343593462641469,
      "loss": 5.2508,
      "mean_token_accuracy": 0.2210957035422325,
      "num_tokens": 54065225.0,
      "step": 29000
    },
    {
      "entropy": 5.9554932594299315,
      "epoch": 2.561374072765807,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004343371133405469,
      "loss": 5.389,
      "mean_token_accuracy": 0.21101811975240709,
      "num_tokens": 54074056.0,
      "step": 29005
    },
    {
      "entropy": 5.9439328670501705,
      "epoch": 2.5618156128576475,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043431487729562345,
      "loss": 5.2505,
      "mean_token_accuracy": 0.21484581828117372,
      "num_tokens": 54083202.0,
      "step": 29010
    },
    {
      "entropy": 5.88990273475647,
      "epoch": 2.562257152949488,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004342926381298122,
      "loss": 5.1793,
      "mean_token_accuracy": 0.2260661691427231,
      "num_tokens": 54092088.0,
      "step": 29015
    },
    {
      "entropy": 5.932715177536011,
      "epoch": 2.562698693041328,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004342703958435487,
      "loss": 5.3819,
      "mean_token_accuracy": 0.2072764277458191,
      "num_tokens": 54101113.0,
      "step": 29020
    },
    {
      "entropy": 5.878082513809204,
      "epoch": 2.5631402331331685,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004342481504372688,
      "loss": 5.2952,
      "mean_token_accuracy": 0.21570076644420624,
      "num_tokens": 54109832.0,
      "step": 29025
    },
    {
      "entropy": 5.935173559188843,
      "epoch": 2.5635817732250086,
      "grad_norm": 1.125,
      "learning_rate": 0.00043422590191140813,
      "loss": 5.2315,
      "mean_token_accuracy": 0.21406220644712448,
      "num_tokens": 54117783.0,
      "step": 29030
    },
    {
      "entropy": 5.961007499694825,
      "epoch": 2.564023313316849,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004342036502664026,
      "loss": 5.3778,
      "mean_token_accuracy": 0.2004447266459465,
      "num_tokens": 54126539.0,
      "step": 29035
    },
    {
      "entropy": 5.914991617202759,
      "epoch": 2.5644648534086896,
      "grad_norm": 1.25,
      "learning_rate": 0.0004341813955026882,
      "loss": 5.207,
      "mean_token_accuracy": 0.21951550245285034,
      "num_tokens": 54135316.0,
      "step": 29040
    },
    {
      "entropy": 5.849492692947388,
      "epoch": 2.5649063935005296,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043415913762070074,
      "loss": 5.2518,
      "mean_token_accuracy": 0.21846236884593964,
      "num_tokens": 54144164.0,
      "step": 29045
    },
    {
      "entropy": 5.930370187759399,
      "epoch": 2.56534793359237,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004341368766208763,
      "loss": 5.3177,
      "mean_token_accuracy": 0.20827536284923553,
      "num_tokens": 54154030.0,
      "step": 29050
    },
    {
      "entropy": 5.864542865753174,
      "epoch": 2.5657894736842106,
      "grad_norm": 1.125,
      "learning_rate": 0.0004341146125036509,
      "loss": 5.2679,
      "mean_token_accuracy": 0.22156967967748642,
      "num_tokens": 54163507.0,
      "step": 29055
    },
    {
      "entropy": 5.83487720489502,
      "epoch": 2.5662310137760507,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043409234526946077,
      "loss": 5.2083,
      "mean_token_accuracy": 0.2218799263238907,
      "num_tokens": 54173016.0,
      "step": 29060
    },
    {
      "entropy": 5.962022876739502,
      "epoch": 2.566672553867891,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043407007491874216,
      "loss": 5.3304,
      "mean_token_accuracy": 0.21463083922863008,
      "num_tokens": 54181965.0,
      "step": 29065
    },
    {
      "entropy": 5.8906818389892575,
      "epoch": 2.5671140939597317,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043404780145193115,
      "loss": 5.2553,
      "mean_token_accuracy": 0.21381769478321075,
      "num_tokens": 54190574.0,
      "step": 29070
    },
    {
      "entropy": 5.962554693222046,
      "epoch": 2.5675556340515717,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043402552486946424,
      "loss": 5.3608,
      "mean_token_accuracy": 0.21075820028781891,
      "num_tokens": 54200105.0,
      "step": 29075
    },
    {
      "entropy": 5.929194450378418,
      "epoch": 2.567997174143412,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043400324517177765,
      "loss": 5.2944,
      "mean_token_accuracy": 0.21126194298267365,
      "num_tokens": 54209247.0,
      "step": 29080
    },
    {
      "entropy": 5.9450929164886475,
      "epoch": 2.5684387142352527,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043398096235930795,
      "loss": 5.2706,
      "mean_token_accuracy": 0.21751275062561035,
      "num_tokens": 54218122.0,
      "step": 29085
    },
    {
      "entropy": 5.860794591903686,
      "epoch": 2.5688802543270928,
      "grad_norm": 1.3125,
      "learning_rate": 0.00043395867643249164,
      "loss": 5.1713,
      "mean_token_accuracy": 0.2269056484103203,
      "num_tokens": 54226706.0,
      "step": 29090
    },
    {
      "entropy": 5.875028944015503,
      "epoch": 2.5693217944189333,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004339363873917652,
      "loss": 5.2673,
      "mean_token_accuracy": 0.21075802743434907,
      "num_tokens": 54235927.0,
      "step": 29095
    },
    {
      "entropy": 5.952543020248413,
      "epoch": 2.5697633345107738,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043391409523756537,
      "loss": 5.3926,
      "mean_token_accuracy": 0.20519386380910873,
      "num_tokens": 54245939.0,
      "step": 29100
    },
    {
      "entropy": 5.933586597442627,
      "epoch": 2.570204874602614,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043389179997032885,
      "loss": 5.3467,
      "mean_token_accuracy": 0.21148675233125686,
      "num_tokens": 54255726.0,
      "step": 29105
    },
    {
      "entropy": 5.876232051849366,
      "epoch": 2.5706464146944543,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004338695015904922,
      "loss": 5.2486,
      "mean_token_accuracy": 0.2137849286198616,
      "num_tokens": 54264485.0,
      "step": 29110
    },
    {
      "entropy": 5.861765909194946,
      "epoch": 2.571087954786295,
      "grad_norm": 1.25,
      "learning_rate": 0.0004338472000984924,
      "loss": 5.2368,
      "mean_token_accuracy": 0.2130070999264717,
      "num_tokens": 54272717.0,
      "step": 29115
    },
    {
      "entropy": 5.89320125579834,
      "epoch": 2.571529494878135,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043382489549476636,
      "loss": 5.1998,
      "mean_token_accuracy": 0.21799689382314683,
      "num_tokens": 54281806.0,
      "step": 29120
    },
    {
      "entropy": 5.903325080871582,
      "epoch": 2.5719710349699754,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000433802587779751,
      "loss": 5.3077,
      "mean_token_accuracy": 0.21344600468873978,
      "num_tokens": 54291147.0,
      "step": 29125
    },
    {
      "entropy": 5.91637921333313,
      "epoch": 2.572412575061816,
      "grad_norm": 1.25,
      "learning_rate": 0.00043378027695388315,
      "loss": 5.3191,
      "mean_token_accuracy": 0.2077517494559288,
      "num_tokens": 54299696.0,
      "step": 29130
    },
    {
      "entropy": 5.980043172836304,
      "epoch": 2.572854115153656,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043375796301760007,
      "loss": 5.2837,
      "mean_token_accuracy": 0.2219064399600029,
      "num_tokens": 54309320.0,
      "step": 29135
    },
    {
      "entropy": 6.008489942550659,
      "epoch": 2.5732956552454964,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004337356459713387,
      "loss": 5.3924,
      "mean_token_accuracy": 0.21088528782129287,
      "num_tokens": 54318794.0,
      "step": 29140
    },
    {
      "entropy": 5.916824674606323,
      "epoch": 2.5737371953373365,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004337133258155363,
      "loss": 5.3094,
      "mean_token_accuracy": 0.2117357909679413,
      "num_tokens": 54327792.0,
      "step": 29145
    },
    {
      "entropy": 5.942623233795166,
      "epoch": 2.574178735429177,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043369100255063016,
      "loss": 5.314,
      "mean_token_accuracy": 0.20946352034807206,
      "num_tokens": 54337524.0,
      "step": 29150
    },
    {
      "entropy": 5.883197069168091,
      "epoch": 2.5746202755210175,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004336686761770575,
      "loss": 5.211,
      "mean_token_accuracy": 0.22219059020280837,
      "num_tokens": 54347409.0,
      "step": 29155
    },
    {
      "entropy": 5.828567123413086,
      "epoch": 2.5750618156128575,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043364634669525577,
      "loss": 5.2041,
      "mean_token_accuracy": 0.2217286169528961,
      "num_tokens": 54356744.0,
      "step": 29160
    },
    {
      "entropy": 5.914896154403687,
      "epoch": 2.575503355704698,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043362401410566224,
      "loss": 5.2973,
      "mean_token_accuracy": 0.20703740566968917,
      "num_tokens": 54365274.0,
      "step": 29165
    },
    {
      "entropy": 5.886529397964478,
      "epoch": 2.575944895796538,
      "grad_norm": 1.125,
      "learning_rate": 0.00043360167840871455,
      "loss": 5.2509,
      "mean_token_accuracy": 0.21746753603219987,
      "num_tokens": 54374448.0,
      "step": 29170
    },
    {
      "entropy": 5.861358594894409,
      "epoch": 2.5763864358883786,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043357933960485017,
      "loss": 5.2734,
      "mean_token_accuracy": 0.2125807672739029,
      "num_tokens": 54383599.0,
      "step": 29175
    },
    {
      "entropy": 5.859539937973023,
      "epoch": 2.576827975980219,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043355699769450674,
      "loss": 5.1758,
      "mean_token_accuracy": 0.22562425583600998,
      "num_tokens": 54392294.0,
      "step": 29180
    },
    {
      "entropy": 5.923239755630493,
      "epoch": 2.577269516072059,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043353465267812185,
      "loss": 5.2839,
      "mean_token_accuracy": 0.21551448553800584,
      "num_tokens": 54401854.0,
      "step": 29185
    },
    {
      "entropy": 5.938776350021362,
      "epoch": 2.5777110561638996,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004335123045561333,
      "loss": 5.3971,
      "mean_token_accuracy": 0.204318805038929,
      "num_tokens": 54411255.0,
      "step": 29190
    },
    {
      "entropy": 5.96297402381897,
      "epoch": 2.57815259625574,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043348995332897887,
      "loss": 5.3533,
      "mean_token_accuracy": 0.19693140387535096,
      "num_tokens": 54420437.0,
      "step": 29195
    },
    {
      "entropy": 5.887395811080933,
      "epoch": 2.57859413634758,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043346759899709643,
      "loss": 5.152,
      "mean_token_accuracy": 0.2253085419535637,
      "num_tokens": 54429166.0,
      "step": 29200
    },
    {
      "entropy": 5.959748935699463,
      "epoch": 2.5790356764394207,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004334452415609238,
      "loss": 5.2624,
      "mean_token_accuracy": 0.20990725457668305,
      "num_tokens": 54438060.0,
      "step": 29205
    },
    {
      "entropy": 5.837468004226684,
      "epoch": 2.579477216531261,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000433422881020899,
      "loss": 5.223,
      "mean_token_accuracy": 0.2167680576443672,
      "num_tokens": 54447503.0,
      "step": 29210
    },
    {
      "entropy": 5.824471759796142,
      "epoch": 2.579918756623101,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043340051737746,
      "loss": 5.235,
      "mean_token_accuracy": 0.21378451734781265,
      "num_tokens": 54456741.0,
      "step": 29215
    },
    {
      "entropy": 5.928438329696656,
      "epoch": 2.5803602967149417,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000433378150631045,
      "loss": 5.3025,
      "mean_token_accuracy": 0.20666094720363617,
      "num_tokens": 54466614.0,
      "step": 29220
    },
    {
      "entropy": 5.99570140838623,
      "epoch": 2.580801836806782,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004333557807820922,
      "loss": 5.2873,
      "mean_token_accuracy": 0.21680670380592346,
      "num_tokens": 54475191.0,
      "step": 29225
    },
    {
      "entropy": 5.9413035869598385,
      "epoch": 2.5812433768986223,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004333334078310396,
      "loss": 5.222,
      "mean_token_accuracy": 0.2192588582634926,
      "num_tokens": 54483924.0,
      "step": 29230
    },
    {
      "entropy": 5.780759525299072,
      "epoch": 2.5816849169904628,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043331103177832566,
      "loss": 5.221,
      "mean_token_accuracy": 0.2163703605532646,
      "num_tokens": 54493739.0,
      "step": 29235
    },
    {
      "entropy": 5.889431142807007,
      "epoch": 2.5821264570823033,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004332886526243886,
      "loss": 5.2689,
      "mean_token_accuracy": 0.21682532727718354,
      "num_tokens": 54503760.0,
      "step": 29240
    },
    {
      "entropy": 5.963654565811157,
      "epoch": 2.5825679971741433,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004332662703696669,
      "loss": 5.2547,
      "mean_token_accuracy": 0.22072449773550035,
      "num_tokens": 54512884.0,
      "step": 29245
    },
    {
      "entropy": 5.944212579727173,
      "epoch": 2.583009537265984,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004332438850145989,
      "loss": 5.3108,
      "mean_token_accuracy": 0.21221732646226882,
      "num_tokens": 54522747.0,
      "step": 29250
    },
    {
      "entropy": 5.9021240234375,
      "epoch": 2.5834510773578243,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004332214965596232,
      "loss": 5.2072,
      "mean_token_accuracy": 0.22079750448465346,
      "num_tokens": 54532109.0,
      "step": 29255
    },
    {
      "entropy": 5.916968774795532,
      "epoch": 2.5838926174496644,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004331991050051785,
      "loss": 5.2669,
      "mean_token_accuracy": 0.220901657640934,
      "num_tokens": 54541160.0,
      "step": 29260
    },
    {
      "entropy": 5.88804988861084,
      "epoch": 2.584334157541505,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043317671035170315,
      "loss": 5.2765,
      "mean_token_accuracy": 0.22038799673318862,
      "num_tokens": 54550162.0,
      "step": 29265
    },
    {
      "entropy": 5.8534423828125,
      "epoch": 2.5847756976333454,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043315431259963606,
      "loss": 5.1366,
      "mean_token_accuracy": 0.22177425771951675,
      "num_tokens": 54559355.0,
      "step": 29270
    },
    {
      "entropy": 5.853288030624389,
      "epoch": 2.5852172377251854,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043313191174941594,
      "loss": 5.1968,
      "mean_token_accuracy": 0.22519367486238479,
      "num_tokens": 54568628.0,
      "step": 29275
    },
    {
      "entropy": 5.906941699981689,
      "epoch": 2.585658777817026,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004331095078014816,
      "loss": 5.2693,
      "mean_token_accuracy": 0.2124227911233902,
      "num_tokens": 54578744.0,
      "step": 29280
    },
    {
      "entropy": 5.934917402267456,
      "epoch": 2.586100317908866,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043308710075627195,
      "loss": 5.309,
      "mean_token_accuracy": 0.21183923929929732,
      "num_tokens": 54588654.0,
      "step": 29285
    },
    {
      "entropy": 5.950028705596924,
      "epoch": 2.5865418580007065,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004330646906142259,
      "loss": 5.2606,
      "mean_token_accuracy": 0.21684273183345795,
      "num_tokens": 54597085.0,
      "step": 29290
    },
    {
      "entropy": 5.938215684890747,
      "epoch": 2.586983398092547,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004330422773757824,
      "loss": 5.3488,
      "mean_token_accuracy": 0.20935806334018708,
      "num_tokens": 54606151.0,
      "step": 29295
    },
    {
      "entropy": 5.905852460861206,
      "epoch": 2.587424938184387,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004330198610413806,
      "loss": 5.2311,
      "mean_token_accuracy": 0.21925524175167083,
      "num_tokens": 54615085.0,
      "step": 29300
    },
    {
      "entropy": 5.9175886631011965,
      "epoch": 2.5878664782762275,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004329974416114596,
      "loss": 5.2188,
      "mean_token_accuracy": 0.21819822043180465,
      "num_tokens": 54623895.0,
      "step": 29305
    },
    {
      "entropy": 5.884753751754761,
      "epoch": 2.5883080183680676,
      "grad_norm": 1.125,
      "learning_rate": 0.0004329750190864586,
      "loss": 5.1798,
      "mean_token_accuracy": 0.22388937324285507,
      "num_tokens": 54633096.0,
      "step": 29310
    },
    {
      "entropy": 5.923306655883789,
      "epoch": 2.588749558459908,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043295259346681677,
      "loss": 5.3423,
      "mean_token_accuracy": 0.20712548941373826,
      "num_tokens": 54642210.0,
      "step": 29315
    },
    {
      "entropy": 5.999179792404175,
      "epoch": 2.5891910985517486,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004329301647529734,
      "loss": 5.4341,
      "mean_token_accuracy": 0.19573965668678284,
      "num_tokens": 54651960.0,
      "step": 29320
    },
    {
      "entropy": 5.97430157661438,
      "epoch": 2.5896326386435886,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043290773294536814,
      "loss": 5.2871,
      "mean_token_accuracy": 0.21185322999954223,
      "num_tokens": 54661560.0,
      "step": 29325
    },
    {
      "entropy": 5.963971519470215,
      "epoch": 2.590074178735429,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043288529804443997,
      "loss": 5.2971,
      "mean_token_accuracy": 0.21004565060138702,
      "num_tokens": 54671302.0,
      "step": 29330
    },
    {
      "entropy": 5.903332710266113,
      "epoch": 2.5905157188272696,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00043286286005062875,
      "loss": 5.3186,
      "mean_token_accuracy": 0.21034468561410904,
      "num_tokens": 54681641.0,
      "step": 29335
    },
    {
      "entropy": 5.901215124130249,
      "epoch": 2.5909572589191097,
      "grad_norm": 1.125,
      "learning_rate": 0.00043284041896437386,
      "loss": 5.2803,
      "mean_token_accuracy": 0.20557858049869537,
      "num_tokens": 54690190.0,
      "step": 29340
    },
    {
      "entropy": 5.968325185775757,
      "epoch": 2.59139879901095,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004328179747861149,
      "loss": 5.3494,
      "mean_token_accuracy": 0.20240288972854614,
      "num_tokens": 54699164.0,
      "step": 29345
    },
    {
      "entropy": 5.821555423736572,
      "epoch": 2.5918403391027907,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043279552751629153,
      "loss": 5.1829,
      "mean_token_accuracy": 0.22319599390029907,
      "num_tokens": 54708509.0,
      "step": 29350
    },
    {
      "entropy": 5.917746925354004,
      "epoch": 2.5922818791946307,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043277307715534365,
      "loss": 5.3118,
      "mean_token_accuracy": 0.21134113818407058,
      "num_tokens": 54717382.0,
      "step": 29355
    },
    {
      "entropy": 5.889475774765015,
      "epoch": 2.592723419286471,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004327506237037109,
      "loss": 5.2895,
      "mean_token_accuracy": 0.2161715880036354,
      "num_tokens": 54726191.0,
      "step": 29360
    },
    {
      "entropy": 5.941412496566772,
      "epoch": 2.5931649593783117,
      "grad_norm": 1.09375,
      "learning_rate": 0.000432728167161833,
      "loss": 5.3027,
      "mean_token_accuracy": 0.21358927339315414,
      "num_tokens": 54736748.0,
      "step": 29365
    },
    {
      "entropy": 5.869662237167359,
      "epoch": 2.5936064994701518,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004327057075301501,
      "loss": 5.2446,
      "mean_token_accuracy": 0.21612711995840073,
      "num_tokens": 54746623.0,
      "step": 29370
    },
    {
      "entropy": 5.904717445373535,
      "epoch": 2.5940480395619923,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043268324480910203,
      "loss": 5.2215,
      "mean_token_accuracy": 0.22099438458681106,
      "num_tokens": 54755778.0,
      "step": 29375
    },
    {
      "entropy": 5.975928544998169,
      "epoch": 2.5944895796538328,
      "grad_norm": 1.0625,
      "learning_rate": 0.000432660778999129,
      "loss": 5.4008,
      "mean_token_accuracy": 0.2065340206027031,
      "num_tokens": 54765410.0,
      "step": 29380
    },
    {
      "entropy": 6.019135046005249,
      "epoch": 2.594931119745673,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004326383101006709,
      "loss": 5.4055,
      "mean_token_accuracy": 0.20079344063997268,
      "num_tokens": 54776115.0,
      "step": 29385
    },
    {
      "entropy": 5.924548768997193,
      "epoch": 2.5953726598375133,
      "grad_norm": 1.28125,
      "learning_rate": 0.000432615838114168,
      "loss": 5.2754,
      "mean_token_accuracy": 0.210279481112957,
      "num_tokens": 54784719.0,
      "step": 29390
    },
    {
      "entropy": 5.968031597137451,
      "epoch": 2.595814199929354,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043259336304006036,
      "loss": 5.222,
      "mean_token_accuracy": 0.21257797628641129,
      "num_tokens": 54794003.0,
      "step": 29395
    },
    {
      "entropy": 5.9461119174957275,
      "epoch": 2.596255740021194,
      "grad_norm": 1.125,
      "learning_rate": 0.0004325708848787884,
      "loss": 5.3138,
      "mean_token_accuracy": 0.21133118867874146,
      "num_tokens": 54803634.0,
      "step": 29400
    },
    {
      "entropy": 5.886228227615357,
      "epoch": 2.5966972801130344,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004325484036307925,
      "loss": 5.3312,
      "mean_token_accuracy": 0.21940651088953017,
      "num_tokens": 54813466.0,
      "step": 29405
    },
    {
      "entropy": 5.97582426071167,
      "epoch": 2.597138820204875,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004325259192965129,
      "loss": 5.2838,
      "mean_token_accuracy": 0.21172297298908233,
      "num_tokens": 54823855.0,
      "step": 29410
    },
    {
      "entropy": 5.989750099182129,
      "epoch": 2.597580360296715,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004325034318763902,
      "loss": 5.3011,
      "mean_token_accuracy": 0.21453199088573455,
      "num_tokens": 54832696.0,
      "step": 29415
    },
    {
      "entropy": 5.924095344543457,
      "epoch": 2.5980219003885554,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004324809413708648,
      "loss": 5.3638,
      "mean_token_accuracy": 0.20289274156093598,
      "num_tokens": 54841552.0,
      "step": 29420
    },
    {
      "entropy": 5.857829236984253,
      "epoch": 2.5984634404803955,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004324584477803773,
      "loss": 5.2531,
      "mean_token_accuracy": 0.21800061762332917,
      "num_tokens": 54851196.0,
      "step": 29425
    },
    {
      "entropy": 5.979892206192017,
      "epoch": 2.598904980572236,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043243595110536846,
      "loss": 5.2284,
      "mean_token_accuracy": 0.2143957331776619,
      "num_tokens": 54860334.0,
      "step": 29430
    },
    {
      "entropy": 5.927029657363891,
      "epoch": 2.5993465206640765,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043241345134627885,
      "loss": 5.2771,
      "mean_token_accuracy": 0.21104304641485214,
      "num_tokens": 54870056.0,
      "step": 29435
    },
    {
      "entropy": 5.907989025115967,
      "epoch": 2.5997880607559165,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043239094850354914,
      "loss": 5.3696,
      "mean_token_accuracy": 0.20459881126880647,
      "num_tokens": 54880788.0,
      "step": 29440
    },
    {
      "entropy": 5.896127653121948,
      "epoch": 2.600229600847757,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004323684425776203,
      "loss": 5.2832,
      "mean_token_accuracy": 0.21032752841711044,
      "num_tokens": 54890021.0,
      "step": 29445
    },
    {
      "entropy": 5.861667776107788,
      "epoch": 2.600671140939597,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043234593356893325,
      "loss": 5.1794,
      "mean_token_accuracy": 0.23029984086751937,
      "num_tokens": 54898884.0,
      "step": 29450
    },
    {
      "entropy": 5.913671684265137,
      "epoch": 2.6011126810314376,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043232342147792887,
      "loss": 5.3434,
      "mean_token_accuracy": 0.20486652702093125,
      "num_tokens": 54907742.0,
      "step": 29455
    },
    {
      "entropy": 5.963853788375855,
      "epoch": 2.601554221123278,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000432300906305048,
      "loss": 5.2981,
      "mean_token_accuracy": 0.2130737066268921,
      "num_tokens": 54916838.0,
      "step": 29460
    },
    {
      "entropy": 5.976876831054687,
      "epoch": 2.601995761215118,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004322783880507319,
      "loss": 5.3764,
      "mean_token_accuracy": 0.2048438623547554,
      "num_tokens": 54925942.0,
      "step": 29465
    },
    {
      "entropy": 5.872803592681885,
      "epoch": 2.6024373013069586,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043225586671542164,
      "loss": 5.2138,
      "mean_token_accuracy": 0.21165077835321428,
      "num_tokens": 54934611.0,
      "step": 29470
    },
    {
      "entropy": 5.9203156471252445,
      "epoch": 2.602878841398799,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004322333422995584,
      "loss": 5.2825,
      "mean_token_accuracy": 0.20769805163145066,
      "num_tokens": 54944354.0,
      "step": 29475
    },
    {
      "entropy": 5.914855194091797,
      "epoch": 2.603320381490639,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043221081480358337,
      "loss": 5.3476,
      "mean_token_accuracy": 0.20324004590511321,
      "num_tokens": 54954070.0,
      "step": 29480
    },
    {
      "entropy": 5.984734201431275,
      "epoch": 2.6037619215824797,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004321882842279379,
      "loss": 5.3708,
      "mean_token_accuracy": 0.20405154675245285,
      "num_tokens": 54963922.0,
      "step": 29485
    },
    {
      "entropy": 5.9289439678192135,
      "epoch": 2.60420346167432,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004321657505730634,
      "loss": 5.2371,
      "mean_token_accuracy": 0.21589307785034179,
      "num_tokens": 54973202.0,
      "step": 29490
    },
    {
      "entropy": 5.966213274002075,
      "epoch": 2.60464500176616,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043214321383940117,
      "loss": 5.2986,
      "mean_token_accuracy": 0.20985390543937682,
      "num_tokens": 54982318.0,
      "step": 29495
    },
    {
      "entropy": 5.843818473815918,
      "epoch": 2.6050865418580007,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043212067402739274,
      "loss": 5.1485,
      "mean_token_accuracy": 0.2278016224503517,
      "num_tokens": 54991397.0,
      "step": 29500
    },
    {
      "entropy": 5.996559906005859,
      "epoch": 2.605528081949841,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004320981311374797,
      "loss": 5.3655,
      "mean_token_accuracy": 0.2080049991607666,
      "num_tokens": 55001160.0,
      "step": 29505
    },
    {
      "entropy": 5.996462774276734,
      "epoch": 2.6059696220416813,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043207558517010364,
      "loss": 5.333,
      "mean_token_accuracy": 0.21404379457235337,
      "num_tokens": 55010311.0,
      "step": 29510
    },
    {
      "entropy": 5.865195035934448,
      "epoch": 2.6064111621335218,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004320530361257061,
      "loss": 5.2856,
      "mean_token_accuracy": 0.2132151708006859,
      "num_tokens": 55020017.0,
      "step": 29515
    },
    {
      "entropy": 5.95326943397522,
      "epoch": 2.6068527022253623,
      "grad_norm": 1.078125,
      "learning_rate": 0.000432030484004729,
      "loss": 5.3971,
      "mean_token_accuracy": 0.20018167942762374,
      "num_tokens": 55029930.0,
      "step": 29520
    },
    {
      "entropy": 5.940564918518066,
      "epoch": 2.6072942423172023,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043200792880761403,
      "loss": 5.2823,
      "mean_token_accuracy": 0.20763902813196183,
      "num_tokens": 55039065.0,
      "step": 29525
    },
    {
      "entropy": 5.905433559417725,
      "epoch": 2.607735782409043,
      "grad_norm": 1.109375,
      "learning_rate": 0.000431985370534803,
      "loss": 5.2079,
      "mean_token_accuracy": 0.21546036899089813,
      "num_tokens": 55047592.0,
      "step": 29530
    },
    {
      "entropy": 5.926668453216553,
      "epoch": 2.6081773225008833,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043196280918673783,
      "loss": 5.3167,
      "mean_token_accuracy": 0.20458817332983018,
      "num_tokens": 55058031.0,
      "step": 29535
    },
    {
      "entropy": 5.9155748844146725,
      "epoch": 2.6086188625927234,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043194024476386045,
      "loss": 5.3562,
      "mean_token_accuracy": 0.2128615066409111,
      "num_tokens": 55068283.0,
      "step": 29540
    },
    {
      "entropy": 5.872447156906128,
      "epoch": 2.609060402684564,
      "grad_norm": 1.296875,
      "learning_rate": 0.000431917677266613,
      "loss": 5.1571,
      "mean_token_accuracy": 0.2176583781838417,
      "num_tokens": 55077480.0,
      "step": 29545
    },
    {
      "entropy": 5.9417072296142575,
      "epoch": 2.6095019427764043,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043189510669543754,
      "loss": 5.3255,
      "mean_token_accuracy": 0.20946847796440124,
      "num_tokens": 55086783.0,
      "step": 29550
    },
    {
      "entropy": 5.924038982391357,
      "epoch": 2.6099434828682444,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004318725330507762,
      "loss": 5.3544,
      "mean_token_accuracy": 0.20497765392065048,
      "num_tokens": 55096861.0,
      "step": 29555
    },
    {
      "entropy": 5.920787191390991,
      "epoch": 2.610385022960085,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043184995633307094,
      "loss": 5.3201,
      "mean_token_accuracy": 0.21674314141273499,
      "num_tokens": 55107504.0,
      "step": 29560
    },
    {
      "entropy": 6.039354467391968,
      "epoch": 2.610826563051925,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004318273765427644,
      "loss": 5.4261,
      "mean_token_accuracy": 0.19989271759986876,
      "num_tokens": 55116620.0,
      "step": 29565
    },
    {
      "entropy": 5.960881471633911,
      "epoch": 2.6112681031437655,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004318047936802987,
      "loss": 5.2858,
      "mean_token_accuracy": 0.2104386046528816,
      "num_tokens": 55126428.0,
      "step": 29570
    },
    {
      "entropy": 5.908171224594116,
      "epoch": 2.611709643235606,
      "grad_norm": 1.125,
      "learning_rate": 0.0004317822077461164,
      "loss": 5.266,
      "mean_token_accuracy": 0.20790134370326996,
      "num_tokens": 55135369.0,
      "step": 29575
    },
    {
      "entropy": 5.942065668106079,
      "epoch": 2.612151183327446,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004317596187406596,
      "loss": 5.3136,
      "mean_token_accuracy": 0.20553473234176636,
      "num_tokens": 55144742.0,
      "step": 29580
    },
    {
      "entropy": 5.918109178543091,
      "epoch": 2.6125927234192865,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043173702666437114,
      "loss": 5.3606,
      "mean_token_accuracy": 0.2065993905067444,
      "num_tokens": 55154665.0,
      "step": 29585
    },
    {
      "entropy": 5.947984313964843,
      "epoch": 2.6130342635111266,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043171443151769347,
      "loss": 5.2821,
      "mean_token_accuracy": 0.21488139629364014,
      "num_tokens": 55163288.0,
      "step": 29590
    },
    {
      "entropy": 5.960472440719604,
      "epoch": 2.613475803602967,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004316918333010692,
      "loss": 5.3307,
      "mean_token_accuracy": 0.20863488763570787,
      "num_tokens": 55172775.0,
      "step": 29595
    },
    {
      "entropy": 5.945137071609497,
      "epoch": 2.6139173436948075,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004316692320149411,
      "loss": 5.3007,
      "mean_token_accuracy": 0.2133328065276146,
      "num_tokens": 55182227.0,
      "step": 29600
    },
    {
      "entropy": 5.975440216064453,
      "epoch": 2.6143588837866476,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004316466276597517,
      "loss": 5.3324,
      "mean_token_accuracy": 0.207251612842083,
      "num_tokens": 55191562.0,
      "step": 29605
    },
    {
      "entropy": 5.856915616989136,
      "epoch": 2.614800423878488,
      "grad_norm": 1.25,
      "learning_rate": 0.000431624020235944,
      "loss": 5.1915,
      "mean_token_accuracy": 0.22718624472618104,
      "num_tokens": 55200439.0,
      "step": 29610
    },
    {
      "entropy": 5.8251519203186035,
      "epoch": 2.6152419639703286,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004316014097439609,
      "loss": 5.2302,
      "mean_token_accuracy": 0.22107379734516144,
      "num_tokens": 55208850.0,
      "step": 29615
    },
    {
      "entropy": 5.949382400512695,
      "epoch": 2.6156835040621687,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004315787961842452,
      "loss": 5.3378,
      "mean_token_accuracy": 0.2109069809317589,
      "num_tokens": 55218364.0,
      "step": 29620
    },
    {
      "entropy": 5.9525651931762695,
      "epoch": 2.616125044154009,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043155617955723994,
      "loss": 5.2885,
      "mean_token_accuracy": 0.20967071056365966,
      "num_tokens": 55227201.0,
      "step": 29625
    },
    {
      "entropy": 5.964637041091919,
      "epoch": 2.6165665842458496,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004315335598633882,
      "loss": 5.3792,
      "mean_token_accuracy": 0.20624406337738038,
      "num_tokens": 55236907.0,
      "step": 29630
    },
    {
      "entropy": 6.009500217437744,
      "epoch": 2.6170081243376897,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043151093710313296,
      "loss": 5.3437,
      "mean_token_accuracy": 0.20457646548748015,
      "num_tokens": 55246795.0,
      "step": 29635
    },
    {
      "entropy": 5.901585578918457,
      "epoch": 2.61744966442953,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043148831127691754,
      "loss": 5.3174,
      "mean_token_accuracy": 0.2102062925696373,
      "num_tokens": 55255313.0,
      "step": 29640
    },
    {
      "entropy": 5.884474039077759,
      "epoch": 2.6178912045213707,
      "grad_norm": 1.125,
      "learning_rate": 0.00043146568238518505,
      "loss": 5.2424,
      "mean_token_accuracy": 0.21274448037147523,
      "num_tokens": 55264698.0,
      "step": 29645
    },
    {
      "entropy": 5.981578254699707,
      "epoch": 2.6183327446132107,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004314430504283789,
      "loss": 5.3535,
      "mean_token_accuracy": 0.20538992881774903,
      "num_tokens": 55274180.0,
      "step": 29650
    },
    {
      "entropy": 6.00784592628479,
      "epoch": 2.6187742847050512,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004314204154069423,
      "loss": 5.3449,
      "mean_token_accuracy": 0.20230282843112946,
      "num_tokens": 55283552.0,
      "step": 29655
    },
    {
      "entropy": 5.966785621643067,
      "epoch": 2.6192158247968917,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004313977773213188,
      "loss": 5.3705,
      "mean_token_accuracy": 0.2034056857228279,
      "num_tokens": 55293929.0,
      "step": 29660
    },
    {
      "entropy": 5.9082115650177,
      "epoch": 2.619657364888732,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004313751361719517,
      "loss": 5.2401,
      "mean_token_accuracy": 0.21597134619951247,
      "num_tokens": 55303586.0,
      "step": 29665
    },
    {
      "entropy": 5.9861283779144285,
      "epoch": 2.6200989049805723,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043135249195928464,
      "loss": 5.2644,
      "mean_token_accuracy": 0.21534239649772643,
      "num_tokens": 55313032.0,
      "step": 29670
    },
    {
      "entropy": 5.902238035202027,
      "epoch": 2.620540445072413,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043132984468376124,
      "loss": 5.354,
      "mean_token_accuracy": 0.20888935327529906,
      "num_tokens": 55323046.0,
      "step": 29675
    },
    {
      "entropy": 5.975273561477661,
      "epoch": 2.620981985164253,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043130719434582505,
      "loss": 5.2709,
      "mean_token_accuracy": 0.21290733069181442,
      "num_tokens": 55331956.0,
      "step": 29680
    },
    {
      "entropy": 5.855353736877442,
      "epoch": 2.6214235252560933,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004312845409459198,
      "loss": 5.2552,
      "mean_token_accuracy": 0.20678339004516602,
      "num_tokens": 55341573.0,
      "step": 29685
    },
    {
      "entropy": 5.916062450408935,
      "epoch": 2.621865065347934,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043126188448448934,
      "loss": 5.2344,
      "mean_token_accuracy": 0.2139099806547165,
      "num_tokens": 55350325.0,
      "step": 29690
    },
    {
      "entropy": 5.979375600814819,
      "epoch": 2.622306605439774,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004312392249619774,
      "loss": 5.3895,
      "mean_token_accuracy": 0.20299292206764222,
      "num_tokens": 55359716.0,
      "step": 29695
    },
    {
      "entropy": 6.049036121368408,
      "epoch": 2.6227481455316144,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043121656237882786,
      "loss": 5.3888,
      "mean_token_accuracy": 0.20373494178056717,
      "num_tokens": 55369443.0,
      "step": 29700
    },
    {
      "entropy": 5.966266632080078,
      "epoch": 2.6231896856234544,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043119389673548477,
      "loss": 5.3363,
      "mean_token_accuracy": 0.20340440273284913,
      "num_tokens": 55379018.0,
      "step": 29705
    },
    {
      "entropy": 5.92928638458252,
      "epoch": 2.623631225715295,
      "grad_norm": 1.171875,
      "learning_rate": 0.000431171228032392,
      "loss": 5.3291,
      "mean_token_accuracy": 0.20667247474193573,
      "num_tokens": 55388681.0,
      "step": 29710
    },
    {
      "entropy": 5.936031484603882,
      "epoch": 2.6240727658071354,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004311485562699938,
      "loss": 5.2594,
      "mean_token_accuracy": 0.20670946985483168,
      "num_tokens": 55398495.0,
      "step": 29715
    },
    {
      "entropy": 5.929808712005615,
      "epoch": 2.6245143058989755,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004311258814487342,
      "loss": 5.3062,
      "mean_token_accuracy": 0.21339736878871918,
      "num_tokens": 55408498.0,
      "step": 29720
    },
    {
      "entropy": 5.947146224975586,
      "epoch": 2.624955845990816,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004311032035690572,
      "loss": 5.3176,
      "mean_token_accuracy": 0.20828582346439362,
      "num_tokens": 55417336.0,
      "step": 29725
    },
    {
      "entropy": 5.961361217498779,
      "epoch": 2.625397386082656,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004310805226314074,
      "loss": 5.2792,
      "mean_token_accuracy": 0.21181860864162444,
      "num_tokens": 55426828.0,
      "step": 29730
    },
    {
      "entropy": 5.961540222167969,
      "epoch": 2.6258389261744965,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004310578386362288,
      "loss": 5.2434,
      "mean_token_accuracy": 0.2107645481824875,
      "num_tokens": 55435325.0,
      "step": 29735
    },
    {
      "entropy": 5.9549354076385494,
      "epoch": 2.626280466266337,
      "grad_norm": 1.0625,
      "learning_rate": 0.000431035151583966,
      "loss": 5.3692,
      "mean_token_accuracy": 0.20667655467987062,
      "num_tokens": 55444892.0,
      "step": 29740
    },
    {
      "entropy": 5.832911682128906,
      "epoch": 2.626722006358177,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004310124614750632,
      "loss": 5.1755,
      "mean_token_accuracy": 0.2199745237827301,
      "num_tokens": 55453816.0,
      "step": 29745
    },
    {
      "entropy": 5.978606367111206,
      "epoch": 2.6271635464500176,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004309897683099652,
      "loss": 5.3564,
      "mean_token_accuracy": 0.20477150231599808,
      "num_tokens": 55463411.0,
      "step": 29750
    },
    {
      "entropy": 6.013540649414063,
      "epoch": 2.627605086541858,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004309670720891162,
      "loss": 5.4364,
      "mean_token_accuracy": 0.19481461644172668,
      "num_tokens": 55472939.0,
      "step": 29755
    },
    {
      "entropy": 5.857368278503418,
      "epoch": 2.628046626633698,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000430944372812961,
      "loss": 5.1585,
      "mean_token_accuracy": 0.21996524631977082,
      "num_tokens": 55481149.0,
      "step": 29760
    },
    {
      "entropy": 5.944967985153198,
      "epoch": 2.6284881667255386,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004309216704819442,
      "loss": 5.2423,
      "mean_token_accuracy": 0.210441492497921,
      "num_tokens": 55489858.0,
      "step": 29765
    },
    {
      "entropy": 5.926961898803711,
      "epoch": 2.628929706817379,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004308989650965106,
      "loss": 5.2848,
      "mean_token_accuracy": 0.21360872834920883,
      "num_tokens": 55498835.0,
      "step": 29770
    },
    {
      "entropy": 5.87429575920105,
      "epoch": 2.629371246909219,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004308762566571049,
      "loss": 5.1928,
      "mean_token_accuracy": 0.22338983565568923,
      "num_tokens": 55507437.0,
      "step": 29775
    },
    {
      "entropy": 5.883915519714355,
      "epoch": 2.6298127870010597,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004308535451641721,
      "loss": 5.3248,
      "mean_token_accuracy": 0.2054891899228096,
      "num_tokens": 55516304.0,
      "step": 29780
    },
    {
      "entropy": 5.915488624572754,
      "epoch": 2.6302543270929,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004308308306181569,
      "loss": 5.1875,
      "mean_token_accuracy": 0.221619713306427,
      "num_tokens": 55525720.0,
      "step": 29785
    },
    {
      "entropy": 5.973190546035767,
      "epoch": 2.6306958671847402,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004308081130195042,
      "loss": 5.3751,
      "mean_token_accuracy": 0.205017751455307,
      "num_tokens": 55534882.0,
      "step": 29790
    },
    {
      "entropy": 5.894999361038208,
      "epoch": 2.6311374072765807,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004307853923686594,
      "loss": 5.2563,
      "mean_token_accuracy": 0.2163546696305275,
      "num_tokens": 55544337.0,
      "step": 29795
    },
    {
      "entropy": 5.938616991043091,
      "epoch": 2.6315789473684212,
      "grad_norm": 1.25,
      "learning_rate": 0.0004307626686660673,
      "loss": 5.2625,
      "mean_token_accuracy": 0.21119942516088486,
      "num_tokens": 55553616.0,
      "step": 29800
    },
    {
      "entropy": 5.957215642929077,
      "epoch": 2.6320204874602613,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00043073994191217307,
      "loss": 5.336,
      "mean_token_accuracy": 0.2119108408689499,
      "num_tokens": 55564066.0,
      "step": 29805
    },
    {
      "entropy": 5.9131467819213865,
      "epoch": 2.632462027552102,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043071721210742196,
      "loss": 5.2158,
      "mean_token_accuracy": 0.2215617388486862,
      "num_tokens": 55573189.0,
      "step": 29810
    },
    {
      "entropy": 5.921234560012818,
      "epoch": 2.6329035676439423,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004306944792522593,
      "loss": 5.2562,
      "mean_token_accuracy": 0.21713759005069733,
      "num_tokens": 55581421.0,
      "step": 29815
    },
    {
      "entropy": 5.882262802124023,
      "epoch": 2.6333451077357823,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043067174334713025,
      "loss": 5.3187,
      "mean_token_accuracy": 0.20058839172124862,
      "num_tokens": 55591323.0,
      "step": 29820
    },
    {
      "entropy": 5.8426234245300295,
      "epoch": 2.633786647827623,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004306490043924803,
      "loss": 5.2346,
      "mean_token_accuracy": 0.22193381041288376,
      "num_tokens": 55600352.0,
      "step": 29825
    },
    {
      "entropy": 5.951696395874023,
      "epoch": 2.6342281879194633,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004306262623887549,
      "loss": 5.3263,
      "mean_token_accuracy": 0.2090954452753067,
      "num_tokens": 55609678.0,
      "step": 29830
    },
    {
      "entropy": 5.962586116790772,
      "epoch": 2.6346697280113034,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004306035173363995,
      "loss": 5.3272,
      "mean_token_accuracy": 0.21333633959293366,
      "num_tokens": 55619401.0,
      "step": 29835
    },
    {
      "entropy": 5.930533409118652,
      "epoch": 2.635111268103144,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043058076923585976,
      "loss": 5.3545,
      "mean_token_accuracy": 0.20419475436210632,
      "num_tokens": 55629208.0,
      "step": 29840
    },
    {
      "entropy": 5.8817723274230955,
      "epoch": 2.635552808194984,
      "grad_norm": 1.125,
      "learning_rate": 0.00043055801808758123,
      "loss": 5.259,
      "mean_token_accuracy": 0.20804973393678666,
      "num_tokens": 55638304.0,
      "step": 29845
    },
    {
      "entropy": 5.902126693725586,
      "epoch": 2.6359943482868244,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043053526389200955,
      "loss": 5.3093,
      "mean_token_accuracy": 0.2068308636546135,
      "num_tokens": 55647982.0,
      "step": 29850
    },
    {
      "entropy": 5.873469018936158,
      "epoch": 2.636435888378665,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043051250664959053,
      "loss": 5.2336,
      "mean_token_accuracy": 0.21230139434337617,
      "num_tokens": 55657088.0,
      "step": 29855
    },
    {
      "entropy": 5.953074312210083,
      "epoch": 2.636877428470505,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004304897463607699,
      "loss": 5.3216,
      "mean_token_accuracy": 0.20843052119016647,
      "num_tokens": 55666451.0,
      "step": 29860
    },
    {
      "entropy": 5.882851266860962,
      "epoch": 2.6373189685623455,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004304669830259936,
      "loss": 5.2346,
      "mean_token_accuracy": 0.21635981500148774,
      "num_tokens": 55675636.0,
      "step": 29865
    },
    {
      "entropy": 5.944692182540893,
      "epoch": 2.6377605086541855,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043044421664570755,
      "loss": 5.3079,
      "mean_token_accuracy": 0.21237703859806062,
      "num_tokens": 55686092.0,
      "step": 29870
    },
    {
      "entropy": 5.872170162200928,
      "epoch": 2.638202048746026,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043042144722035765,
      "loss": 5.2677,
      "mean_token_accuracy": 0.21437085866928102,
      "num_tokens": 55695212.0,
      "step": 29875
    },
    {
      "entropy": 5.8826704025268555,
      "epoch": 2.6386435888378665,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043039867475039,
      "loss": 5.2821,
      "mean_token_accuracy": 0.21473384648561478,
      "num_tokens": 55704471.0,
      "step": 29880
    },
    {
      "entropy": 5.998332166671753,
      "epoch": 2.6390851289297066,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004303758992362507,
      "loss": 5.3086,
      "mean_token_accuracy": 0.20818208903074265,
      "num_tokens": 55712915.0,
      "step": 29885
    },
    {
      "entropy": 5.995214319229126,
      "epoch": 2.639526669021547,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004303531206783859,
      "loss": 5.3295,
      "mean_token_accuracy": 0.20413149148225784,
      "num_tokens": 55722045.0,
      "step": 29890
    },
    {
      "entropy": 5.862912321090699,
      "epoch": 2.6399682091133876,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004303303390772419,
      "loss": 5.2794,
      "mean_token_accuracy": 0.21460001915693283,
      "num_tokens": 55731548.0,
      "step": 29895
    },
    {
      "entropy": 5.977206087112426,
      "epoch": 2.6404097492052276,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00043030755443326476,
      "loss": 5.3708,
      "mean_token_accuracy": 0.21434206366539002,
      "num_tokens": 55741494.0,
      "step": 29900
    },
    {
      "entropy": 5.977902603149414,
      "epoch": 2.640851289297068,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043028476674690107,
      "loss": 5.3127,
      "mean_token_accuracy": 0.2087973788380623,
      "num_tokens": 55750000.0,
      "step": 29905
    },
    {
      "entropy": 5.93075852394104,
      "epoch": 2.6412928293889086,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004302619760185971,
      "loss": 5.3212,
      "mean_token_accuracy": 0.2069429099559784,
      "num_tokens": 55760375.0,
      "step": 29910
    },
    {
      "entropy": 5.970297384262085,
      "epoch": 2.6417343694807487,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004302391822487993,
      "loss": 5.2724,
      "mean_token_accuracy": 0.21319663971662522,
      "num_tokens": 55768726.0,
      "step": 29915
    },
    {
      "entropy": 5.953742742538452,
      "epoch": 2.642175909572589,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004302163854379541,
      "loss": 5.3445,
      "mean_token_accuracy": 0.20997271090745925,
      "num_tokens": 55778203.0,
      "step": 29920
    },
    {
      "entropy": 5.961909198760987,
      "epoch": 2.6426174496644297,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043019358558650837,
      "loss": 5.2321,
      "mean_token_accuracy": 0.21654016077518462,
      "num_tokens": 55787719.0,
      "step": 29925
    },
    {
      "entropy": 5.869709825515747,
      "epoch": 2.6430589897562697,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043017078269490846,
      "loss": 5.2618,
      "mean_token_accuracy": 0.21087992638349534,
      "num_tokens": 55796824.0,
      "step": 29930
    },
    {
      "entropy": 5.962376260757447,
      "epoch": 2.6435005298481102,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043014797676360124,
      "loss": 5.2668,
      "mean_token_accuracy": 0.21910624504089354,
      "num_tokens": 55805460.0,
      "step": 29935
    },
    {
      "entropy": 5.816734027862549,
      "epoch": 2.6439420699399507,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004301251677930332,
      "loss": 5.2641,
      "mean_token_accuracy": 0.21677508503198623,
      "num_tokens": 55815112.0,
      "step": 29940
    },
    {
      "entropy": 5.886136388778686,
      "epoch": 2.644383610031791,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004301023557836516,
      "loss": 5.2668,
      "mean_token_accuracy": 0.21590791642665863,
      "num_tokens": 55823815.0,
      "step": 29945
    },
    {
      "entropy": 5.923751735687256,
      "epoch": 2.6448251501236313,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004300795407359029,
      "loss": 5.2344,
      "mean_token_accuracy": 0.2172545775771141,
      "num_tokens": 55833513.0,
      "step": 29950
    },
    {
      "entropy": 5.930221605300903,
      "epoch": 2.645266690215472,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043005672265023425,
      "loss": 5.2639,
      "mean_token_accuracy": 0.21359017342329026,
      "num_tokens": 55841765.0,
      "step": 29955
    },
    {
      "entropy": 5.901856565475464,
      "epoch": 2.645708230307312,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043003390152709264,
      "loss": 5.3162,
      "mean_token_accuracy": 0.2089465156197548,
      "num_tokens": 55851125.0,
      "step": 29960
    },
    {
      "entropy": 5.9356300830841064,
      "epoch": 2.6461497703991523,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043001107736692504,
      "loss": 5.2738,
      "mean_token_accuracy": 0.21776421666145324,
      "num_tokens": 55861826.0,
      "step": 29965
    },
    {
      "entropy": 6.037865018844604,
      "epoch": 2.646591310490993,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004299882501701785,
      "loss": 5.3104,
      "mean_token_accuracy": 0.21114346385002136,
      "num_tokens": 55870658.0,
      "step": 29970
    },
    {
      "entropy": 5.962369537353515,
      "epoch": 2.647032850582833,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042996541993730035,
      "loss": 5.3047,
      "mean_token_accuracy": 0.20779592096805571,
      "num_tokens": 55880228.0,
      "step": 29975
    },
    {
      "entropy": 5.886042737960816,
      "epoch": 2.6474743906746734,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042994258666873773,
      "loss": 5.3486,
      "mean_token_accuracy": 0.2039686471223831,
      "num_tokens": 55890101.0,
      "step": 29980
    },
    {
      "entropy": 5.973631429672241,
      "epoch": 2.6479159307665134,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000429919750364938,
      "loss": 5.234,
      "mean_token_accuracy": 0.21923320442438127,
      "num_tokens": 55899164.0,
      "step": 29985
    },
    {
      "entropy": 5.969372653961182,
      "epoch": 2.648357470858354,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042989691102634847,
      "loss": 5.2642,
      "mean_token_accuracy": 0.2156035140156746,
      "num_tokens": 55907735.0,
      "step": 29990
    },
    {
      "entropy": 5.943638467788697,
      "epoch": 2.6487990109501944,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042987406865341657,
      "loss": 5.3978,
      "mean_token_accuracy": 0.2047902375459671,
      "num_tokens": 55917374.0,
      "step": 29995
    },
    {
      "entropy": 5.943044424057007,
      "epoch": 2.6492405510420345,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004298512232465897,
      "loss": 5.3515,
      "mean_token_accuracy": 0.20451628565788268,
      "num_tokens": 55926430.0,
      "step": 30000
    },
    {
      "epoch": 2.6492405510420345,
      "eval_entropy": 5.690560141560176,
      "eval_loss": 5.348297595977783,
      "eval_mean_token_accuracy": 0.21809556935223015,
      "eval_num_tokens": 55926430.0,
      "eval_runtime": 20.1402,
      "eval_samples_per_second": 1753.459,
      "eval_steps_per_second": 219.213,
      "step": 30000
    },
    {
      "entropy": 5.948500156402588,
      "epoch": 2.649682091133875,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004298283748063154,
      "loss": 5.2719,
      "mean_token_accuracy": 0.2105682149529457,
      "num_tokens": 55935805.0,
      "step": 30005
    },
    {
      "entropy": 5.953987455368042,
      "epoch": 2.650123631225715,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004298055233330414,
      "loss": 5.2878,
      "mean_token_accuracy": 0.21550484299659728,
      "num_tokens": 55945278.0,
      "step": 30010
    },
    {
      "entropy": 5.931280946731567,
      "epoch": 2.6505651713175555,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004297826688272151,
      "loss": 5.247,
      "mean_token_accuracy": 0.21991252899169922,
      "num_tokens": 55954316.0,
      "step": 30015
    },
    {
      "entropy": 5.879923534393311,
      "epoch": 2.651006711409396,
      "grad_norm": 1.125,
      "learning_rate": 0.0004297598112892845,
      "loss": 5.1914,
      "mean_token_accuracy": 0.22328784465789794,
      "num_tokens": 55963249.0,
      "step": 30020
    },
    {
      "entropy": 5.965307760238647,
      "epoch": 2.651448251501236,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042973695071969715,
      "loss": 5.278,
      "mean_token_accuracy": 0.21676760017871857,
      "num_tokens": 55972010.0,
      "step": 30025
    },
    {
      "entropy": 5.8536773204803465,
      "epoch": 2.6518897915930766,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004297140871189009,
      "loss": 5.2135,
      "mean_token_accuracy": 0.2185996189713478,
      "num_tokens": 55981197.0,
      "step": 30030
    },
    {
      "entropy": 5.958897876739502,
      "epoch": 2.652331331684917,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004296912204873437,
      "loss": 5.3541,
      "mean_token_accuracy": 0.20682618916034698,
      "num_tokens": 55991367.0,
      "step": 30035
    },
    {
      "entropy": 5.961681842803955,
      "epoch": 2.652772871776757,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042966835082547346,
      "loss": 5.2294,
      "mean_token_accuracy": 0.21893686354160308,
      "num_tokens": 56000614.0,
      "step": 30040
    },
    {
      "entropy": 5.832093334197998,
      "epoch": 2.6532144118685976,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004296454781337382,
      "loss": 5.2407,
      "mean_token_accuracy": 0.2201880097389221,
      "num_tokens": 56009786.0,
      "step": 30045
    },
    {
      "entropy": 5.9051965236663815,
      "epoch": 2.653655951960438,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000429622602412586,
      "loss": 5.3078,
      "mean_token_accuracy": 0.21249389201402663,
      "num_tokens": 56019696.0,
      "step": 30050
    },
    {
      "entropy": 5.937598848342896,
      "epoch": 2.654097492052278,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042959972366246493,
      "loss": 5.2697,
      "mean_token_accuracy": 0.216501784324646,
      "num_tokens": 56027960.0,
      "step": 30055
    },
    {
      "entropy": 5.959663009643554,
      "epoch": 2.6545390321441187,
      "grad_norm": 1.125,
      "learning_rate": 0.0004295768418838232,
      "loss": 5.3414,
      "mean_token_accuracy": 0.20403106659650802,
      "num_tokens": 56036875.0,
      "step": 30060
    },
    {
      "entropy": 5.938409233093262,
      "epoch": 2.654980572235959,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004295539570771091,
      "loss": 5.3329,
      "mean_token_accuracy": 0.20948461443185806,
      "num_tokens": 56045916.0,
      "step": 30065
    },
    {
      "entropy": 5.939558362960815,
      "epoch": 2.6554221123277992,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004295310692427708,
      "loss": 5.3232,
      "mean_token_accuracy": 0.21306592375040054,
      "num_tokens": 56056048.0,
      "step": 30070
    },
    {
      "entropy": 5.901064348220825,
      "epoch": 2.6558636524196397,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042950817838125675,
      "loss": 5.2755,
      "mean_token_accuracy": 0.2092679500579834,
      "num_tokens": 56065455.0,
      "step": 30075
    },
    {
      "entropy": 5.939755630493164,
      "epoch": 2.6563051925114802,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004294852844930154,
      "loss": 5.2953,
      "mean_token_accuracy": 0.2144603595137596,
      "num_tokens": 56074070.0,
      "step": 30080
    },
    {
      "entropy": 5.940696430206299,
      "epoch": 2.6567467326033203,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042946238757849515,
      "loss": 5.3325,
      "mean_token_accuracy": 0.21027863025665283,
      "num_tokens": 56083518.0,
      "step": 30085
    },
    {
      "entropy": 5.870160531997681,
      "epoch": 2.657188272695161,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042943948763814465,
      "loss": 5.2829,
      "mean_token_accuracy": 0.2124789074063301,
      "num_tokens": 56092591.0,
      "step": 30090
    },
    {
      "entropy": 6.00924129486084,
      "epoch": 2.6576298127870013,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004294165846724123,
      "loss": 5.4799,
      "mean_token_accuracy": 0.19203050285577775,
      "num_tokens": 56102120.0,
      "step": 30095
    },
    {
      "entropy": 6.038431930541992,
      "epoch": 2.6580713528788413,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000429393678681747,
      "loss": 5.302,
      "mean_token_accuracy": 0.2165894851088524,
      "num_tokens": 56111885.0,
      "step": 30100
    },
    {
      "entropy": 5.90104603767395,
      "epoch": 2.658512892970682,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042937076966659726,
      "loss": 5.2314,
      "mean_token_accuracy": 0.2176900878548622,
      "num_tokens": 56122070.0,
      "step": 30105
    },
    {
      "entropy": 5.872009515762329,
      "epoch": 2.6589544330625223,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042934785762741197,
      "loss": 5.2833,
      "mean_token_accuracy": 0.2178093060851097,
      "num_tokens": 56132142.0,
      "step": 30110
    },
    {
      "entropy": 5.9560322761535645,
      "epoch": 2.6593959731543624,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004293249425646399,
      "loss": 5.3235,
      "mean_token_accuracy": 0.20436921268701552,
      "num_tokens": 56141860.0,
      "step": 30115
    },
    {
      "entropy": 5.913686847686767,
      "epoch": 2.659837513246203,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042930202447873003,
      "loss": 5.2227,
      "mean_token_accuracy": 0.21353253722190857,
      "num_tokens": 56152035.0,
      "step": 30120
    },
    {
      "entropy": 5.970901584625244,
      "epoch": 2.660279053338043,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004292791033701312,
      "loss": 5.3414,
      "mean_token_accuracy": 0.2120572343468666,
      "num_tokens": 56161266.0,
      "step": 30125
    },
    {
      "entropy": 5.899970006942749,
      "epoch": 2.6607205934298834,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004292561792392926,
      "loss": 5.3446,
      "mean_token_accuracy": 0.20971922129392623,
      "num_tokens": 56171860.0,
      "step": 30130
    },
    {
      "entropy": 6.002635049819946,
      "epoch": 2.661162133521724,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042923325208666307,
      "loss": 5.3379,
      "mean_token_accuracy": 0.2130967989563942,
      "num_tokens": 56180492.0,
      "step": 30135
    },
    {
      "entropy": 5.883637714385986,
      "epoch": 2.661603673613564,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042921032191269183,
      "loss": 5.2622,
      "mean_token_accuracy": 0.20988354831933975,
      "num_tokens": 56190292.0,
      "step": 30140
    },
    {
      "entropy": 6.01088547706604,
      "epoch": 2.6620452137054045,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042918738871782814,
      "loss": 5.3998,
      "mean_token_accuracy": 0.20059143751859665,
      "num_tokens": 56200270.0,
      "step": 30145
    },
    {
      "entropy": 5.962783622741699,
      "epoch": 2.6624867537972445,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004291644525025212,
      "loss": 5.346,
      "mean_token_accuracy": 0.2094578802585602,
      "num_tokens": 56210876.0,
      "step": 30150
    },
    {
      "entropy": 5.958168601989746,
      "epoch": 2.662928293889085,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042914151326722025,
      "loss": 5.309,
      "mean_token_accuracy": 0.21021298617124556,
      "num_tokens": 56221100.0,
      "step": 30155
    },
    {
      "entropy": 5.97096848487854,
      "epoch": 2.6633698339809255,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004291185710123748,
      "loss": 5.3339,
      "mean_token_accuracy": 0.20767012238502502,
      "num_tokens": 56230404.0,
      "step": 30160
    },
    {
      "entropy": 5.870409345626831,
      "epoch": 2.6638113740727656,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042909562573843416,
      "loss": 5.171,
      "mean_token_accuracy": 0.23235356211662292,
      "num_tokens": 56239120.0,
      "step": 30165
    },
    {
      "entropy": 5.912965631484985,
      "epoch": 2.664252914164606,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042907267744584775,
      "loss": 5.2973,
      "mean_token_accuracy": 0.21686581522226334,
      "num_tokens": 56248979.0,
      "step": 30170
    },
    {
      "entropy": 5.933758354187011,
      "epoch": 2.6646944542564466,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042904972613506536,
      "loss": 5.3359,
      "mean_token_accuracy": 0.20504150390625,
      "num_tokens": 56258133.0,
      "step": 30175
    },
    {
      "entropy": 6.0130431175231935,
      "epoch": 2.6651359943482866,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042902677180653643,
      "loss": 5.3958,
      "mean_token_accuracy": 0.20684002488851547,
      "num_tokens": 56267806.0,
      "step": 30180
    },
    {
      "entropy": 5.9423415184021,
      "epoch": 2.665577534440127,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004290038144607106,
      "loss": 5.33,
      "mean_token_accuracy": 0.20993110686540603,
      "num_tokens": 56276993.0,
      "step": 30185
    },
    {
      "entropy": 5.767441320419311,
      "epoch": 2.6660190745319676,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042898085409803766,
      "loss": 5.1048,
      "mean_token_accuracy": 0.23139368295669555,
      "num_tokens": 56286716.0,
      "step": 30190
    },
    {
      "entropy": 5.900459575653076,
      "epoch": 2.6664606146238077,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004289578907189673,
      "loss": 5.2287,
      "mean_token_accuracy": 0.22115472555160523,
      "num_tokens": 56296194.0,
      "step": 30195
    },
    {
      "entropy": 5.956528186798096,
      "epoch": 2.666902154715648,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042893492432394953,
      "loss": 5.2404,
      "mean_token_accuracy": 0.2123638093471527,
      "num_tokens": 56304855.0,
      "step": 30200
    },
    {
      "entropy": 5.998675966262818,
      "epoch": 2.6673436948074887,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042891195491343405,
      "loss": 5.3575,
      "mean_token_accuracy": 0.20072102993726731,
      "num_tokens": 56313869.0,
      "step": 30205
    },
    {
      "entropy": 5.936729145050049,
      "epoch": 2.6677852348993287,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042888898248787094,
      "loss": 5.2874,
      "mean_token_accuracy": 0.21168984472751617,
      "num_tokens": 56323551.0,
      "step": 30210
    },
    {
      "entropy": 5.934345293045044,
      "epoch": 2.6682267749911692,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004288660070477101,
      "loss": 5.3054,
      "mean_token_accuracy": 0.20744280964136125,
      "num_tokens": 56333068.0,
      "step": 30215
    },
    {
      "entropy": 5.906696462631226,
      "epoch": 2.6686683150830097,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042884302859340183,
      "loss": 5.2221,
      "mean_token_accuracy": 0.21428233087062837,
      "num_tokens": 56341952.0,
      "step": 30220
    },
    {
      "entropy": 5.855170679092407,
      "epoch": 2.6691098551748498,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042882004712539615,
      "loss": 5.1621,
      "mean_token_accuracy": 0.21444136053323745,
      "num_tokens": 56351829.0,
      "step": 30225
    },
    {
      "entropy": 5.949182224273682,
      "epoch": 2.6695513952666903,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042879706264414304,
      "loss": 5.3427,
      "mean_token_accuracy": 0.20560797154903412,
      "num_tokens": 56361290.0,
      "step": 30230
    },
    {
      "entropy": 5.870204019546509,
      "epoch": 2.6699929353585308,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004287740751500931,
      "loss": 5.3051,
      "mean_token_accuracy": 0.22230759710073472,
      "num_tokens": 56371582.0,
      "step": 30235
    },
    {
      "entropy": 5.862482261657715,
      "epoch": 2.670434475450371,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042875108464369646,
      "loss": 5.1611,
      "mean_token_accuracy": 0.2233293130993843,
      "num_tokens": 56380637.0,
      "step": 30240
    },
    {
      "entropy": 5.924016237258911,
      "epoch": 2.6708760155422113,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042872809112540347,
      "loss": 5.3524,
      "mean_token_accuracy": 0.20527918487787247,
      "num_tokens": 56390332.0,
      "step": 30245
    },
    {
      "entropy": 5.96146879196167,
      "epoch": 2.671317555634052,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004287050945956646,
      "loss": 5.2835,
      "mean_token_accuracy": 0.2051306813955307,
      "num_tokens": 56399379.0,
      "step": 30250
    },
    {
      "entropy": 5.927504634857177,
      "epoch": 2.671759095725892,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042868209505493034,
      "loss": 5.2359,
      "mean_token_accuracy": 0.2169322684407234,
      "num_tokens": 56408645.0,
      "step": 30255
    },
    {
      "entropy": 5.871516180038452,
      "epoch": 2.6722006358177324,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004286590925036513,
      "loss": 5.2345,
      "mean_token_accuracy": 0.21985612362623214,
      "num_tokens": 56417119.0,
      "step": 30260
    },
    {
      "entropy": 5.877132749557495,
      "epoch": 2.6726421759095724,
      "grad_norm": 1.09375,
      "learning_rate": 0.000428636086942278,
      "loss": 5.3146,
      "mean_token_accuracy": 0.20803434550762176,
      "num_tokens": 56426443.0,
      "step": 30265
    },
    {
      "entropy": 5.96612401008606,
      "epoch": 2.673083716001413,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004286130783712611,
      "loss": 5.415,
      "mean_token_accuracy": 0.19638187885284425,
      "num_tokens": 56435675.0,
      "step": 30270
    },
    {
      "entropy": 5.986404895782471,
      "epoch": 2.6735252560932534,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004285900667910514,
      "loss": 5.3104,
      "mean_token_accuracy": 0.2086201086640358,
      "num_tokens": 56445787.0,
      "step": 30275
    },
    {
      "entropy": 5.9748570919036865,
      "epoch": 2.6739667961850935,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004285670522020996,
      "loss": 5.276,
      "mean_token_accuracy": 0.21931329220533372,
      "num_tokens": 56455029.0,
      "step": 30280
    },
    {
      "entropy": 5.998804759979248,
      "epoch": 2.674408336276934,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004285440346048566,
      "loss": 5.305,
      "mean_token_accuracy": 0.20512138158082963,
      "num_tokens": 56464566.0,
      "step": 30285
    },
    {
      "entropy": 5.93938422203064,
      "epoch": 2.674849876368774,
      "grad_norm": 1.125,
      "learning_rate": 0.0004285210139997733,
      "loss": 5.3314,
      "mean_token_accuracy": 0.2115415006875992,
      "num_tokens": 56473350.0,
      "step": 30290
    },
    {
      "entropy": 5.897784900665283,
      "epoch": 2.6752914164606145,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004284979903873007,
      "loss": 5.2774,
      "mean_token_accuracy": 0.21192871630191804,
      "num_tokens": 56482841.0,
      "step": 30295
    },
    {
      "entropy": 5.985298109054566,
      "epoch": 2.675732956552455,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042847496376788964,
      "loss": 5.3916,
      "mean_token_accuracy": 0.2027365431189537,
      "num_tokens": 56492851.0,
      "step": 30300
    },
    {
      "entropy": 6.025664377212524,
      "epoch": 2.676174496644295,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042845193414199146,
      "loss": 5.3757,
      "mean_token_accuracy": 0.20273386538028718,
      "num_tokens": 56501514.0,
      "step": 30305
    },
    {
      "entropy": 6.000804710388183,
      "epoch": 2.6766160367361356,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042842890151005705,
      "loss": 5.2534,
      "mean_token_accuracy": 0.2155844733119011,
      "num_tokens": 56509963.0,
      "step": 30310
    },
    {
      "entropy": 5.888183450698852,
      "epoch": 2.677057576827976,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042840586587253776,
      "loss": 5.3057,
      "mean_token_accuracy": 0.21040222495794297,
      "num_tokens": 56519598.0,
      "step": 30315
    },
    {
      "entropy": 5.828030920028686,
      "epoch": 2.677499116919816,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042838282722988484,
      "loss": 5.2553,
      "mean_token_accuracy": 0.21263096779584884,
      "num_tokens": 56528275.0,
      "step": 30320
    },
    {
      "entropy": 5.981657218933106,
      "epoch": 2.6779406570116566,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042835978558254954,
      "loss": 5.3128,
      "mean_token_accuracy": 0.211962927877903,
      "num_tokens": 56537974.0,
      "step": 30325
    },
    {
      "entropy": 6.008897733688355,
      "epoch": 2.678382197103497,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004283367409309833,
      "loss": 5.3663,
      "mean_token_accuracy": 0.2067365199327469,
      "num_tokens": 56546262.0,
      "step": 30330
    },
    {
      "entropy": 5.986781740188599,
      "epoch": 2.678823737195337,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004283136932756375,
      "loss": 5.2769,
      "mean_token_accuracy": 0.2085500791668892,
      "num_tokens": 56556564.0,
      "step": 30335
    },
    {
      "entropy": 5.853544521331787,
      "epoch": 2.6792652772871777,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004282906426169636,
      "loss": 5.1628,
      "mean_token_accuracy": 0.22233890295028685,
      "num_tokens": 56565693.0,
      "step": 30340
    },
    {
      "entropy": 5.8823119640350345,
      "epoch": 2.679706817379018,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004282675889554132,
      "loss": 5.275,
      "mean_token_accuracy": 0.22001984864473342,
      "num_tokens": 56575693.0,
      "step": 30345
    },
    {
      "entropy": 5.9744363784790036,
      "epoch": 2.6801483574708582,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004282445322914379,
      "loss": 5.2996,
      "mean_token_accuracy": 0.20991754978895188,
      "num_tokens": 56585147.0,
      "step": 30350
    },
    {
      "entropy": 5.969154977798462,
      "epoch": 2.6805898975626987,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042822147262548937,
      "loss": 5.3489,
      "mean_token_accuracy": 0.20819698572158812,
      "num_tokens": 56595841.0,
      "step": 30355
    },
    {
      "entropy": 5.8716662406921385,
      "epoch": 2.681031437654539,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004281984099580194,
      "loss": 5.2982,
      "mean_token_accuracy": 0.20375325083732604,
      "num_tokens": 56606226.0,
      "step": 30360
    },
    {
      "entropy": 5.9745267868042,
      "epoch": 2.6814729777463793,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004281753442894796,
      "loss": 5.3012,
      "mean_token_accuracy": 0.21439256072044371,
      "num_tokens": 56614821.0,
      "step": 30365
    },
    {
      "entropy": 5.902958345413208,
      "epoch": 2.6819145178382198,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000428152275620322,
      "loss": 5.2305,
      "mean_token_accuracy": 0.21982047706842422,
      "num_tokens": 56623489.0,
      "step": 30370
    },
    {
      "entropy": 5.900275373458863,
      "epoch": 2.6823560579300603,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004281292039509984,
      "loss": 5.3132,
      "mean_token_accuracy": 0.21739144772291183,
      "num_tokens": 56633071.0,
      "step": 30375
    },
    {
      "entropy": 5.949919605255127,
      "epoch": 2.6827975980219003,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004281061292819608,
      "loss": 5.3758,
      "mean_token_accuracy": 0.19729256480932236,
      "num_tokens": 56643316.0,
      "step": 30380
    },
    {
      "entropy": 5.914574337005615,
      "epoch": 2.683239138113741,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004280830516136611,
      "loss": 5.2543,
      "mean_token_accuracy": 0.21757240146398543,
      "num_tokens": 56652690.0,
      "step": 30385
    },
    {
      "entropy": 5.79608907699585,
      "epoch": 2.6836806782055813,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042805997094655157,
      "loss": 5.1973,
      "mean_token_accuracy": 0.2194118767976761,
      "num_tokens": 56662442.0,
      "step": 30390
    },
    {
      "entropy": 5.895685482025146,
      "epoch": 2.6841222182974214,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004280368872810843,
      "loss": 5.3091,
      "mean_token_accuracy": 0.2106982573866844,
      "num_tokens": 56672308.0,
      "step": 30395
    },
    {
      "entropy": 5.916185092926026,
      "epoch": 2.684563758389262,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004280138006177114,
      "loss": 5.2922,
      "mean_token_accuracy": 0.21089142113924025,
      "num_tokens": 56682193.0,
      "step": 30400
    },
    {
      "entropy": 6.042352485656738,
      "epoch": 2.685005298481102,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042799071095688516,
      "loss": 5.3204,
      "mean_token_accuracy": 0.2099079057574272,
      "num_tokens": 56690856.0,
      "step": 30405
    },
    {
      "entropy": 5.963382959365845,
      "epoch": 2.6854468385729424,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042796761829905797,
      "loss": 5.3076,
      "mean_token_accuracy": 0.20558902472257615,
      "num_tokens": 56699998.0,
      "step": 30410
    },
    {
      "entropy": 5.857285118103027,
      "epoch": 2.685888378664783,
      "grad_norm": 1.125,
      "learning_rate": 0.000427944522644682,
      "loss": 5.2794,
      "mean_token_accuracy": 0.21322032809257507,
      "num_tokens": 56708977.0,
      "step": 30415
    },
    {
      "entropy": 5.939264154434204,
      "epoch": 2.686329918756623,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004279214239942099,
      "loss": 5.3172,
      "mean_token_accuracy": 0.2166338860988617,
      "num_tokens": 56718141.0,
      "step": 30420
    },
    {
      "entropy": 5.991215515136719,
      "epoch": 2.6867714588484635,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042789832234809414,
      "loss": 5.3369,
      "mean_token_accuracy": 0.20956986546516418,
      "num_tokens": 56727211.0,
      "step": 30425
    },
    {
      "entropy": 5.958644056320191,
      "epoch": 2.6872129989403035,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042787521770678704,
      "loss": 5.377,
      "mean_token_accuracy": 0.2022753909230232,
      "num_tokens": 56737027.0,
      "step": 30430
    },
    {
      "entropy": 5.850568008422852,
      "epoch": 2.687654539032144,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042785211007074147,
      "loss": 5.226,
      "mean_token_accuracy": 0.20625687688589095,
      "num_tokens": 56745957.0,
      "step": 30435
    },
    {
      "entropy": 5.928841876983642,
      "epoch": 2.6880960791239845,
      "grad_norm": 1.125,
      "learning_rate": 0.00042782899944041004,
      "loss": 5.2294,
      "mean_token_accuracy": 0.21447080075740815,
      "num_tokens": 56755487.0,
      "step": 30440
    },
    {
      "entropy": 5.911452579498291,
      "epoch": 2.6885376192158246,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042780588581624535,
      "loss": 5.3342,
      "mean_token_accuracy": 0.2198861539363861,
      "num_tokens": 56765606.0,
      "step": 30445
    },
    {
      "entropy": 5.938370656967163,
      "epoch": 2.688979159307665,
      "grad_norm": 1.125,
      "learning_rate": 0.0004277827691987003,
      "loss": 5.2606,
      "mean_token_accuracy": 0.21484241187572478,
      "num_tokens": 56775015.0,
      "step": 30450
    },
    {
      "entropy": 5.992617082595825,
      "epoch": 2.6894206993995056,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00042775964958822766,
      "loss": 5.4286,
      "mean_token_accuracy": 0.20480500012636185,
      "num_tokens": 56784166.0,
      "step": 30455
    },
    {
      "entropy": 5.9934430599212645,
      "epoch": 2.6898622394913456,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004277365269852804,
      "loss": 5.3576,
      "mean_token_accuracy": 0.20398036390542984,
      "num_tokens": 56793146.0,
      "step": 30460
    },
    {
      "entropy": 5.9004284858703615,
      "epoch": 2.690303779583186,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004277134013903114,
      "loss": 5.1802,
      "mean_token_accuracy": 0.2282739222049713,
      "num_tokens": 56802041.0,
      "step": 30465
    },
    {
      "entropy": 5.953090763092041,
      "epoch": 2.6907453196750266,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042769027280377376,
      "loss": 5.3465,
      "mean_token_accuracy": 0.21032098233699797,
      "num_tokens": 56810981.0,
      "step": 30470
    },
    {
      "entropy": 5.885869312286377,
      "epoch": 2.6911868597668667,
      "grad_norm": 1.125,
      "learning_rate": 0.0004276671412261205,
      "loss": 5.2394,
      "mean_token_accuracy": 0.21631449311971665,
      "num_tokens": 56820591.0,
      "step": 30475
    },
    {
      "entropy": 5.954400110244751,
      "epoch": 2.691628399858707,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004276440066578048,
      "loss": 5.3296,
      "mean_token_accuracy": 0.2122537076473236,
      "num_tokens": 56830376.0,
      "step": 30480
    },
    {
      "entropy": 6.027338171005249,
      "epoch": 2.6920699399505477,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042762086909927977,
      "loss": 5.385,
      "mean_token_accuracy": 0.20671252012252808,
      "num_tokens": 56839487.0,
      "step": 30485
    },
    {
      "entropy": 5.991406726837158,
      "epoch": 2.6925114800423877,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004275977285509987,
      "loss": 5.3504,
      "mean_token_accuracy": 0.216791769862175,
      "num_tokens": 56849570.0,
      "step": 30490
    },
    {
      "entropy": 5.893406057357788,
      "epoch": 2.692953020134228,
      "grad_norm": 1.171875,
      "learning_rate": 0.000427574585013415,
      "loss": 5.302,
      "mean_token_accuracy": 0.21135196685791016,
      "num_tokens": 56858506.0,
      "step": 30495
    },
    {
      "entropy": 5.927654647827149,
      "epoch": 2.6933945602260687,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042755143848698194,
      "loss": 5.2856,
      "mean_token_accuracy": 0.2165857285261154,
      "num_tokens": 56868276.0,
      "step": 30500
    },
    {
      "entropy": 5.945782470703125,
      "epoch": 2.6938361003179088,
      "grad_norm": 1.125,
      "learning_rate": 0.00042752828897215295,
      "loss": 5.2894,
      "mean_token_accuracy": 0.21562058329582215,
      "num_tokens": 56877434.0,
      "step": 30505
    },
    {
      "entropy": 5.966252374649048,
      "epoch": 2.6942776404097493,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004275051364693815,
      "loss": 5.326,
      "mean_token_accuracy": 0.21067022234201432,
      "num_tokens": 56888186.0,
      "step": 30510
    },
    {
      "entropy": 5.9227375984191895,
      "epoch": 2.6947191805015898,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042748198097912113,
      "loss": 5.2145,
      "mean_token_accuracy": 0.2179109588265419,
      "num_tokens": 56897125.0,
      "step": 30515
    },
    {
      "entropy": 5.918860864639282,
      "epoch": 2.69516072059343,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004274588225018255,
      "loss": 5.3009,
      "mean_token_accuracy": 0.2135752335190773,
      "num_tokens": 56906305.0,
      "step": 30520
    },
    {
      "entropy": 5.947680282592773,
      "epoch": 2.6956022606852703,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004274356610379483,
      "loss": 5.2959,
      "mean_token_accuracy": 0.20172210186719894,
      "num_tokens": 56915645.0,
      "step": 30525
    },
    {
      "entropy": 5.89109320640564,
      "epoch": 2.696043800777111,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004274124965879431,
      "loss": 5.256,
      "mean_token_accuracy": 0.21449303925037383,
      "num_tokens": 56924567.0,
      "step": 30530
    },
    {
      "entropy": 5.957662153244018,
      "epoch": 2.696485340868951,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004273893291522638,
      "loss": 5.3256,
      "mean_token_accuracy": 0.20865843892097474,
      "num_tokens": 56933864.0,
      "step": 30535
    },
    {
      "entropy": 5.930090618133545,
      "epoch": 2.6969268809607914,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042736615873136425,
      "loss": 5.2443,
      "mean_token_accuracy": 0.22196718901395798,
      "num_tokens": 56943150.0,
      "step": 30540
    },
    {
      "entropy": 5.864338779449463,
      "epoch": 2.6973684210526314,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042734298532569825,
      "loss": 5.2508,
      "mean_token_accuracy": 0.21782076805830003,
      "num_tokens": 56951775.0,
      "step": 30545
    },
    {
      "entropy": 5.974887943267822,
      "epoch": 2.697809961144472,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042731980893571985,
      "loss": 5.3656,
      "mean_token_accuracy": 0.2078203722834587,
      "num_tokens": 56960827.0,
      "step": 30550
    },
    {
      "entropy": 5.977626323699951,
      "epoch": 2.6982515012363124,
      "grad_norm": 1.171875,
      "learning_rate": 0.000427296629561883,
      "loss": 5.274,
      "mean_token_accuracy": 0.2131943166255951,
      "num_tokens": 56968966.0,
      "step": 30555
    },
    {
      "entropy": 5.926692008972168,
      "epoch": 2.6986930413281525,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004272734472046419,
      "loss": 5.299,
      "mean_token_accuracy": 0.21669170558452605,
      "num_tokens": 56977818.0,
      "step": 30560
    },
    {
      "entropy": 5.925864839553833,
      "epoch": 2.699134581419993,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042725026186445043,
      "loss": 5.2549,
      "mean_token_accuracy": 0.21622505336999892,
      "num_tokens": 56986641.0,
      "step": 30565
    },
    {
      "entropy": 5.986659002304077,
      "epoch": 2.699576121511833,
      "grad_norm": 1.125,
      "learning_rate": 0.00042722707354176303,
      "loss": 5.3339,
      "mean_token_accuracy": 0.20980894118547438,
      "num_tokens": 56995806.0,
      "step": 30570
    },
    {
      "entropy": 5.942989444732666,
      "epoch": 2.7000176616036735,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004272038822370338,
      "loss": 5.3382,
      "mean_token_accuracy": 0.20474520325660706,
      "num_tokens": 57004701.0,
      "step": 30575
    },
    {
      "entropy": 5.917185068130493,
      "epoch": 2.700459201695514,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000427180687950717,
      "loss": 5.2314,
      "mean_token_accuracy": 0.21690067946910857,
      "num_tokens": 57014260.0,
      "step": 30580
    },
    {
      "entropy": 5.9476330280303955,
      "epoch": 2.700900741787354,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042715749068326713,
      "loss": 5.3227,
      "mean_token_accuracy": 0.21114426404237746,
      "num_tokens": 57023057.0,
      "step": 30585
    },
    {
      "entropy": 5.990786027908325,
      "epoch": 2.7013422818791946,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042713429043513863,
      "loss": 5.3014,
      "mean_token_accuracy": 0.21023475080728532,
      "num_tokens": 57032046.0,
      "step": 30590
    },
    {
      "entropy": 5.93529200553894,
      "epoch": 2.701783821971035,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004271110872067858,
      "loss": 5.2511,
      "mean_token_accuracy": 0.21401549726724625,
      "num_tokens": 57041473.0,
      "step": 30595
    },
    {
      "entropy": 5.912738561630249,
      "epoch": 2.702225362062875,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042708788099866333,
      "loss": 5.3194,
      "mean_token_accuracy": 0.21070767045021058,
      "num_tokens": 57050098.0,
      "step": 30600
    },
    {
      "entropy": 5.901719570159912,
      "epoch": 2.7026669021547156,
      "grad_norm": 1.46875,
      "learning_rate": 0.00042706467181122574,
      "loss": 5.2016,
      "mean_token_accuracy": 0.2234045907855034,
      "num_tokens": 57059922.0,
      "step": 30605
    },
    {
      "entropy": 5.9424032211303714,
      "epoch": 2.703108442246556,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042704145964492766,
      "loss": 5.31,
      "mean_token_accuracy": 0.21601886451244354,
      "num_tokens": 57069841.0,
      "step": 30610
    },
    {
      "entropy": 5.939793062210083,
      "epoch": 2.703549982338396,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000427018244500224,
      "loss": 5.3272,
      "mean_token_accuracy": 0.20438270270824432,
      "num_tokens": 57078891.0,
      "step": 30615
    },
    {
      "entropy": 5.9108617305755615,
      "epoch": 2.7039915224302367,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042699502637756924,
      "loss": 5.2168,
      "mean_token_accuracy": 0.22455231845378876,
      "num_tokens": 57087899.0,
      "step": 30620
    },
    {
      "entropy": 5.952547168731689,
      "epoch": 2.704433062522077,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004269718052774185,
      "loss": 5.3544,
      "mean_token_accuracy": 0.20233257114887238,
      "num_tokens": 57096169.0,
      "step": 30625
    },
    {
      "entropy": 5.943078708648682,
      "epoch": 2.704874602613917,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004269485812002264,
      "loss": 5.257,
      "mean_token_accuracy": 0.2138029381632805,
      "num_tokens": 57106423.0,
      "step": 30630
    },
    {
      "entropy": 5.956169605255127,
      "epoch": 2.7053161427057577,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042692535414644804,
      "loss": 5.2783,
      "mean_token_accuracy": 0.2188625678420067,
      "num_tokens": 57115308.0,
      "step": 30635
    },
    {
      "entropy": 5.9547590732574465,
      "epoch": 2.705757682797598,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042690212411653844,
      "loss": 5.32,
      "mean_token_accuracy": 0.21154437065124512,
      "num_tokens": 57125108.0,
      "step": 30640
    },
    {
      "entropy": 5.900923728942871,
      "epoch": 2.7061992228894383,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004268788911109526,
      "loss": 5.2343,
      "mean_token_accuracy": 0.21740791499614714,
      "num_tokens": 57134392.0,
      "step": 30645
    },
    {
      "entropy": 5.907728242874145,
      "epoch": 2.7066407629812788,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004268556551301457,
      "loss": 5.2902,
      "mean_token_accuracy": 0.21421440094709396,
      "num_tokens": 57143245.0,
      "step": 30650
    },
    {
      "entropy": 5.945869302749633,
      "epoch": 2.7070823030731193,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004268324161745728,
      "loss": 5.2512,
      "mean_token_accuracy": 0.22001037746667862,
      "num_tokens": 57151691.0,
      "step": 30655
    },
    {
      "entropy": 5.93883056640625,
      "epoch": 2.7075238431649593,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042680917424468926,
      "loss": 5.277,
      "mean_token_accuracy": 0.21620582938194274,
      "num_tokens": 57161157.0,
      "step": 30660
    },
    {
      "entropy": 6.000668048858643,
      "epoch": 2.7079653832568,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004267859293409504,
      "loss": 5.267,
      "mean_token_accuracy": 0.21450457125902175,
      "num_tokens": 57169599.0,
      "step": 30665
    },
    {
      "entropy": 5.93957839012146,
      "epoch": 2.7084069233486403,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042676268146381134,
      "loss": 5.2638,
      "mean_token_accuracy": 0.21386970281600953,
      "num_tokens": 57178788.0,
      "step": 30670
    },
    {
      "entropy": 5.954804611206055,
      "epoch": 2.7088484634404804,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042673943061372776,
      "loss": 5.3113,
      "mean_token_accuracy": 0.21198065280914308,
      "num_tokens": 57187634.0,
      "step": 30675
    },
    {
      "entropy": 5.936024761199951,
      "epoch": 2.709290003532321,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004267161767911551,
      "loss": 5.2835,
      "mean_token_accuracy": 0.2108532965183258,
      "num_tokens": 57196777.0,
      "step": 30680
    },
    {
      "entropy": 6.014348745346069,
      "epoch": 2.709731543624161,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042669291999654867,
      "loss": 5.3998,
      "mean_token_accuracy": 0.20108452141284944,
      "num_tokens": 57206092.0,
      "step": 30685
    },
    {
      "entropy": 5.9539759159088135,
      "epoch": 2.7101730837160014,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042666966023036443,
      "loss": 5.2723,
      "mean_token_accuracy": 0.21847835034132004,
      "num_tokens": 57215717.0,
      "step": 30690
    },
    {
      "entropy": 5.961084794998169,
      "epoch": 2.710614623807842,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004266463974930576,
      "loss": 5.2982,
      "mean_token_accuracy": 0.20834921896457673,
      "num_tokens": 57224172.0,
      "step": 30695
    },
    {
      "entropy": 5.901815938949585,
      "epoch": 2.711056163899682,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004266231317850842,
      "loss": 5.2383,
      "mean_token_accuracy": 0.21493707746267318,
      "num_tokens": 57233532.0,
      "step": 30700
    },
    {
      "entropy": 5.962484836578369,
      "epoch": 2.7114977039915225,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004265998631068998,
      "loss": 5.2386,
      "mean_token_accuracy": 0.22003307491540908,
      "num_tokens": 57243348.0,
      "step": 30705
    },
    {
      "entropy": 5.979242515563965,
      "epoch": 2.7119392440833625,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042657659145896044,
      "loss": 5.4402,
      "mean_token_accuracy": 0.1984683945775032,
      "num_tokens": 57253028.0,
      "step": 30710
    },
    {
      "entropy": 5.863406848907471,
      "epoch": 2.712380784175203,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042655331684172173,
      "loss": 5.2439,
      "mean_token_accuracy": 0.2142437919974327,
      "num_tokens": 57261843.0,
      "step": 30715
    },
    {
      "entropy": 5.999493408203125,
      "epoch": 2.7128223242670435,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042653003925563975,
      "loss": 5.3421,
      "mean_token_accuracy": 0.20620111674070357,
      "num_tokens": 57271370.0,
      "step": 30720
    },
    {
      "entropy": 5.904567527770996,
      "epoch": 2.7132638643588836,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042650675870117057,
      "loss": 5.2035,
      "mean_token_accuracy": 0.21592871248722076,
      "num_tokens": 57280107.0,
      "step": 30725
    },
    {
      "entropy": 5.848471450805664,
      "epoch": 2.713705404450724,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042648347517877,
      "loss": 5.2101,
      "mean_token_accuracy": 0.22118865251541137,
      "num_tokens": 57289878.0,
      "step": 30730
    },
    {
      "entropy": 5.9242054462432865,
      "epoch": 2.7141469445425646,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042646018868889446,
      "loss": 5.3093,
      "mean_token_accuracy": 0.20534122437238694,
      "num_tokens": 57298591.0,
      "step": 30735
    },
    {
      "entropy": 5.948996543884277,
      "epoch": 2.7145884846344046,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004264368992319999,
      "loss": 5.1981,
      "mean_token_accuracy": 0.2174736201763153,
      "num_tokens": 57307476.0,
      "step": 30740
    },
    {
      "entropy": 5.911013126373291,
      "epoch": 2.715030024726245,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004264136068085426,
      "loss": 5.3341,
      "mean_token_accuracy": 0.20375702530145645,
      "num_tokens": 57316953.0,
      "step": 30745
    },
    {
      "entropy": 5.930068254470825,
      "epoch": 2.7154715648180856,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042639031141897884,
      "loss": 5.2507,
      "mean_token_accuracy": 0.2218890368938446,
      "num_tokens": 57326317.0,
      "step": 30750
    },
    {
      "entropy": 5.901408767700195,
      "epoch": 2.7159131049099257,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000426367013063765,
      "loss": 5.206,
      "mean_token_accuracy": 0.21635644733905793,
      "num_tokens": 57335706.0,
      "step": 30755
    },
    {
      "entropy": 5.9619041919708256,
      "epoch": 2.716354645001766,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042634371174335743,
      "loss": 5.3616,
      "mean_token_accuracy": 0.21109147369861603,
      "num_tokens": 57345491.0,
      "step": 30760
    },
    {
      "entropy": 5.897224378585816,
      "epoch": 2.7167961850936067,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042632040745821273,
      "loss": 5.321,
      "mean_token_accuracy": 0.20174132734537126,
      "num_tokens": 57355158.0,
      "step": 30765
    },
    {
      "entropy": 5.968891143798828,
      "epoch": 2.7172377251854467,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004262971002087872,
      "loss": 5.3587,
      "mean_token_accuracy": 0.2140426903963089,
      "num_tokens": 57364327.0,
      "step": 30770
    },
    {
      "entropy": 5.9693397045135494,
      "epoch": 2.717679265277287,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004262737899955376,
      "loss": 5.3133,
      "mean_token_accuracy": 0.2083237960934639,
      "num_tokens": 57373366.0,
      "step": 30775
    },
    {
      "entropy": 5.9475670337677,
      "epoch": 2.7181208053691277,
      "grad_norm": 1.125,
      "learning_rate": 0.00042625047681892043,
      "loss": 5.2981,
      "mean_token_accuracy": 0.21824266165494918,
      "num_tokens": 57382221.0,
      "step": 30780
    },
    {
      "entropy": 5.972321462631226,
      "epoch": 2.7185623454609678,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004262271606793924,
      "loss": 5.4009,
      "mean_token_accuracy": 0.20402777343988418,
      "num_tokens": 57391937.0,
      "step": 30785
    },
    {
      "entropy": 5.9689031600952145,
      "epoch": 2.7190038855528083,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042620384157741035,
      "loss": 5.3059,
      "mean_token_accuracy": 0.21250333338975907,
      "num_tokens": 57401189.0,
      "step": 30790
    },
    {
      "entropy": 6.005610752105713,
      "epoch": 2.7194454256446488,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042618051951343104,
      "loss": 5.2959,
      "mean_token_accuracy": 0.2173563465476036,
      "num_tokens": 57410248.0,
      "step": 30795
    },
    {
      "entropy": 6.012479543685913,
      "epoch": 2.719886965736489,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004261571944879114,
      "loss": 5.3751,
      "mean_token_accuracy": 0.20168106108903885,
      "num_tokens": 57420450.0,
      "step": 30800
    },
    {
      "entropy": 5.948327112197876,
      "epoch": 2.7203285058283293,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004261338665013082,
      "loss": 5.3371,
      "mean_token_accuracy": 0.20244843661785125,
      "num_tokens": 57430381.0,
      "step": 30805
    },
    {
      "entropy": 6.002298021316529,
      "epoch": 2.72077004592017,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004261105355540785,
      "loss": 5.3913,
      "mean_token_accuracy": 0.2103773683309555,
      "num_tokens": 57439252.0,
      "step": 30810
    },
    {
      "entropy": 6.052457809448242,
      "epoch": 2.72121158601201,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004260872016466794,
      "loss": 5.492,
      "mean_token_accuracy": 0.19243690520524978,
      "num_tokens": 57449143.0,
      "step": 30815
    },
    {
      "entropy": 5.970091962814331,
      "epoch": 2.7216531261038504,
      "grad_norm": 1.125,
      "learning_rate": 0.00042606386477956796,
      "loss": 5.3184,
      "mean_token_accuracy": 0.2108898788690567,
      "num_tokens": 57458608.0,
      "step": 30820
    },
    {
      "entropy": 6.005902576446533,
      "epoch": 2.7220946661956904,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004260405249532013,
      "loss": 5.3659,
      "mean_token_accuracy": 0.20499761551618575,
      "num_tokens": 57467616.0,
      "step": 30825
    },
    {
      "entropy": 6.007714796066284,
      "epoch": 2.722536206287531,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042601718216803657,
      "loss": 5.3165,
      "mean_token_accuracy": 0.20665229260921478,
      "num_tokens": 57477460.0,
      "step": 30830
    },
    {
      "entropy": 6.007090520858765,
      "epoch": 2.7229777463793714,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042599383642453125,
      "loss": 5.3792,
      "mean_token_accuracy": 0.20567699819803237,
      "num_tokens": 57486287.0,
      "step": 30835
    },
    {
      "entropy": 5.9473639011383055,
      "epoch": 2.7234192864712115,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004259704877231425,
      "loss": 5.2991,
      "mean_token_accuracy": 0.21077223271131515,
      "num_tokens": 57495891.0,
      "step": 30840
    },
    {
      "entropy": 5.935214614868164,
      "epoch": 2.723860826563052,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042594713606432775,
      "loss": 5.3272,
      "mean_token_accuracy": 0.2111702784895897,
      "num_tokens": 57505384.0,
      "step": 30845
    },
    {
      "entropy": 5.994154167175293,
      "epoch": 2.724302366654892,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042592378144854447,
      "loss": 5.3378,
      "mean_token_accuracy": 0.20696205794811248,
      "num_tokens": 57514106.0,
      "step": 30850
    },
    {
      "entropy": 5.884390497207642,
      "epoch": 2.7247439067467325,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042590042387625023,
      "loss": 5.2434,
      "mean_token_accuracy": 0.21500762850046157,
      "num_tokens": 57524018.0,
      "step": 30855
    },
    {
      "entropy": 5.950527143478394,
      "epoch": 2.725185446838573,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004258770633479024,
      "loss": 5.2441,
      "mean_token_accuracy": 0.21619687378406524,
      "num_tokens": 57533393.0,
      "step": 30860
    },
    {
      "entropy": 5.818268489837647,
      "epoch": 2.725626986930413,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004258536998639588,
      "loss": 5.1748,
      "mean_token_accuracy": 0.22946167290210723,
      "num_tokens": 57542442.0,
      "step": 30865
    },
    {
      "entropy": 5.941956233978272,
      "epoch": 2.7260685270222536,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004258303334248769,
      "loss": 5.2837,
      "mean_token_accuracy": 0.2163831487298012,
      "num_tokens": 57552367.0,
      "step": 30870
    },
    {
      "entropy": 5.94041256904602,
      "epoch": 2.726510067114094,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004258069640311146,
      "loss": 5.345,
      "mean_token_accuracy": 0.20353465974330903,
      "num_tokens": 57562291.0,
      "step": 30875
    },
    {
      "entropy": 6.085751533508301,
      "epoch": 2.726951607205934,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004257835916831296,
      "loss": 5.4054,
      "mean_token_accuracy": 0.2067608654499054,
      "num_tokens": 57571490.0,
      "step": 30880
    },
    {
      "entropy": 5.962321090698242,
      "epoch": 2.7273931472977746,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042576021638137984,
      "loss": 5.3557,
      "mean_token_accuracy": 0.20606574416160583,
      "num_tokens": 57580010.0,
      "step": 30885
    },
    {
      "entropy": 5.9557130336761475,
      "epoch": 2.727834687389615,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004257368381263232,
      "loss": 5.3595,
      "mean_token_accuracy": 0.2090591385960579,
      "num_tokens": 57589773.0,
      "step": 30890
    },
    {
      "entropy": 5.964918375015259,
      "epoch": 2.728276227481455,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042571345691841763,
      "loss": 5.3,
      "mean_token_accuracy": 0.21055728644132615,
      "num_tokens": 57598358.0,
      "step": 30895
    },
    {
      "entropy": 5.925642633438111,
      "epoch": 2.7287177675732956,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042569007275812103,
      "loss": 5.2838,
      "mean_token_accuracy": 0.21277842223644255,
      "num_tokens": 57607447.0,
      "step": 30900
    },
    {
      "entropy": 5.865810298919678,
      "epoch": 2.729159307665136,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004256666856458918,
      "loss": 5.2622,
      "mean_token_accuracy": 0.2157633438706398,
      "num_tokens": 57617253.0,
      "step": 30905
    },
    {
      "entropy": 5.9832172870635985,
      "epoch": 2.729600847756976,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042564329558218776,
      "loss": 5.3593,
      "mean_token_accuracy": 0.20532226860523223,
      "num_tokens": 57627304.0,
      "step": 30910
    },
    {
      "entropy": 5.990463447570801,
      "epoch": 2.7300423878488167,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004256199025674673,
      "loss": 5.3166,
      "mean_token_accuracy": 0.2117823228240013,
      "num_tokens": 57635639.0,
      "step": 30915
    },
    {
      "entropy": 5.9731245040893555,
      "epoch": 2.730483927940657,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042559650660218857,
      "loss": 5.3342,
      "mean_token_accuracy": 0.20868618786334991,
      "num_tokens": 57644986.0,
      "step": 30920
    },
    {
      "entropy": 5.913240051269531,
      "epoch": 2.7309254680324972,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042557310768680994,
      "loss": 5.2579,
      "mean_token_accuracy": 0.2060508206486702,
      "num_tokens": 57654506.0,
      "step": 30925
    },
    {
      "entropy": 5.957946252822876,
      "epoch": 2.7313670081243377,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004255497058217898,
      "loss": 5.3055,
      "mean_token_accuracy": 0.20932869017124175,
      "num_tokens": 57663075.0,
      "step": 30930
    },
    {
      "entropy": 5.984567022323608,
      "epoch": 2.7318085482161782,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004255263010075865,
      "loss": 5.3359,
      "mean_token_accuracy": 0.21470957696437837,
      "num_tokens": 57672483.0,
      "step": 30935
    },
    {
      "entropy": 5.978055095672607,
      "epoch": 2.7322500883080183,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004255028932446586,
      "loss": 5.3526,
      "mean_token_accuracy": 0.20980747044086456,
      "num_tokens": 57681624.0,
      "step": 30940
    },
    {
      "entropy": 5.984005641937256,
      "epoch": 2.732691628399859,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004254794825334645,
      "loss": 5.3371,
      "mean_token_accuracy": 0.20743974596261977,
      "num_tokens": 57692166.0,
      "step": 30945
    },
    {
      "entropy": 6.018848609924317,
      "epoch": 2.7331331684916993,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004254560688744631,
      "loss": 5.3919,
      "mean_token_accuracy": 0.20627410113811492,
      "num_tokens": 57701984.0,
      "step": 30950
    },
    {
      "entropy": 5.903823041915894,
      "epoch": 2.7335747085835393,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042543265226811273,
      "loss": 5.1537,
      "mean_token_accuracy": 0.2270692214369774,
      "num_tokens": 57710870.0,
      "step": 30955
    },
    {
      "entropy": 5.944604730606079,
      "epoch": 2.73401624867538,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004254092327148723,
      "loss": 5.3409,
      "mean_token_accuracy": 0.20840571373701094,
      "num_tokens": 57719381.0,
      "step": 30960
    },
    {
      "entropy": 5.969965600967408,
      "epoch": 2.73445778876722,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042538581021520056,
      "loss": 5.335,
      "mean_token_accuracy": 0.20673130750656127,
      "num_tokens": 57728551.0,
      "step": 30965
    },
    {
      "entropy": 5.895312356948852,
      "epoch": 2.7348993288590604,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042536238476955637,
      "loss": 5.2777,
      "mean_token_accuracy": 0.21009668409824372,
      "num_tokens": 57737532.0,
      "step": 30970
    },
    {
      "entropy": 5.907098054885864,
      "epoch": 2.735340868950901,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004253389563783986,
      "loss": 5.2333,
      "mean_token_accuracy": 0.2116685152053833,
      "num_tokens": 57746207.0,
      "step": 30975
    },
    {
      "entropy": 5.94420428276062,
      "epoch": 2.735782409042741,
      "grad_norm": 1.125,
      "learning_rate": 0.00042531552504218605,
      "loss": 5.3798,
      "mean_token_accuracy": 0.2035423547029495,
      "num_tokens": 57755876.0,
      "step": 30980
    },
    {
      "entropy": 5.9688966274261475,
      "epoch": 2.7362239491345814,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004252920907613779,
      "loss": 5.2712,
      "mean_token_accuracy": 0.21347110867500305,
      "num_tokens": 57764862.0,
      "step": 30985
    },
    {
      "entropy": 5.944360685348511,
      "epoch": 2.7366654892264215,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042526865353643316,
      "loss": 5.274,
      "mean_token_accuracy": 0.21115680634975434,
      "num_tokens": 57773910.0,
      "step": 30990
    },
    {
      "entropy": 5.984404373168945,
      "epoch": 2.737107029318262,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004252452133678111,
      "loss": 5.3673,
      "mean_token_accuracy": 0.20226613879203797,
      "num_tokens": 57783373.0,
      "step": 30995
    },
    {
      "entropy": 5.990754127502441,
      "epoch": 2.7375485694101025,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042522177025597065,
      "loss": 5.2828,
      "mean_token_accuracy": 0.2196209192276001,
      "num_tokens": 57792214.0,
      "step": 31000
    },
    {
      "entropy": 5.944048643112183,
      "epoch": 2.7379901095019425,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004251983242013711,
      "loss": 5.2755,
      "mean_token_accuracy": 0.21414539366960525,
      "num_tokens": 57801250.0,
      "step": 31005
    },
    {
      "entropy": 5.891736078262329,
      "epoch": 2.738431649593783,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042517487520447183,
      "loss": 5.2373,
      "mean_token_accuracy": 0.21856038570404052,
      "num_tokens": 57810814.0,
      "step": 31010
    },
    {
      "entropy": 5.916189908981323,
      "epoch": 2.7388731896856235,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004251514232657323,
      "loss": 5.251,
      "mean_token_accuracy": 0.21274961531162262,
      "num_tokens": 57819804.0,
      "step": 31015
    },
    {
      "entropy": 5.930148410797119,
      "epoch": 2.7393147297774636,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042512796838561163,
      "loss": 5.3075,
      "mean_token_accuracy": 0.20902360677719117,
      "num_tokens": 57829321.0,
      "step": 31020
    },
    {
      "entropy": 5.908321475982666,
      "epoch": 2.739756269869304,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004251045105645695,
      "loss": 5.2577,
      "mean_token_accuracy": 0.21154509782791137,
      "num_tokens": 57838885.0,
      "step": 31025
    },
    {
      "entropy": 5.9552881717681885,
      "epoch": 2.7401978099611446,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042508104980306527,
      "loss": 5.2798,
      "mean_token_accuracy": 0.2047302708029747,
      "num_tokens": 57848856.0,
      "step": 31030
    },
    {
      "entropy": 5.942824125289917,
      "epoch": 2.7406393500529846,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042505758610155873,
      "loss": 5.2924,
      "mean_token_accuracy": 0.21159034073352814,
      "num_tokens": 57857527.0,
      "step": 31035
    },
    {
      "entropy": 5.920636940002441,
      "epoch": 2.741080890144825,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042503411946050943,
      "loss": 5.3262,
      "mean_token_accuracy": 0.20624795109033583,
      "num_tokens": 57866946.0,
      "step": 31040
    },
    {
      "entropy": 6.027286767959595,
      "epoch": 2.7415224302366656,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004250106498803771,
      "loss": 5.3524,
      "mean_token_accuracy": 0.20834006518125534,
      "num_tokens": 57876400.0,
      "step": 31045
    },
    {
      "entropy": 5.8422223091125485,
      "epoch": 2.7419639703285057,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004249871773616214,
      "loss": 5.1792,
      "mean_token_accuracy": 0.2230550765991211,
      "num_tokens": 57885224.0,
      "step": 31050
    },
    {
      "entropy": 5.92926516532898,
      "epoch": 2.742405510420346,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042496370190470216,
      "loss": 5.3203,
      "mean_token_accuracy": 0.21049046963453294,
      "num_tokens": 57894379.0,
      "step": 31055
    },
    {
      "entropy": 5.945979595184326,
      "epoch": 2.7428470505121867,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004249402235100793,
      "loss": 5.3315,
      "mean_token_accuracy": 0.21014340668916703,
      "num_tokens": 57903851.0,
      "step": 31060
    },
    {
      "entropy": 6.017545127868653,
      "epoch": 2.7432885906040267,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004249167421782127,
      "loss": 5.4111,
      "mean_token_accuracy": 0.1962215632200241,
      "num_tokens": 57912705.0,
      "step": 31065
    },
    {
      "entropy": 5.931126308441162,
      "epoch": 2.7437301306958672,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004248932579095625,
      "loss": 5.225,
      "mean_token_accuracy": 0.2194824904203415,
      "num_tokens": 57921623.0,
      "step": 31070
    },
    {
      "entropy": 5.870238018035889,
      "epoch": 2.7441716707877077,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042486977070458856,
      "loss": 5.2576,
      "mean_token_accuracy": 0.21097144335508347,
      "num_tokens": 57930673.0,
      "step": 31075
    },
    {
      "entropy": 5.876417589187622,
      "epoch": 2.744613210879548,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004248462805637511,
      "loss": 5.2782,
      "mean_token_accuracy": 0.2173558384180069,
      "num_tokens": 57939727.0,
      "step": 31080
    },
    {
      "entropy": 5.953379058837891,
      "epoch": 2.7450547509713883,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042482278748751014,
      "loss": 5.2886,
      "mean_token_accuracy": 0.2087733581662178,
      "num_tokens": 57949830.0,
      "step": 31085
    },
    {
      "entropy": 5.949925231933594,
      "epoch": 2.745496291063229,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000424799291476326,
      "loss": 5.3246,
      "mean_token_accuracy": 0.2165609210729599,
      "num_tokens": 57959577.0,
      "step": 31090
    },
    {
      "entropy": 5.872238683700561,
      "epoch": 2.745937831155069,
      "grad_norm": 1.109375,
      "learning_rate": 0.000424775792530659,
      "loss": 5.2637,
      "mean_token_accuracy": 0.21949288696050645,
      "num_tokens": 57968569.0,
      "step": 31095
    },
    {
      "entropy": 5.960246753692627,
      "epoch": 2.7463793712469093,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004247522906509693,
      "loss": 5.3111,
      "mean_token_accuracy": 0.2066404178738594,
      "num_tokens": 57978285.0,
      "step": 31100
    },
    {
      "entropy": 5.932980918884278,
      "epoch": 2.7468209113387494,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004247287858377176,
      "loss": 5.343,
      "mean_token_accuracy": 0.21259113252162934,
      "num_tokens": 57987561.0,
      "step": 31105
    },
    {
      "entropy": 5.907671356201172,
      "epoch": 2.74726245143059,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000424705278091364,
      "loss": 5.2748,
      "mean_token_accuracy": 0.21183860599994658,
      "num_tokens": 57997999.0,
      "step": 31110
    },
    {
      "entropy": 5.929106855392456,
      "epoch": 2.7477039915224304,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004246817674123691,
      "loss": 5.3422,
      "mean_token_accuracy": 0.20681235790252686,
      "num_tokens": 58007655.0,
      "step": 31115
    },
    {
      "entropy": 6.001876354217529,
      "epoch": 2.7481455316142704,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004246582538011935,
      "loss": 5.3052,
      "mean_token_accuracy": 0.2133523628115654,
      "num_tokens": 58016542.0,
      "step": 31120
    },
    {
      "entropy": 5.90468397140503,
      "epoch": 2.748587071706111,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004246347372582979,
      "loss": 5.2779,
      "mean_token_accuracy": 0.21416825354099273,
      "num_tokens": 58026169.0,
      "step": 31125
    },
    {
      "entropy": 5.908491325378418,
      "epoch": 2.749028611797951,
      "grad_norm": 1.125,
      "learning_rate": 0.00042461121778414286,
      "loss": 5.2872,
      "mean_token_accuracy": 0.21004119962453843,
      "num_tokens": 58036006.0,
      "step": 31130
    },
    {
      "entropy": 5.952009868621826,
      "epoch": 2.7494701518897915,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042458769537918916,
      "loss": 5.2513,
      "mean_token_accuracy": 0.21239465326070786,
      "num_tokens": 58044601.0,
      "step": 31135
    },
    {
      "entropy": 5.91278305053711,
      "epoch": 2.749911691981632,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004245641700438976,
      "loss": 5.2748,
      "mean_token_accuracy": 0.2158896803855896,
      "num_tokens": 58053492.0,
      "step": 31140
    },
    {
      "entropy": 5.918777227401733,
      "epoch": 2.750353232073472,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000424540641778729,
      "loss": 5.2764,
      "mean_token_accuracy": 0.22039460986852646,
      "num_tokens": 58062457.0,
      "step": 31145
    },
    {
      "entropy": 5.845924425125122,
      "epoch": 2.7507947721653125,
      "grad_norm": 1.125,
      "learning_rate": 0.0004245171105841443,
      "loss": 5.2248,
      "mean_token_accuracy": 0.22352837324142455,
      "num_tokens": 58071432.0,
      "step": 31150
    },
    {
      "entropy": 5.936443281173706,
      "epoch": 2.751236312257153,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042449357646060437,
      "loss": 5.4049,
      "mean_token_accuracy": 0.20769146978855133,
      "num_tokens": 58080588.0,
      "step": 31155
    },
    {
      "entropy": 5.929170370101929,
      "epoch": 2.751677852348993,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004244700394085704,
      "loss": 5.1851,
      "mean_token_accuracy": 0.2184702441096306,
      "num_tokens": 58090467.0,
      "step": 31160
    },
    {
      "entropy": 5.986742925643921,
      "epoch": 2.7521193924408336,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042444649942850325,
      "loss": 5.362,
      "mean_token_accuracy": 0.20194568037986754,
      "num_tokens": 58099746.0,
      "step": 31165
    },
    {
      "entropy": 5.889603614807129,
      "epoch": 2.752560932532674,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042442295652086434,
      "loss": 5.3117,
      "mean_token_accuracy": 0.21422937214374543,
      "num_tokens": 58109039.0,
      "step": 31170
    },
    {
      "entropy": 5.909534692764282,
      "epoch": 2.753002472624514,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004243994106861145,
      "loss": 5.2178,
      "mean_token_accuracy": 0.21855864226818084,
      "num_tokens": 58117567.0,
      "step": 31175
    },
    {
      "entropy": 5.890165853500366,
      "epoch": 2.7534440127163546,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004243758619247153,
      "loss": 5.2137,
      "mean_token_accuracy": 0.21214772164821624,
      "num_tokens": 58125918.0,
      "step": 31180
    },
    {
      "entropy": 5.919575357437134,
      "epoch": 2.753885552808195,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00042435231023712784,
      "loss": 5.3626,
      "mean_token_accuracy": 0.20211314260959626,
      "num_tokens": 58135862.0,
      "step": 31185
    },
    {
      "entropy": 5.978096437454224,
      "epoch": 2.754327092900035,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042432875562381354,
      "loss": 5.4119,
      "mean_token_accuracy": 0.1984200194478035,
      "num_tokens": 58145182.0,
      "step": 31190
    },
    {
      "entropy": 5.961599254608155,
      "epoch": 2.7547686329918757,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004243051980852339,
      "loss": 5.2981,
      "mean_token_accuracy": 0.21272313296794892,
      "num_tokens": 58153764.0,
      "step": 31195
    },
    {
      "entropy": 5.953351402282715,
      "epoch": 2.755210173083716,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004242816376218504,
      "loss": 5.2459,
      "mean_token_accuracy": 0.21474723219871522,
      "num_tokens": 58162982.0,
      "step": 31200
    },
    {
      "entropy": 5.920239877700806,
      "epoch": 2.7556517131755562,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042425807423412444,
      "loss": 5.3001,
      "mean_token_accuracy": 0.2045394718647003,
      "num_tokens": 58172300.0,
      "step": 31205
    },
    {
      "entropy": 5.940036630630493,
      "epoch": 2.7560932532673967,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042423450792251774,
      "loss": 5.2725,
      "mean_token_accuracy": 0.21447224617004396,
      "num_tokens": 58181494.0,
      "step": 31210
    },
    {
      "entropy": 5.943058204650879,
      "epoch": 2.7565347933592372,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042421093868749195,
      "loss": 5.3286,
      "mean_token_accuracy": 0.20875296592712403,
      "num_tokens": 58191500.0,
      "step": 31215
    },
    {
      "entropy": 5.929742383956909,
      "epoch": 2.7569763334510773,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004241873665295086,
      "loss": 5.3061,
      "mean_token_accuracy": 0.2188175231218338,
      "num_tokens": 58201012.0,
      "step": 31220
    },
    {
      "entropy": 5.9187782287597654,
      "epoch": 2.757417873542918,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004241637914490297,
      "loss": 5.2578,
      "mean_token_accuracy": 0.21392053067684175,
      "num_tokens": 58209901.0,
      "step": 31225
    },
    {
      "entropy": 5.981106853485107,
      "epoch": 2.7578594136347583,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042414021344651685,
      "loss": 5.3355,
      "mean_token_accuracy": 0.20548055320978165,
      "num_tokens": 58219809.0,
      "step": 31230
    },
    {
      "entropy": 5.932482671737671,
      "epoch": 2.7583009537265983,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004241166325224321,
      "loss": 5.2589,
      "mean_token_accuracy": 0.21328631937503814,
      "num_tokens": 58229626.0,
      "step": 31235
    },
    {
      "entropy": 5.949576950073242,
      "epoch": 2.758742493818439,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004240930486772374,
      "loss": 5.2836,
      "mean_token_accuracy": 0.21364456564188003,
      "num_tokens": 58239533.0,
      "step": 31240
    },
    {
      "entropy": 5.8957963466644285,
      "epoch": 2.759184033910279,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042406946191139456,
      "loss": 5.2423,
      "mean_token_accuracy": 0.21442670077085496,
      "num_tokens": 58248998.0,
      "step": 31245
    },
    {
      "entropy": 5.939557266235352,
      "epoch": 2.7596255740021194,
      "grad_norm": 1.125,
      "learning_rate": 0.00042404587222536573,
      "loss": 5.2362,
      "mean_token_accuracy": 0.2129012018442154,
      "num_tokens": 58258024.0,
      "step": 31250
    },
    {
      "entropy": 5.940679502487183,
      "epoch": 2.76006711409396,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00042402227961961314,
      "loss": 5.3187,
      "mean_token_accuracy": 0.20848273485898972,
      "num_tokens": 58267937.0,
      "step": 31255
    },
    {
      "entropy": 5.9519702911376955,
      "epoch": 2.7605086541858,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004239986840945988,
      "loss": 5.3519,
      "mean_token_accuracy": 0.20805926322937013,
      "num_tokens": 58276834.0,
      "step": 31260
    },
    {
      "entropy": 5.932507467269898,
      "epoch": 2.7609501942776404,
      "grad_norm": 1.125,
      "learning_rate": 0.0004239750856507849,
      "loss": 5.3552,
      "mean_token_accuracy": 0.20837851315736772,
      "num_tokens": 58286816.0,
      "step": 31265
    },
    {
      "entropy": 5.906787729263305,
      "epoch": 2.7613917343694805,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004239514842886339,
      "loss": 5.2591,
      "mean_token_accuracy": 0.2226272478699684,
      "num_tokens": 58296086.0,
      "step": 31270
    },
    {
      "entropy": 5.955360984802246,
      "epoch": 2.761833274461321,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042392788000860797,
      "loss": 5.3283,
      "mean_token_accuracy": 0.20577773451805115,
      "num_tokens": 58305851.0,
      "step": 31275
    },
    {
      "entropy": 5.827650547027588,
      "epoch": 2.7622748145531615,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042390427281116966,
      "loss": 5.192,
      "mean_token_accuracy": 0.21920021772384643,
      "num_tokens": 58314693.0,
      "step": 31280
    },
    {
      "entropy": 5.879519081115722,
      "epoch": 2.7627163546450015,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042388066269678115,
      "loss": 5.2628,
      "mean_token_accuracy": 0.21621132344007493,
      "num_tokens": 58323425.0,
      "step": 31285
    },
    {
      "entropy": 5.926635980606079,
      "epoch": 2.763157894736842,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042385704966590537,
      "loss": 5.2395,
      "mean_token_accuracy": 0.2183494135737419,
      "num_tokens": 58332651.0,
      "step": 31290
    },
    {
      "entropy": 5.938325071334839,
      "epoch": 2.7635994348286825,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004238334337190045,
      "loss": 5.3198,
      "mean_token_accuracy": 0.20766104608774186,
      "num_tokens": 58342366.0,
      "step": 31295
    },
    {
      "entropy": 5.923809289932251,
      "epoch": 2.7640409749205226,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042380981485654127,
      "loss": 5.2869,
      "mean_token_accuracy": 0.21947538405656813,
      "num_tokens": 58352370.0,
      "step": 31300
    },
    {
      "entropy": 5.990560531616211,
      "epoch": 2.764482515012363,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042378619307897853,
      "loss": 5.3764,
      "mean_token_accuracy": 0.21409311890602112,
      "num_tokens": 58360819.0,
      "step": 31305
    },
    {
      "entropy": 6.011930751800537,
      "epoch": 2.7649240551042036,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004237625683867788,
      "loss": 5.3939,
      "mean_token_accuracy": 0.20437879264354705,
      "num_tokens": 58370744.0,
      "step": 31310
    },
    {
      "entropy": 5.995243263244629,
      "epoch": 2.7653655951960436,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042373894078040497,
      "loss": 5.3034,
      "mean_token_accuracy": 0.21344417482614517,
      "num_tokens": 58380965.0,
      "step": 31315
    },
    {
      "entropy": 5.967277002334595,
      "epoch": 2.765807135287884,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042371531026031995,
      "loss": 5.293,
      "mean_token_accuracy": 0.2086782142519951,
      "num_tokens": 58390386.0,
      "step": 31320
    },
    {
      "entropy": 5.852947568893432,
      "epoch": 2.7662486753797246,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004236916768269865,
      "loss": 5.2138,
      "mean_token_accuracy": 0.2166598841547966,
      "num_tokens": 58399603.0,
      "step": 31325
    },
    {
      "entropy": 5.945249080657959,
      "epoch": 2.7666902154715647,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004236680404808677,
      "loss": 5.262,
      "mean_token_accuracy": 0.2166685938835144,
      "num_tokens": 58408627.0,
      "step": 31330
    },
    {
      "entropy": 5.962018632888794,
      "epoch": 2.767131755563405,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004236444012224266,
      "loss": 5.2974,
      "mean_token_accuracy": 0.2189066082239151,
      "num_tokens": 58418132.0,
      "step": 31335
    },
    {
      "entropy": 5.883098602294922,
      "epoch": 2.7675732956552457,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004236207590521261,
      "loss": 5.3267,
      "mean_token_accuracy": 0.21218224614858627,
      "num_tokens": 58427298.0,
      "step": 31340
    },
    {
      "entropy": 5.862355661392212,
      "epoch": 2.7680148357470857,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004235971139704295,
      "loss": 5.2777,
      "mean_token_accuracy": 0.22102488726377487,
      "num_tokens": 58436977.0,
      "step": 31345
    },
    {
      "entropy": 5.966272783279419,
      "epoch": 2.7684563758389262,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004235734659777999,
      "loss": 5.3793,
      "mean_token_accuracy": 0.20386277586221696,
      "num_tokens": 58446801.0,
      "step": 31350
    },
    {
      "entropy": 5.976469612121582,
      "epoch": 2.7688979159307667,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004235498150747007,
      "loss": 5.2562,
      "mean_token_accuracy": 0.21505590975284578,
      "num_tokens": 58456625.0,
      "step": 31355
    },
    {
      "entropy": 6.012414646148682,
      "epoch": 2.769339456022607,
      "grad_norm": 1.125,
      "learning_rate": 0.0004235261612615951,
      "loss": 5.2773,
      "mean_token_accuracy": 0.20657749325037003,
      "num_tokens": 58465989.0,
      "step": 31360
    },
    {
      "entropy": 5.88777232170105,
      "epoch": 2.7697809961144473,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004235025045389464,
      "loss": 5.3045,
      "mean_token_accuracy": 0.21589075028896332,
      "num_tokens": 58475703.0,
      "step": 31365
    },
    {
      "entropy": 5.950919342041016,
      "epoch": 2.7702225362062878,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004234788449072181,
      "loss": 5.2309,
      "mean_token_accuracy": 0.21906739771366118,
      "num_tokens": 58485192.0,
      "step": 31370
    },
    {
      "entropy": 6.014928960800171,
      "epoch": 2.770664076298128,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042345518236687376,
      "loss": 5.3423,
      "mean_token_accuracy": 0.20608380436897278,
      "num_tokens": 58494275.0,
      "step": 31375
    },
    {
      "entropy": 5.869299221038818,
      "epoch": 2.7711056163899683,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042343151691837675,
      "loss": 5.2259,
      "mean_token_accuracy": 0.2223943680524826,
      "num_tokens": 58504513.0,
      "step": 31380
    },
    {
      "entropy": 5.938451099395752,
      "epoch": 2.7715471564818084,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004234078485621907,
      "loss": 5.3418,
      "mean_token_accuracy": 0.20863808542490006,
      "num_tokens": 58514190.0,
      "step": 31385
    },
    {
      "entropy": 5.858577537536621,
      "epoch": 2.771988696573649,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004233841772987794,
      "loss": 5.1547,
      "mean_token_accuracy": 0.23686596602201462,
      "num_tokens": 58524189.0,
      "step": 31390
    },
    {
      "entropy": 5.957919120788574,
      "epoch": 2.7724302366654894,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004233605031286065,
      "loss": 5.2808,
      "mean_token_accuracy": 0.2033315047621727,
      "num_tokens": 58533899.0,
      "step": 31395
    },
    {
      "entropy": 5.956649494171143,
      "epoch": 2.7728717767573294,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042333682605213556,
      "loss": 5.265,
      "mean_token_accuracy": 0.2159938707947731,
      "num_tokens": 58543523.0,
      "step": 31400
    },
    {
      "entropy": 5.878533506393433,
      "epoch": 2.77331331684917,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042331314606983063,
      "loss": 5.2402,
      "mean_token_accuracy": 0.21454541832208635,
      "num_tokens": 58552623.0,
      "step": 31405
    },
    {
      "entropy": 5.922834205627441,
      "epoch": 2.77375485694101,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004232894631821556,
      "loss": 5.3555,
      "mean_token_accuracy": 0.20867476612329483,
      "num_tokens": 58562194.0,
      "step": 31410
    },
    {
      "entropy": 5.915790176391601,
      "epoch": 2.7741963970328505,
      "grad_norm": 1.125,
      "learning_rate": 0.00042326577738957424,
      "loss": 5.2521,
      "mean_token_accuracy": 0.21416175365447998,
      "num_tokens": 58571715.0,
      "step": 31415
    },
    {
      "entropy": 5.9360757827758786,
      "epoch": 2.774637937124691,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042324208869255064,
      "loss": 5.2795,
      "mean_token_accuracy": 0.21246531009674072,
      "num_tokens": 58582305.0,
      "step": 31420
    },
    {
      "entropy": 5.86460313796997,
      "epoch": 2.775079477216531,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004232183970915489,
      "loss": 5.3337,
      "mean_token_accuracy": 0.20862710028886794,
      "num_tokens": 58592825.0,
      "step": 31425
    },
    {
      "entropy": 5.912860965728759,
      "epoch": 2.7755210173083715,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042319470258703293,
      "loss": 5.2427,
      "mean_token_accuracy": 0.21249168664216994,
      "num_tokens": 58601985.0,
      "step": 31430
    },
    {
      "entropy": 5.895394563674927,
      "epoch": 2.775962557400212,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042317100517946706,
      "loss": 5.2181,
      "mean_token_accuracy": 0.21889678090810777,
      "num_tokens": 58611390.0,
      "step": 31435
    },
    {
      "entropy": 6.026500463485718,
      "epoch": 2.776404097492052,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004231473048693155,
      "loss": 5.2893,
      "mean_token_accuracy": 0.21472128927707673,
      "num_tokens": 58621154.0,
      "step": 31440
    },
    {
      "entropy": 5.869780492782593,
      "epoch": 2.7768456375838926,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042312360165704246,
      "loss": 5.1418,
      "mean_token_accuracy": 0.2290532946586609,
      "num_tokens": 58629346.0,
      "step": 31445
    },
    {
      "entropy": 5.898691892623901,
      "epoch": 2.777287177675733,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004230998955431124,
      "loss": 5.3444,
      "mean_token_accuracy": 0.2128448337316513,
      "num_tokens": 58638472.0,
      "step": 31450
    },
    {
      "entropy": 5.934980010986328,
      "epoch": 2.777728717767573,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004230761865279895,
      "loss": 5.2936,
      "mean_token_accuracy": 0.2216157928109169,
      "num_tokens": 58648016.0,
      "step": 31455
    },
    {
      "entropy": 5.950721406936646,
      "epoch": 2.7781702578594136,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042305247461213837,
      "loss": 5.3242,
      "mean_token_accuracy": 0.2066814810037613,
      "num_tokens": 58657667.0,
      "step": 31460
    },
    {
      "entropy": 5.9454700469970705,
      "epoch": 2.778611797951254,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042302875979602345,
      "loss": 5.3633,
      "mean_token_accuracy": 0.20417210161685945,
      "num_tokens": 58667465.0,
      "step": 31465
    },
    {
      "entropy": 5.89295072555542,
      "epoch": 2.779053338043094,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004230050420801093,
      "loss": 5.2262,
      "mean_token_accuracy": 0.2204824835062027,
      "num_tokens": 58676037.0,
      "step": 31470
    },
    {
      "entropy": 5.816945123672485,
      "epoch": 2.7794948781349347,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042298132146486054,
      "loss": 5.2308,
      "mean_token_accuracy": 0.22598933428525925,
      "num_tokens": 58684807.0,
      "step": 31475
    },
    {
      "entropy": 5.893632698059082,
      "epoch": 2.779936418226775,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004229575979507419,
      "loss": 5.2827,
      "mean_token_accuracy": 0.21020852029323578,
      "num_tokens": 58694013.0,
      "step": 31480
    },
    {
      "entropy": 5.951111793518066,
      "epoch": 2.7803779583186152,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042293387153821807,
      "loss": 5.2848,
      "mean_token_accuracy": 0.2186296060681343,
      "num_tokens": 58703328.0,
      "step": 31485
    },
    {
      "entropy": 5.936185312271118,
      "epoch": 2.7808194984104557,
      "grad_norm": 1.125,
      "learning_rate": 0.0004229101422277538,
      "loss": 5.2739,
      "mean_token_accuracy": 0.21518463939428328,
      "num_tokens": 58712130.0,
      "step": 31490
    },
    {
      "entropy": 5.963828372955322,
      "epoch": 2.7812610385022962,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000422886410019814,
      "loss": 5.3256,
      "mean_token_accuracy": 0.20925518721342087,
      "num_tokens": 58721386.0,
      "step": 31495
    },
    {
      "entropy": 5.943588542938232,
      "epoch": 2.7817025785941363,
      "grad_norm": 1.3125,
      "learning_rate": 0.00042286267491486345,
      "loss": 5.2436,
      "mean_token_accuracy": 0.21797479838132858,
      "num_tokens": 58730572.0,
      "step": 31500
    },
    {
      "entropy": 5.928125667572021,
      "epoch": 2.7821441186859768,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004228389369133673,
      "loss": 5.2013,
      "mean_token_accuracy": 0.2167012795805931,
      "num_tokens": 58738605.0,
      "step": 31505
    },
    {
      "entropy": 5.911075115203857,
      "epoch": 2.7825856587778173,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042281519601579035,
      "loss": 5.3285,
      "mean_token_accuracy": 0.20831088572740555,
      "num_tokens": 58747429.0,
      "step": 31510
    },
    {
      "entropy": 5.853201246261596,
      "epoch": 2.7830271988696573,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042279145222259785,
      "loss": 5.2363,
      "mean_token_accuracy": 0.21640751808881759,
      "num_tokens": 58756213.0,
      "step": 31515
    },
    {
      "entropy": 5.940379095077515,
      "epoch": 2.783468738961498,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042276770553425493,
      "loss": 5.2851,
      "mean_token_accuracy": 0.21725283116102218,
      "num_tokens": 58765025.0,
      "step": 31520
    },
    {
      "entropy": 6.01012659072876,
      "epoch": 2.783910279053338,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042274395595122654,
      "loss": 5.3476,
      "mean_token_accuracy": 0.20616718083620073,
      "num_tokens": 58775237.0,
      "step": 31525
    },
    {
      "entropy": 5.904255771636963,
      "epoch": 2.7843518191451784,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004227202034739782,
      "loss": 5.247,
      "mean_token_accuracy": 0.21534183323383332,
      "num_tokens": 58784555.0,
      "step": 31530
    },
    {
      "entropy": 5.92126955986023,
      "epoch": 2.784793359237019,
      "grad_norm": 1.265625,
      "learning_rate": 0.000422696448102975,
      "loss": 5.3102,
      "mean_token_accuracy": 0.20822891891002654,
      "num_tokens": 58793277.0,
      "step": 31535
    },
    {
      "entropy": 5.930372047424316,
      "epoch": 2.785234899328859,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042267268983868246,
      "loss": 5.2713,
      "mean_token_accuracy": 0.21565667390823365,
      "num_tokens": 58802872.0,
      "step": 31540
    },
    {
      "entropy": 5.862581396102906,
      "epoch": 2.7856764394206994,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004226489286815658,
      "loss": 5.261,
      "mean_token_accuracy": 0.21842146068811416,
      "num_tokens": 58811728.0,
      "step": 31545
    },
    {
      "entropy": 5.9424182891845705,
      "epoch": 2.7861179795125395,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004226251646320907,
      "loss": 5.3967,
      "mean_token_accuracy": 0.20806865841150285,
      "num_tokens": 58820689.0,
      "step": 31550
    },
    {
      "entropy": 5.969481468200684,
      "epoch": 2.78655951960438,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004226013976907225,
      "loss": 5.3082,
      "mean_token_accuracy": 0.22166915833950043,
      "num_tokens": 58829062.0,
      "step": 31555
    },
    {
      "entropy": 5.9655712127685545,
      "epoch": 2.7870010596962205,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042257762785792693,
      "loss": 5.2793,
      "mean_token_accuracy": 0.20906503349542618,
      "num_tokens": 58838183.0,
      "step": 31560
    },
    {
      "entropy": 5.858804845809937,
      "epoch": 2.7874425997880605,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042255385513416963,
      "loss": 5.2393,
      "mean_token_accuracy": 0.21823868304491043,
      "num_tokens": 58847305.0,
      "step": 31565
    },
    {
      "entropy": 5.844763326644897,
      "epoch": 2.787884139879901,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004225300795199161,
      "loss": 5.2502,
      "mean_token_accuracy": 0.21357858031988144,
      "num_tokens": 58856161.0,
      "step": 31570
    },
    {
      "entropy": 5.926393318176269,
      "epoch": 2.7883256799717415,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042250630101563224,
      "loss": 5.2865,
      "mean_token_accuracy": 0.21666061580181123,
      "num_tokens": 58865620.0,
      "step": 31575
    },
    {
      "entropy": 5.96656436920166,
      "epoch": 2.7887672200635816,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042248251962178387,
      "loss": 5.3032,
      "mean_token_accuracy": 0.21764093190431594,
      "num_tokens": 58873997.0,
      "step": 31580
    },
    {
      "entropy": 5.916681814193725,
      "epoch": 2.789208760155422,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004224587353388368,
      "loss": 5.3289,
      "mean_token_accuracy": 0.21716069132089616,
      "num_tokens": 58882905.0,
      "step": 31585
    },
    {
      "entropy": 5.954072141647339,
      "epoch": 2.7896503002472626,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00042243494816725696,
      "loss": 5.265,
      "mean_token_accuracy": 0.21108698546886445,
      "num_tokens": 58892551.0,
      "step": 31590
    },
    {
      "entropy": 5.824707317352295,
      "epoch": 2.7900918403391026,
      "grad_norm": 1.125,
      "learning_rate": 0.0004224111581075104,
      "loss": 5.172,
      "mean_token_accuracy": 0.21573340743780137,
      "num_tokens": 58902195.0,
      "step": 31595
    },
    {
      "entropy": 5.9339234828948975,
      "epoch": 2.790533380430943,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042238736516006303,
      "loss": 5.2789,
      "mean_token_accuracy": 0.21933909952640535,
      "num_tokens": 58911055.0,
      "step": 31600
    },
    {
      "entropy": 5.9168304920196535,
      "epoch": 2.7909749205227836,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000422363569325381,
      "loss": 5.2641,
      "mean_token_accuracy": 0.21943263262510299,
      "num_tokens": 58920931.0,
      "step": 31605
    },
    {
      "entropy": 5.986640405654907,
      "epoch": 2.7914164606146237,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004223397706039304,
      "loss": 5.4171,
      "mean_token_accuracy": 0.19969285577535628,
      "num_tokens": 58931571.0,
      "step": 31610
    },
    {
      "entropy": 6.06803822517395,
      "epoch": 2.791858000706464,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004223159689961777,
      "loss": 5.362,
      "mean_token_accuracy": 0.20287423580884933,
      "num_tokens": 58940323.0,
      "step": 31615
    },
    {
      "entropy": 5.970644140243531,
      "epoch": 2.7922995407983047,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042229216450258874,
      "loss": 5.3186,
      "mean_token_accuracy": 0.20651263892650604,
      "num_tokens": 58949865.0,
      "step": 31620
    },
    {
      "entropy": 5.853204250335693,
      "epoch": 2.7927410808901447,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042226835712363015,
      "loss": 5.273,
      "mean_token_accuracy": 0.21790147125720977,
      "num_tokens": 58958075.0,
      "step": 31625
    },
    {
      "entropy": 5.978744983673096,
      "epoch": 2.793182620981985,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004222445468597681,
      "loss": 5.3388,
      "mean_token_accuracy": 0.20886678695678712,
      "num_tokens": 58966985.0,
      "step": 31630
    },
    {
      "entropy": 5.948934841156006,
      "epoch": 2.7936241610738257,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042222073371146925,
      "loss": 5.2234,
      "mean_token_accuracy": 0.21364858746528625,
      "num_tokens": 58976668.0,
      "step": 31635
    },
    {
      "entropy": 5.985609531402588,
      "epoch": 2.7940657011656658,
      "grad_norm": 1.125,
      "learning_rate": 0.0004221969176791999,
      "loss": 5.3635,
      "mean_token_accuracy": 0.205069100856781,
      "num_tokens": 58986867.0,
      "step": 31640
    },
    {
      "entropy": 5.942461252212524,
      "epoch": 2.7945072412575063,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042217309876342655,
      "loss": 5.3282,
      "mean_token_accuracy": 0.20330387204885483,
      "num_tokens": 58996626.0,
      "step": 31645
    },
    {
      "entropy": 5.880248022079468,
      "epoch": 2.7949487813493468,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042214927696461604,
      "loss": 5.2863,
      "mean_token_accuracy": 0.2126637727022171,
      "num_tokens": 59005291.0,
      "step": 31650
    },
    {
      "entropy": 5.9715722560882565,
      "epoch": 2.795390321441187,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042212545228323476,
      "loss": 5.3324,
      "mean_token_accuracy": 0.20894817560911177,
      "num_tokens": 59016308.0,
      "step": 31655
    },
    {
      "entropy": 5.972396421432495,
      "epoch": 2.7958318615330273,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004221016247197497,
      "loss": 5.316,
      "mean_token_accuracy": 0.2057036817073822,
      "num_tokens": 59026069.0,
      "step": 31660
    },
    {
      "entropy": 5.965813016891479,
      "epoch": 2.7962734016248674,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042207779427462733,
      "loss": 5.3363,
      "mean_token_accuracy": 0.20383412390947342,
      "num_tokens": 59036163.0,
      "step": 31665
    },
    {
      "entropy": 5.906683492660522,
      "epoch": 2.796714941716708,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004220539609483347,
      "loss": 5.2754,
      "mean_token_accuracy": 0.21972087174654006,
      "num_tokens": 59045752.0,
      "step": 31670
    },
    {
      "entropy": 5.918186187744141,
      "epoch": 2.7971564818085484,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042203012474133857,
      "loss": 5.217,
      "mean_token_accuracy": 0.216997729241848,
      "num_tokens": 59054515.0,
      "step": 31675
    },
    {
      "entropy": 5.934774780273438,
      "epoch": 2.7975980219003884,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042200628565410594,
      "loss": 5.309,
      "mean_token_accuracy": 0.20888220518827438,
      "num_tokens": 59063166.0,
      "step": 31680
    },
    {
      "entropy": 5.99329571723938,
      "epoch": 2.798039561992229,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042198244368710377,
      "loss": 5.3963,
      "mean_token_accuracy": 0.1937592402100563,
      "num_tokens": 59072133.0,
      "step": 31685
    },
    {
      "entropy": 5.998527574539184,
      "epoch": 2.798481102084069,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042195859884079905,
      "loss": 5.3348,
      "mean_token_accuracy": 0.2098817065358162,
      "num_tokens": 59081968.0,
      "step": 31690
    },
    {
      "entropy": 6.024462413787842,
      "epoch": 2.7989226421759095,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042193475111565915,
      "loss": 5.3865,
      "mean_token_accuracy": 0.1961662381887436,
      "num_tokens": 59091606.0,
      "step": 31695
    },
    {
      "entropy": 5.966808271408081,
      "epoch": 2.79936418226775,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042191090051215085,
      "loss": 5.2311,
      "mean_token_accuracy": 0.21852320432662964,
      "num_tokens": 59099768.0,
      "step": 31700
    },
    {
      "entropy": 5.971528339385986,
      "epoch": 2.79980572235959,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004218870470307417,
      "loss": 5.356,
      "mean_token_accuracy": 0.2078592449426651,
      "num_tokens": 59109071.0,
      "step": 31705
    },
    {
      "entropy": 5.936318635940552,
      "epoch": 2.8002472624514305,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004218631906718987,
      "loss": 5.3443,
      "mean_token_accuracy": 0.20808715969324112,
      "num_tokens": 59117634.0,
      "step": 31710
    },
    {
      "entropy": 5.971993780136108,
      "epoch": 2.800688802543271,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042183933143608943,
      "loss": 5.3378,
      "mean_token_accuracy": 0.20795061439275742,
      "num_tokens": 59127785.0,
      "step": 31715
    },
    {
      "entropy": 5.994748640060425,
      "epoch": 2.801130342635111,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042181546932378114,
      "loss": 5.3362,
      "mean_token_accuracy": 0.21449251621961593,
      "num_tokens": 59137599.0,
      "step": 31720
    },
    {
      "entropy": 5.974083662033081,
      "epoch": 2.8015718827269516,
      "grad_norm": 1.296875,
      "learning_rate": 0.00042179160433544126,
      "loss": 5.3327,
      "mean_token_accuracy": 0.21628940850496292,
      "num_tokens": 59146089.0,
      "step": 31725
    },
    {
      "entropy": 5.9688294410705565,
      "epoch": 2.802013422818792,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004217677364715373,
      "loss": 5.3559,
      "mean_token_accuracy": 0.204853293299675,
      "num_tokens": 59155749.0,
      "step": 31730
    },
    {
      "entropy": 5.971048974990845,
      "epoch": 2.802454962910632,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042174386573253696,
      "loss": 5.3418,
      "mean_token_accuracy": 0.2011984258890152,
      "num_tokens": 59165714.0,
      "step": 31735
    },
    {
      "entropy": 6.039644479751587,
      "epoch": 2.8028965030024726,
      "grad_norm": 1.125,
      "learning_rate": 0.00042171999211890773,
      "loss": 5.3989,
      "mean_token_accuracy": 0.2003950446844101,
      "num_tokens": 59174738.0,
      "step": 31740
    },
    {
      "entropy": 5.90334734916687,
      "epoch": 2.803338043094313,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004216961156311173,
      "loss": 5.2128,
      "mean_token_accuracy": 0.22732720226049424,
      "num_tokens": 59183380.0,
      "step": 31745
    },
    {
      "entropy": 5.874258422851563,
      "epoch": 2.803779583186153,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004216722362696334,
      "loss": 5.2077,
      "mean_token_accuracy": 0.22047798633575438,
      "num_tokens": 59193265.0,
      "step": 31750
    },
    {
      "entropy": 5.920547962188721,
      "epoch": 2.8042211232779937,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042164835403492377,
      "loss": 5.317,
      "mean_token_accuracy": 0.20925978273153306,
      "num_tokens": 59202543.0,
      "step": 31755
    },
    {
      "entropy": 5.98650541305542,
      "epoch": 2.804662663369834,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042162446892745633,
      "loss": 5.3194,
      "mean_token_accuracy": 0.20957317650318147,
      "num_tokens": 59211564.0,
      "step": 31760
    },
    {
      "entropy": 5.949907016754151,
      "epoch": 2.805104203461674,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042160058094769886,
      "loss": 5.3019,
      "mean_token_accuracy": 0.21308634132146836,
      "num_tokens": 59220908.0,
      "step": 31765
    },
    {
      "entropy": 5.998842191696167,
      "epoch": 2.8055457435535147,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042157669009611947,
      "loss": 5.3431,
      "mean_token_accuracy": 0.20331145823001862,
      "num_tokens": 59231230.0,
      "step": 31770
    },
    {
      "entropy": 5.815077543258667,
      "epoch": 2.805987283645355,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042155279637318604,
      "loss": 5.1327,
      "mean_token_accuracy": 0.2285885587334633,
      "num_tokens": 59240356.0,
      "step": 31775
    },
    {
      "entropy": 5.8510082244873045,
      "epoch": 2.8064288237371953,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042152889977936664,
      "loss": 5.2446,
      "mean_token_accuracy": 0.22071040272712708,
      "num_tokens": 59248834.0,
      "step": 31780
    },
    {
      "entropy": 6.046728658676147,
      "epoch": 2.8068703638290358,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004215050003151295,
      "loss": 5.3719,
      "mean_token_accuracy": 0.20741645395755767,
      "num_tokens": 59258683.0,
      "step": 31785
    },
    {
      "entropy": 6.045284605026245,
      "epoch": 2.8073119039208763,
      "grad_norm": 1.125,
      "learning_rate": 0.0004214810979809427,
      "loss": 5.4049,
      "mean_token_accuracy": 0.20199019610881805,
      "num_tokens": 59268291.0,
      "step": 31790
    },
    {
      "entropy": 6.0107080936431885,
      "epoch": 2.8077534440127163,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042145719277727453,
      "loss": 5.4378,
      "mean_token_accuracy": 0.20028250515460969,
      "num_tokens": 59278608.0,
      "step": 31795
    },
    {
      "entropy": 5.904757356643676,
      "epoch": 2.808194984104557,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004214332847045932,
      "loss": 5.2187,
      "mean_token_accuracy": 0.22178030014038086,
      "num_tokens": 59287859.0,
      "step": 31800
    },
    {
      "entropy": 5.845541906356812,
      "epoch": 2.808636524196397,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004214093737633671,
      "loss": 5.1881,
      "mean_token_accuracy": 0.2215498685836792,
      "num_tokens": 59296996.0,
      "step": 31805
    },
    {
      "entropy": 5.872626399993896,
      "epoch": 2.8090780642882374,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042138545995406467,
      "loss": 5.267,
      "mean_token_accuracy": 0.21419845819473265,
      "num_tokens": 59306776.0,
      "step": 31810
    },
    {
      "entropy": 5.8948812007904055,
      "epoch": 2.809519604380078,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004213615432771543,
      "loss": 5.2366,
      "mean_token_accuracy": 0.20906531512737275,
      "num_tokens": 59316271.0,
      "step": 31815
    },
    {
      "entropy": 5.838790464401245,
      "epoch": 2.809961144471918,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004213376237331046,
      "loss": 5.1407,
      "mean_token_accuracy": 0.2263694241642952,
      "num_tokens": 59324654.0,
      "step": 31820
    },
    {
      "entropy": 5.922581005096435,
      "epoch": 2.8104026845637584,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000421313701322384,
      "loss": 5.3423,
      "mean_token_accuracy": 0.20775478184223176,
      "num_tokens": 59333681.0,
      "step": 31825
    },
    {
      "entropy": 5.929658603668213,
      "epoch": 2.8108442246555985,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004212897760454613,
      "loss": 5.3536,
      "mean_token_accuracy": 0.20740125328302383,
      "num_tokens": 59342455.0,
      "step": 31830
    },
    {
      "entropy": 5.930131626129151,
      "epoch": 2.811285764747439,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000421265847902805,
      "loss": 5.2911,
      "mean_token_accuracy": 0.2120332419872284,
      "num_tokens": 59351435.0,
      "step": 31835
    },
    {
      "entropy": 5.983516359329224,
      "epoch": 2.8117273048392795,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042124191689488395,
      "loss": 5.3491,
      "mean_token_accuracy": 0.20689472258090974,
      "num_tokens": 59360832.0,
      "step": 31840
    },
    {
      "entropy": 5.9682066440582275,
      "epoch": 2.8121688449311195,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004212179830221669,
      "loss": 5.2402,
      "mean_token_accuracy": 0.21010085195302963,
      "num_tokens": 59368898.0,
      "step": 31845
    },
    {
      "entropy": 5.896722459793091,
      "epoch": 2.81261038502296,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004211940462851227,
      "loss": 5.2116,
      "mean_token_accuracy": 0.22176071554422377,
      "num_tokens": 59377951.0,
      "step": 31850
    },
    {
      "entropy": 5.887892055511474,
      "epoch": 2.8130519251148005,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042117010668422033,
      "loss": 5.2773,
      "mean_token_accuracy": 0.2162030354142189,
      "num_tokens": 59386988.0,
      "step": 31855
    },
    {
      "entropy": 5.948774957656861,
      "epoch": 2.8134934652066406,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004211461642199287,
      "loss": 5.2129,
      "mean_token_accuracy": 0.22258210331201553,
      "num_tokens": 59396658.0,
      "step": 31860
    },
    {
      "entropy": 5.884593820571899,
      "epoch": 2.813935005298481,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004211222188927168,
      "loss": 5.2316,
      "mean_token_accuracy": 0.214429771900177,
      "num_tokens": 59407086.0,
      "step": 31865
    },
    {
      "entropy": 5.906793022155762,
      "epoch": 2.8143765453903216,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004210982707030538,
      "loss": 5.3272,
      "mean_token_accuracy": 0.20892880856990814,
      "num_tokens": 59416141.0,
      "step": 31870
    },
    {
      "entropy": 5.9471817970275875,
      "epoch": 2.8148180854821616,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042107431965140876,
      "loss": 5.301,
      "mean_token_accuracy": 0.21642542034387588,
      "num_tokens": 59425856.0,
      "step": 31875
    },
    {
      "entropy": 6.007299470901489,
      "epoch": 2.815259625574002,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004210503657382508,
      "loss": 5.3907,
      "mean_token_accuracy": 0.21014566868543624,
      "num_tokens": 59435338.0,
      "step": 31880
    },
    {
      "entropy": 5.9929742336273195,
      "epoch": 2.8157011656658426,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004210264089640493,
      "loss": 5.2648,
      "mean_token_accuracy": 0.2113248273730278,
      "num_tokens": 59444453.0,
      "step": 31885
    },
    {
      "entropy": 5.883859062194825,
      "epoch": 2.8161427057576827,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042100244932927346,
      "loss": 5.189,
      "mean_token_accuracy": 0.22442772686481477,
      "num_tokens": 59453342.0,
      "step": 31890
    },
    {
      "entropy": 5.883956432342529,
      "epoch": 2.816584245849523,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004209784868343927,
      "loss": 5.3092,
      "mean_token_accuracy": 0.2094373270869255,
      "num_tokens": 59462669.0,
      "step": 31895
    },
    {
      "entropy": 5.943943452835083,
      "epoch": 2.8170257859413637,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004209545214798764,
      "loss": 5.3252,
      "mean_token_accuracy": 0.20627207607030867,
      "num_tokens": 59472400.0,
      "step": 31900
    },
    {
      "entropy": 5.9868871688842775,
      "epoch": 2.8174673260332037,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000420930553266194,
      "loss": 5.2398,
      "mean_token_accuracy": 0.21819192469120025,
      "num_tokens": 59482306.0,
      "step": 31905
    },
    {
      "entropy": 5.946674346923828,
      "epoch": 2.817908866125044,
      "grad_norm": 1.125,
      "learning_rate": 0.0004209065821938151,
      "loss": 5.2979,
      "mean_token_accuracy": 0.21458456218242644,
      "num_tokens": 59491876.0,
      "step": 31910
    },
    {
      "entropy": 5.864655876159668,
      "epoch": 2.8183504062168847,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004208826082632092,
      "loss": 5.3581,
      "mean_token_accuracy": 0.20974958688020706,
      "num_tokens": 59502007.0,
      "step": 31915
    },
    {
      "entropy": 5.934280300140381,
      "epoch": 2.8187919463087248,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000420858631474846,
      "loss": 5.3646,
      "mean_token_accuracy": 0.21038123369216918,
      "num_tokens": 59511770.0,
      "step": 31920
    },
    {
      "entropy": 6.094281482696533,
      "epoch": 2.8192334864005653,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004208346518291952,
      "loss": 5.4393,
      "mean_token_accuracy": 0.2028668150305748,
      "num_tokens": 59521303.0,
      "step": 31925
    },
    {
      "entropy": 6.106580495834351,
      "epoch": 2.8196750264924058,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042081066932672645,
      "loss": 5.443,
      "mean_token_accuracy": 0.19938048869371414,
      "num_tokens": 59532104.0,
      "step": 31930
    },
    {
      "entropy": 5.9644392967224125,
      "epoch": 2.820116566584246,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042078668396790963,
      "loss": 5.2235,
      "mean_token_accuracy": 0.22054282277822496,
      "num_tokens": 59542163.0,
      "step": 31935
    },
    {
      "entropy": 5.965909194946289,
      "epoch": 2.8205581066760863,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042076269575321457,
      "loss": 5.3129,
      "mean_token_accuracy": 0.21089100539684297,
      "num_tokens": 59552588.0,
      "step": 31940
    },
    {
      "entropy": 5.873945808410644,
      "epoch": 2.820999646767927,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004207387046831112,
      "loss": 5.3345,
      "mean_token_accuracy": 0.2155772402882576,
      "num_tokens": 59561531.0,
      "step": 31945
    },
    {
      "entropy": 5.889024829864502,
      "epoch": 2.821441186859767,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004207147107580695,
      "loss": 5.2098,
      "mean_token_accuracy": 0.2144917517900467,
      "num_tokens": 59570960.0,
      "step": 31950
    },
    {
      "entropy": 6.046193647384643,
      "epoch": 2.8218827269516074,
      "grad_norm": 1.34375,
      "learning_rate": 0.00042069071397855953,
      "loss": 5.379,
      "mean_token_accuracy": 0.20773628503084182,
      "num_tokens": 59579734.0,
      "step": 31955
    },
    {
      "entropy": 5.963562822341919,
      "epoch": 2.8223242670434474,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042066671434505125,
      "loss": 5.3692,
      "mean_token_accuracy": 0.21307939887046815,
      "num_tokens": 59589260.0,
      "step": 31960
    },
    {
      "entropy": 5.939743375778198,
      "epoch": 2.822765807135288,
      "grad_norm": 1.265625,
      "learning_rate": 0.000420642711858015,
      "loss": 5.2561,
      "mean_token_accuracy": 0.22415608465671538,
      "num_tokens": 59597972.0,
      "step": 31965
    },
    {
      "entropy": 5.8874000072479244,
      "epoch": 2.823207347227128,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042061870651792076,
      "loss": 5.2209,
      "mean_token_accuracy": 0.21569012254476547,
      "num_tokens": 59607645.0,
      "step": 31970
    },
    {
      "entropy": 6.035185146331787,
      "epoch": 2.8236488873189685,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042059469832523894,
      "loss": 5.357,
      "mean_token_accuracy": 0.2009819582104683,
      "num_tokens": 59616405.0,
      "step": 31975
    },
    {
      "entropy": 5.860014867782593,
      "epoch": 2.824090427410809,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004205706872804397,
      "loss": 5.2328,
      "mean_token_accuracy": 0.22294441908597945,
      "num_tokens": 59625635.0,
      "step": 31980
    },
    {
      "entropy": 5.8008184909820555,
      "epoch": 2.824531967502649,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004205466733839935,
      "loss": 5.0785,
      "mean_token_accuracy": 0.23370221257209778,
      "num_tokens": 59635115.0,
      "step": 31985
    },
    {
      "entropy": 5.975692558288574,
      "epoch": 2.8249735075944895,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004205226566363708,
      "loss": 5.3456,
      "mean_token_accuracy": 0.21454449295997619,
      "num_tokens": 59644249.0,
      "step": 31990
    },
    {
      "entropy": 5.970183086395264,
      "epoch": 2.82541504768633,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042049863703804205,
      "loss": 5.2742,
      "mean_token_accuracy": 0.21755435466766357,
      "num_tokens": 59653190.0,
      "step": 31995
    },
    {
      "entropy": 5.879968118667603,
      "epoch": 2.82585658777817,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004204746145894777,
      "loss": 5.1962,
      "mean_token_accuracy": 0.21631181985139847,
      "num_tokens": 59662181.0,
      "step": 32000
    },
    {
      "entropy": 5.9312104225158695,
      "epoch": 2.8262981278700106,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004204505892911484,
      "loss": 5.2856,
      "mean_token_accuracy": 0.21385398954153062,
      "num_tokens": 59670949.0,
      "step": 32005
    },
    {
      "entropy": 5.955525159835815,
      "epoch": 2.826739667961851,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004204265611435247,
      "loss": 5.3505,
      "mean_token_accuracy": 0.20739322751760483,
      "num_tokens": 59680112.0,
      "step": 32010
    },
    {
      "entropy": 5.992297029495239,
      "epoch": 2.827181208053691,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042040253014707745,
      "loss": 5.2888,
      "mean_token_accuracy": 0.20599354952573776,
      "num_tokens": 59688910.0,
      "step": 32015
    },
    {
      "entropy": 5.996553087234497,
      "epoch": 2.8276227481455316,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004203784963022774,
      "loss": 5.3242,
      "mean_token_accuracy": 0.2114655926823616,
      "num_tokens": 59699047.0,
      "step": 32020
    },
    {
      "entropy": 5.973946380615234,
      "epoch": 2.828064288237372,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042035445960959513,
      "loss": 5.3681,
      "mean_token_accuracy": 0.21329355090856553,
      "num_tokens": 59708545.0,
      "step": 32025
    },
    {
      "entropy": 6.025304222106934,
      "epoch": 2.828505828329212,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004203304200695017,
      "loss": 5.4184,
      "mean_token_accuracy": 0.20704313218593598,
      "num_tokens": 59718118.0,
      "step": 32030
    },
    {
      "entropy": 5.978156709671021,
      "epoch": 2.8289473684210527,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004203063776824681,
      "loss": 5.2623,
      "mean_token_accuracy": 0.2110673353075981,
      "num_tokens": 59726465.0,
      "step": 32035
    },
    {
      "entropy": 5.972930669784546,
      "epoch": 2.829388908512893,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042028233244896507,
      "loss": 5.308,
      "mean_token_accuracy": 0.2177726462483406,
      "num_tokens": 59735399.0,
      "step": 32040
    },
    {
      "entropy": 5.851134777069092,
      "epoch": 2.829830448604733,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042025828436946386,
      "loss": 5.2144,
      "mean_token_accuracy": 0.22057828158140183,
      "num_tokens": 59744725.0,
      "step": 32045
    },
    {
      "entropy": 5.852007961273193,
      "epoch": 2.8302719886965737,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004202342334444355,
      "loss": 5.2514,
      "mean_token_accuracy": 0.21748136729002,
      "num_tokens": 59754317.0,
      "step": 32050
    },
    {
      "entropy": 6.021510171890259,
      "epoch": 2.830713528788414,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042021017967435104,
      "loss": 5.371,
      "mean_token_accuracy": 0.20439515113830567,
      "num_tokens": 59764951.0,
      "step": 32055
    },
    {
      "entropy": 6.005381298065186,
      "epoch": 2.8311550688802543,
      "grad_norm": 1.125,
      "learning_rate": 0.0004201861230596817,
      "loss": 5.3528,
      "mean_token_accuracy": 0.20550714433193207,
      "num_tokens": 59775116.0,
      "step": 32060
    },
    {
      "entropy": 5.903173542022705,
      "epoch": 2.8315966089720948,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042016206360089893,
      "loss": 5.1664,
      "mean_token_accuracy": 0.22703022360801697,
      "num_tokens": 59783253.0,
      "step": 32065
    },
    {
      "entropy": 5.892591381072998,
      "epoch": 2.8320381490639353,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004201380012984738,
      "loss": 5.2829,
      "mean_token_accuracy": 0.21633019000291825,
      "num_tokens": 59793020.0,
      "step": 32070
    },
    {
      "entropy": 5.977093076705932,
      "epoch": 2.8324796891557753,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004201139361528777,
      "loss": 5.3625,
      "mean_token_accuracy": 0.21012691110372544,
      "num_tokens": 59803860.0,
      "step": 32075
    },
    {
      "entropy": 5.934565210342408,
      "epoch": 2.832921229247616,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004200898681645822,
      "loss": 5.1667,
      "mean_token_accuracy": 0.22635160386562347,
      "num_tokens": 59813695.0,
      "step": 32080
    },
    {
      "entropy": 6.014615058898926,
      "epoch": 2.8333627693394563,
      "grad_norm": 1.125,
      "learning_rate": 0.0004200657973340588,
      "loss": 5.4111,
      "mean_token_accuracy": 0.19774722009897233,
      "num_tokens": 59823445.0,
      "step": 32085
    },
    {
      "entropy": 5.904189538955689,
      "epoch": 2.8338043094312964,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004200417236617788,
      "loss": 5.2703,
      "mean_token_accuracy": 0.21776919662952424,
      "num_tokens": 59833242.0,
      "step": 32090
    },
    {
      "entropy": 5.913767623901367,
      "epoch": 2.834245849523137,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042001764714821394,
      "loss": 5.2174,
      "mean_token_accuracy": 0.2160519018769264,
      "num_tokens": 59843364.0,
      "step": 32095
    },
    {
      "entropy": 5.919834899902344,
      "epoch": 2.834687389614977,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004199935677938359,
      "loss": 5.2499,
      "mean_token_accuracy": 0.22058711051940919,
      "num_tokens": 59852940.0,
      "step": 32100
    },
    {
      "entropy": 5.933721446990967,
      "epoch": 2.8351289297068174,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004199694855991162,
      "loss": 5.3353,
      "mean_token_accuracy": 0.20889459252357484,
      "num_tokens": 59863313.0,
      "step": 32105
    },
    {
      "entropy": 6.05092716217041,
      "epoch": 2.8355704697986575,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004199454005645269,
      "loss": 5.3945,
      "mean_token_accuracy": 0.20801225900650025,
      "num_tokens": 59873307.0,
      "step": 32110
    },
    {
      "entropy": 5.937729787826538,
      "epoch": 2.836012009890498,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004199213126905395,
      "loss": 5.3027,
      "mean_token_accuracy": 0.22189847826957704,
      "num_tokens": 59882958.0,
      "step": 32115
    },
    {
      "entropy": 5.90982084274292,
      "epoch": 2.8364535499823385,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004198972219776261,
      "loss": 5.274,
      "mean_token_accuracy": 0.2194146767258644,
      "num_tokens": 59892259.0,
      "step": 32120
    },
    {
      "entropy": 5.924500894546509,
      "epoch": 2.8368950900741785,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004198731284262585,
      "loss": 5.3122,
      "mean_token_accuracy": 0.21165468394756318,
      "num_tokens": 59901836.0,
      "step": 32125
    },
    {
      "entropy": 5.943654775619507,
      "epoch": 2.837336630166019,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041984903203690874,
      "loss": 5.1813,
      "mean_token_accuracy": 0.21496229916810988,
      "num_tokens": 59910751.0,
      "step": 32130
    },
    {
      "entropy": 5.951314878463745,
      "epoch": 2.8377781702578595,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004198249328100487,
      "loss": 5.2179,
      "mean_token_accuracy": 0.22602489739656448,
      "num_tokens": 59920101.0,
      "step": 32135
    },
    {
      "entropy": 5.947711181640625,
      "epoch": 2.8382197103496996,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004198008307461507,
      "loss": 5.3027,
      "mean_token_accuracy": 0.21083392053842545,
      "num_tokens": 59928710.0,
      "step": 32140
    },
    {
      "entropy": 5.941885137557984,
      "epoch": 2.83866125044154,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004197767258456868,
      "loss": 5.3539,
      "mean_token_accuracy": 0.20655761063098907,
      "num_tokens": 59939277.0,
      "step": 32145
    },
    {
      "entropy": 6.072238779067993,
      "epoch": 2.8391027905333805,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041975261810912905,
      "loss": 5.3958,
      "mean_token_accuracy": 0.20390612930059432,
      "num_tokens": 59949093.0,
      "step": 32150
    },
    {
      "entropy": 5.997541236877441,
      "epoch": 2.8395443306252206,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041972850753695,
      "loss": 5.3021,
      "mean_token_accuracy": 0.2130661278963089,
      "num_tokens": 59958528.0,
      "step": 32155
    },
    {
      "entropy": 5.967082214355469,
      "epoch": 2.839985870717061,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004197043941296217,
      "loss": 5.3264,
      "mean_token_accuracy": 0.2103780746459961,
      "num_tokens": 59967444.0,
      "step": 32160
    },
    {
      "entropy": 5.948641395568847,
      "epoch": 2.8404274108089016,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004196802778876166,
      "loss": 5.2975,
      "mean_token_accuracy": 0.21682139337062836,
      "num_tokens": 59976997.0,
      "step": 32165
    },
    {
      "entropy": 5.891337537765503,
      "epoch": 2.8408689509007417,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004196561588114072,
      "loss": 5.1981,
      "mean_token_accuracy": 0.21537546068429947,
      "num_tokens": 59986352.0,
      "step": 32170
    },
    {
      "entropy": 6.007836580276489,
      "epoch": 2.841310490992582,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041963203690146585,
      "loss": 5.4091,
      "mean_token_accuracy": 0.20063070356845855,
      "num_tokens": 59995345.0,
      "step": 32175
    },
    {
      "entropy": 5.979075384140015,
      "epoch": 2.8417520310844226,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004196079121582652,
      "loss": 5.2871,
      "mean_token_accuracy": 0.21629167795181276,
      "num_tokens": 60004792.0,
      "step": 32180
    },
    {
      "entropy": 5.954074335098267,
      "epoch": 2.8421935711762627,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004195837845822778,
      "loss": 5.2826,
      "mean_token_accuracy": 0.21123869270086287,
      "num_tokens": 60013472.0,
      "step": 32185
    },
    {
      "entropy": 5.960694980621338,
      "epoch": 2.842635111268103,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004195596541739763,
      "loss": 5.3075,
      "mean_token_accuracy": 0.21322947442531587,
      "num_tokens": 60023753.0,
      "step": 32190
    },
    {
      "entropy": 5.982694435119629,
      "epoch": 2.8430766513599437,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041953552093383336,
      "loss": 5.2641,
      "mean_token_accuracy": 0.21019338220357894,
      "num_tokens": 60032850.0,
      "step": 32195
    },
    {
      "entropy": 5.946872520446777,
      "epoch": 2.8435181914517837,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041951138486232177,
      "loss": 5.2627,
      "mean_token_accuracy": 0.20759363174438478,
      "num_tokens": 60041644.0,
      "step": 32200
    },
    {
      "entropy": 5.954316854476929,
      "epoch": 2.8439597315436242,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004194872459599144,
      "loss": 5.3235,
      "mean_token_accuracy": 0.21594562232494355,
      "num_tokens": 60051453.0,
      "step": 32205
    },
    {
      "entropy": 5.97462420463562,
      "epoch": 2.8444012716354647,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041946310422708396,
      "loss": 5.3496,
      "mean_token_accuracy": 0.20043090283870696,
      "num_tokens": 60060408.0,
      "step": 32210
    },
    {
      "entropy": 5.966831398010254,
      "epoch": 2.844842811727305,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004194389596643035,
      "loss": 5.3516,
      "mean_token_accuracy": 0.21770361959934234,
      "num_tokens": 60069981.0,
      "step": 32215
    },
    {
      "entropy": 5.998292112350464,
      "epoch": 2.8452843518191453,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041941481227204596,
      "loss": 5.4025,
      "mean_token_accuracy": 0.20718762576580046,
      "num_tokens": 60078785.0,
      "step": 32220
    },
    {
      "entropy": 5.992251014709472,
      "epoch": 2.845725891910986,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041939066205078443,
      "loss": 5.238,
      "mean_token_accuracy": 0.21242726743221282,
      "num_tokens": 60087639.0,
      "step": 32225
    },
    {
      "entropy": 5.931231069564819,
      "epoch": 2.846167432002826,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041936650900099186,
      "loss": 5.2613,
      "mean_token_accuracy": 0.20926441699266435,
      "num_tokens": 60096385.0,
      "step": 32230
    },
    {
      "entropy": 5.9711589336395265,
      "epoch": 2.8466089720946663,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041934235312314163,
      "loss": 5.297,
      "mean_token_accuracy": 0.2124288097023964,
      "num_tokens": 60105449.0,
      "step": 32235
    },
    {
      "entropy": 5.905494546890258,
      "epoch": 2.8470505121865064,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004193181944177066,
      "loss": 5.3233,
      "mean_token_accuracy": 0.21355954110622405,
      "num_tokens": 60115060.0,
      "step": 32240
    },
    {
      "entropy": 5.978457403182984,
      "epoch": 2.847492052278347,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004192940328851604,
      "loss": 5.3113,
      "mean_token_accuracy": 0.21436705440282822,
      "num_tokens": 60124737.0,
      "step": 32245
    },
    {
      "entropy": 5.941097688674927,
      "epoch": 2.847933592370187,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004192698685259761,
      "loss": 5.3311,
      "mean_token_accuracy": 0.2052650049328804,
      "num_tokens": 60133964.0,
      "step": 32250
    },
    {
      "entropy": 5.920156860351563,
      "epoch": 2.8483751324620274,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041924570134062707,
      "loss": 5.2261,
      "mean_token_accuracy": 0.2189095601439476,
      "num_tokens": 60143777.0,
      "step": 32255
    },
    {
      "entropy": 6.001034450531006,
      "epoch": 2.848816672553868,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004192215313295868,
      "loss": 5.2869,
      "mean_token_accuracy": 0.21675652414560317,
      "num_tokens": 60152813.0,
      "step": 32260
    },
    {
      "entropy": 5.99021430015564,
      "epoch": 2.849258212645708,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004191973584933288,
      "loss": 5.3243,
      "mean_token_accuracy": 0.2069294720888138,
      "num_tokens": 60161789.0,
      "step": 32265
    },
    {
      "entropy": 5.946095705032349,
      "epoch": 2.8496997527375485,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041917318283232654,
      "loss": 5.3259,
      "mean_token_accuracy": 0.20061446875333785,
      "num_tokens": 60171171.0,
      "step": 32270
    },
    {
      "entropy": 5.984044790267944,
      "epoch": 2.850141292829389,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004191490043470536,
      "loss": 5.3332,
      "mean_token_accuracy": 0.21522440165281295,
      "num_tokens": 60180512.0,
      "step": 32275
    },
    {
      "entropy": 6.063876104354859,
      "epoch": 2.850582832921229,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041912482303798373,
      "loss": 5.357,
      "mean_token_accuracy": 0.2064753830432892,
      "num_tokens": 60189675.0,
      "step": 32280
    },
    {
      "entropy": 5.969464540481567,
      "epoch": 2.8510243730130695,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041910063890559043,
      "loss": 5.2532,
      "mean_token_accuracy": 0.21294936984777452,
      "num_tokens": 60199037.0,
      "step": 32285
    },
    {
      "entropy": 5.968156003952027,
      "epoch": 2.85146591310491,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041907645195034764,
      "loss": 5.2117,
      "mean_token_accuracy": 0.2191062182188034,
      "num_tokens": 60208631.0,
      "step": 32290
    },
    {
      "entropy": 5.958714771270752,
      "epoch": 2.85190745319675,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004190522621727291,
      "loss": 5.2204,
      "mean_token_accuracy": 0.2181653529405594,
      "num_tokens": 60217977.0,
      "step": 32295
    },
    {
      "entropy": 5.92012128829956,
      "epoch": 2.8523489932885906,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004190280695732086,
      "loss": 5.2787,
      "mean_token_accuracy": 0.2131631404161453,
      "num_tokens": 60227264.0,
      "step": 32300
    },
    {
      "entropy": 5.932190370559693,
      "epoch": 2.852790533380431,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004190038741522602,
      "loss": 5.3358,
      "mean_token_accuracy": 0.20650791972875596,
      "num_tokens": 60236355.0,
      "step": 32305
    },
    {
      "entropy": 5.879115343093872,
      "epoch": 2.853232073472271,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041897967591035784,
      "loss": 5.291,
      "mean_token_accuracy": 0.2074032709002495,
      "num_tokens": 60245993.0,
      "step": 32310
    },
    {
      "entropy": 5.939132022857666,
      "epoch": 2.8536736135641116,
      "grad_norm": 1.25,
      "learning_rate": 0.0004189554748479755,
      "loss": 5.2602,
      "mean_token_accuracy": 0.22078985422849656,
      "num_tokens": 60255449.0,
      "step": 32315
    },
    {
      "entropy": 5.970338964462281,
      "epoch": 2.854115153655952,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041893127096558714,
      "loss": 5.2754,
      "mean_token_accuracy": 0.21330875307321548,
      "num_tokens": 60264708.0,
      "step": 32320
    },
    {
      "entropy": 5.982881832122803,
      "epoch": 2.854556693747792,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004189070642636672,
      "loss": 5.3726,
      "mean_token_accuracy": 0.20378067642450332,
      "num_tokens": 60273654.0,
      "step": 32325
    },
    {
      "entropy": 5.932777261734008,
      "epoch": 2.8549982338396327,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041888285474268964,
      "loss": 5.1884,
      "mean_token_accuracy": 0.22532227337360383,
      "num_tokens": 60282479.0,
      "step": 32330
    },
    {
      "entropy": 5.886355590820313,
      "epoch": 2.855439773931473,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041885864240312883,
      "loss": 5.2551,
      "mean_token_accuracy": 0.21790775805711746,
      "num_tokens": 60292622.0,
      "step": 32335
    },
    {
      "entropy": 5.951969957351684,
      "epoch": 2.8558813140233132,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041883442724545894,
      "loss": 5.2741,
      "mean_token_accuracy": 0.21242440044879912,
      "num_tokens": 60302137.0,
      "step": 32340
    },
    {
      "entropy": 5.953062248229981,
      "epoch": 2.8563228541151537,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004188102092701546,
      "loss": 5.3459,
      "mean_token_accuracy": 0.19969377666711807,
      "num_tokens": 60312253.0,
      "step": 32345
    },
    {
      "entropy": 5.972516441345215,
      "epoch": 2.8567643942069942,
      "grad_norm": 1.125,
      "learning_rate": 0.00041878598847768986,
      "loss": 5.3422,
      "mean_token_accuracy": 0.20701395273208617,
      "num_tokens": 60321486.0,
      "step": 32350
    },
    {
      "entropy": 6.004119110107422,
      "epoch": 2.8572059342988343,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041876176486853946,
      "loss": 5.2661,
      "mean_token_accuracy": 0.2116925448179245,
      "num_tokens": 60330279.0,
      "step": 32355
    },
    {
      "entropy": 5.957367610931397,
      "epoch": 2.857647474390675,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004187375384431779,
      "loss": 5.3138,
      "mean_token_accuracy": 0.20705816894769669,
      "num_tokens": 60340915.0,
      "step": 32360
    },
    {
      "entropy": 5.891090154647827,
      "epoch": 2.8580890144825153,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004187133092020796,
      "loss": 5.3173,
      "mean_token_accuracy": 0.20469451397657396,
      "num_tokens": 60349849.0,
      "step": 32365
    },
    {
      "entropy": 5.962789916992188,
      "epoch": 2.8585305545743553,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004186890771457194,
      "loss": 5.3462,
      "mean_token_accuracy": 0.20347629934549333,
      "num_tokens": 60360072.0,
      "step": 32370
    },
    {
      "entropy": 6.04688868522644,
      "epoch": 2.858972094666196,
      "grad_norm": 1.125,
      "learning_rate": 0.0004186648422745718,
      "loss": 5.3961,
      "mean_token_accuracy": 0.20392194986343384,
      "num_tokens": 60369898.0,
      "step": 32375
    },
    {
      "entropy": 5.921404790878296,
      "epoch": 2.859413634758036,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041864060458911176,
      "loss": 5.2489,
      "mean_token_accuracy": 0.21967290043830873,
      "num_tokens": 60378573.0,
      "step": 32380
    },
    {
      "entropy": 6.009613084793091,
      "epoch": 2.8598551748498764,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004186163640898139,
      "loss": 5.3883,
      "mean_token_accuracy": 0.2004510760307312,
      "num_tokens": 60388479.0,
      "step": 32385
    },
    {
      "entropy": 5.880601263046264,
      "epoch": 2.8602967149417164,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041859212077715316,
      "loss": 5.213,
      "mean_token_accuracy": 0.2176972284913063,
      "num_tokens": 60397854.0,
      "step": 32390
    },
    {
      "entropy": 5.965158462524414,
      "epoch": 2.860738255033557,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041856787465160446,
      "loss": 5.3366,
      "mean_token_accuracy": 0.2137874037027359,
      "num_tokens": 60407464.0,
      "step": 32395
    },
    {
      "entropy": 5.914599990844726,
      "epoch": 2.8611797951253974,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041854362571364274,
      "loss": 5.1397,
      "mean_token_accuracy": 0.23089986741542817,
      "num_tokens": 60415463.0,
      "step": 32400
    },
    {
      "entropy": 5.8744346618652346,
      "epoch": 2.8616213352172375,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041851937396374296,
      "loss": 5.2348,
      "mean_token_accuracy": 0.21944186985492706,
      "num_tokens": 60424927.0,
      "step": 32405
    },
    {
      "entropy": 5.954278087615966,
      "epoch": 2.862062875309078,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004184951194023804,
      "loss": 5.3284,
      "mean_token_accuracy": 0.21354325413703917,
      "num_tokens": 60435353.0,
      "step": 32410
    },
    {
      "entropy": 5.944235515594483,
      "epoch": 2.8625044154009185,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041847086203003,
      "loss": 5.2679,
      "mean_token_accuracy": 0.21343024224042892,
      "num_tokens": 60445271.0,
      "step": 32415
    },
    {
      "entropy": 5.8921088695526125,
      "epoch": 2.8629459554927585,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041844660184716694,
      "loss": 5.17,
      "mean_token_accuracy": 0.21748588383197784,
      "num_tokens": 60454817.0,
      "step": 32420
    },
    {
      "entropy": 5.996472978591919,
      "epoch": 2.863387495584599,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041842233885426664,
      "loss": 5.3168,
      "mean_token_accuracy": 0.20651798844337463,
      "num_tokens": 60464796.0,
      "step": 32425
    },
    {
      "entropy": 5.945512866973877,
      "epoch": 2.8638290356764395,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041839807305180424,
      "loss": 5.2385,
      "mean_token_accuracy": 0.21819022297859192,
      "num_tokens": 60473766.0,
      "step": 32430
    },
    {
      "entropy": 6.04214916229248,
      "epoch": 2.8642705757682796,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004183738044402551,
      "loss": 5.3962,
      "mean_token_accuracy": 0.20452522933483125,
      "num_tokens": 60483564.0,
      "step": 32435
    },
    {
      "entropy": 5.9896753311157225,
      "epoch": 2.86471211586012,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004183495330200947,
      "loss": 5.3636,
      "mean_token_accuracy": 0.21063577830791474,
      "num_tokens": 60492495.0,
      "step": 32440
    },
    {
      "entropy": 5.974906635284424,
      "epoch": 2.8651536559519606,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041832525879179853,
      "loss": 5.2533,
      "mean_token_accuracy": 0.2176944360136986,
      "num_tokens": 60502406.0,
      "step": 32445
    },
    {
      "entropy": 5.897673273086548,
      "epoch": 2.8655951960438006,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041830098175584196,
      "loss": 5.2394,
      "mean_token_accuracy": 0.21584665179252624,
      "num_tokens": 60511477.0,
      "step": 32450
    },
    {
      "entropy": 5.935951614379883,
      "epoch": 2.866036736135641,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041827670191270074,
      "loss": 5.2956,
      "mean_token_accuracy": 0.21654515117406845,
      "num_tokens": 60520709.0,
      "step": 32455
    },
    {
      "entropy": 5.986697626113892,
      "epoch": 2.8664782762274816,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041825241926285033,
      "loss": 5.3063,
      "mean_token_accuracy": 0.2122371792793274,
      "num_tokens": 60529831.0,
      "step": 32460
    },
    {
      "entropy": 5.975148630142212,
      "epoch": 2.8669198163193217,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041822813380676665,
      "loss": 5.3361,
      "mean_token_accuracy": 0.20963315665721893,
      "num_tokens": 60539501.0,
      "step": 32465
    },
    {
      "entropy": 5.9620496273040775,
      "epoch": 2.867361356411162,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041820384554492517,
      "loss": 5.2934,
      "mean_token_accuracy": 0.2152968928217888,
      "num_tokens": 60547533.0,
      "step": 32470
    },
    {
      "entropy": 5.919245386123658,
      "epoch": 2.8678028965030027,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004181795544778018,
      "loss": 5.3104,
      "mean_token_accuracy": 0.213161002099514,
      "num_tokens": 60557047.0,
      "step": 32475
    },
    {
      "entropy": 6.042220258712769,
      "epoch": 2.8682444365948427,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004181552606058725,
      "loss": 5.3023,
      "mean_token_accuracy": 0.21146691888570784,
      "num_tokens": 60566333.0,
      "step": 32480
    },
    {
      "entropy": 5.975422668457031,
      "epoch": 2.8686859766866832,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004181309639296129,
      "loss": 5.2734,
      "mean_token_accuracy": 0.21941187530755996,
      "num_tokens": 60575090.0,
      "step": 32485
    },
    {
      "entropy": 5.859653663635254,
      "epoch": 2.8691275167785237,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004181066644494992,
      "loss": 5.2459,
      "mean_token_accuracy": 0.21270665526390076,
      "num_tokens": 60584000.0,
      "step": 32490
    },
    {
      "entropy": 5.834622001647949,
      "epoch": 2.869569056870364,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004180823621660073,
      "loss": 5.1858,
      "mean_token_accuracy": 0.219612056016922,
      "num_tokens": 60593239.0,
      "step": 32495
    },
    {
      "entropy": 5.984924745559693,
      "epoch": 2.8700105969622043,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004180580570796134,
      "loss": 5.3314,
      "mean_token_accuracy": 0.20796655118465424,
      "num_tokens": 60601850.0,
      "step": 32500
    },
    {
      "entropy": 6.025399303436279,
      "epoch": 2.870452137054045,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004180337491907934,
      "loss": 5.3927,
      "mean_token_accuracy": 0.2044035851955414,
      "num_tokens": 60611280.0,
      "step": 32505
    },
    {
      "entropy": 5.947235631942749,
      "epoch": 2.870893677145885,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004180094385000237,
      "loss": 5.3015,
      "mean_token_accuracy": 0.2131294548511505,
      "num_tokens": 60620738.0,
      "step": 32510
    },
    {
      "entropy": 5.943355560302734,
      "epoch": 2.8713352172377253,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041798512500778034,
      "loss": 5.3512,
      "mean_token_accuracy": 0.20940739512443543,
      "num_tokens": 60630472.0,
      "step": 32515
    },
    {
      "entropy": 5.991823625564575,
      "epoch": 2.8717767573295654,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004179608087145397,
      "loss": 5.2843,
      "mean_token_accuracy": 0.2128814473748207,
      "num_tokens": 60640555.0,
      "step": 32520
    },
    {
      "entropy": 6.069680261611938,
      "epoch": 2.872218297421406,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041793648962077816,
      "loss": 5.3998,
      "mean_token_accuracy": 0.20682356953620912,
      "num_tokens": 60649941.0,
      "step": 32525
    },
    {
      "entropy": 6.003444719314575,
      "epoch": 2.872659837513246,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041791216772697217,
      "loss": 5.3395,
      "mean_token_accuracy": 0.20861465036869048,
      "num_tokens": 60659883.0,
      "step": 32530
    },
    {
      "entropy": 6.0305358409881595,
      "epoch": 2.8731013776050864,
      "grad_norm": 1.265625,
      "learning_rate": 0.000417887843033598,
      "loss": 5.371,
      "mean_token_accuracy": 0.20884274691343307,
      "num_tokens": 60669323.0,
      "step": 32535
    },
    {
      "entropy": 5.99633469581604,
      "epoch": 2.873542917696927,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004178635155411323,
      "loss": 5.286,
      "mean_token_accuracy": 0.2144134983420372,
      "num_tokens": 60678497.0,
      "step": 32540
    },
    {
      "entropy": 5.9284483909606935,
      "epoch": 2.873984457788767,
      "grad_norm": 1.296875,
      "learning_rate": 0.00041783918525005155,
      "loss": 5.2689,
      "mean_token_accuracy": 0.21025389283895493,
      "num_tokens": 60687312.0,
      "step": 32545
    },
    {
      "entropy": 5.93980860710144,
      "epoch": 2.8744259978806075,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041781485216083237,
      "loss": 5.2867,
      "mean_token_accuracy": 0.21085347682237626,
      "num_tokens": 60696765.0,
      "step": 32550
    },
    {
      "entropy": 5.93997631072998,
      "epoch": 2.874867537972448,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041779051627395146,
      "loss": 5.3362,
      "mean_token_accuracy": 0.22017936706542968,
      "num_tokens": 60705909.0,
      "step": 32555
    },
    {
      "entropy": 5.949883079528808,
      "epoch": 2.875309078064288,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041776617758988553,
      "loss": 5.1888,
      "mean_token_accuracy": 0.22455387711524963,
      "num_tokens": 60715367.0,
      "step": 32560
    },
    {
      "entropy": 5.874039220809936,
      "epoch": 2.8757506181561285,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041774183610911143,
      "loss": 5.2954,
      "mean_token_accuracy": 0.21020587533712387,
      "num_tokens": 60724588.0,
      "step": 32565
    },
    {
      "entropy": 5.964269161224365,
      "epoch": 2.876192158247969,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004177174918321059,
      "loss": 5.2993,
      "mean_token_accuracy": 0.22094751000404358,
      "num_tokens": 60732943.0,
      "step": 32570
    },
    {
      "entropy": 5.8986553192138675,
      "epoch": 2.876633698339809,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041769314475934596,
      "loss": 5.2246,
      "mean_token_accuracy": 0.21871429681777954,
      "num_tokens": 60742378.0,
      "step": 32575
    },
    {
      "entropy": 5.931229925155639,
      "epoch": 2.8770752384316496,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004176687948913084,
      "loss": 5.2674,
      "mean_token_accuracy": 0.21834666281938553,
      "num_tokens": 60751175.0,
      "step": 32580
    },
    {
      "entropy": 6.036599206924438,
      "epoch": 2.87751677852349,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004176444422284703,
      "loss": 5.3437,
      "mean_token_accuracy": 0.2120925545692444,
      "num_tokens": 60760484.0,
      "step": 32585
    },
    {
      "entropy": 5.972375774383545,
      "epoch": 2.87795831861533,
      "grad_norm": 1.125,
      "learning_rate": 0.0004176200867713086,
      "loss": 5.2816,
      "mean_token_accuracy": 0.21512831151485443,
      "num_tokens": 60770055.0,
      "step": 32590
    },
    {
      "entropy": 5.952885389328003,
      "epoch": 2.8783998587071706,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004175957285203007,
      "loss": 5.3703,
      "mean_token_accuracy": 0.2073764607310295,
      "num_tokens": 60780347.0,
      "step": 32595
    },
    {
      "entropy": 5.9564845085144045,
      "epoch": 2.878841398799011,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004175713674759235,
      "loss": 5.2326,
      "mean_token_accuracy": 0.21895605325698853,
      "num_tokens": 60789489.0,
      "step": 32600
    },
    {
      "entropy": 5.952539300918579,
      "epoch": 2.879282938890851,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041754700363865426,
      "loss": 5.3294,
      "mean_token_accuracy": 0.20608580857515335,
      "num_tokens": 60797838.0,
      "step": 32605
    },
    {
      "entropy": 5.907006168365479,
      "epoch": 2.8797244789826917,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004175226370089703,
      "loss": 5.2422,
      "mean_token_accuracy": 0.21867429614067077,
      "num_tokens": 60808119.0,
      "step": 32610
    },
    {
      "entropy": 5.967578554153443,
      "epoch": 2.880166019074532,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000417498267587349,
      "loss": 5.2359,
      "mean_token_accuracy": 0.213094362616539,
      "num_tokens": 60815905.0,
      "step": 32615
    },
    {
      "entropy": 5.973251628875732,
      "epoch": 2.8806075591663722,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004174738953742676,
      "loss": 5.3143,
      "mean_token_accuracy": 0.20762235373258592,
      "num_tokens": 60826258.0,
      "step": 32620
    },
    {
      "entropy": 5.891865396499634,
      "epoch": 2.8810490992582127,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004174495203702036,
      "loss": 5.2319,
      "mean_token_accuracy": 0.21924026012420655,
      "num_tokens": 60834627.0,
      "step": 32625
    },
    {
      "entropy": 5.955694770812988,
      "epoch": 2.8814906393500532,
      "grad_norm": 1.125,
      "learning_rate": 0.00041742514257563464,
      "loss": 5.3466,
      "mean_token_accuracy": 0.20750416666269303,
      "num_tokens": 60844399.0,
      "step": 32630
    },
    {
      "entropy": 5.980674362182617,
      "epoch": 2.8819321794418933,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000417400761991038,
      "loss": 5.2837,
      "mean_token_accuracy": 0.21381688565015794,
      "num_tokens": 60852964.0,
      "step": 32635
    },
    {
      "entropy": 5.932396316528321,
      "epoch": 2.882373719533734,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041737637861689154,
      "loss": 5.2712,
      "mean_token_accuracy": 0.2126356452703476,
      "num_tokens": 60862100.0,
      "step": 32640
    },
    {
      "entropy": 6.0393232822418215,
      "epoch": 2.8828152596255743,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004173519924536727,
      "loss": 5.3806,
      "mean_token_accuracy": 0.20414622724056244,
      "num_tokens": 60871761.0,
      "step": 32645
    },
    {
      "entropy": 5.928535461425781,
      "epoch": 2.8832567997174143,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004173276035018593,
      "loss": 5.2608,
      "mean_token_accuracy": 0.20985836237668992,
      "num_tokens": 60880644.0,
      "step": 32650
    },
    {
      "entropy": 6.025102376937866,
      "epoch": 2.883698339809255,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041730321176192917,
      "loss": 5.3276,
      "mean_token_accuracy": 0.21123459488153457,
      "num_tokens": 60889566.0,
      "step": 32655
    },
    {
      "entropy": 6.0256476402282715,
      "epoch": 2.884139879901095,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041727881723436,
      "loss": 5.4329,
      "mean_token_accuracy": 0.19794102758169174,
      "num_tokens": 60899591.0,
      "step": 32660
    },
    {
      "entropy": 5.976695013046265,
      "epoch": 2.8845814199929354,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041725441991962965,
      "loss": 5.3554,
      "mean_token_accuracy": 0.20184616148471832,
      "num_tokens": 60908928.0,
      "step": 32665
    },
    {
      "entropy": 5.9616045475006105,
      "epoch": 2.8850229600847754,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004172300198182162,
      "loss": 5.3109,
      "mean_token_accuracy": 0.2136756584048271,
      "num_tokens": 60918105.0,
      "step": 32670
    },
    {
      "entropy": 6.002583646774292,
      "epoch": 2.885464500176616,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041720561693059756,
      "loss": 5.2994,
      "mean_token_accuracy": 0.21371701806783677,
      "num_tokens": 60927760.0,
      "step": 32675
    },
    {
      "entropy": 6.0610716342926025,
      "epoch": 2.8859060402684564,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041718121125725174,
      "loss": 5.3888,
      "mean_token_accuracy": 0.19848044514656066,
      "num_tokens": 60938551.0,
      "step": 32680
    },
    {
      "entropy": 5.963662815093994,
      "epoch": 2.8863475803602965,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041715680279865694,
      "loss": 5.2673,
      "mean_token_accuracy": 0.2172220304608345,
      "num_tokens": 60947457.0,
      "step": 32685
    },
    {
      "entropy": 5.882833051681518,
      "epoch": 2.886789120452137,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004171323915552911,
      "loss": 5.2384,
      "mean_token_accuracy": 0.21748886555433272,
      "num_tokens": 60957459.0,
      "step": 32690
    },
    {
      "entropy": 5.9063458919525145,
      "epoch": 2.8872306605439775,
      "grad_norm": 1.125,
      "learning_rate": 0.0004171079775276326,
      "loss": 5.2595,
      "mean_token_accuracy": 0.21364516019821167,
      "num_tokens": 60966348.0,
      "step": 32695
    },
    {
      "entropy": 6.04410719871521,
      "epoch": 2.8876722006358175,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041708356071615954,
      "loss": 5.2778,
      "mean_token_accuracy": 0.21313629001379014,
      "num_tokens": 60974778.0,
      "step": 32700
    },
    {
      "entropy": 6.005597925186157,
      "epoch": 2.888113740727658,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004170591411213504,
      "loss": 5.3551,
      "mean_token_accuracy": 0.208540441095829,
      "num_tokens": 60984816.0,
      "step": 32705
    },
    {
      "entropy": 5.923843240737915,
      "epoch": 2.8885552808194985,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041703471874368354,
      "loss": 5.2738,
      "mean_token_accuracy": 0.21120763570070267,
      "num_tokens": 60993723.0,
      "step": 32710
    },
    {
      "entropy": 5.940958786010742,
      "epoch": 2.8889968209113386,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041701029358363727,
      "loss": 5.3081,
      "mean_token_accuracy": 0.2226558431982994,
      "num_tokens": 61002363.0,
      "step": 32715
    },
    {
      "entropy": 5.997938728332519,
      "epoch": 2.889438361003179,
      "grad_norm": 1.25,
      "learning_rate": 0.00041698586564169014,
      "loss": 5.2988,
      "mean_token_accuracy": 0.20929040014743805,
      "num_tokens": 61011178.0,
      "step": 32720
    },
    {
      "entropy": 5.924110174179077,
      "epoch": 2.8898799010950196,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041696143491832063,
      "loss": 5.2741,
      "mean_token_accuracy": 0.22066835612058638,
      "num_tokens": 61020415.0,
      "step": 32725
    },
    {
      "entropy": 5.993039608001709,
      "epoch": 2.8903214411868596,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004169370014140073,
      "loss": 5.3471,
      "mean_token_accuracy": 0.20404120534658432,
      "num_tokens": 61030537.0,
      "step": 32730
    },
    {
      "entropy": 6.012492227554321,
      "epoch": 2.8907629812787,
      "grad_norm": 1.125,
      "learning_rate": 0.00041691256512922897,
      "loss": 5.2776,
      "mean_token_accuracy": 0.21234608441591263,
      "num_tokens": 61039666.0,
      "step": 32735
    },
    {
      "entropy": 5.998287105560303,
      "epoch": 2.8912045213705406,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041688812606446407,
      "loss": 5.3282,
      "mean_token_accuracy": 0.20995556563138962,
      "num_tokens": 61049066.0,
      "step": 32740
    },
    {
      "entropy": 5.9680304527282715,
      "epoch": 2.8916460614623807,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041686368422019153,
      "loss": 5.3588,
      "mean_token_accuracy": 0.20372347831726073,
      "num_tokens": 61059647.0,
      "step": 32745
    },
    {
      "entropy": 5.927970457077026,
      "epoch": 2.892087601554221,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004168392395968901,
      "loss": 5.2873,
      "mean_token_accuracy": 0.20796165913343428,
      "num_tokens": 61068814.0,
      "step": 32750
    },
    {
      "entropy": 5.9989465236663815,
      "epoch": 2.8925291416460617,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041681479219503857,
      "loss": 5.3197,
      "mean_token_accuracy": 0.20789508670568466,
      "num_tokens": 61078480.0,
      "step": 32755
    },
    {
      "entropy": 5.952192878723144,
      "epoch": 2.8929706817379017,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000416790342015116,
      "loss": 5.3773,
      "mean_token_accuracy": 0.21339216083288193,
      "num_tokens": 61087830.0,
      "step": 32760
    },
    {
      "entropy": 5.90556902885437,
      "epoch": 2.8934122218297422,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041676588905760123,
      "loss": 5.2454,
      "mean_token_accuracy": 0.22311301231384278,
      "num_tokens": 61096675.0,
      "step": 32765
    },
    {
      "entropy": 5.953578138351441,
      "epoch": 2.8938537619215827,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004167414333229733,
      "loss": 5.3099,
      "mean_token_accuracy": 0.2187599554657936,
      "num_tokens": 61106526.0,
      "step": 32770
    },
    {
      "entropy": 6.014493799209594,
      "epoch": 2.8942953020134228,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004167169748117114,
      "loss": 5.3397,
      "mean_token_accuracy": 0.20880998224020003,
      "num_tokens": 61115851.0,
      "step": 32775
    },
    {
      "entropy": 5.912729024887085,
      "epoch": 2.8947368421052633,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004166925135242945,
      "loss": 5.2534,
      "mean_token_accuracy": 0.2137616530060768,
      "num_tokens": 61125222.0,
      "step": 32780
    },
    {
      "entropy": 5.982861757278442,
      "epoch": 2.8951783821971038,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041666804946120193,
      "loss": 5.3015,
      "mean_token_accuracy": 0.20894471257925035,
      "num_tokens": 61134672.0,
      "step": 32785
    },
    {
      "entropy": 5.940854120254516,
      "epoch": 2.895619922288944,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041664358262291266,
      "loss": 5.2552,
      "mean_token_accuracy": 0.21937933713197708,
      "num_tokens": 61143809.0,
      "step": 32790
    },
    {
      "entropy": 5.901933670043945,
      "epoch": 2.8960614623807843,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041661911300990625,
      "loss": 5.2713,
      "mean_token_accuracy": 0.21068245470523833,
      "num_tokens": 61153362.0,
      "step": 32795
    },
    {
      "entropy": 5.899621820449829,
      "epoch": 2.8965030024726244,
      "grad_norm": 1.109375,
      "learning_rate": 0.000416594640622662,
      "loss": 5.1698,
      "mean_token_accuracy": 0.2282852977514267,
      "num_tokens": 61162291.0,
      "step": 32800
    },
    {
      "entropy": 6.008292770385742,
      "epoch": 2.896944542564465,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041657016546165914,
      "loss": 5.332,
      "mean_token_accuracy": 0.20897233188152314,
      "num_tokens": 61171335.0,
      "step": 32805
    },
    {
      "entropy": 5.9471124649047855,
      "epoch": 2.897386082656305,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041654568752737736,
      "loss": 5.3964,
      "mean_token_accuracy": 0.20811207592487335,
      "num_tokens": 61181137.0,
      "step": 32810
    },
    {
      "entropy": 6.0250598907470705,
      "epoch": 2.8978276227481454,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041652120682029607,
      "loss": 5.3814,
      "mean_token_accuracy": 0.21138982325792313,
      "num_tokens": 61190743.0,
      "step": 32815
    },
    {
      "entropy": 5.958170557022095,
      "epoch": 2.898269162839986,
      "grad_norm": 1.25,
      "learning_rate": 0.0004164967233408948,
      "loss": 5.2146,
      "mean_token_accuracy": 0.21247186213731767,
      "num_tokens": 61200168.0,
      "step": 32820
    },
    {
      "entropy": 6.062820148468018,
      "epoch": 2.898710702931826,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004164722370896533,
      "loss": 5.3339,
      "mean_token_accuracy": 0.20227923542261123,
      "num_tokens": 61209930.0,
      "step": 32825
    },
    {
      "entropy": 5.908832454681397,
      "epoch": 2.8991522430236665,
      "grad_norm": 1.15625,
      "learning_rate": 0.000416447748067051,
      "loss": 5.2293,
      "mean_token_accuracy": 0.21685363352298737,
      "num_tokens": 61219028.0,
      "step": 32830
    },
    {
      "entropy": 5.883637285232544,
      "epoch": 2.899593783115507,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004164232562735678,
      "loss": 5.2398,
      "mean_token_accuracy": 0.2217300772666931,
      "num_tokens": 61228824.0,
      "step": 32835
    },
    {
      "entropy": 6.022135591506958,
      "epoch": 2.900035323207347,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004163987617096834,
      "loss": 5.3385,
      "mean_token_accuracy": 0.2045088604092598,
      "num_tokens": 61237862.0,
      "step": 32840
    },
    {
      "entropy": 5.917687273025512,
      "epoch": 2.9004768632991875,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004163742643758777,
      "loss": 5.2653,
      "mean_token_accuracy": 0.21598423272371292,
      "num_tokens": 61248779.0,
      "step": 32845
    },
    {
      "entropy": 5.9876265048980715,
      "epoch": 2.900918403391028,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041634976427263064,
      "loss": 5.3069,
      "mean_token_accuracy": 0.2128102794289589,
      "num_tokens": 61259385.0,
      "step": 32850
    },
    {
      "entropy": 6.017366933822632,
      "epoch": 2.901359943482868,
      "grad_norm": 1.3125,
      "learning_rate": 0.000416325261400422,
      "loss": 5.3131,
      "mean_token_accuracy": 0.21162403523921966,
      "num_tokens": 61269789.0,
      "step": 32855
    },
    {
      "entropy": 6.042626619338989,
      "epoch": 2.9018014835747086,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004163007557597319,
      "loss": 5.4101,
      "mean_token_accuracy": 0.20214769691228868,
      "num_tokens": 61279136.0,
      "step": 32860
    },
    {
      "entropy": 6.010868263244629,
      "epoch": 2.902243023666549,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004162762473510403,
      "loss": 5.2397,
      "mean_token_accuracy": 0.22023361474275588,
      "num_tokens": 61287448.0,
      "step": 32865
    },
    {
      "entropy": 5.989743089675903,
      "epoch": 2.902684563758389,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004162517361748275,
      "loss": 5.2571,
      "mean_token_accuracy": 0.21418891996145248,
      "num_tokens": 61296006.0,
      "step": 32870
    },
    {
      "entropy": 5.9708963394165036,
      "epoch": 2.9031261038502296,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004162272222315734,
      "loss": 5.281,
      "mean_token_accuracy": 0.21047419160604477,
      "num_tokens": 61304420.0,
      "step": 32875
    },
    {
      "entropy": 6.017952013015747,
      "epoch": 2.90356764394207,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041620270552175843,
      "loss": 5.3658,
      "mean_token_accuracy": 0.20505814999341965,
      "num_tokens": 61314299.0,
      "step": 32880
    },
    {
      "entropy": 5.891271114349365,
      "epoch": 2.90400918403391,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041617818604586265,
      "loss": 5.2286,
      "mean_token_accuracy": 0.21501036137342452,
      "num_tokens": 61323015.0,
      "step": 32885
    },
    {
      "entropy": 5.958874988555908,
      "epoch": 2.9044507241257507,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004161536638043665,
      "loss": 5.3314,
      "mean_token_accuracy": 0.20483235120773316,
      "num_tokens": 61332950.0,
      "step": 32890
    },
    {
      "entropy": 6.007175636291504,
      "epoch": 2.904892264217591,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004161291387977504,
      "loss": 5.3104,
      "mean_token_accuracy": 0.21861246973276138,
      "num_tokens": 61342212.0,
      "step": 32895
    },
    {
      "entropy": 5.966858434677124,
      "epoch": 2.9053338043094312,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004161046110264948,
      "loss": 5.2589,
      "mean_token_accuracy": 0.22563037723302842,
      "num_tokens": 61350786.0,
      "step": 32900
    },
    {
      "entropy": 5.930511808395385,
      "epoch": 2.9057753444012717,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004160800804910801,
      "loss": 5.2752,
      "mean_token_accuracy": 0.22184980660676956,
      "num_tokens": 61360603.0,
      "step": 32905
    },
    {
      "entropy": 5.990667295455933,
      "epoch": 2.906216884493112,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041605554719198674,
      "loss": 5.2766,
      "mean_token_accuracy": 0.216735278069973,
      "num_tokens": 61368552.0,
      "step": 32910
    },
    {
      "entropy": 5.90898060798645,
      "epoch": 2.9066584245849523,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004160310111296956,
      "loss": 5.1885,
      "mean_token_accuracy": 0.22330650687217712,
      "num_tokens": 61377833.0,
      "step": 32915
    },
    {
      "entropy": 6.025966501235962,
      "epoch": 2.9070999646767928,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004160064723046871,
      "loss": 5.3446,
      "mean_token_accuracy": 0.20599410384893418,
      "num_tokens": 61387002.0,
      "step": 32920
    },
    {
      "entropy": 5.9803859233856205,
      "epoch": 2.9075415047686333,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041598193071744197,
      "loss": 5.2735,
      "mean_token_accuracy": 0.21708758622407914,
      "num_tokens": 61396502.0,
      "step": 32925
    },
    {
      "entropy": 5.9604761600494385,
      "epoch": 2.9079830448604733,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041595738636844093,
      "loss": 5.4363,
      "mean_token_accuracy": 0.20484960451722145,
      "num_tokens": 61405937.0,
      "step": 32930
    },
    {
      "entropy": 5.9960390567779545,
      "epoch": 2.908424584952314,
      "grad_norm": 1.140625,
      "learning_rate": 0.000415932839258165,
      "loss": 5.2526,
      "mean_token_accuracy": 0.21366470456123351,
      "num_tokens": 61415504.0,
      "step": 32935
    },
    {
      "entropy": 5.930758810043335,
      "epoch": 2.908866125044154,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004159082893870948,
      "loss": 5.2886,
      "mean_token_accuracy": 0.21381103992462158,
      "num_tokens": 61424370.0,
      "step": 32940
    },
    {
      "entropy": 5.977843856811523,
      "epoch": 2.9093076651359944,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004158837367557114,
      "loss": 5.3063,
      "mean_token_accuracy": 0.20714256018400193,
      "num_tokens": 61433934.0,
      "step": 32945
    },
    {
      "entropy": 5.96506700515747,
      "epoch": 2.9097492052278344,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041585918136449566,
      "loss": 5.2408,
      "mean_token_accuracy": 0.21880008429288864,
      "num_tokens": 61443749.0,
      "step": 32950
    },
    {
      "entropy": 5.965908861160278,
      "epoch": 2.910190745319675,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004158346232139287,
      "loss": 5.315,
      "mean_token_accuracy": 0.20896364897489547,
      "num_tokens": 61452939.0,
      "step": 32955
    },
    {
      "entropy": 5.929996347427368,
      "epoch": 2.9106322854115154,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004158100623044917,
      "loss": 5.2124,
      "mean_token_accuracy": 0.21667844504117967,
      "num_tokens": 61462037.0,
      "step": 32960
    },
    {
      "entropy": 5.981244230270386,
      "epoch": 2.9110738255033555,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00041578549863666555,
      "loss": 5.2615,
      "mean_token_accuracy": 0.21276430487632753,
      "num_tokens": 61470153.0,
      "step": 32965
    },
    {
      "entropy": 6.01048698425293,
      "epoch": 2.911515365595196,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004157609322109315,
      "loss": 5.4273,
      "mean_token_accuracy": 0.19962445944547652,
      "num_tokens": 61479834.0,
      "step": 32970
    },
    {
      "entropy": 6.0238356590271,
      "epoch": 2.9119569056870365,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004157363630277709,
      "loss": 5.3352,
      "mean_token_accuracy": 0.21159088760614395,
      "num_tokens": 61488929.0,
      "step": 32975
    },
    {
      "entropy": 5.9931848526000975,
      "epoch": 2.9123984457788765,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004157117910876649,
      "loss": 5.3063,
      "mean_token_accuracy": 0.21208125203847886,
      "num_tokens": 61498318.0,
      "step": 32980
    },
    {
      "entropy": 5.960239267349243,
      "epoch": 2.912839985870717,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041568721639109507,
      "loss": 5.2541,
      "mean_token_accuracy": 0.21571986377239227,
      "num_tokens": 61507898.0,
      "step": 32985
    },
    {
      "entropy": 5.914579248428344,
      "epoch": 2.9132815259625575,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004156626389385426,
      "loss": 5.2709,
      "mean_token_accuracy": 0.21690115481615066,
      "num_tokens": 61516275.0,
      "step": 32990
    },
    {
      "entropy": 5.924242210388184,
      "epoch": 2.9137230660543976,
      "grad_norm": 1.21875,
      "learning_rate": 0.000415638058730489,
      "loss": 5.2121,
      "mean_token_accuracy": 0.21886899918317795,
      "num_tokens": 61525077.0,
      "step": 32995
    },
    {
      "entropy": 5.8625260353088375,
      "epoch": 2.914164606146238,
      "grad_norm": 1.203125,
      "learning_rate": 0.000415613475767416,
      "loss": 5.1406,
      "mean_token_accuracy": 0.23153869956731796,
      "num_tokens": 61533801.0,
      "step": 33000
    },
    {
      "epoch": 2.914164606146238,
      "eval_entropy": 5.605634628146853,
      "eval_loss": 5.341355800628662,
      "eval_mean_token_accuracy": 0.21904433943660315,
      "eval_num_tokens": 61533801.0,
      "eval_runtime": 20.3224,
      "eval_samples_per_second": 1737.735,
      "eval_steps_per_second": 217.248,
      "step": 33000
    },
    {
      "entropy": 5.848126411437988,
      "epoch": 2.9146061462380786,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041558889004980485,
      "loss": 5.2187,
      "mean_token_accuracy": 0.21683698445558547,
      "num_tokens": 61541865.0,
      "step": 33005
    },
    {
      "entropy": 5.956819152832031,
      "epoch": 2.9150476863299186,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004155643015781373,
      "loss": 5.237,
      "mean_token_accuracy": 0.21869431287050248,
      "num_tokens": 61551035.0,
      "step": 33010
    },
    {
      "entropy": 5.896478080749512,
      "epoch": 2.915489226421759,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004155397103528951,
      "loss": 5.2311,
      "mean_token_accuracy": 0.22540023922920227,
      "num_tokens": 61560135.0,
      "step": 33015
    },
    {
      "entropy": 5.881282567977905,
      "epoch": 2.9159307665135996,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004155151163745598,
      "loss": 5.2711,
      "mean_token_accuracy": 0.21984147131443024,
      "num_tokens": 61569841.0,
      "step": 33020
    },
    {
      "entropy": 5.942872619628906,
      "epoch": 2.9163723066054397,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004154905196436133,
      "loss": 5.2757,
      "mean_token_accuracy": 0.21476173847913743,
      "num_tokens": 61578373.0,
      "step": 33025
    },
    {
      "entropy": 5.942779541015625,
      "epoch": 2.91681384669728,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041546592016053745,
      "loss": 5.207,
      "mean_token_accuracy": 0.2238152801990509,
      "num_tokens": 61587238.0,
      "step": 33030
    },
    {
      "entropy": 5.97689299583435,
      "epoch": 2.9172553867891207,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004154413179258142,
      "loss": 5.34,
      "mean_token_accuracy": 0.2101318359375,
      "num_tokens": 61596045.0,
      "step": 33035
    },
    {
      "entropy": 5.927169370651245,
      "epoch": 2.9176969268809607,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041541671293992524,
      "loss": 5.3011,
      "mean_token_accuracy": 0.21486708968877793,
      "num_tokens": 61604795.0,
      "step": 33040
    },
    {
      "entropy": 5.922966003417969,
      "epoch": 2.918138466972801,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004153921052033528,
      "loss": 5.2599,
      "mean_token_accuracy": 0.2223320946097374,
      "num_tokens": 61613385.0,
      "step": 33045
    },
    {
      "entropy": 5.9611352443695065,
      "epoch": 2.9185800070646417,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041536749471657894,
      "loss": 5.2421,
      "mean_token_accuracy": 0.22085767835378647,
      "num_tokens": 61622664.0,
      "step": 33050
    },
    {
      "entropy": 6.018079328536987,
      "epoch": 2.9190215471564818,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004153428814800856,
      "loss": 5.2941,
      "mean_token_accuracy": 0.20675090700387955,
      "num_tokens": 61631832.0,
      "step": 33055
    },
    {
      "entropy": 6.024705600738526,
      "epoch": 2.9194630872483223,
      "grad_norm": 1.1875,
      "learning_rate": 0.000415318265494355,
      "loss": 5.2885,
      "mean_token_accuracy": 0.21558437645435333,
      "num_tokens": 61640775.0,
      "step": 33060
    },
    {
      "entropy": 5.991046333312989,
      "epoch": 2.9199046273401628,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041529364675986947,
      "loss": 5.288,
      "mean_token_accuracy": 0.20911226123571397,
      "num_tokens": 61650274.0,
      "step": 33065
    },
    {
      "entropy": 5.930509042739868,
      "epoch": 2.920346167432003,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041526902527711114,
      "loss": 5.2283,
      "mean_token_accuracy": 0.21720846593379975,
      "num_tokens": 61659071.0,
      "step": 33070
    },
    {
      "entropy": 5.941888523101807,
      "epoch": 2.9207877075238433,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004152444010465624,
      "loss": 5.302,
      "mean_token_accuracy": 0.2128773108124733,
      "num_tokens": 61668290.0,
      "step": 33075
    },
    {
      "entropy": 6.014091348648071,
      "epoch": 2.9212292476156834,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004152197740687056,
      "loss": 5.3228,
      "mean_token_accuracy": 0.2096525937318802,
      "num_tokens": 61677706.0,
      "step": 33080
    },
    {
      "entropy": 6.006648349761963,
      "epoch": 2.921670787707524,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004151951443440231,
      "loss": 5.2987,
      "mean_token_accuracy": 0.2142626315355301,
      "num_tokens": 61687406.0,
      "step": 33085
    },
    {
      "entropy": 5.976965045928955,
      "epoch": 2.922112327799364,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041517051187299757,
      "loss": 5.3264,
      "mean_token_accuracy": 0.2052757352590561,
      "num_tokens": 61696638.0,
      "step": 33090
    },
    {
      "entropy": 5.957806205749511,
      "epoch": 2.9225538678912044,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041514587665611143,
      "loss": 5.3116,
      "mean_token_accuracy": 0.20364835411310195,
      "num_tokens": 61706478.0,
      "step": 33095
    },
    {
      "entropy": 6.032657146453857,
      "epoch": 2.922995407983045,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004151212386938472,
      "loss": 5.3259,
      "mean_token_accuracy": 0.2137014612555504,
      "num_tokens": 61714480.0,
      "step": 33100
    },
    {
      "entropy": 5.988548851013183,
      "epoch": 2.923436948074885,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041509659798668756,
      "loss": 5.3069,
      "mean_token_accuracy": 0.20939673781394957,
      "num_tokens": 61723569.0,
      "step": 33105
    },
    {
      "entropy": 5.888975667953491,
      "epoch": 2.9238784881667255,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041507195453511543,
      "loss": 5.2667,
      "mean_token_accuracy": 0.21466677486896515,
      "num_tokens": 61732280.0,
      "step": 33110
    },
    {
      "entropy": 5.954014825820923,
      "epoch": 2.924320028258566,
      "grad_norm": 1.125,
      "learning_rate": 0.00041504730833961316,
      "loss": 5.2434,
      "mean_token_accuracy": 0.2179855927824974,
      "num_tokens": 61741556.0,
      "step": 33115
    },
    {
      "entropy": 6.022669410705566,
      "epoch": 2.924761568350406,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004150226594006639,
      "loss": 5.3512,
      "mean_token_accuracy": 0.20539992600679396,
      "num_tokens": 61751629.0,
      "step": 33120
    },
    {
      "entropy": 6.042569780349732,
      "epoch": 2.9252031084422465,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041499800771875036,
      "loss": 5.3758,
      "mean_token_accuracy": 0.20932677537202835,
      "num_tokens": 61761098.0,
      "step": 33125
    },
    {
      "entropy": 5.929853010177612,
      "epoch": 2.925644648534087,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004149733532943554,
      "loss": 5.2106,
      "mean_token_accuracy": 0.2129356786608696,
      "num_tokens": 61769901.0,
      "step": 33130
    },
    {
      "entropy": 5.951848840713501,
      "epoch": 2.926086188625927,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041494869612796213,
      "loss": 5.3142,
      "mean_token_accuracy": 0.20984726548194885,
      "num_tokens": 61777939.0,
      "step": 33135
    },
    {
      "entropy": 5.895539093017578,
      "epoch": 2.9265277287177676,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004149240362200535,
      "loss": 5.2561,
      "mean_token_accuracy": 0.20951841324567794,
      "num_tokens": 61786790.0,
      "step": 33140
    },
    {
      "entropy": 6.04380464553833,
      "epoch": 2.926969268809608,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004148993735711125,
      "loss": 5.4048,
      "mean_token_accuracy": 0.20662365704774857,
      "num_tokens": 61796976.0,
      "step": 33145
    },
    {
      "entropy": 5.957627534866333,
      "epoch": 2.927410808901448,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004148747081816224,
      "loss": 5.2274,
      "mean_token_accuracy": 0.21569125205278397,
      "num_tokens": 61806833.0,
      "step": 33150
    },
    {
      "entropy": 6.027373600006103,
      "epoch": 2.9278523489932886,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004148500400520663,
      "loss": 5.3574,
      "mean_token_accuracy": 0.20671485513448715,
      "num_tokens": 61816121.0,
      "step": 33155
    },
    {
      "entropy": 5.9682845115661625,
      "epoch": 2.928293889085129,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004148253691829275,
      "loss": 5.3785,
      "mean_token_accuracy": 0.20570259988307954,
      "num_tokens": 61825795.0,
      "step": 33160
    },
    {
      "entropy": 6.043085384368896,
      "epoch": 2.928735429176969,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041480069557468915,
      "loss": 5.3839,
      "mean_token_accuracy": 0.20582212507724762,
      "num_tokens": 61835330.0,
      "step": 33165
    },
    {
      "entropy": 5.97907452583313,
      "epoch": 2.9291769692688097,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041477601922783477,
      "loss": 5.3185,
      "mean_token_accuracy": 0.20903859287500381,
      "num_tokens": 61845147.0,
      "step": 33170
    },
    {
      "entropy": 5.952480697631836,
      "epoch": 2.92961850936065,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004147513401428475,
      "loss": 5.3086,
      "mean_token_accuracy": 0.2046406477689743,
      "num_tokens": 61853601.0,
      "step": 33175
    },
    {
      "entropy": 5.982382297515869,
      "epoch": 2.93006004945249,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004147266583202112,
      "loss": 5.3489,
      "mean_token_accuracy": 0.20819096714258195,
      "num_tokens": 61863012.0,
      "step": 33180
    },
    {
      "entropy": 5.9187584400177,
      "epoch": 2.9305015895443307,
      "grad_norm": 1.03125,
      "learning_rate": 0.000414701973760409,
      "loss": 5.2212,
      "mean_token_accuracy": 0.22323649674654006,
      "num_tokens": 61872436.0,
      "step": 33185
    },
    {
      "entropy": 5.987424993515015,
      "epoch": 2.930943129636171,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041467728646392454,
      "loss": 5.2521,
      "mean_token_accuracy": 0.21535383611917497,
      "num_tokens": 61881290.0,
      "step": 33190
    },
    {
      "entropy": 5.964848899841309,
      "epoch": 2.9313846697280113,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041465259643124163,
      "loss": 5.3287,
      "mean_token_accuracy": 0.2071266546845436,
      "num_tokens": 61891051.0,
      "step": 33195
    },
    {
      "entropy": 5.9450860023498535,
      "epoch": 2.9318262098198518,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004146279036628437,
      "loss": 5.2503,
      "mean_token_accuracy": 0.21206975132226943,
      "num_tokens": 61900422.0,
      "step": 33200
    },
    {
      "entropy": 5.960373783111573,
      "epoch": 2.9322677499116923,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041460320815921446,
      "loss": 5.3049,
      "mean_token_accuracy": 0.21078019291162492,
      "num_tokens": 61910181.0,
      "step": 33205
    },
    {
      "entropy": 6.014093494415283,
      "epoch": 2.9327092900035323,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004145785099208379,
      "loss": 5.3965,
      "mean_token_accuracy": 0.20506855398416518,
      "num_tokens": 61919672.0,
      "step": 33210
    },
    {
      "entropy": 5.99896502494812,
      "epoch": 2.933150830095373,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004145538089481976,
      "loss": 5.2119,
      "mean_token_accuracy": 0.22375087738037108,
      "num_tokens": 61928484.0,
      "step": 33215
    },
    {
      "entropy": 6.003696584701538,
      "epoch": 2.933592370187213,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041452910524177766,
      "loss": 5.3255,
      "mean_token_accuracy": 0.20821850299835204,
      "num_tokens": 61936888.0,
      "step": 33220
    },
    {
      "entropy": 5.946455812454223,
      "epoch": 2.9340339102790534,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004145043988020618,
      "loss": 5.2249,
      "mean_token_accuracy": 0.2222824677824974,
      "num_tokens": 61946642.0,
      "step": 33225
    },
    {
      "entropy": 5.968644523620606,
      "epoch": 2.9344754503708934,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004144796896295342,
      "loss": 5.3461,
      "mean_token_accuracy": 0.21595007330179214,
      "num_tokens": 61956273.0,
      "step": 33230
    },
    {
      "entropy": 5.971418285369873,
      "epoch": 2.934916990462734,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041445497772467875,
      "loss": 5.255,
      "mean_token_accuracy": 0.2156722992658615,
      "num_tokens": 61964432.0,
      "step": 33235
    },
    {
      "entropy": 5.942224740982056,
      "epoch": 2.9353585305545744,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041443026308797965,
      "loss": 5.29,
      "mean_token_accuracy": 0.2062464326620102,
      "num_tokens": 61974632.0,
      "step": 33240
    },
    {
      "entropy": 5.951422309875488,
      "epoch": 2.9358000706464145,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004144055457199209,
      "loss": 5.2605,
      "mean_token_accuracy": 0.21223241239786148,
      "num_tokens": 61984239.0,
      "step": 33245
    },
    {
      "entropy": 5.946776056289673,
      "epoch": 2.936241610738255,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041438082562098694,
      "loss": 5.2494,
      "mean_token_accuracy": 0.22748554199934007,
      "num_tokens": 61994049.0,
      "step": 33250
    },
    {
      "entropy": 5.976417207717896,
      "epoch": 2.9366831508300955,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041435610279166174,
      "loss": 5.2656,
      "mean_token_accuracy": 0.2085379794239998,
      "num_tokens": 62003779.0,
      "step": 33255
    },
    {
      "entropy": 5.984280395507812,
      "epoch": 2.9371246909219355,
      "grad_norm": 1.125,
      "learning_rate": 0.0004143313772324298,
      "loss": 5.3159,
      "mean_token_accuracy": 0.20695887953042985,
      "num_tokens": 62012613.0,
      "step": 33260
    },
    {
      "entropy": 5.862727546691895,
      "epoch": 2.937566231013776,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004143066489437755,
      "loss": 5.2085,
      "mean_token_accuracy": 0.2272448718547821,
      "num_tokens": 62022369.0,
      "step": 33265
    },
    {
      "entropy": 5.997770977020264,
      "epoch": 2.9380077711056165,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004142819179261831,
      "loss": 5.38,
      "mean_token_accuracy": 0.2076144352555275,
      "num_tokens": 62032571.0,
      "step": 33270
    },
    {
      "entropy": 5.989183807373047,
      "epoch": 2.9384493111974566,
      "grad_norm": 1.25,
      "learning_rate": 0.0004142571841801371,
      "loss": 5.2352,
      "mean_token_accuracy": 0.22022099047899246,
      "num_tokens": 62041485.0,
      "step": 33275
    },
    {
      "entropy": 5.965821599960327,
      "epoch": 2.938890851289297,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041423244770612207,
      "loss": 5.3386,
      "mean_token_accuracy": 0.2085998311638832,
      "num_tokens": 62050477.0,
      "step": 33280
    },
    {
      "entropy": 5.899396657943726,
      "epoch": 2.9393323913811376,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004142077085046226,
      "loss": 5.2256,
      "mean_token_accuracy": 0.2194635510444641,
      "num_tokens": 62059873.0,
      "step": 33285
    },
    {
      "entropy": 6.029371833801269,
      "epoch": 2.9397739314729776,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004141829665761233,
      "loss": 5.3453,
      "mean_token_accuracy": 0.20906711667776107,
      "num_tokens": 62068639.0,
      "step": 33290
    },
    {
      "entropy": 5.894618892669678,
      "epoch": 2.940215471564818,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004141582219211088,
      "loss": 5.2073,
      "mean_token_accuracy": 0.21941822916269302,
      "num_tokens": 62077637.0,
      "step": 33295
    },
    {
      "entropy": 5.994197130203247,
      "epoch": 2.9406570116566586,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004141334745400639,
      "loss": 5.3763,
      "mean_token_accuracy": 0.20090678781270982,
      "num_tokens": 62087087.0,
      "step": 33300
    },
    {
      "entropy": 5.955516004562378,
      "epoch": 2.9410985517484987,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004141087244334734,
      "loss": 5.2388,
      "mean_token_accuracy": 0.21885283291339874,
      "num_tokens": 62096104.0,
      "step": 33305
    },
    {
      "entropy": 5.963524627685547,
      "epoch": 2.941540091840339,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041408397160182204,
      "loss": 5.3392,
      "mean_token_accuracy": 0.21034057587385177,
      "num_tokens": 62105398.0,
      "step": 33310
    },
    {
      "entropy": 5.933338356018067,
      "epoch": 2.9419816319321797,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041405921604559487,
      "loss": 5.3371,
      "mean_token_accuracy": 0.21207985430955886,
      "num_tokens": 62114444.0,
      "step": 33315
    },
    {
      "entropy": 6.004371643066406,
      "epoch": 2.9424231720240197,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041403445776527667,
      "loss": 5.2548,
      "mean_token_accuracy": 0.21765215396881105,
      "num_tokens": 62123201.0,
      "step": 33320
    },
    {
      "entropy": 6.02733039855957,
      "epoch": 2.94286471211586,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004140096967613525,
      "loss": 5.2868,
      "mean_token_accuracy": 0.21435454338788987,
      "num_tokens": 62131944.0,
      "step": 33325
    },
    {
      "entropy": 5.867361402511596,
      "epoch": 2.9433062522077007,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041398493303430753,
      "loss": 5.22,
      "mean_token_accuracy": 0.2238285079598427,
      "num_tokens": 62141764.0,
      "step": 33330
    },
    {
      "entropy": 6.016649198532105,
      "epoch": 2.9437477922995408,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004139601665846267,
      "loss": 5.3448,
      "mean_token_accuracy": 0.20862661451101303,
      "num_tokens": 62150651.0,
      "step": 33335
    },
    {
      "entropy": 5.961826658248901,
      "epoch": 2.9441893323913813,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004139353974127952,
      "loss": 5.3007,
      "mean_token_accuracy": 0.20901115983724594,
      "num_tokens": 62160713.0,
      "step": 33340
    },
    {
      "entropy": 5.920734214782715,
      "epoch": 2.9446308724832218,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041391062551929833,
      "loss": 5.236,
      "mean_token_accuracy": 0.21742633134126663,
      "num_tokens": 62170443.0,
      "step": 33345
    },
    {
      "entropy": 5.97745680809021,
      "epoch": 2.945072412575062,
      "grad_norm": 1.125,
      "learning_rate": 0.0004138858509046213,
      "loss": 5.3732,
      "mean_token_accuracy": 0.19893016964197158,
      "num_tokens": 62180345.0,
      "step": 33350
    },
    {
      "entropy": 6.069124794006347,
      "epoch": 2.9455139526669023,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041386107356924945,
      "loss": 5.3065,
      "mean_token_accuracy": 0.20820972919464112,
      "num_tokens": 62188649.0,
      "step": 33355
    },
    {
      "entropy": 5.992939424514771,
      "epoch": 2.9459554927587424,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004138362935136681,
      "loss": 5.3148,
      "mean_token_accuracy": 0.20824109762907028,
      "num_tokens": 62197652.0,
      "step": 33360
    },
    {
      "entropy": 5.968065500259399,
      "epoch": 2.946397032850583,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004138115107383627,
      "loss": 5.3858,
      "mean_token_accuracy": 0.19955867081880568,
      "num_tokens": 62207107.0,
      "step": 33365
    },
    {
      "entropy": 6.0243696689605715,
      "epoch": 2.946838572942423,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041378672524381886,
      "loss": 5.3506,
      "mean_token_accuracy": 0.21011788100004197,
      "num_tokens": 62215700.0,
      "step": 33370
    },
    {
      "entropy": 5.879125690460205,
      "epoch": 2.9472801130342634,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004137619370305219,
      "loss": 5.2327,
      "mean_token_accuracy": 0.22927466928958892,
      "num_tokens": 62225084.0,
      "step": 33375
    },
    {
      "entropy": 5.996356534957886,
      "epoch": 2.947721653126104,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041373714609895765,
      "loss": 5.3473,
      "mean_token_accuracy": 0.20724458992481232,
      "num_tokens": 62234052.0,
      "step": 33380
    },
    {
      "entropy": 5.993090677261352,
      "epoch": 2.948163193217944,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004137123524496115,
      "loss": 5.4195,
      "mean_token_accuracy": 0.21151808202266692,
      "num_tokens": 62242772.0,
      "step": 33385
    },
    {
      "entropy": 5.960429763793945,
      "epoch": 2.9486047333097845,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041368755608296926,
      "loss": 5.2722,
      "mean_token_accuracy": 0.212148779630661,
      "num_tokens": 62252216.0,
      "step": 33390
    },
    {
      "entropy": 5.930487489700317,
      "epoch": 2.949046273401625,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004136627569995167,
      "loss": 5.2464,
      "mean_token_accuracy": 0.21862360388040541,
      "num_tokens": 62260820.0,
      "step": 33395
    },
    {
      "entropy": 5.998747491836548,
      "epoch": 2.949487813493465,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004136379551997396,
      "loss": 5.314,
      "mean_token_accuracy": 0.2146909087896347,
      "num_tokens": 62270152.0,
      "step": 33400
    },
    {
      "entropy": 5.992115020751953,
      "epoch": 2.9499293535853055,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004136131506841238,
      "loss": 5.2951,
      "mean_token_accuracy": 0.21286371648311614,
      "num_tokens": 62279027.0,
      "step": 33405
    },
    {
      "entropy": 5.964709901809693,
      "epoch": 2.950370893677146,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041358834345315513,
      "loss": 5.286,
      "mean_token_accuracy": 0.2124534711241722,
      "num_tokens": 62287653.0,
      "step": 33410
    },
    {
      "entropy": 5.933633947372437,
      "epoch": 2.950812433768986,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041356353350731974,
      "loss": 5.2994,
      "mean_token_accuracy": 0.21332424730062485,
      "num_tokens": 62297786.0,
      "step": 33415
    },
    {
      "entropy": 5.983594608306885,
      "epoch": 2.9512539738608266,
      "grad_norm": 1.25,
      "learning_rate": 0.00041353872084710344,
      "loss": 5.2953,
      "mean_token_accuracy": 0.21160662323236465,
      "num_tokens": 62306393.0,
      "step": 33420
    },
    {
      "entropy": 5.984302139282226,
      "epoch": 2.951695513952667,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041351390547299244,
      "loss": 5.2487,
      "mean_token_accuracy": 0.22228976786136628,
      "num_tokens": 62315771.0,
      "step": 33425
    },
    {
      "entropy": 6.005036020278931,
      "epoch": 2.952137054044507,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041348908738547284,
      "loss": 5.3821,
      "mean_token_accuracy": 0.20792028903961182,
      "num_tokens": 62325159.0,
      "step": 33430
    },
    {
      "entropy": 5.997981309890747,
      "epoch": 2.9525785941363476,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041346426658503074,
      "loss": 5.2686,
      "mean_token_accuracy": 0.21698518097400665,
      "num_tokens": 62333953.0,
      "step": 33435
    },
    {
      "entropy": 5.969851446151734,
      "epoch": 2.953020134228188,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004134394430721524,
      "loss": 5.3044,
      "mean_token_accuracy": 0.2114560902118683,
      "num_tokens": 62343149.0,
      "step": 33440
    },
    {
      "entropy": 6.00349235534668,
      "epoch": 2.953461674320028,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004134146168473241,
      "loss": 5.2945,
      "mean_token_accuracy": 0.2001165896654129,
      "num_tokens": 62352618.0,
      "step": 33445
    },
    {
      "entropy": 5.996891498565674,
      "epoch": 2.9539032144118686,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041338978791103227,
      "loss": 5.2403,
      "mean_token_accuracy": 0.2155925527215004,
      "num_tokens": 62361511.0,
      "step": 33450
    },
    {
      "entropy": 5.855793762207031,
      "epoch": 2.954344754503709,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004133649562637631,
      "loss": 5.2154,
      "mean_token_accuracy": 0.21882383972406388,
      "num_tokens": 62369954.0,
      "step": 33455
    },
    {
      "entropy": 5.864848184585571,
      "epoch": 2.954786294595549,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004133401219060032,
      "loss": 5.1783,
      "mean_token_accuracy": 0.21564210206270218,
      "num_tokens": 62378850.0,
      "step": 33460
    },
    {
      "entropy": 5.913133382797241,
      "epoch": 2.9552278346873897,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004133152848382389,
      "loss": 5.2445,
      "mean_token_accuracy": 0.22219915986061095,
      "num_tokens": 62387249.0,
      "step": 33465
    },
    {
      "entropy": 5.966622972488404,
      "epoch": 2.95566937477923,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041329044506095686,
      "loss": 5.2846,
      "mean_token_accuracy": 0.2121021956205368,
      "num_tokens": 62396874.0,
      "step": 33470
    },
    {
      "entropy": 5.973487091064453,
      "epoch": 2.9561109148710702,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041326560257464373,
      "loss": 5.3387,
      "mean_token_accuracy": 0.21368352323770523,
      "num_tokens": 62407336.0,
      "step": 33475
    },
    {
      "entropy": 6.017908382415771,
      "epoch": 2.9565524549629107,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041324075737978605,
      "loss": 5.2988,
      "mean_token_accuracy": 0.20807872265577315,
      "num_tokens": 62416714.0,
      "step": 33480
    },
    {
      "entropy": 6.017075634002685,
      "epoch": 2.9569939950547512,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041321590947687053,
      "loss": 5.3802,
      "mean_token_accuracy": 0.20856108963489534,
      "num_tokens": 62425494.0,
      "step": 33485
    },
    {
      "entropy": 6.0064185619354244,
      "epoch": 2.9574355351465913,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004131910588663839,
      "loss": 5.3103,
      "mean_token_accuracy": 0.2162251815199852,
      "num_tokens": 62433990.0,
      "step": 33490
    },
    {
      "entropy": 6.000263452529907,
      "epoch": 2.957877075238432,
      "grad_norm": 1.25,
      "learning_rate": 0.0004131662055488132,
      "loss": 5.2983,
      "mean_token_accuracy": 0.20888970494270326,
      "num_tokens": 62443193.0,
      "step": 33495
    },
    {
      "entropy": 5.874657344818115,
      "epoch": 2.958318615330272,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041314134952464486,
      "loss": 5.0975,
      "mean_token_accuracy": 0.22865756452083588,
      "num_tokens": 62452161.0,
      "step": 33500
    },
    {
      "entropy": 5.94887523651123,
      "epoch": 2.9587601554221123,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004131164907943662,
      "loss": 5.3212,
      "mean_token_accuracy": 0.21152315735816957,
      "num_tokens": 62461475.0,
      "step": 33505
    },
    {
      "entropy": 5.881084394454956,
      "epoch": 2.9592016955139524,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004130916293584641,
      "loss": 5.1625,
      "mean_token_accuracy": 0.22531914263963698,
      "num_tokens": 62471394.0,
      "step": 33510
    },
    {
      "entropy": 5.953015422821045,
      "epoch": 2.959643235605793,
      "grad_norm": 1.359375,
      "learning_rate": 0.00041306676521742536,
      "loss": 5.2797,
      "mean_token_accuracy": 0.21366693079471588,
      "num_tokens": 62479871.0,
      "step": 33515
    },
    {
      "entropy": 5.9829267978668215,
      "epoch": 2.9600847756976334,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004130418983717373,
      "loss": 5.3358,
      "mean_token_accuracy": 0.20558838695287704,
      "num_tokens": 62489825.0,
      "step": 33520
    },
    {
      "entropy": 5.996622848510742,
      "epoch": 2.9605263157894735,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004130170288218869,
      "loss": 5.3107,
      "mean_token_accuracy": 0.20877299159765245,
      "num_tokens": 62499513.0,
      "step": 33525
    },
    {
      "entropy": 5.9408656597137455,
      "epoch": 2.960967855881314,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004129921565683614,
      "loss": 5.2932,
      "mean_token_accuracy": 0.21151324957609177,
      "num_tokens": 62509841.0,
      "step": 33530
    },
    {
      "entropy": 5.927914953231811,
      "epoch": 2.9614093959731544,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041296728161164815,
      "loss": 5.273,
      "mean_token_accuracy": 0.20789711326360702,
      "num_tokens": 62518374.0,
      "step": 33535
    },
    {
      "entropy": 5.948477792739868,
      "epoch": 2.9618509360649945,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004129424039522342,
      "loss": 5.2079,
      "mean_token_accuracy": 0.22081878036260605,
      "num_tokens": 62526570.0,
      "step": 33540
    },
    {
      "entropy": 5.941392326354981,
      "epoch": 2.962292476156835,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004129175235906071,
      "loss": 5.3215,
      "mean_token_accuracy": 0.2066471129655838,
      "num_tokens": 62535525.0,
      "step": 33545
    },
    {
      "entropy": 6.025748634338379,
      "epoch": 2.9627340162486755,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000412892640527254,
      "loss": 5.296,
      "mean_token_accuracy": 0.2130693718791008,
      "num_tokens": 62545083.0,
      "step": 33550
    },
    {
      "entropy": 5.950818872451782,
      "epoch": 2.9631755563405155,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004128677547626627,
      "loss": 5.1833,
      "mean_token_accuracy": 0.22046451419591903,
      "num_tokens": 62554156.0,
      "step": 33555
    },
    {
      "entropy": 5.933926296234131,
      "epoch": 2.963617096432356,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004128428662973203,
      "loss": 5.2428,
      "mean_token_accuracy": 0.22864995449781417,
      "num_tokens": 62562696.0,
      "step": 33560
    },
    {
      "entropy": 5.937869596481323,
      "epoch": 2.9640586365241965,
      "grad_norm": 1.125,
      "learning_rate": 0.00041281797513171464,
      "loss": 5.2364,
      "mean_token_accuracy": 0.22044832408428192,
      "num_tokens": 62572337.0,
      "step": 33565
    },
    {
      "entropy": 5.984912157058716,
      "epoch": 2.9645001766160366,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004127930812663333,
      "loss": 5.2929,
      "mean_token_accuracy": 0.21413194686174392,
      "num_tokens": 62581716.0,
      "step": 33570
    },
    {
      "entropy": 5.951999092102051,
      "epoch": 2.964941716707877,
      "grad_norm": 1.125,
      "learning_rate": 0.00041276818470166373,
      "loss": 5.2608,
      "mean_token_accuracy": 0.21675093322992325,
      "num_tokens": 62590270.0,
      "step": 33575
    },
    {
      "entropy": 5.959853029251098,
      "epoch": 2.9653832567997176,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004127432854381939,
      "loss": 5.3356,
      "mean_token_accuracy": 0.21121472120285034,
      "num_tokens": 62600391.0,
      "step": 33580
    },
    {
      "entropy": 5.977387428283691,
      "epoch": 2.9658247968915576,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004127183834764114,
      "loss": 5.3473,
      "mean_token_accuracy": 0.21483975052833557,
      "num_tokens": 62610873.0,
      "step": 33585
    },
    {
      "entropy": 5.988357448577881,
      "epoch": 2.966266336983398,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000412693478816804,
      "loss": 5.2957,
      "mean_token_accuracy": 0.21181966960430146,
      "num_tokens": 62619565.0,
      "step": 33590
    },
    {
      "entropy": 5.972347545623779,
      "epoch": 2.9667078770752386,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004126685714598598,
      "loss": 5.3023,
      "mean_token_accuracy": 0.20779698342084885,
      "num_tokens": 62628303.0,
      "step": 33595
    },
    {
      "entropy": 6.070357847213745,
      "epoch": 2.9671494171670787,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041264366140606644,
      "loss": 5.3913,
      "mean_token_accuracy": 0.20448958426713942,
      "num_tokens": 62638026.0,
      "step": 33600
    },
    {
      "entropy": 6.010159683227539,
      "epoch": 2.967590957258919,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004126187486559121,
      "loss": 5.3242,
      "mean_token_accuracy": 0.2160753697156906,
      "num_tokens": 62647016.0,
      "step": 33605
    },
    {
      "entropy": 5.988988256454467,
      "epoch": 2.9680324973507597,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004125938332098848,
      "loss": 5.3407,
      "mean_token_accuracy": 0.205265636742115,
      "num_tokens": 62657238.0,
      "step": 33610
    },
    {
      "entropy": 6.013646125793457,
      "epoch": 2.9684740374425997,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041256891506847253,
      "loss": 5.3281,
      "mean_token_accuracy": 0.2084490329027176,
      "num_tokens": 62666256.0,
      "step": 33615
    },
    {
      "entropy": 6.085491895675659,
      "epoch": 2.9689155775344402,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004125439942321634,
      "loss": 5.3722,
      "mean_token_accuracy": 0.20452325493097306,
      "num_tokens": 62675859.0,
      "step": 33620
    },
    {
      "entropy": 6.068053960800171,
      "epoch": 2.9693571176262807,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041251907070144575,
      "loss": 5.4144,
      "mean_token_accuracy": 0.20870960950851442,
      "num_tokens": 62685768.0,
      "step": 33625
    },
    {
      "entropy": 5.864530181884765,
      "epoch": 2.969798657718121,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004124941444768076,
      "loss": 5.1403,
      "mean_token_accuracy": 0.22727875262498856,
      "num_tokens": 62694729.0,
      "step": 33630
    },
    {
      "entropy": 5.950620985031128,
      "epoch": 2.9702401978099613,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041246921555873747,
      "loss": 5.2313,
      "mean_token_accuracy": 0.2144442155957222,
      "num_tokens": 62704347.0,
      "step": 33635
    },
    {
      "entropy": 5.97704668045044,
      "epoch": 2.9706817379018013,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004124442839477235,
      "loss": 5.3209,
      "mean_token_accuracy": 0.20537052750587464,
      "num_tokens": 62713627.0,
      "step": 33640
    },
    {
      "entropy": 5.976431560516358,
      "epoch": 2.971123277993642,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004124193496442543,
      "loss": 5.311,
      "mean_token_accuracy": 0.2133747473359108,
      "num_tokens": 62723023.0,
      "step": 33645
    },
    {
      "entropy": 5.89295015335083,
      "epoch": 2.971564818085482,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000412394412648818,
      "loss": 5.2098,
      "mean_token_accuracy": 0.23134251534938813,
      "num_tokens": 62732370.0,
      "step": 33650
    },
    {
      "entropy": 5.999881076812744,
      "epoch": 2.9720063581773224,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004123694729619034,
      "loss": 5.3587,
      "mean_token_accuracy": 0.2051795795559883,
      "num_tokens": 62741751.0,
      "step": 33655
    },
    {
      "entropy": 6.032847213745117,
      "epoch": 2.972447898269163,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041234453058399897,
      "loss": 5.4117,
      "mean_token_accuracy": 0.20182975381612778,
      "num_tokens": 62752052.0,
      "step": 33660
    },
    {
      "entropy": 5.9571309089660645,
      "epoch": 2.972889438361003,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004123195855155932,
      "loss": 5.2868,
      "mean_token_accuracy": 0.21524550914764404,
      "num_tokens": 62762179.0,
      "step": 33665
    },
    {
      "entropy": 5.925134754180908,
      "epoch": 2.9733309784528434,
      "grad_norm": 1.171875,
      "learning_rate": 0.000412294637757175,
      "loss": 5.2124,
      "mean_token_accuracy": 0.2176145002245903,
      "num_tokens": 62771226.0,
      "step": 33670
    },
    {
      "entropy": 5.927850437164307,
      "epoch": 2.973772518544684,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004122696873092328,
      "loss": 5.3187,
      "mean_token_accuracy": 0.20631390810012817,
      "num_tokens": 62780241.0,
      "step": 33675
    },
    {
      "entropy": 6.055274438858032,
      "epoch": 2.974214058636524,
      "grad_norm": 1.25,
      "learning_rate": 0.0004122447341722556,
      "loss": 5.432,
      "mean_token_accuracy": 0.20450659841299057,
      "num_tokens": 62790142.0,
      "step": 33680
    },
    {
      "entropy": 6.019876146316529,
      "epoch": 2.9746555987283645,
      "grad_norm": 1.125,
      "learning_rate": 0.00041221977834673203,
      "loss": 5.3432,
      "mean_token_accuracy": 0.2119385913014412,
      "num_tokens": 62799923.0,
      "step": 33685
    },
    {
      "entropy": 5.999455404281616,
      "epoch": 2.975097138820205,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004121948198331511,
      "loss": 5.2877,
      "mean_token_accuracy": 0.21602463126182556,
      "num_tokens": 62809727.0,
      "step": 33690
    },
    {
      "entropy": 5.995938110351562,
      "epoch": 2.975538678912045,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041216985863200164,
      "loss": 5.2903,
      "mean_token_accuracy": 0.21311983317136765,
      "num_tokens": 62819069.0,
      "step": 33695
    },
    {
      "entropy": 5.866266202926636,
      "epoch": 2.9759802190038855,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041214489474377265,
      "loss": 5.1985,
      "mean_token_accuracy": 0.22503309547901154,
      "num_tokens": 62829683.0,
      "step": 33700
    },
    {
      "entropy": 5.915395259857178,
      "epoch": 2.976421759095726,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004121199281689532,
      "loss": 5.2579,
      "mean_token_accuracy": 0.2168501377105713,
      "num_tokens": 62838824.0,
      "step": 33705
    },
    {
      "entropy": 5.940398931503296,
      "epoch": 2.976863299187566,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041209495890803235,
      "loss": 5.2151,
      "mean_token_accuracy": 0.21379316300153733,
      "num_tokens": 62847649.0,
      "step": 33710
    },
    {
      "entropy": 6.007437372207642,
      "epoch": 2.9773048392794066,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004120699869614992,
      "loss": 5.3767,
      "mean_token_accuracy": 0.20809174031019212,
      "num_tokens": 62857495.0,
      "step": 33715
    },
    {
      "entropy": 5.967010545730591,
      "epoch": 2.977746379371247,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000412045012329843,
      "loss": 5.309,
      "mean_token_accuracy": 0.2092619091272354,
      "num_tokens": 62867344.0,
      "step": 33720
    },
    {
      "entropy": 5.940288352966308,
      "epoch": 2.978187919463087,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004120200350135529,
      "loss": 5.2391,
      "mean_token_accuracy": 0.21734571009874343,
      "num_tokens": 62876770.0,
      "step": 33725
    },
    {
      "entropy": 5.985115623474121,
      "epoch": 2.9786294595549276,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004119950550131183,
      "loss": 5.2992,
      "mean_token_accuracy": 0.21499156951904297,
      "num_tokens": 62886249.0,
      "step": 33730
    },
    {
      "entropy": 6.03442063331604,
      "epoch": 2.979070999646768,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004119700723290284,
      "loss": 5.3353,
      "mean_token_accuracy": 0.20380435436964034,
      "num_tokens": 62896112.0,
      "step": 33735
    },
    {
      "entropy": 5.9233780860900875,
      "epoch": 2.979512539738608,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041194508696177273,
      "loss": 5.2924,
      "mean_token_accuracy": 0.21528191715478898,
      "num_tokens": 62906350.0,
      "step": 33740
    },
    {
      "entropy": 5.907672786712647,
      "epoch": 2.9799540798304487,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004119200989118407,
      "loss": 5.3209,
      "mean_token_accuracy": 0.206858828663826,
      "num_tokens": 62914766.0,
      "step": 33745
    },
    {
      "entropy": 5.943101644515991,
      "epoch": 2.980395619922289,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041189510817972176,
      "loss": 5.235,
      "mean_token_accuracy": 0.21948982328176497,
      "num_tokens": 62924292.0,
      "step": 33750
    },
    {
      "entropy": 6.033412027359009,
      "epoch": 2.9808371600141292,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004118701147659056,
      "loss": 5.3027,
      "mean_token_accuracy": 0.21298626065254211,
      "num_tokens": 62934020.0,
      "step": 33755
    },
    {
      "entropy": 5.9926306247711185,
      "epoch": 2.9812787001059697,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041184511867088165,
      "loss": 5.2251,
      "mean_token_accuracy": 0.2162170112133026,
      "num_tokens": 62943830.0,
      "step": 33760
    },
    {
      "entropy": 5.974293279647827,
      "epoch": 2.9817202401978102,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004118201198951397,
      "loss": 5.3369,
      "mean_token_accuracy": 0.20811415314674378,
      "num_tokens": 62953333.0,
      "step": 33765
    },
    {
      "entropy": 5.884397172927857,
      "epoch": 2.9821617802896503,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004117951184391694,
      "loss": 5.2267,
      "mean_token_accuracy": 0.21925292313098907,
      "num_tokens": 62961950.0,
      "step": 33770
    },
    {
      "entropy": 5.953878355026245,
      "epoch": 2.982603320381491,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004117701143034605,
      "loss": 5.3022,
      "mean_token_accuracy": 0.21787810921669007,
      "num_tokens": 62971006.0,
      "step": 33775
    },
    {
      "entropy": 5.9135661125183105,
      "epoch": 2.983044860473331,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041174510748850284,
      "loss": 5.2216,
      "mean_token_accuracy": 0.226550655066967,
      "num_tokens": 62980477.0,
      "step": 33780
    },
    {
      "entropy": 5.914917135238648,
      "epoch": 2.9834864005651713,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004117200979947864,
      "loss": 5.2442,
      "mean_token_accuracy": 0.21435294598340987,
      "num_tokens": 62990609.0,
      "step": 33785
    },
    {
      "entropy": 6.051311683654785,
      "epoch": 2.9839279406570114,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041169508582280086,
      "loss": 5.2834,
      "mean_token_accuracy": 0.20375465154647826,
      "num_tokens": 62999771.0,
      "step": 33790
    },
    {
      "entropy": 5.94813871383667,
      "epoch": 2.984369480748852,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004116700709730364,
      "loss": 5.2549,
      "mean_token_accuracy": 0.21622627675533296,
      "num_tokens": 63008939.0,
      "step": 33795
    },
    {
      "entropy": 5.9333044528961185,
      "epoch": 2.9848110208406924,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000411645053445983,
      "loss": 5.3146,
      "mean_token_accuracy": 0.21375755816698075,
      "num_tokens": 63018302.0,
      "step": 33800
    },
    {
      "entropy": 5.954346179962158,
      "epoch": 2.9852525609325324,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004116200332421307,
      "loss": 5.3251,
      "mean_token_accuracy": 0.21477386951446534,
      "num_tokens": 63027372.0,
      "step": 33805
    },
    {
      "entropy": 5.925436305999756,
      "epoch": 2.985694101024373,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004115950103619696,
      "loss": 5.3403,
      "mean_token_accuracy": 0.20973036140203477,
      "num_tokens": 63036777.0,
      "step": 33810
    },
    {
      "entropy": 5.983143615722656,
      "epoch": 2.9861356411162134,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041156998480598993,
      "loss": 5.2422,
      "mean_token_accuracy": 0.21419114023447036,
      "num_tokens": 63045819.0,
      "step": 33815
    },
    {
      "entropy": 5.942198038101196,
      "epoch": 2.9865771812080535,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000411544956574682,
      "loss": 5.2731,
      "mean_token_accuracy": 0.21681995838880538,
      "num_tokens": 63054896.0,
      "step": 33820
    },
    {
      "entropy": 5.897773790359497,
      "epoch": 2.987018721299894,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000411519925668536,
      "loss": 5.2214,
      "mean_token_accuracy": 0.22561139166355132,
      "num_tokens": 63063741.0,
      "step": 33825
    },
    {
      "entropy": 5.955629825592041,
      "epoch": 2.9874602613917345,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041149489208804237,
      "loss": 5.2311,
      "mean_token_accuracy": 0.2185116410255432,
      "num_tokens": 63071886.0,
      "step": 33830
    },
    {
      "entropy": 5.976857471466064,
      "epoch": 2.9879018014835745,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041146985583369127,
      "loss": 5.3974,
      "mean_token_accuracy": 0.20522364526987075,
      "num_tokens": 63082162.0,
      "step": 33835
    },
    {
      "entropy": 5.962367534637451,
      "epoch": 2.988343341575415,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004114448169059735,
      "loss": 5.3269,
      "mean_token_accuracy": 0.20938832610845565,
      "num_tokens": 63091894.0,
      "step": 33840
    },
    {
      "entropy": 6.012951135635376,
      "epoch": 2.9887848816672555,
      "grad_norm": 1.125,
      "learning_rate": 0.0004114197753053792,
      "loss": 5.3636,
      "mean_token_accuracy": 0.20533599704504013,
      "num_tokens": 63101446.0,
      "step": 33845
    },
    {
      "entropy": 6.026219701766967,
      "epoch": 2.9892264217590956,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041139473103239915,
      "loss": 5.275,
      "mean_token_accuracy": 0.2175838530063629,
      "num_tokens": 63109672.0,
      "step": 33850
    },
    {
      "entropy": 5.978659725189209,
      "epoch": 2.989667961850936,
      "grad_norm": 1.25,
      "learning_rate": 0.00041136968408752396,
      "loss": 5.3162,
      "mean_token_accuracy": 0.21113641560077667,
      "num_tokens": 63119286.0,
      "step": 33855
    },
    {
      "entropy": 5.897288036346436,
      "epoch": 2.9901095019427766,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004113446344712441,
      "loss": 5.223,
      "mean_token_accuracy": 0.21831308007240297,
      "num_tokens": 63127594.0,
      "step": 33860
    },
    {
      "entropy": 5.994102144241333,
      "epoch": 2.9905510420346166,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004113195821840505,
      "loss": 5.3104,
      "mean_token_accuracy": 0.20955484211444855,
      "num_tokens": 63136552.0,
      "step": 33865
    },
    {
      "entropy": 6.060847377777099,
      "epoch": 2.990992582126457,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004112945272264337,
      "loss": 5.4028,
      "mean_token_accuracy": 0.20642692744731903,
      "num_tokens": 63146085.0,
      "step": 33870
    },
    {
      "entropy": 6.04271969795227,
      "epoch": 2.9914341222182976,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004112694695988847,
      "loss": 5.2427,
      "mean_token_accuracy": 0.21444784849882126,
      "num_tokens": 63154100.0,
      "step": 33875
    },
    {
      "entropy": 6.053703260421753,
      "epoch": 2.9918756623101377,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041124440930189425,
      "loss": 5.3421,
      "mean_token_accuracy": 0.20836732238531114,
      "num_tokens": 63164057.0,
      "step": 33880
    },
    {
      "entropy": 5.974147891998291,
      "epoch": 2.992317202401978,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004112193463359533,
      "loss": 5.3676,
      "mean_token_accuracy": 0.20720436573028564,
      "num_tokens": 63173248.0,
      "step": 33885
    },
    {
      "entropy": 6.007970714569092,
      "epoch": 2.9927587424938187,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004111942807015529,
      "loss": 5.2978,
      "mean_token_accuracy": 0.2120947688817978,
      "num_tokens": 63183391.0,
      "step": 33890
    },
    {
      "entropy": 6.031949806213379,
      "epoch": 2.9932002825856587,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000411169212399184,
      "loss": 5.3209,
      "mean_token_accuracy": 0.2093517765402794,
      "num_tokens": 63192611.0,
      "step": 33895
    },
    {
      "entropy": 5.977518701553345,
      "epoch": 2.9936418226774992,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004111441414293376,
      "loss": 5.3097,
      "mean_token_accuracy": 0.21206195950508117,
      "num_tokens": 63202201.0,
      "step": 33900
    },
    {
      "entropy": 5.970065784454346,
      "epoch": 2.9940833627693397,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004111190677925049,
      "loss": 5.3055,
      "mean_token_accuracy": 0.221853369474411,
      "num_tokens": 63212267.0,
      "step": 33905
    },
    {
      "entropy": 5.91898136138916,
      "epoch": 2.99452490286118,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004110939914891771,
      "loss": 5.1896,
      "mean_token_accuracy": 0.22679060250520705,
      "num_tokens": 63222218.0,
      "step": 33910
    },
    {
      "entropy": 5.98836841583252,
      "epoch": 2.9949664429530203,
      "grad_norm": 1.25,
      "learning_rate": 0.00041106891251984544,
      "loss": 5.3633,
      "mean_token_accuracy": 0.20119740068912506,
      "num_tokens": 63231929.0,
      "step": 33915
    },
    {
      "entropy": 6.022898817062378,
      "epoch": 2.9954079830448603,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004110438308850012,
      "loss": 5.3029,
      "mean_token_accuracy": 0.21307262927293777,
      "num_tokens": 63240971.0,
      "step": 33920
    },
    {
      "entropy": 5.9929345607757565,
      "epoch": 2.995849523136701,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004110187465851356,
      "loss": 5.3554,
      "mean_token_accuracy": 0.20902592092752456,
      "num_tokens": 63251364.0,
      "step": 33925
    },
    {
      "entropy": 5.910732412338257,
      "epoch": 2.996291063228541,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041099365962074015,
      "loss": 5.222,
      "mean_token_accuracy": 0.21579815894365312,
      "num_tokens": 63260568.0,
      "step": 33930
    },
    {
      "entropy": 5.9213865280151365,
      "epoch": 2.9967326033203814,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041096856999230623,
      "loss": 5.2379,
      "mean_token_accuracy": 0.21303392499685286,
      "num_tokens": 63270634.0,
      "step": 33935
    },
    {
      "entropy": 5.915092658996582,
      "epoch": 2.997174143412222,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004109434777003254,
      "loss": 5.1649,
      "mean_token_accuracy": 0.22761210054159164,
      "num_tokens": 63280776.0,
      "step": 33940
    },
    {
      "entropy": 5.981702899932861,
      "epoch": 2.997615683504062,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004109183827452891,
      "loss": 5.368,
      "mean_token_accuracy": 0.2033235639333725,
      "num_tokens": 63290251.0,
      "step": 33945
    },
    {
      "entropy": 5.971038103103638,
      "epoch": 2.9980572235959024,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041089328512768905,
      "loss": 5.2762,
      "mean_token_accuracy": 0.21902882754802705,
      "num_tokens": 63298836.0,
      "step": 33950
    },
    {
      "entropy": 5.8676042556762695,
      "epoch": 2.998498763687743,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004108681848480168,
      "loss": 5.2148,
      "mean_token_accuracy": 0.21841268092393876,
      "num_tokens": 63308506.0,
      "step": 33955
    },
    {
      "entropy": 5.85659441947937,
      "epoch": 2.998940303779583,
      "grad_norm": 1.125,
      "learning_rate": 0.000410843081906764,
      "loss": 5.1865,
      "mean_token_accuracy": 0.22328765988349913,
      "num_tokens": 63317530.0,
      "step": 33960
    },
    {
      "entropy": 6.014948463439941,
      "epoch": 2.9993818438714235,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041081797630442257,
      "loss": 5.2996,
      "mean_token_accuracy": 0.2138907089829445,
      "num_tokens": 63326521.0,
      "step": 33965
    },
    {
      "entropy": 5.976866817474365,
      "epoch": 2.999823383963264,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041079286804148417,
      "loss": 5.3511,
      "mean_token_accuracy": 0.2093915969133377,
      "num_tokens": 63336199.0,
      "step": 33970
    },
    {
      "entropy": 5.98206844329834,
      "epoch": 3.000264924055104,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041076775711844084,
      "loss": 5.3365,
      "mean_token_accuracy": 0.20290817320346832,
      "num_tokens": 63345249.0,
      "step": 33975
    },
    {
      "entropy": 6.068708610534668,
      "epoch": 3.0007064641469445,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041074264353578425,
      "loss": 5.2752,
      "mean_token_accuracy": 0.21479835063219072,
      "num_tokens": 63353655.0,
      "step": 33980
    },
    {
      "entropy": 6.014402580261231,
      "epoch": 3.001148004238785,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041071752729400653,
      "loss": 5.2187,
      "mean_token_accuracy": 0.22262188643217087,
      "num_tokens": 63362984.0,
      "step": 33985
    },
    {
      "entropy": 5.971528339385986,
      "epoch": 3.001589544330625,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041069240839359964,
      "loss": 5.2479,
      "mean_token_accuracy": 0.21528228223323823,
      "num_tokens": 63372078.0,
      "step": 33990
    },
    {
      "entropy": 5.98816442489624,
      "epoch": 3.0020310844224656,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041066728683505563,
      "loss": 5.3836,
      "mean_token_accuracy": 0.2081477850675583,
      "num_tokens": 63382446.0,
      "step": 33995
    },
    {
      "entropy": 6.001004552841186,
      "epoch": 3.002472624514306,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004106421626188666,
      "loss": 5.2843,
      "mean_token_accuracy": 0.21576034277677536,
      "num_tokens": 63391890.0,
      "step": 34000
    },
    {
      "entropy": 6.044013929367066,
      "epoch": 3.002914164606146,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004106170357455247,
      "loss": 5.2659,
      "mean_token_accuracy": 0.2130102962255478,
      "num_tokens": 63401306.0,
      "step": 34005
    },
    {
      "entropy": 5.9939796924591064,
      "epoch": 3.0033557046979866,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004105919062155223,
      "loss": 5.2651,
      "mean_token_accuracy": 0.21635436415672302,
      "num_tokens": 63411037.0,
      "step": 34010
    },
    {
      "entropy": 5.966242265701294,
      "epoch": 3.003797244789827,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041056677402935153,
      "loss": 5.2165,
      "mean_token_accuracy": 0.2229934260249138,
      "num_tokens": 63419010.0,
      "step": 34015
    },
    {
      "entropy": 5.932560825347901,
      "epoch": 3.004238784881667,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041054163918750474,
      "loss": 5.2173,
      "mean_token_accuracy": 0.21623583883047104,
      "num_tokens": 63427868.0,
      "step": 34020
    },
    {
      "entropy": 5.973716402053833,
      "epoch": 3.0046803249735077,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004105165016904744,
      "loss": 5.2554,
      "mean_token_accuracy": 0.21892638504505157,
      "num_tokens": 63436405.0,
      "step": 34025
    },
    {
      "entropy": 5.938160991668701,
      "epoch": 3.0051218650653477,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004104913615387528,
      "loss": 5.2192,
      "mean_token_accuracy": 0.21997667402029036,
      "num_tokens": 63446334.0,
      "step": 34030
    },
    {
      "entropy": 5.9472270011901855,
      "epoch": 3.0055634051571882,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004104662187328324,
      "loss": 5.2394,
      "mean_token_accuracy": 0.22054296731948853,
      "num_tokens": 63455253.0,
      "step": 34035
    },
    {
      "entropy": 5.956081914901733,
      "epoch": 3.0060049452490287,
      "grad_norm": 1.25,
      "learning_rate": 0.000410441073273206,
      "loss": 5.1869,
      "mean_token_accuracy": 0.21009813547134398,
      "num_tokens": 63464120.0,
      "step": 34040
    },
    {
      "entropy": 5.959663009643554,
      "epoch": 3.006446485340869,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041041592516036587,
      "loss": 5.2604,
      "mean_token_accuracy": 0.21461089849472045,
      "num_tokens": 63474049.0,
      "step": 34045
    },
    {
      "entropy": 5.947968626022339,
      "epoch": 3.0068880254327093,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041039077439480485,
      "loss": 5.1857,
      "mean_token_accuracy": 0.22555974125862122,
      "num_tokens": 63482970.0,
      "step": 34050
    },
    {
      "entropy": 5.927834177017212,
      "epoch": 3.0073295655245498,
      "grad_norm": 1.125,
      "learning_rate": 0.0004103656209770155,
      "loss": 5.2449,
      "mean_token_accuracy": 0.20734198689460753,
      "num_tokens": 63492162.0,
      "step": 34055
    },
    {
      "entropy": 5.951849365234375,
      "epoch": 3.00777110561639,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004103404649074908,
      "loss": 5.2722,
      "mean_token_accuracy": 0.2094929948449135,
      "num_tokens": 63501531.0,
      "step": 34060
    },
    {
      "entropy": 5.917407655715943,
      "epoch": 3.0082126457082303,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004103153061867232,
      "loss": 5.256,
      "mean_token_accuracy": 0.2199586570262909,
      "num_tokens": 63510156.0,
      "step": 34065
    },
    {
      "entropy": 5.984799003601074,
      "epoch": 3.008654185800071,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004102901448152057,
      "loss": 5.2476,
      "mean_token_accuracy": 0.21316035985946655,
      "num_tokens": 63520758.0,
      "step": 34070
    },
    {
      "entropy": 6.086665868759155,
      "epoch": 3.009095725891911,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004102649807934312,
      "loss": 5.4163,
      "mean_token_accuracy": 0.20389041006565095,
      "num_tokens": 63529500.0,
      "step": 34075
    },
    {
      "entropy": 5.927806329727173,
      "epoch": 3.0095372659837514,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004102398141218927,
      "loss": 5.1669,
      "mean_token_accuracy": 0.21653734147548676,
      "num_tokens": 63537473.0,
      "step": 34080
    },
    {
      "entropy": 5.907670593261718,
      "epoch": 3.009978806075592,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004102146448010832,
      "loss": 5.2815,
      "mean_token_accuracy": 0.21350571662187576,
      "num_tokens": 63547168.0,
      "step": 34085
    },
    {
      "entropy": 6.024657440185547,
      "epoch": 3.010420346167432,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004101894728314956,
      "loss": 5.3134,
      "mean_token_accuracy": 0.2089492440223694,
      "num_tokens": 63556776.0,
      "step": 34090
    },
    {
      "entropy": 6.016443824768066,
      "epoch": 3.0108618862592724,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004101642982136232,
      "loss": 5.3135,
      "mean_token_accuracy": 0.21499282568693162,
      "num_tokens": 63567097.0,
      "step": 34095
    },
    {
      "entropy": 5.984295606613159,
      "epoch": 3.0113034263511125,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041013912094795907,
      "loss": 5.2063,
      "mean_token_accuracy": 0.21374855041503907,
      "num_tokens": 63575047.0,
      "step": 34100
    },
    {
      "entropy": 5.87528133392334,
      "epoch": 3.011744966442953,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004101139410349964,
      "loss": 5.1542,
      "mean_token_accuracy": 0.22261232435703276,
      "num_tokens": 63584629.0,
      "step": 34105
    },
    {
      "entropy": 5.99114727973938,
      "epoch": 3.0121865065347935,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004100887584752284,
      "loss": 5.2804,
      "mean_token_accuracy": 0.21452885121107101,
      "num_tokens": 63593739.0,
      "step": 34110
    },
    {
      "entropy": 5.952167320251465,
      "epoch": 3.0126280466266335,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041006357326914863,
      "loss": 5.2579,
      "mean_token_accuracy": 0.21160849928855896,
      "num_tokens": 63602783.0,
      "step": 34115
    },
    {
      "entropy": 5.972410106658936,
      "epoch": 3.013069586718474,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004100383854172501,
      "loss": 5.2144,
      "mean_token_accuracy": 0.2196671262383461,
      "num_tokens": 63612292.0,
      "step": 34120
    },
    {
      "entropy": 5.941990232467651,
      "epoch": 3.0135111268103145,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004100131949200265,
      "loss": 5.1911,
      "mean_token_accuracy": 0.21857593804597855,
      "num_tokens": 63621656.0,
      "step": 34125
    },
    {
      "entropy": 5.962312364578247,
      "epoch": 3.0139526669021546,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004099880017779711,
      "loss": 5.2355,
      "mean_token_accuracy": 0.2136356234550476,
      "num_tokens": 63630320.0,
      "step": 34130
    },
    {
      "entropy": 5.927200841903686,
      "epoch": 3.014394206993995,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004099628059915776,
      "loss": 5.2113,
      "mean_token_accuracy": 0.22035278826951982,
      "num_tokens": 63639155.0,
      "step": 34135
    },
    {
      "entropy": 5.950505685806275,
      "epoch": 3.0148357470858356,
      "grad_norm": 1.125,
      "learning_rate": 0.0004099376075613395,
      "loss": 5.2384,
      "mean_token_accuracy": 0.21180692911148072,
      "num_tokens": 63648461.0,
      "step": 34140
    },
    {
      "entropy": 5.969177103042602,
      "epoch": 3.0152772871776756,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004099124064877504,
      "loss": 5.2405,
      "mean_token_accuracy": 0.21531456112861633,
      "num_tokens": 63657287.0,
      "step": 34145
    },
    {
      "entropy": 6.035009145736694,
      "epoch": 3.015718827269516,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040988720277130387,
      "loss": 5.2761,
      "mean_token_accuracy": 0.2170203760266304,
      "num_tokens": 63667263.0,
      "step": 34150
    },
    {
      "entropy": 5.909566020965576,
      "epoch": 3.0161603673613566,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004098619964124939,
      "loss": 5.2433,
      "mean_token_accuracy": 0.21483051031827927,
      "num_tokens": 63677470.0,
      "step": 34155
    },
    {
      "entropy": 5.981174612045288,
      "epoch": 3.0166019074531967,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004098367874118141,
      "loss": 5.2551,
      "mean_token_accuracy": 0.21248051524162292,
      "num_tokens": 63685897.0,
      "step": 34160
    },
    {
      "entropy": 5.982457065582276,
      "epoch": 3.017043447545037,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004098115757697582,
      "loss": 5.2333,
      "mean_token_accuracy": 0.21918827295303345,
      "num_tokens": 63694740.0,
      "step": 34165
    },
    {
      "entropy": 5.838617134094238,
      "epoch": 3.0174849876368772,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004097863614868203,
      "loss": 5.093,
      "mean_token_accuracy": 0.2311730057001114,
      "num_tokens": 63703551.0,
      "step": 34170
    },
    {
      "entropy": 5.969901180267334,
      "epoch": 3.0179265277287177,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040976114456349423,
      "loss": 5.239,
      "mean_token_accuracy": 0.2184781551361084,
      "num_tokens": 63712956.0,
      "step": 34175
    },
    {
      "entropy": 5.97595911026001,
      "epoch": 3.018368067820558,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040973592500027396,
      "loss": 5.3349,
      "mean_token_accuracy": 0.2028307870030403,
      "num_tokens": 63722942.0,
      "step": 34180
    },
    {
      "entropy": 5.933911943435669,
      "epoch": 3.0188096079123983,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004097107027976536,
      "loss": 5.2181,
      "mean_token_accuracy": 0.22548166811466216,
      "num_tokens": 63732261.0,
      "step": 34185
    },
    {
      "entropy": 6.031429052352905,
      "epoch": 3.0192511480042388,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040968547795612717,
      "loss": 5.3028,
      "mean_token_accuracy": 0.21193269789218902,
      "num_tokens": 63741220.0,
      "step": 34190
    },
    {
      "entropy": 6.111148977279663,
      "epoch": 3.0196926880960793,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040966025047618873,
      "loss": 5.3493,
      "mean_token_accuracy": 0.2129879578948021,
      "num_tokens": 63750595.0,
      "step": 34195
    },
    {
      "entropy": 5.940471172332764,
      "epoch": 3.0201342281879193,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004096350203583327,
      "loss": 5.2666,
      "mean_token_accuracy": 0.21332412362098693,
      "num_tokens": 63760409.0,
      "step": 34200
    },
    {
      "entropy": 5.919585943222046,
      "epoch": 3.02057576827976,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040960978760305305,
      "loss": 5.2091,
      "mean_token_accuracy": 0.22509117424488068,
      "num_tokens": 63769035.0,
      "step": 34205
    },
    {
      "entropy": 5.995185327529907,
      "epoch": 3.0210173083716003,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004095845522108443,
      "loss": 5.2834,
      "mean_token_accuracy": 0.21177609413862228,
      "num_tokens": 63778917.0,
      "step": 34210
    },
    {
      "entropy": 5.926572799682617,
      "epoch": 3.0214588484634404,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040955931418220075,
      "loss": 5.1652,
      "mean_token_accuracy": 0.22267415821552278,
      "num_tokens": 63787909.0,
      "step": 34215
    },
    {
      "entropy": 5.93312292098999,
      "epoch": 3.021900388555281,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040953407351761677,
      "loss": 5.1497,
      "mean_token_accuracy": 0.2238768681883812,
      "num_tokens": 63796757.0,
      "step": 34220
    },
    {
      "entropy": 5.951229858398437,
      "epoch": 3.0223419286471214,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040950883021758673,
      "loss": 5.2045,
      "mean_token_accuracy": 0.21904694586992263,
      "num_tokens": 63805403.0,
      "step": 34225
    },
    {
      "entropy": 5.937988042831421,
      "epoch": 3.0227834687389614,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040948358428260525,
      "loss": 5.3052,
      "mean_token_accuracy": 0.21112036108970642,
      "num_tokens": 63814966.0,
      "step": 34230
    },
    {
      "entropy": 6.0439372062683105,
      "epoch": 3.023225008830802,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004094583357131668,
      "loss": 5.2723,
      "mean_token_accuracy": 0.21672303527593612,
      "num_tokens": 63824451.0,
      "step": 34235
    },
    {
      "entropy": 5.9255513668060305,
      "epoch": 3.023666548922642,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000409433084509766,
      "loss": 5.1921,
      "mean_token_accuracy": 0.22517396807670592,
      "num_tokens": 63833690.0,
      "step": 34240
    },
    {
      "entropy": 5.891075134277344,
      "epoch": 3.0241080890144825,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004094078306728977,
      "loss": 5.1449,
      "mean_token_accuracy": 0.22333415746688842,
      "num_tokens": 63842729.0,
      "step": 34245
    },
    {
      "entropy": 5.968945693969727,
      "epoch": 3.024549629106323,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004093825742030563,
      "loss": 5.2327,
      "mean_token_accuracy": 0.21951612383127211,
      "num_tokens": 63852173.0,
      "step": 34250
    },
    {
      "entropy": 5.913188552856445,
      "epoch": 3.024991169198163,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040935731510073663,
      "loss": 5.2394,
      "mean_token_accuracy": 0.2187245547771454,
      "num_tokens": 63862363.0,
      "step": 34255
    },
    {
      "entropy": 5.933097267150879,
      "epoch": 3.0254327092900035,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040933205336643366,
      "loss": 5.3071,
      "mean_token_accuracy": 0.21326765716075896,
      "num_tokens": 63871840.0,
      "step": 34260
    },
    {
      "entropy": 6.033309984207153,
      "epoch": 3.025874249381844,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040930678900064223,
      "loss": 5.3318,
      "mean_token_accuracy": 0.2117711365222931,
      "num_tokens": 63881111.0,
      "step": 34265
    },
    {
      "entropy": 5.9264411449432375,
      "epoch": 3.026315789473684,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000409281522003857,
      "loss": 5.1911,
      "mean_token_accuracy": 0.22492024749517442,
      "num_tokens": 63890218.0,
      "step": 34270
    },
    {
      "entropy": 5.949359512329101,
      "epoch": 3.0267573295655246,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040925625237657335,
      "loss": 5.2231,
      "mean_token_accuracy": 0.22081731110811234,
      "num_tokens": 63899435.0,
      "step": 34275
    },
    {
      "entropy": 5.879389667510987,
      "epoch": 3.027198869657365,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004092309801192859,
      "loss": 5.1803,
      "mean_token_accuracy": 0.2202072709798813,
      "num_tokens": 63908981.0,
      "step": 34280
    },
    {
      "entropy": 5.948145055770874,
      "epoch": 3.027640409749205,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040920570523249,
      "loss": 5.2509,
      "mean_token_accuracy": 0.21969990879297258,
      "num_tokens": 63918436.0,
      "step": 34285
    },
    {
      "entropy": 5.990373802185059,
      "epoch": 3.0280819498410456,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040918042771668063,
      "loss": 5.2079,
      "mean_token_accuracy": 0.225028957426548,
      "num_tokens": 63927730.0,
      "step": 34290
    },
    {
      "entropy": 5.913103485107422,
      "epoch": 3.028523489932886,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040915514757235296,
      "loss": 5.2696,
      "mean_token_accuracy": 0.21096449196338654,
      "num_tokens": 63937231.0,
      "step": 34295
    },
    {
      "entropy": 5.849359607696533,
      "epoch": 3.028965030024726,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004091298648000023,
      "loss": 5.1707,
      "mean_token_accuracy": 0.2255462020635605,
      "num_tokens": 63946194.0,
      "step": 34300
    },
    {
      "entropy": 5.990848112106323,
      "epoch": 3.0294065701165667,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040910457940012383,
      "loss": 5.3257,
      "mean_token_accuracy": 0.20817954689264298,
      "num_tokens": 63956363.0,
      "step": 34305
    },
    {
      "entropy": 6.006012344360352,
      "epoch": 3.0298481102084067,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004090792913732129,
      "loss": 5.2682,
      "mean_token_accuracy": 0.21870879977941513,
      "num_tokens": 63965192.0,
      "step": 34310
    },
    {
      "entropy": 5.986861371994019,
      "epoch": 3.030289650300247,
      "grad_norm": 1.328125,
      "learning_rate": 0.00040905400071976496,
      "loss": 5.2956,
      "mean_token_accuracy": 0.20901987701654434,
      "num_tokens": 63974102.0,
      "step": 34315
    },
    {
      "entropy": 5.931967544555664,
      "epoch": 3.0307311903920877,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040902870744027533,
      "loss": 5.2624,
      "mean_token_accuracy": 0.2132463812828064,
      "num_tokens": 63982981.0,
      "step": 34320
    },
    {
      "entropy": 5.996710681915284,
      "epoch": 3.0311727304839278,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040900341153523955,
      "loss": 5.3831,
      "mean_token_accuracy": 0.20630093812942504,
      "num_tokens": 63991824.0,
      "step": 34325
    },
    {
      "entropy": 6.003992652893066,
      "epoch": 3.0316142705757683,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040897811300515325,
      "loss": 5.3291,
      "mean_token_accuracy": 0.21952647417783738,
      "num_tokens": 64001753.0,
      "step": 34330
    },
    {
      "entropy": 5.934638547897339,
      "epoch": 3.0320558106676088,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040895281185051186,
      "loss": 5.2612,
      "mean_token_accuracy": 0.21682929247617722,
      "num_tokens": 64011399.0,
      "step": 34335
    },
    {
      "entropy": 5.911697244644165,
      "epoch": 3.032497350759449,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000408927508071811,
      "loss": 5.1746,
      "mean_token_accuracy": 0.2306449979543686,
      "num_tokens": 64021190.0,
      "step": 34340
    },
    {
      "entropy": 6.00191912651062,
      "epoch": 3.0329388908512893,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004089022016695465,
      "loss": 5.2779,
      "mean_token_accuracy": 0.2101811170578003,
      "num_tokens": 64030352.0,
      "step": 34345
    },
    {
      "entropy": 5.995956754684448,
      "epoch": 3.03338043094313,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000408876892644214,
      "loss": 5.2416,
      "mean_token_accuracy": 0.20811799466609954,
      "num_tokens": 64039330.0,
      "step": 34350
    },
    {
      "entropy": 5.945193481445313,
      "epoch": 3.03382197103497,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004088515809963094,
      "loss": 5.2454,
      "mean_token_accuracy": 0.21634680181741714,
      "num_tokens": 64048272.0,
      "step": 34355
    },
    {
      "entropy": 5.9219019412994385,
      "epoch": 3.0342635111268104,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004088262667263283,
      "loss": 5.2331,
      "mean_token_accuracy": 0.21878008544445038,
      "num_tokens": 64056376.0,
      "step": 34360
    },
    {
      "entropy": 6.001427698135376,
      "epoch": 3.034705051218651,
      "grad_norm": 1.125,
      "learning_rate": 0.0004088009498347668,
      "loss": 5.3061,
      "mean_token_accuracy": 0.21336586326360701,
      "num_tokens": 64066870.0,
      "step": 34365
    },
    {
      "entropy": 6.027231597900391,
      "epoch": 3.035146591310491,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004087756303221208,
      "loss": 5.2891,
      "mean_token_accuracy": 0.20900773555040358,
      "num_tokens": 64077055.0,
      "step": 34370
    },
    {
      "entropy": 5.949713277816772,
      "epoch": 3.0355881314023314,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004087503081888863,
      "loss": 5.2284,
      "mean_token_accuracy": 0.21694404929876326,
      "num_tokens": 64086954.0,
      "step": 34375
    },
    {
      "entropy": 5.8808691024780275,
      "epoch": 3.0360296714941715,
      "grad_norm": 1.25,
      "learning_rate": 0.0004087249834355593,
      "loss": 5.184,
      "mean_token_accuracy": 0.2226566568017006,
      "num_tokens": 64096341.0,
      "step": 34380
    },
    {
      "entropy": 5.97212815284729,
      "epoch": 3.036471211586012,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004086996560626359,
      "loss": 5.276,
      "mean_token_accuracy": 0.2153841719031334,
      "num_tokens": 64106354.0,
      "step": 34385
    },
    {
      "entropy": 6.025452947616577,
      "epoch": 3.0369127516778525,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040867432607061234,
      "loss": 5.319,
      "mean_token_accuracy": 0.2128889724612236,
      "num_tokens": 64115669.0,
      "step": 34390
    },
    {
      "entropy": 5.915193128585815,
      "epoch": 3.0373542917696925,
      "grad_norm": 1.25,
      "learning_rate": 0.00040864899345998475,
      "loss": 5.1393,
      "mean_token_accuracy": 0.22553676664829253,
      "num_tokens": 64124383.0,
      "step": 34395
    },
    {
      "entropy": 5.901161003112793,
      "epoch": 3.037795831861533,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004086236582312494,
      "loss": 5.1643,
      "mean_token_accuracy": 0.2234511211514473,
      "num_tokens": 64133154.0,
      "step": 34400
    },
    {
      "entropy": 5.968749952316284,
      "epoch": 3.0382373719533735,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004085983203849025,
      "loss": 5.3152,
      "mean_token_accuracy": 0.2163180023431778,
      "num_tokens": 64142894.0,
      "step": 34405
    },
    {
      "entropy": 5.980237197875977,
      "epoch": 3.0386789120452136,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004085729799214406,
      "loss": 5.2871,
      "mean_token_accuracy": 0.21642522811889647,
      "num_tokens": 64152123.0,
      "step": 34410
    },
    {
      "entropy": 5.967230701446534,
      "epoch": 3.039120452137054,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004085476368413599,
      "loss": 5.315,
      "mean_token_accuracy": 0.2108997017145157,
      "num_tokens": 64160943.0,
      "step": 34415
    },
    {
      "entropy": 5.94762544631958,
      "epoch": 3.0395619922288946,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040852229114515706,
      "loss": 5.1972,
      "mean_token_accuracy": 0.21803293824195863,
      "num_tokens": 64170629.0,
      "step": 34420
    },
    {
      "entropy": 5.984078884124756,
      "epoch": 3.0400035323207346,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004084969428333284,
      "loss": 5.3487,
      "mean_token_accuracy": 0.20963981598615647,
      "num_tokens": 64180398.0,
      "step": 34425
    },
    {
      "entropy": 5.9646899700164795,
      "epoch": 3.040445072412575,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040847159190637053,
      "loss": 5.2793,
      "mean_token_accuracy": 0.2198530614376068,
      "num_tokens": 64190604.0,
      "step": 34430
    },
    {
      "entropy": 5.9121702194213865,
      "epoch": 3.0408866125044156,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004084462383647801,
      "loss": 5.2404,
      "mean_token_accuracy": 0.21956140398979188,
      "num_tokens": 64199796.0,
      "step": 34435
    },
    {
      "entropy": 5.9377399444580075,
      "epoch": 3.0413281525962557,
      "grad_norm": 1.3125,
      "learning_rate": 0.00040842088220905374,
      "loss": 5.203,
      "mean_token_accuracy": 0.21755778938531875,
      "num_tokens": 64208032.0,
      "step": 34440
    },
    {
      "entropy": 5.977966833114624,
      "epoch": 3.041769692688096,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004083955234396882,
      "loss": 5.2521,
      "mean_token_accuracy": 0.2204464465379715,
      "num_tokens": 64217091.0,
      "step": 34445
    },
    {
      "entropy": 5.9232676982879635,
      "epoch": 3.042211232779936,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004083701620571802,
      "loss": 5.2247,
      "mean_token_accuracy": 0.21766875684261322,
      "num_tokens": 64226904.0,
      "step": 34450
    },
    {
      "entropy": 5.939728736877441,
      "epoch": 3.0426527728717767,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004083447980620266,
      "loss": 5.2578,
      "mean_token_accuracy": 0.22116239815950395,
      "num_tokens": 64236730.0,
      "step": 34455
    },
    {
      "entropy": 5.80568380355835,
      "epoch": 3.043094312963617,
      "grad_norm": 1.25,
      "learning_rate": 0.0004083194314547243,
      "loss": 5.0559,
      "mean_token_accuracy": 0.22889765501022338,
      "num_tokens": 64245679.0,
      "step": 34460
    },
    {
      "entropy": 5.941674661636353,
      "epoch": 3.0435358530554573,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004082940622357701,
      "loss": 5.2492,
      "mean_token_accuracy": 0.21693167090415955,
      "num_tokens": 64254966.0,
      "step": 34465
    },
    {
      "entropy": 5.931897783279419,
      "epoch": 3.0439773931472978,
      "grad_norm": 1.234375,
      "learning_rate": 0.000408268690405661,
      "loss": 5.2799,
      "mean_token_accuracy": 0.2159496784210205,
      "num_tokens": 64263477.0,
      "step": 34470
    },
    {
      "entropy": 5.9883520126342775,
      "epoch": 3.0444189332391383,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040824331596489403,
      "loss": 5.2634,
      "mean_token_accuracy": 0.21713244467973708,
      "num_tokens": 64272577.0,
      "step": 34475
    },
    {
      "entropy": 5.91998519897461,
      "epoch": 3.0448604733309783,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004082179389139663,
      "loss": 5.1216,
      "mean_token_accuracy": 0.23039422184228897,
      "num_tokens": 64281453.0,
      "step": 34480
    },
    {
      "entropy": 5.92945499420166,
      "epoch": 3.045302013422819,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000408192559253375,
      "loss": 5.1634,
      "mean_token_accuracy": 0.22226074784994126,
      "num_tokens": 64290093.0,
      "step": 34485
    },
    {
      "entropy": 5.918104076385498,
      "epoch": 3.0457435535146593,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004081671769836172,
      "loss": 5.299,
      "mean_token_accuracy": 0.21564594954252242,
      "num_tokens": 64299360.0,
      "step": 34490
    },
    {
      "entropy": 5.992335844039917,
      "epoch": 3.0461850936064994,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004081417921051901,
      "loss": 5.3042,
      "mean_token_accuracy": 0.2122478261590004,
      "num_tokens": 64308391.0,
      "step": 34495
    },
    {
      "entropy": 5.996554613113403,
      "epoch": 3.04662663369834,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040811640461859095,
      "loss": 5.2789,
      "mean_token_accuracy": 0.21082865297794343,
      "num_tokens": 64318306.0,
      "step": 34500
    },
    {
      "entropy": 5.898817873001098,
      "epoch": 3.0470681737901804,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004080910145243172,
      "loss": 5.1547,
      "mean_token_accuracy": 0.22540477365255357,
      "num_tokens": 64327014.0,
      "step": 34505
    },
    {
      "entropy": 5.969620943069458,
      "epoch": 3.0475097138820204,
      "grad_norm": 1.125,
      "learning_rate": 0.00040806562182286614,
      "loss": 5.3058,
      "mean_token_accuracy": 0.2121325433254242,
      "num_tokens": 64337757.0,
      "step": 34510
    },
    {
      "entropy": 5.9647115707397464,
      "epoch": 3.047951253973861,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004080402265147353,
      "loss": 5.2343,
      "mean_token_accuracy": 0.22078128159046173,
      "num_tokens": 64346661.0,
      "step": 34515
    },
    {
      "entropy": 5.998860168457031,
      "epoch": 3.048392794065701,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040801482860042205,
      "loss": 5.3072,
      "mean_token_accuracy": 0.2088911786675453,
      "num_tokens": 64356198.0,
      "step": 34520
    },
    {
      "entropy": 5.9976073741912845,
      "epoch": 3.0488343341575415,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040798942808042396,
      "loss": 5.2908,
      "mean_token_accuracy": 0.21309721767902373,
      "num_tokens": 64365646.0,
      "step": 34525
    },
    {
      "entropy": 6.012217903137207,
      "epoch": 3.049275874249382,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040796402495523853,
      "loss": 5.3068,
      "mean_token_accuracy": 0.2063884288072586,
      "num_tokens": 64373864.0,
      "step": 34530
    },
    {
      "entropy": 5.9652704238891605,
      "epoch": 3.049717414341222,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004079386192253636,
      "loss": 5.3184,
      "mean_token_accuracy": 0.2084166169166565,
      "num_tokens": 64383153.0,
      "step": 34535
    },
    {
      "entropy": 5.958429336547852,
      "epoch": 3.0501589544330625,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004079132108912967,
      "loss": 5.1944,
      "mean_token_accuracy": 0.22619785815477372,
      "num_tokens": 64392012.0,
      "step": 34540
    },
    {
      "entropy": 5.954658460617066,
      "epoch": 3.050600494524903,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004078877999535355,
      "loss": 5.2797,
      "mean_token_accuracy": 0.21397661566734313,
      "num_tokens": 64402353.0,
      "step": 34545
    },
    {
      "entropy": 5.978918790817261,
      "epoch": 3.051042034616743,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004078623864125779,
      "loss": 5.3072,
      "mean_token_accuracy": 0.21435216665267945,
      "num_tokens": 64411967.0,
      "step": 34550
    },
    {
      "entropy": 5.964757061004638,
      "epoch": 3.0514835747085836,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040783697026892176,
      "loss": 5.2675,
      "mean_token_accuracy": 0.21239093393087388,
      "num_tokens": 64420683.0,
      "step": 34555
    },
    {
      "entropy": 6.01400785446167,
      "epoch": 3.051925114800424,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040781155152306485,
      "loss": 5.2297,
      "mean_token_accuracy": 0.219720658659935,
      "num_tokens": 64429434.0,
      "step": 34560
    },
    {
      "entropy": 5.858343124389648,
      "epoch": 3.052366654892264,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004077861301755053,
      "loss": 5.0668,
      "mean_token_accuracy": 0.2409309670329094,
      "num_tokens": 64437907.0,
      "step": 34565
    },
    {
      "entropy": 5.906284284591675,
      "epoch": 3.0528081949841046,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040776070622674095,
      "loss": 5.2855,
      "mean_token_accuracy": 0.2099296972155571,
      "num_tokens": 64446808.0,
      "step": 34570
    },
    {
      "entropy": 5.9049125671386715,
      "epoch": 3.053249735075945,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004077352796772698,
      "loss": 5.2385,
      "mean_token_accuracy": 0.21133791655302048,
      "num_tokens": 64455493.0,
      "step": 34575
    },
    {
      "entropy": 5.970285177230835,
      "epoch": 3.053691275167785,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040770985052759006,
      "loss": 5.2794,
      "mean_token_accuracy": 0.21169459968805313,
      "num_tokens": 64465077.0,
      "step": 34580
    },
    {
      "entropy": 6.065244054794311,
      "epoch": 3.0541328152596257,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004076844187781998,
      "loss": 5.3074,
      "mean_token_accuracy": 0.2107689544558525,
      "num_tokens": 64474624.0,
      "step": 34585
    },
    {
      "entropy": 5.977415466308594,
      "epoch": 3.0545743553514657,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004076589844295972,
      "loss": 5.241,
      "mean_token_accuracy": 0.20941822081804276,
      "num_tokens": 64483804.0,
      "step": 34590
    },
    {
      "entropy": 5.9221851348876955,
      "epoch": 3.055015895443306,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004076335474822806,
      "loss": 5.2814,
      "mean_token_accuracy": 0.21961967647075653,
      "num_tokens": 64493131.0,
      "step": 34595
    },
    {
      "entropy": 5.850303602218628,
      "epoch": 3.0554574355351467,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004076081079367483,
      "loss": 5.1533,
      "mean_token_accuracy": 0.23221734315156936,
      "num_tokens": 64502410.0,
      "step": 34600
    },
    {
      "entropy": 5.901896286010742,
      "epoch": 3.0558989756269868,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004075826657934985,
      "loss": 5.207,
      "mean_token_accuracy": 0.21530976444482802,
      "num_tokens": 64511042.0,
      "step": 34605
    },
    {
      "entropy": 6.037807512283325,
      "epoch": 3.0563405157188273,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040755722105302963,
      "loss": 5.3533,
      "mean_token_accuracy": 0.20797429531812667,
      "num_tokens": 64520168.0,
      "step": 34610
    },
    {
      "entropy": 5.98173885345459,
      "epoch": 3.0567820558106678,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004075317737158403,
      "loss": 5.2225,
      "mean_token_accuracy": 0.22368698120117186,
      "num_tokens": 64529650.0,
      "step": 34615
    },
    {
      "entropy": 5.903033876419068,
      "epoch": 3.057223595902508,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040750632378242884,
      "loss": 5.2046,
      "mean_token_accuracy": 0.22155518531799318,
      "num_tokens": 64539002.0,
      "step": 34620
    },
    {
      "entropy": 5.9084679126739506,
      "epoch": 3.0576651359943483,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040748087125329384,
      "loss": 5.2678,
      "mean_token_accuracy": 0.21785784959793092,
      "num_tokens": 64548522.0,
      "step": 34625
    },
    {
      "entropy": 5.999771165847778,
      "epoch": 3.058106676086189,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004074554161289339,
      "loss": 5.2646,
      "mean_token_accuracy": 0.2169452577829361,
      "num_tokens": 64558357.0,
      "step": 34630
    },
    {
      "entropy": 5.981771850585938,
      "epoch": 3.058548216178029,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040742995840984785,
      "loss": 5.1891,
      "mean_token_accuracy": 0.21935179829597473,
      "num_tokens": 64566922.0,
      "step": 34635
    },
    {
      "entropy": 5.9267106533050535,
      "epoch": 3.0589897562698694,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040740449809653406,
      "loss": 5.231,
      "mean_token_accuracy": 0.2201274126768112,
      "num_tokens": 64576330.0,
      "step": 34640
    },
    {
      "entropy": 5.942563104629516,
      "epoch": 3.05943129636171,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004073790351894915,
      "loss": 5.2424,
      "mean_token_accuracy": 0.21959538161754608,
      "num_tokens": 64586562.0,
      "step": 34645
    },
    {
      "entropy": 5.940783166885376,
      "epoch": 3.05987283645355,
      "grad_norm": 1.09375,
      "learning_rate": 0.000407353569689219,
      "loss": 5.2218,
      "mean_token_accuracy": 0.222295081615448,
      "num_tokens": 64595582.0,
      "step": 34650
    },
    {
      "entropy": 5.985535430908203,
      "epoch": 3.0603143765453904,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040732810159621526,
      "loss": 5.342,
      "mean_token_accuracy": 0.20134981572628022,
      "num_tokens": 64606807.0,
      "step": 34655
    },
    {
      "entropy": 6.02361912727356,
      "epoch": 3.0607559166372305,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004073026309109793,
      "loss": 5.3073,
      "mean_token_accuracy": 0.20744801610708236,
      "num_tokens": 64617451.0,
      "step": 34660
    },
    {
      "entropy": 6.007956075668335,
      "epoch": 3.061197456729071,
      "grad_norm": 1.125,
      "learning_rate": 0.00040727715763401,
      "loss": 5.3292,
      "mean_token_accuracy": 0.21119798868894576,
      "num_tokens": 64627721.0,
      "step": 34665
    },
    {
      "entropy": 6.002080965042114,
      "epoch": 3.0616389968209115,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004072516817658065,
      "loss": 5.2795,
      "mean_token_accuracy": 0.21419495046138765,
      "num_tokens": 64638363.0,
      "step": 34670
    },
    {
      "entropy": 5.943088436126709,
      "epoch": 3.0620805369127515,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040722620330686766,
      "loss": 5.2025,
      "mean_token_accuracy": 0.21840236485004424,
      "num_tokens": 64647539.0,
      "step": 34675
    },
    {
      "entropy": 5.873691606521606,
      "epoch": 3.062522077004592,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004072007222576928,
      "loss": 5.0997,
      "mean_token_accuracy": 0.22575951516628265,
      "num_tokens": 64657034.0,
      "step": 34680
    },
    {
      "entropy": 5.956696844100952,
      "epoch": 3.0629636170964325,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004071752386187809,
      "loss": 5.3058,
      "mean_token_accuracy": 0.21452290266752244,
      "num_tokens": 64666676.0,
      "step": 34685
    },
    {
      "entropy": 5.963718605041504,
      "epoch": 3.0634051571882726,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040714975239063134,
      "loss": 5.1885,
      "mean_token_accuracy": 0.22212553024291992,
      "num_tokens": 64677207.0,
      "step": 34690
    },
    {
      "entropy": 5.955222177505493,
      "epoch": 3.063846697280113,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004071242635737432,
      "loss": 5.2387,
      "mean_token_accuracy": 0.21355517953634262,
      "num_tokens": 64686523.0,
      "step": 34695
    },
    {
      "entropy": 5.919836187362671,
      "epoch": 3.0642882373719535,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004070987721686159,
      "loss": 5.3348,
      "mean_token_accuracy": 0.20658704340457917,
      "num_tokens": 64696646.0,
      "step": 34700
    },
    {
      "entropy": 5.887861633300782,
      "epoch": 3.0647297774637936,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004070732781757488,
      "loss": 5.2266,
      "mean_token_accuracy": 0.22574645429849624,
      "num_tokens": 64705643.0,
      "step": 34705
    },
    {
      "entropy": 6.016112041473389,
      "epoch": 3.065171317555634,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040704778159564124,
      "loss": 5.3601,
      "mean_token_accuracy": 0.20549359768629075,
      "num_tokens": 64715728.0,
      "step": 34710
    },
    {
      "entropy": 6.028974151611328,
      "epoch": 3.0656128576474746,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040702228242879286,
      "loss": 5.2769,
      "mean_token_accuracy": 0.21301494240760804,
      "num_tokens": 64725233.0,
      "step": 34715
    },
    {
      "entropy": 5.998831033706665,
      "epoch": 3.0660543977393147,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004069967806757029,
      "loss": 5.2775,
      "mean_token_accuracy": 0.21853265762329102,
      "num_tokens": 64735350.0,
      "step": 34720
    },
    {
      "entropy": 5.968993043899536,
      "epoch": 3.066495937831155,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004069712763368712,
      "loss": 5.2412,
      "mean_token_accuracy": 0.217703378200531,
      "num_tokens": 64744435.0,
      "step": 34725
    },
    {
      "entropy": 6.06793475151062,
      "epoch": 3.066937477922995,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004069457694127972,
      "loss": 5.2901,
      "mean_token_accuracy": 0.2126305341720581,
      "num_tokens": 64753112.0,
      "step": 34730
    },
    {
      "entropy": 6.003526878356934,
      "epoch": 3.0673790180148357,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004069202599039806,
      "loss": 5.3198,
      "mean_token_accuracy": 0.20982678681612016,
      "num_tokens": 64763191.0,
      "step": 34735
    },
    {
      "entropy": 5.918735027313232,
      "epoch": 3.067820558106676,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004068947478109212,
      "loss": 5.2212,
      "mean_token_accuracy": 0.22309835404157638,
      "num_tokens": 64772159.0,
      "step": 34740
    },
    {
      "entropy": 5.89440279006958,
      "epoch": 3.0682620981985163,
      "grad_norm": 1.25,
      "learning_rate": 0.00040686923313411864,
      "loss": 5.1878,
      "mean_token_accuracy": 0.22066576778888702,
      "num_tokens": 64781118.0,
      "step": 34745
    },
    {
      "entropy": 5.950133419036865,
      "epoch": 3.0687036382903567,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00040684371587407285,
      "loss": 5.1987,
      "mean_token_accuracy": 0.22130415439605713,
      "num_tokens": 64789866.0,
      "step": 34750
    },
    {
      "entropy": 5.971149873733521,
      "epoch": 3.0691451783821972,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040681819603128363,
      "loss": 5.2149,
      "mean_token_accuracy": 0.22262748926877976,
      "num_tokens": 64799511.0,
      "step": 34755
    },
    {
      "entropy": 5.940932989120483,
      "epoch": 3.0695867184740373,
      "grad_norm": 1.328125,
      "learning_rate": 0.00040679267360625086,
      "loss": 5.2156,
      "mean_token_accuracy": 0.2189225897192955,
      "num_tokens": 64808042.0,
      "step": 34760
    },
    {
      "entropy": 5.947498464584351,
      "epoch": 3.070028258565878,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004067671485994747,
      "loss": 5.2006,
      "mean_token_accuracy": 0.21812765151262284,
      "num_tokens": 64816866.0,
      "step": 34765
    },
    {
      "entropy": 5.893246841430664,
      "epoch": 3.0704697986577183,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004067416210114549,
      "loss": 5.1962,
      "mean_token_accuracy": 0.21930197924375533,
      "num_tokens": 64825923.0,
      "step": 34770
    },
    {
      "entropy": 6.013867664337158,
      "epoch": 3.0709113387495584,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040671609084269177,
      "loss": 5.1912,
      "mean_token_accuracy": 0.21633997112512587,
      "num_tokens": 64835029.0,
      "step": 34775
    },
    {
      "entropy": 6.007920455932617,
      "epoch": 3.071352878841399,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004066905580936853,
      "loss": 5.342,
      "mean_token_accuracy": 0.20872714072465898,
      "num_tokens": 64845209.0,
      "step": 34780
    },
    {
      "entropy": 5.9022692203521725,
      "epoch": 3.0717944189332393,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040666502276493567,
      "loss": 5.1794,
      "mean_token_accuracy": 0.2225068524479866,
      "num_tokens": 64854157.0,
      "step": 34785
    },
    {
      "entropy": 6.015286827087403,
      "epoch": 3.0722359590250794,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040663948485694315,
      "loss": 5.2789,
      "mean_token_accuracy": 0.21583382934331893,
      "num_tokens": 64865021.0,
      "step": 34790
    },
    {
      "entropy": 6.0305760383605955,
      "epoch": 3.07267749911692,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004066139443702079,
      "loss": 5.2676,
      "mean_token_accuracy": 0.21030874401330948,
      "num_tokens": 64873800.0,
      "step": 34795
    },
    {
      "entropy": 5.9839025974273685,
      "epoch": 3.07311903920876,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004065884013052305,
      "loss": 5.2376,
      "mean_token_accuracy": 0.21132558733224868,
      "num_tokens": 64883176.0,
      "step": 34800
    },
    {
      "entropy": 5.884747838973999,
      "epoch": 3.0735605793006004,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004065628556625111,
      "loss": 5.2298,
      "mean_token_accuracy": 0.22229802757501602,
      "num_tokens": 64892694.0,
      "step": 34805
    },
    {
      "entropy": 6.007223081588745,
      "epoch": 3.074002119392441,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004065373074425501,
      "loss": 5.3616,
      "mean_token_accuracy": 0.20674777030944824,
      "num_tokens": 64904182.0,
      "step": 34810
    },
    {
      "entropy": 5.9950135231018065,
      "epoch": 3.074443659484281,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004065117566458481,
      "loss": 5.3324,
      "mean_token_accuracy": 0.20557309985160827,
      "num_tokens": 64914120.0,
      "step": 34815
    },
    {
      "entropy": 5.99551157951355,
      "epoch": 3.0748851995761215,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040648620327290557,
      "loss": 5.2406,
      "mean_token_accuracy": 0.21155576556921005,
      "num_tokens": 64922781.0,
      "step": 34820
    },
    {
      "entropy": 5.932155561447144,
      "epoch": 3.075326739667962,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004064606473242232,
      "loss": 5.2236,
      "mean_token_accuracy": 0.22552135437726975,
      "num_tokens": 64932076.0,
      "step": 34825
    },
    {
      "entropy": 5.8804388523101805,
      "epoch": 3.075768279759802,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004064350888003014,
      "loss": 5.2197,
      "mean_token_accuracy": 0.2190702497959137,
      "num_tokens": 64941848.0,
      "step": 34830
    },
    {
      "entropy": 5.919737958908081,
      "epoch": 3.0762098198516425,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040640952770164095,
      "loss": 5.2138,
      "mean_token_accuracy": 0.2190563201904297,
      "num_tokens": 64951045.0,
      "step": 34835
    },
    {
      "entropy": 6.041483211517334,
      "epoch": 3.076651359943483,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040638396402874265,
      "loss": 5.277,
      "mean_token_accuracy": 0.2179775521159172,
      "num_tokens": 64959919.0,
      "step": 34840
    },
    {
      "entropy": 5.889916944503784,
      "epoch": 3.077092900035323,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004063583977821072,
      "loss": 5.1599,
      "mean_token_accuracy": 0.23563415557146072,
      "num_tokens": 64969358.0,
      "step": 34845
    },
    {
      "entropy": 5.941536331176758,
      "epoch": 3.0775344401271636,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040633282896223547,
      "loss": 5.2409,
      "mean_token_accuracy": 0.20615866482257844,
      "num_tokens": 64977441.0,
      "step": 34850
    },
    {
      "entropy": 6.002431488037109,
      "epoch": 3.077975980219004,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040630725756962823,
      "loss": 5.3069,
      "mean_token_accuracy": 0.20509794503450393,
      "num_tokens": 64986762.0,
      "step": 34855
    },
    {
      "entropy": 5.895196151733399,
      "epoch": 3.078417520310844,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004062816836047866,
      "loss": 5.2334,
      "mean_token_accuracy": 0.22274239212274552,
      "num_tokens": 64994958.0,
      "step": 34860
    },
    {
      "entropy": 6.015530252456665,
      "epoch": 3.0788590604026846,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040625610706821136,
      "loss": 5.3465,
      "mean_token_accuracy": 0.2029055118560791,
      "num_tokens": 65003387.0,
      "step": 34865
    },
    {
      "entropy": 5.9654356956481935,
      "epoch": 3.0793006004945247,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040623052796040374,
      "loss": 5.3135,
      "mean_token_accuracy": 0.2103275552392006,
      "num_tokens": 65013182.0,
      "step": 34870
    },
    {
      "entropy": 5.978297472000122,
      "epoch": 3.079742140586365,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004062049462818646,
      "loss": 5.2483,
      "mean_token_accuracy": 0.2194433867931366,
      "num_tokens": 65023073.0,
      "step": 34875
    },
    {
      "entropy": 5.946377801895141,
      "epoch": 3.0801836806782057,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004061793620330953,
      "loss": 5.2633,
      "mean_token_accuracy": 0.2114816576242447,
      "num_tokens": 65032488.0,
      "step": 34880
    },
    {
      "entropy": 5.99887261390686,
      "epoch": 3.0806252207700457,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004061537752145968,
      "loss": 5.2587,
      "mean_token_accuracy": 0.2156369611620903,
      "num_tokens": 65041689.0,
      "step": 34885
    },
    {
      "entropy": 5.908077001571655,
      "epoch": 3.0810667608618862,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040612818582687045,
      "loss": 5.1768,
      "mean_token_accuracy": 0.2234656259417534,
      "num_tokens": 65050028.0,
      "step": 34890
    },
    {
      "entropy": 5.920657539367676,
      "epoch": 3.0815083009537267,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040610259387041756,
      "loss": 5.309,
      "mean_token_accuracy": 0.2137081056833267,
      "num_tokens": 65060192.0,
      "step": 34895
    },
    {
      "entropy": 5.973893404006958,
      "epoch": 3.081949841045567,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040607699934573936,
      "loss": 5.2641,
      "mean_token_accuracy": 0.21460136771202087,
      "num_tokens": 65069648.0,
      "step": 34900
    },
    {
      "entropy": 5.921839809417724,
      "epoch": 3.0823913811374073,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040605140225333724,
      "loss": 5.1427,
      "mean_token_accuracy": 0.228104892373085,
      "num_tokens": 65077950.0,
      "step": 34905
    },
    {
      "entropy": 6.006782054901123,
      "epoch": 3.082832921229248,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040602580259371286,
      "loss": 5.3634,
      "mean_token_accuracy": 0.20800017416477204,
      "num_tokens": 65087361.0,
      "step": 34910
    },
    {
      "entropy": 5.938420677185059,
      "epoch": 3.083274461321088,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004060002003673674,
      "loss": 5.2089,
      "mean_token_accuracy": 0.21650345921516417,
      "num_tokens": 65097179.0,
      "step": 34915
    },
    {
      "entropy": 6.0303075313568115,
      "epoch": 3.0837160014129283,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040597459557480256,
      "loss": 5.278,
      "mean_token_accuracy": 0.21337927430868148,
      "num_tokens": 65106227.0,
      "step": 34920
    },
    {
      "entropy": 6.020399188995361,
      "epoch": 3.084157541504769,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004059489882165199,
      "loss": 5.2903,
      "mean_token_accuracy": 0.21189258694648744,
      "num_tokens": 65115703.0,
      "step": 34925
    },
    {
      "entropy": 6.013534593582153,
      "epoch": 3.084599081596609,
      "grad_norm": 1.28125,
      "learning_rate": 0.000405923378293021,
      "loss": 5.3203,
      "mean_token_accuracy": 0.20288024991750717,
      "num_tokens": 65125358.0,
      "step": 34930
    },
    {
      "entropy": 5.971079254150391,
      "epoch": 3.0850406216884494,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004058977658048076,
      "loss": 5.2899,
      "mean_token_accuracy": 0.21539922803640366,
      "num_tokens": 65134357.0,
      "step": 34935
    },
    {
      "entropy": 6.002318954467773,
      "epoch": 3.0854821617802894,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040587215075238133,
      "loss": 5.2513,
      "mean_token_accuracy": 0.21391943842172623,
      "num_tokens": 65143720.0,
      "step": 34940
    },
    {
      "entropy": 6.034972715377807,
      "epoch": 3.08592370187213,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004058465331362442,
      "loss": 5.3167,
      "mean_token_accuracy": 0.20776463598012923,
      "num_tokens": 65153896.0,
      "step": 34945
    },
    {
      "entropy": 5.987048673629761,
      "epoch": 3.0863652419639704,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040582091295689784,
      "loss": 5.2594,
      "mean_token_accuracy": 0.22706868052482604,
      "num_tokens": 65162595.0,
      "step": 34950
    },
    {
      "entropy": 6.049841690063476,
      "epoch": 3.0868067820558105,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004057952902148441,
      "loss": 5.4282,
      "mean_token_accuracy": 0.20377825647592546,
      "num_tokens": 65172186.0,
      "step": 34955
    },
    {
      "entropy": 5.929776668548584,
      "epoch": 3.087248322147651,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040576966491058514,
      "loss": 5.2429,
      "mean_token_accuracy": 0.21957921385765075,
      "num_tokens": 65181977.0,
      "step": 34960
    },
    {
      "entropy": 5.952157926559448,
      "epoch": 3.0876898622394915,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040574403704462283,
      "loss": 5.2169,
      "mean_token_accuracy": 0.22835930287837983,
      "num_tokens": 65190914.0,
      "step": 34965
    },
    {
      "entropy": 6.0708190441131595,
      "epoch": 3.0881314023313315,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040571840661745906,
      "loss": 5.3577,
      "mean_token_accuracy": 0.20792041569948197,
      "num_tokens": 65200716.0,
      "step": 34970
    },
    {
      "entropy": 5.980419111251831,
      "epoch": 3.088572942423172,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004056927736295962,
      "loss": 5.3128,
      "mean_token_accuracy": 0.21233796030282975,
      "num_tokens": 65209557.0,
      "step": 34975
    },
    {
      "entropy": 6.024065399169922,
      "epoch": 3.0890144825150125,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004056671380815361,
      "loss": 5.3107,
      "mean_token_accuracy": 0.20826441198587417,
      "num_tokens": 65220405.0,
      "step": 34980
    },
    {
      "entropy": 5.966082191467285,
      "epoch": 3.0894560226068526,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004056414999737811,
      "loss": 5.2565,
      "mean_token_accuracy": 0.21960753947496414,
      "num_tokens": 65229773.0,
      "step": 34985
    },
    {
      "entropy": 6.036549854278564,
      "epoch": 3.089897562698693,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004056158593068335,
      "loss": 5.2064,
      "mean_token_accuracy": 0.2209862768650055,
      "num_tokens": 65239235.0,
      "step": 34990
    },
    {
      "entropy": 5.963653898239135,
      "epoch": 3.0903391027905336,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040559021608119547,
      "loss": 5.2673,
      "mean_token_accuracy": 0.21439759135246278,
      "num_tokens": 65247918.0,
      "step": 34995
    },
    {
      "entropy": 5.922471380233764,
      "epoch": 3.0907806428823736,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040556457029736936,
      "loss": 5.301,
      "mean_token_accuracy": 0.20823077261447906,
      "num_tokens": 65257571.0,
      "step": 35000
    },
    {
      "entropy": 5.927800512313842,
      "epoch": 3.091222182974214,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004055389219558575,
      "loss": 5.1543,
      "mean_token_accuracy": 0.22476459592580794,
      "num_tokens": 65266579.0,
      "step": 35005
    },
    {
      "entropy": 6.0286789894104,
      "epoch": 3.091663723066054,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004055132710571626,
      "loss": 5.3245,
      "mean_token_accuracy": 0.20461165010929108,
      "num_tokens": 65275196.0,
      "step": 35010
    },
    {
      "entropy": 5.963628435134888,
      "epoch": 3.0921052631578947,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040548761760178677,
      "loss": 5.3075,
      "mean_token_accuracy": 0.21006533205509187,
      "num_tokens": 65285011.0,
      "step": 35015
    },
    {
      "entropy": 6.016395473480225,
      "epoch": 3.092546803249735,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004054619615902328,
      "loss": 5.3369,
      "mean_token_accuracy": 0.2072808936238289,
      "num_tokens": 65294032.0,
      "step": 35020
    },
    {
      "entropy": 5.969661903381348,
      "epoch": 3.0929883433415752,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004054363030230032,
      "loss": 5.2651,
      "mean_token_accuracy": 0.21246125400066376,
      "num_tokens": 65302478.0,
      "step": 35025
    },
    {
      "entropy": 5.95814323425293,
      "epoch": 3.0934298834334157,
      "grad_norm": 1.125,
      "learning_rate": 0.0004054106419006006,
      "loss": 5.2691,
      "mean_token_accuracy": 0.21430668085813523,
      "num_tokens": 65312078.0,
      "step": 35030
    },
    {
      "entropy": 5.937908411026001,
      "epoch": 3.0938714235252562,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040538497822352763,
      "loss": 5.2262,
      "mean_token_accuracy": 0.2226658508181572,
      "num_tokens": 65321125.0,
      "step": 35035
    },
    {
      "entropy": 5.988195705413818,
      "epoch": 3.0943129636170963,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004053593119922871,
      "loss": 5.3188,
      "mean_token_accuracy": 0.21325269490480422,
      "num_tokens": 65330168.0,
      "step": 35040
    },
    {
      "entropy": 6.027047538757325,
      "epoch": 3.094754503708937,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004053336432073818,
      "loss": 5.2147,
      "mean_token_accuracy": 0.21485222727060319,
      "num_tokens": 65338715.0,
      "step": 35045
    },
    {
      "entropy": 5.895285987854004,
      "epoch": 3.0951960438007773,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004053079718693146,
      "loss": 5.1549,
      "mean_token_accuracy": 0.22595864087343215,
      "num_tokens": 65349100.0,
      "step": 35050
    },
    {
      "entropy": 5.973459243774414,
      "epoch": 3.0956375838926173,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004052822979785884,
      "loss": 5.3171,
      "mean_token_accuracy": 0.21196908354759217,
      "num_tokens": 65358629.0,
      "step": 35055
    },
    {
      "entropy": 6.108779239654541,
      "epoch": 3.096079123984458,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00040525662153570593,
      "loss": 5.3387,
      "mean_token_accuracy": 0.20668396651744841,
      "num_tokens": 65368300.0,
      "step": 35060
    },
    {
      "entropy": 5.985944223403931,
      "epoch": 3.0965206640762983,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004052309425411705,
      "loss": 5.2604,
      "mean_token_accuracy": 0.21512964963912964,
      "num_tokens": 65377501.0,
      "step": 35065
    },
    {
      "entropy": 5.94781265258789,
      "epoch": 3.0969622041681384,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040520526099548483,
      "loss": 5.2998,
      "mean_token_accuracy": 0.2111433655023575,
      "num_tokens": 65386398.0,
      "step": 35070
    },
    {
      "entropy": 5.870036363601685,
      "epoch": 3.097403744259979,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004051795768991522,
      "loss": 5.1598,
      "mean_token_accuracy": 0.2255345031619072,
      "num_tokens": 65395735.0,
      "step": 35075
    },
    {
      "entropy": 5.89941439628601,
      "epoch": 3.097845284351819,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004051538902526756,
      "loss": 5.2259,
      "mean_token_accuracy": 0.21168812215328217,
      "num_tokens": 65405080.0,
      "step": 35080
    },
    {
      "entropy": 6.087805986404419,
      "epoch": 3.0982868244436594,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040512820105655843,
      "loss": 5.3311,
      "mean_token_accuracy": 0.2086866691708565,
      "num_tokens": 65414811.0,
      "step": 35085
    },
    {
      "entropy": 5.96954083442688,
      "epoch": 3.0987283645355,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040510250931130376,
      "loss": 5.2687,
      "mean_token_accuracy": 0.2128230407834053,
      "num_tokens": 65424479.0,
      "step": 35090
    },
    {
      "entropy": 5.915002727508545,
      "epoch": 3.09916990462734,
      "grad_norm": 1.125,
      "learning_rate": 0.0004050768150174149,
      "loss": 5.1966,
      "mean_token_accuracy": 0.2226760670542717,
      "num_tokens": 65434503.0,
      "step": 35095
    },
    {
      "entropy": 6.003227090835571,
      "epoch": 3.0996114447191805,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040505111817539527,
      "loss": 5.3023,
      "mean_token_accuracy": 0.21791571378707886,
      "num_tokens": 65443895.0,
      "step": 35100
    },
    {
      "entropy": 6.0093639373779295,
      "epoch": 3.100052984811021,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040502541878574815,
      "loss": 5.3402,
      "mean_token_accuracy": 0.20968252569437026,
      "num_tokens": 65453575.0,
      "step": 35105
    },
    {
      "entropy": 6.050988578796387,
      "epoch": 3.100494524902861,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004049997168489771,
      "loss": 5.3608,
      "mean_token_accuracy": 0.20611459165811538,
      "num_tokens": 65463084.0,
      "step": 35110
    },
    {
      "entropy": 5.974347734451294,
      "epoch": 3.1009360649947015,
      "grad_norm": 1.125,
      "learning_rate": 0.00040497401236558545,
      "loss": 5.2267,
      "mean_token_accuracy": 0.21829117387533187,
      "num_tokens": 65472767.0,
      "step": 35115
    },
    {
      "entropy": 6.014629650115967,
      "epoch": 3.101377605086542,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004049483053360769,
      "loss": 5.3727,
      "mean_token_accuracy": 0.21655728965997695,
      "num_tokens": 65483407.0,
      "step": 35120
    },
    {
      "entropy": 6.021605205535889,
      "epoch": 3.101819145178382,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004049225957609549,
      "loss": 5.23,
      "mean_token_accuracy": 0.2164325386285782,
      "num_tokens": 65493494.0,
      "step": 35125
    },
    {
      "entropy": 5.979679822921753,
      "epoch": 3.1022606852702226,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004048968836407231,
      "loss": 5.2276,
      "mean_token_accuracy": 0.21828143447637557,
      "num_tokens": 65502195.0,
      "step": 35130
    },
    {
      "entropy": 6.022274494171143,
      "epoch": 3.102702225362063,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004048711689758853,
      "loss": 5.3379,
      "mean_token_accuracy": 0.2087246984243393,
      "num_tokens": 65511783.0,
      "step": 35135
    },
    {
      "entropy": 5.949586057662964,
      "epoch": 3.103143765453903,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040484545176694523,
      "loss": 5.2242,
      "mean_token_accuracy": 0.21637108325958251,
      "num_tokens": 65521400.0,
      "step": 35140
    },
    {
      "entropy": 5.980377435684204,
      "epoch": 3.1035853055457436,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00040481973201440645,
      "loss": 5.1988,
      "mean_token_accuracy": 0.21779425740242003,
      "num_tokens": 65529442.0,
      "step": 35145
    },
    {
      "entropy": 5.946694993972779,
      "epoch": 3.1040268456375837,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040479400971877303,
      "loss": 5.2536,
      "mean_token_accuracy": 0.21888154447078706,
      "num_tokens": 65538522.0,
      "step": 35150
    },
    {
      "entropy": 5.99483585357666,
      "epoch": 3.104468385729424,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004047682848805488,
      "loss": 5.3165,
      "mean_token_accuracy": 0.21063343137502671,
      "num_tokens": 65547693.0,
      "step": 35155
    },
    {
      "entropy": 5.931551504135132,
      "epoch": 3.1049099258212647,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040474255750023763,
      "loss": 5.2472,
      "mean_token_accuracy": 0.21219490021467208,
      "num_tokens": 65557147.0,
      "step": 35160
    },
    {
      "entropy": 5.9499188423156735,
      "epoch": 3.1053514659131047,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040471682757834353,
      "loss": 5.2565,
      "mean_token_accuracy": 0.21762852370738983,
      "num_tokens": 65566925.0,
      "step": 35165
    },
    {
      "entropy": 5.934571695327759,
      "epoch": 3.1057930060049452,
      "grad_norm": 1.125,
      "learning_rate": 0.0004046910951153706,
      "loss": 5.2975,
      "mean_token_accuracy": 0.21622540354728698,
      "num_tokens": 65577390.0,
      "step": 35170
    },
    {
      "entropy": 6.028803300857544,
      "epoch": 3.1062345460967857,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004046653601118228,
      "loss": 5.231,
      "mean_token_accuracy": 0.2199520856142044,
      "num_tokens": 65586604.0,
      "step": 35175
    },
    {
      "entropy": 6.0009983539581295,
      "epoch": 3.106676086188626,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040463962256820433,
      "loss": 5.3637,
      "mean_token_accuracy": 0.20666112005710602,
      "num_tokens": 65595428.0,
      "step": 35180
    },
    {
      "entropy": 5.8569660663604735,
      "epoch": 3.1071176262804663,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004046138824850196,
      "loss": 5.2094,
      "mean_token_accuracy": 0.2212355300784111,
      "num_tokens": 65603784.0,
      "step": 35185
    },
    {
      "entropy": 5.937604284286499,
      "epoch": 3.107559166372307,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040458813986277246,
      "loss": 5.2327,
      "mean_token_accuracy": 0.21204868704080582,
      "num_tokens": 65611917.0,
      "step": 35190
    },
    {
      "entropy": 6.010827493667603,
      "epoch": 3.108000706464147,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040456239470196726,
      "loss": 5.2729,
      "mean_token_accuracy": 0.20837569534778594,
      "num_tokens": 65621908.0,
      "step": 35195
    },
    {
      "entropy": 6.044777154922485,
      "epoch": 3.1084422465559873,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004045366470031086,
      "loss": 5.3445,
      "mean_token_accuracy": 0.2131248503923416,
      "num_tokens": 65631489.0,
      "step": 35200
    },
    {
      "entropy": 5.974296951293946,
      "epoch": 3.108883786647828,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004045108967667006,
      "loss": 5.233,
      "mean_token_accuracy": 0.22050707191228866,
      "num_tokens": 65640562.0,
      "step": 35205
    },
    {
      "entropy": 6.043453550338745,
      "epoch": 3.109325326739668,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040448514399324783,
      "loss": 5.3694,
      "mean_token_accuracy": 0.2037214994430542,
      "num_tokens": 65650297.0,
      "step": 35210
    },
    {
      "entropy": 5.919959306716919,
      "epoch": 3.1097668668315084,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040445938868325477,
      "loss": 5.1665,
      "mean_token_accuracy": 0.22563135027885436,
      "num_tokens": 65660080.0,
      "step": 35215
    },
    {
      "entropy": 6.026842927932739,
      "epoch": 3.1102084069233484,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004044336308372259,
      "loss": 5.2795,
      "mean_token_accuracy": 0.21585960686206818,
      "num_tokens": 65669167.0,
      "step": 35220
    },
    {
      "entropy": 5.89007306098938,
      "epoch": 3.110649947015189,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004044078704556658,
      "loss": 5.2368,
      "mean_token_accuracy": 0.22106262296438217,
      "num_tokens": 65678993.0,
      "step": 35225
    },
    {
      "entropy": 5.947437763214111,
      "epoch": 3.1110914871070294,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040438210753907905,
      "loss": 5.3106,
      "mean_token_accuracy": 0.20525835305452347,
      "num_tokens": 65688118.0,
      "step": 35230
    },
    {
      "entropy": 5.981254434585571,
      "epoch": 3.1115330271988695,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040435634208797045,
      "loss": 5.2363,
      "mean_token_accuracy": 0.21657833606004714,
      "num_tokens": 65696559.0,
      "step": 35235
    },
    {
      "entropy": 5.953590965270996,
      "epoch": 3.11197456729071,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040433057410284474,
      "loss": 5.2854,
      "mean_token_accuracy": 0.2100257620215416,
      "num_tokens": 65705969.0,
      "step": 35240
    },
    {
      "entropy": 5.987623929977417,
      "epoch": 3.1124161073825505,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004043048035842066,
      "loss": 5.2769,
      "mean_token_accuracy": 0.21246804893016816,
      "num_tokens": 65714867.0,
      "step": 35245
    },
    {
      "entropy": 6.023465871810913,
      "epoch": 3.1128576474743905,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004042790305325608,
      "loss": 5.3698,
      "mean_token_accuracy": 0.2050655573606491,
      "num_tokens": 65724619.0,
      "step": 35250
    },
    {
      "entropy": 6.087150573730469,
      "epoch": 3.113299187566231,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004042532549484124,
      "loss": 5.3252,
      "mean_token_accuracy": 0.20590299963951111,
      "num_tokens": 65733273.0,
      "step": 35255
    },
    {
      "entropy": 5.924673843383789,
      "epoch": 3.1137407276580715,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004042274768322662,
      "loss": 5.1988,
      "mean_token_accuracy": 0.21902862638235093,
      "num_tokens": 65741428.0,
      "step": 35260
    },
    {
      "entropy": 5.8111350536346436,
      "epoch": 3.1141822677499116,
      "grad_norm": 1.25,
      "learning_rate": 0.0004042016961846274,
      "loss": 5.2231,
      "mean_token_accuracy": 0.2235541895031929,
      "num_tokens": 65750195.0,
      "step": 35265
    },
    {
      "entropy": 6.029531764984131,
      "epoch": 3.114623807841752,
      "grad_norm": 1.125,
      "learning_rate": 0.00040417591300600066,
      "loss": 5.3292,
      "mean_token_accuracy": 0.20809134393930434,
      "num_tokens": 65759723.0,
      "step": 35270
    },
    {
      "entropy": 6.0239075183868405,
      "epoch": 3.1150653479335926,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040415012729689133,
      "loss": 5.3335,
      "mean_token_accuracy": 0.21320871263742447,
      "num_tokens": 65769066.0,
      "step": 35275
    },
    {
      "entropy": 6.042795991897583,
      "epoch": 3.1155068880254326,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004041243390578044,
      "loss": 5.3361,
      "mean_token_accuracy": 0.20419751554727555,
      "num_tokens": 65778736.0,
      "step": 35280
    },
    {
      "entropy": 5.983734846115112,
      "epoch": 3.115948428117273,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040409854828924515,
      "loss": 5.2793,
      "mean_token_accuracy": 0.2119748204946518,
      "num_tokens": 65787568.0,
      "step": 35285
    },
    {
      "entropy": 6.045874786376953,
      "epoch": 3.116389968209113,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004040727549917187,
      "loss": 5.2988,
      "mean_token_accuracy": 0.21043696403503417,
      "num_tokens": 65797008.0,
      "step": 35290
    },
    {
      "entropy": 5.99157452583313,
      "epoch": 3.1168315083009537,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004040469591657305,
      "loss": 5.161,
      "mean_token_accuracy": 0.22307967990636826,
      "num_tokens": 65806080.0,
      "step": 35295
    },
    {
      "entropy": 5.918934059143067,
      "epoch": 3.117273048392794,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040402116081178567,
      "loss": 5.2785,
      "mean_token_accuracy": 0.21396734863519667,
      "num_tokens": 65815292.0,
      "step": 35300
    },
    {
      "entropy": 5.895604515075684,
      "epoch": 3.1177145884846342,
      "grad_norm": 1.125,
      "learning_rate": 0.0004039953599303898,
      "loss": 5.1404,
      "mean_token_accuracy": 0.2250533789396286,
      "num_tokens": 65824559.0,
      "step": 35305
    },
    {
      "entropy": 5.9900233268737795,
      "epoch": 3.1181561285764747,
      "grad_norm": 1.28125,
      "learning_rate": 0.000403969556522048,
      "loss": 5.3336,
      "mean_token_accuracy": 0.2050386920571327,
      "num_tokens": 65833556.0,
      "step": 35310
    },
    {
      "entropy": 5.950572538375854,
      "epoch": 3.1185976686683152,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004039437505872661,
      "loss": 5.2615,
      "mean_token_accuracy": 0.20990122556686402,
      "num_tokens": 65843199.0,
      "step": 35315
    },
    {
      "entropy": 5.94022126197815,
      "epoch": 3.1190392087601553,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004039179421265493,
      "loss": 5.2706,
      "mean_token_accuracy": 0.2063291609287262,
      "num_tokens": 65852902.0,
      "step": 35320
    },
    {
      "entropy": 6.02375373840332,
      "epoch": 3.1194807488519958,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040389213114040354,
      "loss": 5.2679,
      "mean_token_accuracy": 0.20937047749757767,
      "num_tokens": 65862213.0,
      "step": 35325
    },
    {
      "entropy": 5.940439510345459,
      "epoch": 3.1199222889438363,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004038663176293341,
      "loss": 5.1864,
      "mean_token_accuracy": 0.22516971975564956,
      "num_tokens": 65872436.0,
      "step": 35330
    },
    {
      "entropy": 5.9768421173095705,
      "epoch": 3.1203638290356763,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004038405015938468,
      "loss": 5.2797,
      "mean_token_accuracy": 0.20612494945526122,
      "num_tokens": 65881261.0,
      "step": 35335
    },
    {
      "entropy": 6.0063135623931885,
      "epoch": 3.120805369127517,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040381468303444737,
      "loss": 5.2635,
      "mean_token_accuracy": 0.21074342876672744,
      "num_tokens": 65890934.0,
      "step": 35340
    },
    {
      "entropy": 5.908429431915283,
      "epoch": 3.1212469092193573,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040378886195164155,
      "loss": 5.2002,
      "mean_token_accuracy": 0.2192688211798668,
      "num_tokens": 65900565.0,
      "step": 35345
    },
    {
      "entropy": 5.937917613983155,
      "epoch": 3.1216884493111974,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004037630383459351,
      "loss": 5.1707,
      "mean_token_accuracy": 0.2193541407585144,
      "num_tokens": 65910134.0,
      "step": 35350
    },
    {
      "entropy": 6.037173557281494,
      "epoch": 3.122129989403038,
      "grad_norm": 1.25,
      "learning_rate": 0.0004037372122178341,
      "loss": 5.3127,
      "mean_token_accuracy": 0.20929723083972931,
      "num_tokens": 65919403.0,
      "step": 35355
    },
    {
      "entropy": 5.952218294143677,
      "epoch": 3.122571529494878,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004037113835678443,
      "loss": 5.2028,
      "mean_token_accuracy": 0.22391830384731293,
      "num_tokens": 65928001.0,
      "step": 35360
    },
    {
      "entropy": 5.913156890869141,
      "epoch": 3.1230130695867184,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040368555239647165,
      "loss": 5.2745,
      "mean_token_accuracy": 0.2189257860183716,
      "num_tokens": 65936842.0,
      "step": 35365
    },
    {
      "entropy": 5.933640241622925,
      "epoch": 3.123454609678559,
      "grad_norm": 1.359375,
      "learning_rate": 0.00040365971870422217,
      "loss": 5.17,
      "mean_token_accuracy": 0.22514431327581405,
      "num_tokens": 65944842.0,
      "step": 35370
    },
    {
      "entropy": 5.907202243804932,
      "epoch": 3.123896149770399,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004036338824916021,
      "loss": 5.1896,
      "mean_token_accuracy": 0.22471880316734313,
      "num_tokens": 65954065.0,
      "step": 35375
    },
    {
      "entropy": 5.930830669403076,
      "epoch": 3.1243376898622395,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004036080437591174,
      "loss": 5.2826,
      "mean_token_accuracy": 0.21088252067565919,
      "num_tokens": 65963665.0,
      "step": 35380
    },
    {
      "entropy": 5.978958988189698,
      "epoch": 3.12477922995408,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004035822025072742,
      "loss": 5.3375,
      "mean_token_accuracy": 0.20842091292142867,
      "num_tokens": 65972889.0,
      "step": 35385
    },
    {
      "entropy": 5.970526027679443,
      "epoch": 3.12522077004592,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004035563587365788,
      "loss": 5.2616,
      "mean_token_accuracy": 0.2153450667858124,
      "num_tokens": 65981907.0,
      "step": 35390
    },
    {
      "entropy": 6.029620409011841,
      "epoch": 3.1256623101377605,
      "grad_norm": 1.25,
      "learning_rate": 0.0004035305124475374,
      "loss": 5.2698,
      "mean_token_accuracy": 0.20566887855529786,
      "num_tokens": 65990816.0,
      "step": 35395
    },
    {
      "entropy": 5.964464902877808,
      "epoch": 3.126103850229601,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004035046636406565,
      "loss": 5.2302,
      "mean_token_accuracy": 0.2222845658659935,
      "num_tokens": 65999992.0,
      "step": 35400
    },
    {
      "entropy": 5.971085882186889,
      "epoch": 3.126545390321441,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040347881231644217,
      "loss": 5.251,
      "mean_token_accuracy": 0.21816364377737046,
      "num_tokens": 66008597.0,
      "step": 35405
    },
    {
      "entropy": 5.96643385887146,
      "epoch": 3.1269869304132816,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040345295847540104,
      "loss": 5.2526,
      "mean_token_accuracy": 0.21106654405593872,
      "num_tokens": 66017803.0,
      "step": 35410
    },
    {
      "entropy": 5.944544839859009,
      "epoch": 3.127428470505122,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004034271021180396,
      "loss": 5.2352,
      "mean_token_accuracy": 0.21866973638534545,
      "num_tokens": 66026853.0,
      "step": 35415
    },
    {
      "entropy": 5.904330205917359,
      "epoch": 3.127870010596962,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004034012432448641,
      "loss": 5.1844,
      "mean_token_accuracy": 0.22119740396738052,
      "num_tokens": 66035149.0,
      "step": 35420
    },
    {
      "entropy": 5.977722692489624,
      "epoch": 3.1283115506888026,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004033753818563814,
      "loss": 5.2144,
      "mean_token_accuracy": 0.22734625786542892,
      "num_tokens": 66044039.0,
      "step": 35425
    },
    {
      "entropy": 5.954975700378418,
      "epoch": 3.1287530907806427,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040334951795309795,
      "loss": 5.2883,
      "mean_token_accuracy": 0.21879783868789673,
      "num_tokens": 66053413.0,
      "step": 35430
    },
    {
      "entropy": 5.998906183242798,
      "epoch": 3.129194630872483,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004033236515355204,
      "loss": 5.2682,
      "mean_token_accuracy": 0.21531322598457336,
      "num_tokens": 66062323.0,
      "step": 35435
    },
    {
      "entropy": 6.051701259613037,
      "epoch": 3.1296361709643237,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004032977826041555,
      "loss": 5.359,
      "mean_token_accuracy": 0.2043073445558548,
      "num_tokens": 66072682.0,
      "step": 35440
    },
    {
      "entropy": 5.98878846168518,
      "epoch": 3.1300777110561637,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040327191115951006,
      "loss": 5.2267,
      "mean_token_accuracy": 0.2159598633646965,
      "num_tokens": 66081208.0,
      "step": 35445
    },
    {
      "entropy": 5.9804079055786135,
      "epoch": 3.1305192511480042,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004032460372020908,
      "loss": 5.3015,
      "mean_token_accuracy": 0.21097974926233293,
      "num_tokens": 66090332.0,
      "step": 35450
    },
    {
      "entropy": 5.958581018447876,
      "epoch": 3.1309607912398447,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040322016073240463,
      "loss": 5.2129,
      "mean_token_accuracy": 0.22206753492355347,
      "num_tokens": 66099313.0,
      "step": 35455
    },
    {
      "entropy": 5.936326122283935,
      "epoch": 3.1314023313316848,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040319428175095834,
      "loss": 5.2808,
      "mean_token_accuracy": 0.2140596941113472,
      "num_tokens": 66109469.0,
      "step": 35460
    },
    {
      "entropy": 6.069693183898925,
      "epoch": 3.1318438714235253,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004031684002582591,
      "loss": 5.3983,
      "mean_token_accuracy": 0.20282348692417146,
      "num_tokens": 66119115.0,
      "step": 35465
    },
    {
      "entropy": 5.9983374118804935,
      "epoch": 3.1322854115153658,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040314251625481374,
      "loss": 5.31,
      "mean_token_accuracy": 0.21477123200893403,
      "num_tokens": 66129914.0,
      "step": 35470
    },
    {
      "entropy": 5.987498092651367,
      "epoch": 3.132726951607206,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004031166297411294,
      "loss": 5.2631,
      "mean_token_accuracy": 0.2216813877224922,
      "num_tokens": 66139612.0,
      "step": 35475
    },
    {
      "entropy": 6.026372480392456,
      "epoch": 3.1331684916990463,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004030907407177131,
      "loss": 5.3311,
      "mean_token_accuracy": 0.21816493570804596,
      "num_tokens": 66149152.0,
      "step": 35480
    },
    {
      "entropy": 6.047735404968262,
      "epoch": 3.133610031790887,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004030648491850721,
      "loss": 5.2885,
      "mean_token_accuracy": 0.21446291357278824,
      "num_tokens": 66160392.0,
      "step": 35485
    },
    {
      "entropy": 5.9087149143219,
      "epoch": 3.134051571882727,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040303895514371346,
      "loss": 5.1718,
      "mean_token_accuracy": 0.2195432275533676,
      "num_tokens": 66170433.0,
      "step": 35490
    },
    {
      "entropy": 5.89405198097229,
      "epoch": 3.1344931119745674,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040301305859414454,
      "loss": 5.2305,
      "mean_token_accuracy": 0.23111464381217955,
      "num_tokens": 66180321.0,
      "step": 35495
    },
    {
      "entropy": 5.995358657836914,
      "epoch": 3.1349346520664074,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040298715953687264,
      "loss": 5.246,
      "mean_token_accuracy": 0.21837465614080429,
      "num_tokens": 66188633.0,
      "step": 35500
    },
    {
      "entropy": 6.040960454940796,
      "epoch": 3.135376192158248,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000402961257972405,
      "loss": 5.3026,
      "mean_token_accuracy": 0.20509057492017746,
      "num_tokens": 66198351.0,
      "step": 35505
    },
    {
      "entropy": 5.904315233230591,
      "epoch": 3.1358177322500884,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004029353539012491,
      "loss": 5.2157,
      "mean_token_accuracy": 0.22054242938756943,
      "num_tokens": 66207992.0,
      "step": 35510
    },
    {
      "entropy": 5.992284965515137,
      "epoch": 3.1362592723419285,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040290944732391246,
      "loss": 5.3599,
      "mean_token_accuracy": 0.2065111517906189,
      "num_tokens": 66217770.0,
      "step": 35515
    },
    {
      "entropy": 6.043676376342773,
      "epoch": 3.136700812433769,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004028835382409025,
      "loss": 5.3213,
      "mean_token_accuracy": 0.2125895783305168,
      "num_tokens": 66227225.0,
      "step": 35520
    },
    {
      "entropy": 5.9913835525512695,
      "epoch": 3.1371423525256095,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004028576266527266,
      "loss": 5.1815,
      "mean_token_accuracy": 0.22165956795215608,
      "num_tokens": 66236129.0,
      "step": 35525
    },
    {
      "entropy": 5.943367052078247,
      "epoch": 3.1375838926174495,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004028317125598927,
      "loss": 5.2114,
      "mean_token_accuracy": 0.22930195182561874,
      "num_tokens": 66246453.0,
      "step": 35530
    },
    {
      "entropy": 5.880483865737915,
      "epoch": 3.13802543270929,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004028057959629081,
      "loss": 5.1795,
      "mean_token_accuracy": 0.2241520330309868,
      "num_tokens": 66255952.0,
      "step": 35535
    },
    {
      "entropy": 5.933552551269531,
      "epoch": 3.1384669728011305,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004027798768622808,
      "loss": 5.2772,
      "mean_token_accuracy": 0.21671657264232635,
      "num_tokens": 66264225.0,
      "step": 35540
    },
    {
      "entropy": 5.90667495727539,
      "epoch": 3.1389085128929706,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040275395525851834,
      "loss": 5.1084,
      "mean_token_accuracy": 0.23253557235002517,
      "num_tokens": 66272287.0,
      "step": 35545
    },
    {
      "entropy": 6.032442092895508,
      "epoch": 3.139350052984811,
      "grad_norm": 1.328125,
      "learning_rate": 0.00040272803115212844,
      "loss": 5.3592,
      "mean_token_accuracy": 0.2082750380039215,
      "num_tokens": 66280586.0,
      "step": 35550
    },
    {
      "entropy": 5.984931612014771,
      "epoch": 3.1397915930766516,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040270210454361916,
      "loss": 5.3277,
      "mean_token_accuracy": 0.21256031095981598,
      "num_tokens": 66290432.0,
      "step": 35555
    },
    {
      "entropy": 6.0167069911956785,
      "epoch": 3.1402331331684916,
      "grad_norm": 1.25,
      "learning_rate": 0.00040267617543349816,
      "loss": 5.2249,
      "mean_token_accuracy": 0.22220764607191085,
      "num_tokens": 66299543.0,
      "step": 35560
    },
    {
      "entropy": 6.020388269424439,
      "epoch": 3.140674673260332,
      "grad_norm": 1.25,
      "learning_rate": 0.0004026502438222736,
      "loss": 5.3058,
      "mean_token_accuracy": 0.21319346278905868,
      "num_tokens": 66308368.0,
      "step": 35565
    },
    {
      "entropy": 5.874803161621093,
      "epoch": 3.141116213352172,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040262430971045333,
      "loss": 5.196,
      "mean_token_accuracy": 0.2249472126364708,
      "num_tokens": 66318003.0,
      "step": 35570
    },
    {
      "entropy": 5.976338577270508,
      "epoch": 3.1415577534440127,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004025983730985455,
      "loss": 5.2116,
      "mean_token_accuracy": 0.2259576678276062,
      "num_tokens": 66327437.0,
      "step": 35575
    },
    {
      "entropy": 6.02508134841919,
      "epoch": 3.141999293535853,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000402572433987058,
      "loss": 5.27,
      "mean_token_accuracy": 0.20934971272945405,
      "num_tokens": 66336819.0,
      "step": 35580
    },
    {
      "entropy": 5.976150846481323,
      "epoch": 3.142440833627693,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040254649237649916,
      "loss": 5.2522,
      "mean_token_accuracy": 0.21966279447078704,
      "num_tokens": 66345738.0,
      "step": 35585
    },
    {
      "entropy": 5.952600765228271,
      "epoch": 3.1428823737195337,
      "grad_norm": 1.28125,
      "learning_rate": 0.000402520548267377,
      "loss": 5.1846,
      "mean_token_accuracy": 0.2122950181365013,
      "num_tokens": 66355017.0,
      "step": 35590
    },
    {
      "entropy": 6.023689079284668,
      "epoch": 3.143323913811374,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004024946016601998,
      "loss": 5.2876,
      "mean_token_accuracy": 0.21243601739406587,
      "num_tokens": 66363675.0,
      "step": 35595
    },
    {
      "entropy": 5.972366189956665,
      "epoch": 3.1437654539032143,
      "grad_norm": 1.25,
      "learning_rate": 0.000402468652555476,
      "loss": 5.2676,
      "mean_token_accuracy": 0.21809042245149612,
      "num_tokens": 66373452.0,
      "step": 35600
    },
    {
      "entropy": 5.948892211914062,
      "epoch": 3.1442069939950548,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040244270095371367,
      "loss": 5.2967,
      "mean_token_accuracy": 0.21336278021335603,
      "num_tokens": 66382381.0,
      "step": 35605
    },
    {
      "entropy": 5.941742324829102,
      "epoch": 3.1446485340868953,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040241674685542137,
      "loss": 5.2217,
      "mean_token_accuracy": 0.22216498702764512,
      "num_tokens": 66391914.0,
      "step": 35610
    },
    {
      "entropy": 6.0449535846710205,
      "epoch": 3.1450900741787353,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040239079026110734,
      "loss": 5.281,
      "mean_token_accuracy": 0.21253487914800645,
      "num_tokens": 66400774.0,
      "step": 35615
    },
    {
      "entropy": 5.965809726715088,
      "epoch": 3.145531614270576,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040236483117128035,
      "loss": 5.3022,
      "mean_token_accuracy": 0.2129663646221161,
      "num_tokens": 66410369.0,
      "step": 35620
    },
    {
      "entropy": 5.961913824081421,
      "epoch": 3.1459731543624163,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040233886958644866,
      "loss": 5.2344,
      "mean_token_accuracy": 0.2202942416071892,
      "num_tokens": 66419622.0,
      "step": 35625
    },
    {
      "entropy": 5.893232631683349,
      "epoch": 3.1464146944542564,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000402312905507121,
      "loss": 5.2285,
      "mean_token_accuracy": 0.22061611264944075,
      "num_tokens": 66429159.0,
      "step": 35630
    },
    {
      "entropy": 6.064135265350342,
      "epoch": 3.146856234546097,
      "grad_norm": 1.25,
      "learning_rate": 0.00040228693893380583,
      "loss": 5.2938,
      "mean_token_accuracy": 0.2084350109100342,
      "num_tokens": 66437696.0,
      "step": 35635
    },
    {
      "entropy": 6.045545053482056,
      "epoch": 3.147297774637937,
      "grad_norm": 1.234375,
      "learning_rate": 0.000402260969867012,
      "loss": 5.3666,
      "mean_token_accuracy": 0.208215694129467,
      "num_tokens": 66447039.0,
      "step": 35640
    },
    {
      "entropy": 5.9741322040557865,
      "epoch": 3.1477393147297774,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040223499830724813,
      "loss": 5.2241,
      "mean_token_accuracy": 0.22045084983110427,
      "num_tokens": 66456566.0,
      "step": 35645
    },
    {
      "entropy": 5.908289003372192,
      "epoch": 3.148180854821618,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040220902425502296,
      "loss": 5.1887,
      "mean_token_accuracy": 0.22524297684431077,
      "num_tokens": 66466494.0,
      "step": 35650
    },
    {
      "entropy": 5.836434078216553,
      "epoch": 3.148622394913458,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00040218304771084544,
      "loss": 5.0652,
      "mean_token_accuracy": 0.23962082266807555,
      "num_tokens": 66475760.0,
      "step": 35655
    },
    {
      "entropy": 5.967698669433593,
      "epoch": 3.1490639350052985,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040215706867522435,
      "loss": 5.2981,
      "mean_token_accuracy": 0.21352590918540953,
      "num_tokens": 66484239.0,
      "step": 35660
    },
    {
      "entropy": 6.013189315795898,
      "epoch": 3.149505475097139,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004021310871486685,
      "loss": 5.2514,
      "mean_token_accuracy": 0.21402086913585663,
      "num_tokens": 66493513.0,
      "step": 35665
    },
    {
      "entropy": 6.017643070220947,
      "epoch": 3.149947015188979,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000402105103131687,
      "loss": 5.2504,
      "mean_token_accuracy": 0.2187243089079857,
      "num_tokens": 66502491.0,
      "step": 35670
    },
    {
      "entropy": 5.992544078826905,
      "epoch": 3.1503885552808195,
      "grad_norm": 1.25,
      "learning_rate": 0.0004020791166247888,
      "loss": 5.3062,
      "mean_token_accuracy": 0.20516340285539628,
      "num_tokens": 66512405.0,
      "step": 35675
    },
    {
      "entropy": 6.013747501373291,
      "epoch": 3.15083009537266,
      "grad_norm": 1.09375,
      "learning_rate": 0.000402053127628483,
      "loss": 5.2868,
      "mean_token_accuracy": 0.2174329325556755,
      "num_tokens": 66521274.0,
      "step": 35680
    },
    {
      "entropy": 5.964573764801026,
      "epoch": 3.1512716354645,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004020271361432787,
      "loss": 5.2686,
      "mean_token_accuracy": 0.22114374190568925,
      "num_tokens": 66531049.0,
      "step": 35685
    },
    {
      "entropy": 5.944308948516846,
      "epoch": 3.1517131755563406,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000402001142169685,
      "loss": 5.2429,
      "mean_token_accuracy": 0.21849891990423204,
      "num_tokens": 66539866.0,
      "step": 35690
    },
    {
      "entropy": 5.935992097854614,
      "epoch": 3.152154715648181,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040197514570821123,
      "loss": 5.2457,
      "mean_token_accuracy": 0.21384139806032182,
      "num_tokens": 66549147.0,
      "step": 35695
    },
    {
      "entropy": 5.966046524047852,
      "epoch": 3.152596255740021,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040194914675936656,
      "loss": 5.2464,
      "mean_token_accuracy": 0.2152002677321434,
      "num_tokens": 66558855.0,
      "step": 35700
    },
    {
      "entropy": 6.003921937942505,
      "epoch": 3.1530377958318616,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040192314532366027,
      "loss": 5.3108,
      "mean_token_accuracy": 0.2084399938583374,
      "num_tokens": 66569549.0,
      "step": 35705
    },
    {
      "entropy": 6.041668891906738,
      "epoch": 3.1534793359237017,
      "grad_norm": 1.125,
      "learning_rate": 0.0004018971414016017,
      "loss": 5.3535,
      "mean_token_accuracy": 0.21263765543699265,
      "num_tokens": 66578070.0,
      "step": 35710
    },
    {
      "entropy": 5.9120635986328125,
      "epoch": 3.153920876015542,
      "grad_norm": 1.3125,
      "learning_rate": 0.00040187113499370045,
      "loss": 5.1554,
      "mean_token_accuracy": 0.22117463797330855,
      "num_tokens": 66586933.0,
      "step": 35715
    },
    {
      "entropy": 5.900077629089355,
      "epoch": 3.1543624161073827,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040184512610046575,
      "loss": 5.2306,
      "mean_token_accuracy": 0.21457630693912505,
      "num_tokens": 66596767.0,
      "step": 35720
    },
    {
      "entropy": 6.074354457855224,
      "epoch": 3.1548039561992227,
      "grad_norm": 1.125,
      "learning_rate": 0.0004018191147224071,
      "loss": 5.427,
      "mean_token_accuracy": 0.19755344837903976,
      "num_tokens": 66606425.0,
      "step": 35725
    },
    {
      "entropy": 6.017080879211425,
      "epoch": 3.155245496291063,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004017931008600342,
      "loss": 5.2363,
      "mean_token_accuracy": 0.21313409358263016,
      "num_tokens": 66616470.0,
      "step": 35730
    },
    {
      "entropy": 5.847880411148071,
      "epoch": 3.1556870363829037,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004017670845138565,
      "loss": 5.1246,
      "mean_token_accuracy": 0.22500652223825454,
      "num_tokens": 66625726.0,
      "step": 35735
    },
    {
      "entropy": 5.966418743133545,
      "epoch": 3.1561285764747438,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040174106568438377,
      "loss": 5.2379,
      "mean_token_accuracy": 0.21345657706260682,
      "num_tokens": 66634253.0,
      "step": 35740
    },
    {
      "entropy": 6.016397285461426,
      "epoch": 3.1565701165665843,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004017150443721256,
      "loss": 5.2695,
      "mean_token_accuracy": 0.21078983545303345,
      "num_tokens": 66643037.0,
      "step": 35745
    },
    {
      "entropy": 5.883148050308227,
      "epoch": 3.1570116566584248,
      "grad_norm": 1.4375,
      "learning_rate": 0.00040168902057759176,
      "loss": 5.0878,
      "mean_token_accuracy": 0.22314192801713945,
      "num_tokens": 66651650.0,
      "step": 35750
    },
    {
      "entropy": 5.915228796005249,
      "epoch": 3.157453196750265,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040166299430129213,
      "loss": 5.2841,
      "mean_token_accuracy": 0.21825626194477082,
      "num_tokens": 66660538.0,
      "step": 35755
    },
    {
      "entropy": 5.970098686218262,
      "epoch": 3.1578947368421053,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040163696554373646,
      "loss": 5.2696,
      "mean_token_accuracy": 0.2130484402179718,
      "num_tokens": 66669245.0,
      "step": 35760
    },
    {
      "entropy": 6.059433269500732,
      "epoch": 3.158336276933946,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040161093430543467,
      "loss": 5.3275,
      "mean_token_accuracy": 0.21086563915014267,
      "num_tokens": 66678805.0,
      "step": 35765
    },
    {
      "entropy": 5.9946290969848635,
      "epoch": 3.158777817025786,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040158490058689664,
      "loss": 5.2598,
      "mean_token_accuracy": 0.2087426319718361,
      "num_tokens": 66687289.0,
      "step": 35770
    },
    {
      "entropy": 5.948067712783813,
      "epoch": 3.1592193571176264,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004015588643886324,
      "loss": 5.2236,
      "mean_token_accuracy": 0.2202102616429329,
      "num_tokens": 66696266.0,
      "step": 35775
    },
    {
      "entropy": 5.922720003128052,
      "epoch": 3.1596608972094664,
      "grad_norm": 1.234375,
      "learning_rate": 0.000401532825711152,
      "loss": 5.281,
      "mean_token_accuracy": 0.21559077501296997,
      "num_tokens": 66705679.0,
      "step": 35780
    },
    {
      "entropy": 5.974360656738281,
      "epoch": 3.160102437301307,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004015067845549655,
      "loss": 5.2513,
      "mean_token_accuracy": 0.20952336341142655,
      "num_tokens": 66714740.0,
      "step": 35785
    },
    {
      "entropy": 5.969404888153076,
      "epoch": 3.1605439773931474,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004014807409205831,
      "loss": 5.2659,
      "mean_token_accuracy": 0.2158701926469803,
      "num_tokens": 66724403.0,
      "step": 35790
    },
    {
      "entropy": 5.997127294540405,
      "epoch": 3.1609855174849875,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004014546948085148,
      "loss": 5.2317,
      "mean_token_accuracy": 0.21663298159837724,
      "num_tokens": 66733447.0,
      "step": 35795
    },
    {
      "entropy": 5.949308300018311,
      "epoch": 3.161427057576828,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040142864621927097,
      "loss": 5.1726,
      "mean_token_accuracy": 0.22248048782348634,
      "num_tokens": 66742404.0,
      "step": 35800
    },
    {
      "entropy": 5.938191080093384,
      "epoch": 3.1618685976686685,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040140259515336195,
      "loss": 5.2982,
      "mean_token_accuracy": 0.21851206570863724,
      "num_tokens": 66752173.0,
      "step": 35805
    },
    {
      "entropy": 5.920646715164184,
      "epoch": 3.1623101377605085,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00040137654161129797,
      "loss": 5.1668,
      "mean_token_accuracy": 0.22067183256149292,
      "num_tokens": 66761000.0,
      "step": 35810
    },
    {
      "entropy": 5.956452655792236,
      "epoch": 3.162751677852349,
      "grad_norm": 1.125,
      "learning_rate": 0.00040135048559358934,
      "loss": 5.2416,
      "mean_token_accuracy": 0.21302293241024017,
      "num_tokens": 66769655.0,
      "step": 35815
    },
    {
      "entropy": 5.925367832183838,
      "epoch": 3.1631932179441895,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040132442710074664,
      "loss": 5.2034,
      "mean_token_accuracy": 0.22790869176387787,
      "num_tokens": 66777927.0,
      "step": 35820
    },
    {
      "entropy": 5.959603643417358,
      "epoch": 3.1636347580360296,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004012983661332802,
      "loss": 5.2952,
      "mean_token_accuracy": 0.2160269796848297,
      "num_tokens": 66788131.0,
      "step": 35825
    },
    {
      "entropy": 5.868001985549927,
      "epoch": 3.16407629812787,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040127230269170063,
      "loss": 5.256,
      "mean_token_accuracy": 0.21841289550065995,
      "num_tokens": 66797961.0,
      "step": 35830
    },
    {
      "entropy": 6.020868158340454,
      "epoch": 3.1645178382197106,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040124623677651844,
      "loss": 5.3211,
      "mean_token_accuracy": 0.20845720171928406,
      "num_tokens": 66807718.0,
      "step": 35835
    },
    {
      "entropy": 6.038587045669556,
      "epoch": 3.1649593783115506,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004012201683882443,
      "loss": 5.2665,
      "mean_token_accuracy": 0.2091606602072716,
      "num_tokens": 66816907.0,
      "step": 35840
    },
    {
      "entropy": 5.928565502166748,
      "epoch": 3.165400918403391,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004011940975273888,
      "loss": 5.1436,
      "mean_token_accuracy": 0.22637107968330383,
      "num_tokens": 66825528.0,
      "step": 35845
    },
    {
      "entropy": 5.979663133621216,
      "epoch": 3.165842458495231,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040116802419446276,
      "loss": 5.3422,
      "mean_token_accuracy": 0.2111227884888649,
      "num_tokens": 66835511.0,
      "step": 35850
    },
    {
      "entropy": 5.958665466308593,
      "epoch": 3.1662839985870717,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004011419483899768,
      "loss": 5.2221,
      "mean_token_accuracy": 0.21386638134717942,
      "num_tokens": 66844904.0,
      "step": 35855
    },
    {
      "entropy": 6.022878122329712,
      "epoch": 3.166725538678912,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004011158701144419,
      "loss": 5.2942,
      "mean_token_accuracy": 0.2117481127381325,
      "num_tokens": 66853517.0,
      "step": 35860
    },
    {
      "entropy": 5.9849365711212155,
      "epoch": 3.167167078770752,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004010897893683688,
      "loss": 5.2846,
      "mean_token_accuracy": 0.21321359425783157,
      "num_tokens": 66863369.0,
      "step": 35865
    },
    {
      "entropy": 5.907566213607788,
      "epoch": 3.1676086188625927,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004010637061522684,
      "loss": 5.1806,
      "mean_token_accuracy": 0.22405079752206802,
      "num_tokens": 66872801.0,
      "step": 35870
    },
    {
      "entropy": 5.971067571640015,
      "epoch": 3.168050158954433,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004010376204666517,
      "loss": 5.1769,
      "mean_token_accuracy": 0.2177545666694641,
      "num_tokens": 66881276.0,
      "step": 35875
    },
    {
      "entropy": 6.03500452041626,
      "epoch": 3.1684916990462733,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040101153231202974,
      "loss": 5.3206,
      "mean_token_accuracy": 0.2085520803928375,
      "num_tokens": 66889924.0,
      "step": 35880
    },
    {
      "entropy": 5.991970205307007,
      "epoch": 3.1689332391381138,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004009854416889135,
      "loss": 5.2357,
      "mean_token_accuracy": 0.21958472579717636,
      "num_tokens": 66898350.0,
      "step": 35885
    },
    {
      "entropy": 5.982799530029297,
      "epoch": 3.1693747792299543,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040095934859781413,
      "loss": 5.3064,
      "mean_token_accuracy": 0.21486769765615463,
      "num_tokens": 66908416.0,
      "step": 35890
    },
    {
      "entropy": 6.025751399993896,
      "epoch": 3.1698163193217943,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00040093325303924274,
      "loss": 5.2869,
      "mean_token_accuracy": 0.2162501856684685,
      "num_tokens": 66917944.0,
      "step": 35895
    },
    {
      "entropy": 5.955474138259888,
      "epoch": 3.170257859413635,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004009071550137105,
      "loss": 5.2382,
      "mean_token_accuracy": 0.21961467564105988,
      "num_tokens": 66926932.0,
      "step": 35900
    },
    {
      "entropy": 5.837101888656616,
      "epoch": 3.1706993995054753,
      "grad_norm": 1.25,
      "learning_rate": 0.0004008810545217288,
      "loss": 5.1696,
      "mean_token_accuracy": 0.2119014024734497,
      "num_tokens": 66935912.0,
      "step": 35905
    },
    {
      "entropy": 5.874984788894653,
      "epoch": 3.1711409395973154,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004008549515638087,
      "loss": 5.1481,
      "mean_token_accuracy": 0.23322298377752304,
      "num_tokens": 66945314.0,
      "step": 35910
    },
    {
      "entropy": 5.974461793899536,
      "epoch": 3.171582479689156,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040082884614046183,
      "loss": 5.1741,
      "mean_token_accuracy": 0.2250753328204155,
      "num_tokens": 66953080.0,
      "step": 35915
    },
    {
      "entropy": 5.928352403640747,
      "epoch": 3.172024019780996,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004008027382521993,
      "loss": 5.2783,
      "mean_token_accuracy": 0.2217516243457794,
      "num_tokens": 66964982.0,
      "step": 35920
    },
    {
      "entropy": 6.014183616638183,
      "epoch": 3.1724655598728364,
      "grad_norm": 1.375,
      "learning_rate": 0.00040077662789953286,
      "loss": 5.2927,
      "mean_token_accuracy": 0.21558413952589034,
      "num_tokens": 66974225.0,
      "step": 35925
    },
    {
      "entropy": 5.921043777465821,
      "epoch": 3.172907099964677,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004007505150829736,
      "loss": 5.1471,
      "mean_token_accuracy": 0.23298854678869246,
      "num_tokens": 66982272.0,
      "step": 35930
    },
    {
      "entropy": 5.916000175476074,
      "epoch": 3.173348640056517,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004007243998030333,
      "loss": 5.1575,
      "mean_token_accuracy": 0.2270924210548401,
      "num_tokens": 66992207.0,
      "step": 35935
    },
    {
      "entropy": 6.006545162200927,
      "epoch": 3.1737901801483575,
      "grad_norm": 1.25,
      "learning_rate": 0.00040069828206022365,
      "loss": 5.3583,
      "mean_token_accuracy": 0.20841998606920242,
      "num_tokens": 67001301.0,
      "step": 35940
    },
    {
      "entropy": 5.954947519302368,
      "epoch": 3.174231720240198,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000400672161855056,
      "loss": 5.2257,
      "mean_token_accuracy": 0.21857340782880783,
      "num_tokens": 67009899.0,
      "step": 35945
    },
    {
      "entropy": 5.986636066436768,
      "epoch": 3.174673260332038,
      "grad_norm": 1.375,
      "learning_rate": 0.0004006460391880422,
      "loss": 5.2695,
      "mean_token_accuracy": 0.20743793100118638,
      "num_tokens": 67018093.0,
      "step": 35950
    },
    {
      "entropy": 6.013485097885132,
      "epoch": 3.1751148004238785,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040061991405969404,
      "loss": 5.3208,
      "mean_token_accuracy": 0.21072499006986617,
      "num_tokens": 67027731.0,
      "step": 35955
    },
    {
      "entropy": 6.010189390182495,
      "epoch": 3.175556340515719,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004005937864705231,
      "loss": 5.2705,
      "mean_token_accuracy": 0.20772038102149964,
      "num_tokens": 67037278.0,
      "step": 35960
    },
    {
      "entropy": 6.052437114715576,
      "epoch": 3.175997880607559,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040056765642104144,
      "loss": 5.3761,
      "mean_token_accuracy": 0.21287937313318253,
      "num_tokens": 67047288.0,
      "step": 35965
    },
    {
      "entropy": 5.888608455657959,
      "epoch": 3.1764394206993996,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004005415239117607,
      "loss": 5.15,
      "mean_token_accuracy": 0.22371659129858018,
      "num_tokens": 67056131.0,
      "step": 35970
    },
    {
      "entropy": 6.027485942840576,
      "epoch": 3.17688096079124,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004005153889431929,
      "loss": 5.304,
      "mean_token_accuracy": 0.205876824259758,
      "num_tokens": 67065587.0,
      "step": 35975
    },
    {
      "entropy": 5.973563241958618,
      "epoch": 3.17732250088308,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004004892515158501,
      "loss": 5.2021,
      "mean_token_accuracy": 0.22333188503980636,
      "num_tokens": 67074146.0,
      "step": 35980
    },
    {
      "entropy": 5.984386920928955,
      "epoch": 3.1777640409749206,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004004631116302441,
      "loss": 5.2899,
      "mean_token_accuracy": 0.21070636808872223,
      "num_tokens": 67082927.0,
      "step": 35985
    },
    {
      "entropy": 5.88962140083313,
      "epoch": 3.1782055810667607,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004004369692868872,
      "loss": 5.1108,
      "mean_token_accuracy": 0.23677016347646712,
      "num_tokens": 67091293.0,
      "step": 35990
    },
    {
      "entropy": 5.903986835479737,
      "epoch": 3.178647121158601,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004004108244862914,
      "loss": 5.197,
      "mean_token_accuracy": 0.22178069800138472,
      "num_tokens": 67100719.0,
      "step": 35995
    },
    {
      "entropy": 5.995440435409546,
      "epoch": 3.1790886612504416,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004003846772289688,
      "loss": 5.237,
      "mean_token_accuracy": 0.21319113820791244,
      "num_tokens": 67109891.0,
      "step": 36000
    },
    {
      "epoch": 3.1790886612504416,
      "eval_entropy": 5.706552520343698,
      "eval_loss": 5.33390998840332,
      "eval_mean_token_accuracy": 0.2209036310769761,
      "eval_num_tokens": 67109891.0,
      "eval_runtime": 20.1085,
      "eval_samples_per_second": 1756.222,
      "eval_steps_per_second": 219.559,
      "step": 36000
    },
    {
      "entropy": 5.877378845214844,
      "epoch": 3.1795302013422817,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040035852751543175,
      "loss": 5.1003,
      "mean_token_accuracy": 0.2390286833047867,
      "num_tokens": 67118491.0,
      "step": 36005
    },
    {
      "entropy": 5.928313779830932,
      "epoch": 3.179971741434122,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004003323753461925,
      "loss": 5.3619,
      "mean_token_accuracy": 0.20519149601459502,
      "num_tokens": 67128444.0,
      "step": 36010
    },
    {
      "entropy": 6.004872035980225,
      "epoch": 3.1804132815259627,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040030622072176325,
      "loss": 5.3114,
      "mean_token_accuracy": 0.20799928903579712,
      "num_tokens": 67138114.0,
      "step": 36015
    },
    {
      "entropy": 5.988152933120728,
      "epoch": 3.1808548216178028,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040028006364265634,
      "loss": 5.2788,
      "mean_token_accuracy": 0.20913151800632476,
      "num_tokens": 67147373.0,
      "step": 36020
    },
    {
      "entropy": 5.918635845184326,
      "epoch": 3.1812963617096433,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004002539041093843,
      "loss": 5.2268,
      "mean_token_accuracy": 0.2165445417165756,
      "num_tokens": 67156731.0,
      "step": 36025
    },
    {
      "entropy": 6.026432847976684,
      "epoch": 3.1817379018014837,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004002277421224594,
      "loss": 5.265,
      "mean_token_accuracy": 0.21933830231428147,
      "num_tokens": 67166185.0,
      "step": 36030
    },
    {
      "entropy": 5.993581342697143,
      "epoch": 3.182179441893324,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004002015776823944,
      "loss": 5.2544,
      "mean_token_accuracy": 0.21201379150152205,
      "num_tokens": 67175255.0,
      "step": 36035
    },
    {
      "entropy": 5.9310565948486325,
      "epoch": 3.1826209819851643,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040017541078970164,
      "loss": 5.2198,
      "mean_token_accuracy": 0.2119176283478737,
      "num_tokens": 67184543.0,
      "step": 36040
    },
    {
      "entropy": 5.998438978195191,
      "epoch": 3.183062522077005,
      "grad_norm": 1.25,
      "learning_rate": 0.00040014924144489384,
      "loss": 5.1845,
      "mean_token_accuracy": 0.2176115781068802,
      "num_tokens": 67193802.0,
      "step": 36045
    },
    {
      "entropy": 5.977354621887207,
      "epoch": 3.183504062168845,
      "grad_norm": 1.3125,
      "learning_rate": 0.00040012306964848344,
      "loss": 5.3071,
      "mean_token_accuracy": 0.2086343064904213,
      "num_tokens": 67202514.0,
      "step": 36050
    },
    {
      "entropy": 6.038415241241455,
      "epoch": 3.1839456022606853,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040009689540098335,
      "loss": 5.3053,
      "mean_token_accuracy": 0.21248068958520888,
      "num_tokens": 67213304.0,
      "step": 36055
    },
    {
      "entropy": 5.99135799407959,
      "epoch": 3.1843871423525254,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040007071870290625,
      "loss": 5.3022,
      "mean_token_accuracy": 0.21189354807138444,
      "num_tokens": 67222397.0,
      "step": 36060
    },
    {
      "entropy": 5.891929769515992,
      "epoch": 3.184828682444366,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004000445395547648,
      "loss": 5.2494,
      "mean_token_accuracy": 0.22094108164310455,
      "num_tokens": 67230584.0,
      "step": 36065
    },
    {
      "entropy": 6.019092178344726,
      "epoch": 3.1852702225362064,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040001835795707206,
      "loss": 5.2844,
      "mean_token_accuracy": 0.2146849364042282,
      "num_tokens": 67239851.0,
      "step": 36070
    },
    {
      "entropy": 6.017898845672607,
      "epoch": 3.1857117626280465,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003999921739103408,
      "loss": 5.3044,
      "mean_token_accuracy": 0.21099983006715775,
      "num_tokens": 67249461.0,
      "step": 36075
    },
    {
      "entropy": 5.9445007801055905,
      "epoch": 3.186153302719887,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039996598741508384,
      "loss": 5.2707,
      "mean_token_accuracy": 0.21596881300210952,
      "num_tokens": 67258339.0,
      "step": 36080
    },
    {
      "entropy": 5.990455961227417,
      "epoch": 3.1865948428117274,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003999397984718142,
      "loss": 5.2408,
      "mean_token_accuracy": 0.21395591497421265,
      "num_tokens": 67267247.0,
      "step": 36085
    },
    {
      "entropy": 5.879807043075561,
      "epoch": 3.1870363829035675,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003999136070810451,
      "loss": 5.1074,
      "mean_token_accuracy": 0.23284451365470887,
      "num_tokens": 67275710.0,
      "step": 36090
    },
    {
      "entropy": 5.975707864761352,
      "epoch": 3.187477922995408,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039988741324328944,
      "loss": 5.3043,
      "mean_token_accuracy": 0.21191399693489074,
      "num_tokens": 67285085.0,
      "step": 36095
    },
    {
      "entropy": 6.029934597015381,
      "epoch": 3.1879194630872485,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039986121695906036,
      "loss": 5.3769,
      "mean_token_accuracy": 0.20195440500974654,
      "num_tokens": 67295602.0,
      "step": 36100
    },
    {
      "entropy": 6.079214668273925,
      "epoch": 3.1883610031790885,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039983501822887107,
      "loss": 5.3153,
      "mean_token_accuracy": 0.21555040180683135,
      "num_tokens": 67304230.0,
      "step": 36105
    },
    {
      "entropy": 6.026354074478149,
      "epoch": 3.188802543270929,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003998088170532348,
      "loss": 5.251,
      "mean_token_accuracy": 0.2223515585064888,
      "num_tokens": 67312688.0,
      "step": 36110
    },
    {
      "entropy": 5.982703781127929,
      "epoch": 3.1892440833627695,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003997826134326647,
      "loss": 5.2948,
      "mean_token_accuracy": 0.21109363585710525,
      "num_tokens": 67321623.0,
      "step": 36115
    },
    {
      "entropy": 5.957826852798462,
      "epoch": 3.1896856234546096,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003997564073676743,
      "loss": 5.2896,
      "mean_token_accuracy": 0.20783807784318925,
      "num_tokens": 67330825.0,
      "step": 36120
    },
    {
      "entropy": 5.986899185180664,
      "epoch": 3.19012716354645,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039973019885877677,
      "loss": 5.2853,
      "mean_token_accuracy": 0.2171356350183487,
      "num_tokens": 67339859.0,
      "step": 36125
    },
    {
      "entropy": 6.0292041301727295,
      "epoch": 3.19056870363829,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039970398790648557,
      "loss": 5.2942,
      "mean_token_accuracy": 0.2209121748805046,
      "num_tokens": 67350367.0,
      "step": 36130
    },
    {
      "entropy": 6.013728952407837,
      "epoch": 3.1910102437301306,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039967777451131416,
      "loss": 5.3682,
      "mean_token_accuracy": 0.20223139822483063,
      "num_tokens": 67360264.0,
      "step": 36135
    },
    {
      "entropy": 6.035822343826294,
      "epoch": 3.191451783821971,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000399651558673776,
      "loss": 5.2654,
      "mean_token_accuracy": 0.21454998701810837,
      "num_tokens": 67369713.0,
      "step": 36140
    },
    {
      "entropy": 5.915282726287842,
      "epoch": 3.191893323913811,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003996253403943848,
      "loss": 5.1899,
      "mean_token_accuracy": 0.2218659445643425,
      "num_tokens": 67378817.0,
      "step": 36145
    },
    {
      "entropy": 6.0022704124450685,
      "epoch": 3.1923348640056517,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000399599119673654,
      "loss": 5.3836,
      "mean_token_accuracy": 0.20979859083890914,
      "num_tokens": 67389098.0,
      "step": 36150
    },
    {
      "entropy": 6.006583642959595,
      "epoch": 3.192776404097492,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039957289651209736,
      "loss": 5.2078,
      "mean_token_accuracy": 0.22113976627588272,
      "num_tokens": 67397974.0,
      "step": 36155
    },
    {
      "entropy": 5.959064149856568,
      "epoch": 3.1932179441893322,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039954667091022844,
      "loss": 5.2707,
      "mean_token_accuracy": 0.2154268816113472,
      "num_tokens": 67407629.0,
      "step": 36160
    },
    {
      "entropy": 5.956231117248535,
      "epoch": 3.1936594842811727,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003995204428685612,
      "loss": 5.2878,
      "mean_token_accuracy": 0.2175666481256485,
      "num_tokens": 67416395.0,
      "step": 36165
    },
    {
      "entropy": 5.932162046432495,
      "epoch": 3.1941010243730132,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003994942123876092,
      "loss": 5.2241,
      "mean_token_accuracy": 0.2184286206960678,
      "num_tokens": 67426370.0,
      "step": 36170
    },
    {
      "entropy": 5.94841775894165,
      "epoch": 3.1945425644648533,
      "grad_norm": 1.375,
      "learning_rate": 0.0003994679794678865,
      "loss": 5.0941,
      "mean_token_accuracy": 0.22304990142583847,
      "num_tokens": 67435586.0,
      "step": 36175
    },
    {
      "entropy": 5.989915561676026,
      "epoch": 3.194984104556694,
      "grad_norm": 1.25,
      "learning_rate": 0.00039944174410990667,
      "loss": 5.3154,
      "mean_token_accuracy": 0.21420885175466536,
      "num_tokens": 67444807.0,
      "step": 36180
    },
    {
      "entropy": 5.957475900650024,
      "epoch": 3.1954256446485343,
      "grad_norm": 1.265625,
      "learning_rate": 0.000399415506314184,
      "loss": 5.2578,
      "mean_token_accuracy": 0.21722030639648438,
      "num_tokens": 67454954.0,
      "step": 36185
    },
    {
      "entropy": 5.985105991363525,
      "epoch": 3.1958671847403743,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039938926608123235,
      "loss": 5.3012,
      "mean_token_accuracy": 0.21422597765922546,
      "num_tokens": 67464600.0,
      "step": 36190
    },
    {
      "entropy": 5.99470157623291,
      "epoch": 3.196308724832215,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039936302341156556,
      "loss": 5.3654,
      "mean_token_accuracy": 0.2123486191034317,
      "num_tokens": 67473970.0,
      "step": 36195
    },
    {
      "entropy": 5.969054889678955,
      "epoch": 3.196750264924055,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00039933677830569797,
      "loss": 5.2055,
      "mean_token_accuracy": 0.222606560587883,
      "num_tokens": 67481984.0,
      "step": 36200
    },
    {
      "entropy": 5.943986225128174,
      "epoch": 3.1971918050158954,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003993105307641437,
      "loss": 5.2312,
      "mean_token_accuracy": 0.22472152411937713,
      "num_tokens": 67490988.0,
      "step": 36205
    },
    {
      "entropy": 5.998425245285034,
      "epoch": 3.197633345107736,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039928428078741667,
      "loss": 5.2269,
      "mean_token_accuracy": 0.21568318009376525,
      "num_tokens": 67500091.0,
      "step": 36210
    },
    {
      "entropy": 5.981442785263061,
      "epoch": 3.198074885199576,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039925802837603133,
      "loss": 5.2429,
      "mean_token_accuracy": 0.21227175146341323,
      "num_tokens": 67510237.0,
      "step": 36215
    },
    {
      "entropy": 5.916167545318603,
      "epoch": 3.1985164252914164,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039923177353050186,
      "loss": 5.2369,
      "mean_token_accuracy": 0.21085310727357864,
      "num_tokens": 67520218.0,
      "step": 36220
    },
    {
      "entropy": 5.894588947296143,
      "epoch": 3.198957965383257,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003992055162513426,
      "loss": 5.2205,
      "mean_token_accuracy": 0.21777189075946807,
      "num_tokens": 67528980.0,
      "step": 36225
    },
    {
      "entropy": 5.9884827613830565,
      "epoch": 3.199399505475097,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003991792565390679,
      "loss": 5.3428,
      "mean_token_accuracy": 0.20801330506801605,
      "num_tokens": 67539829.0,
      "step": 36230
    },
    {
      "entropy": 6.0701251983642575,
      "epoch": 3.1998410455669375,
      "grad_norm": 1.25,
      "learning_rate": 0.00039915299439419226,
      "loss": 5.2776,
      "mean_token_accuracy": 0.21249369233846666,
      "num_tokens": 67549643.0,
      "step": 36235
    },
    {
      "entropy": 6.0223612785339355,
      "epoch": 3.200282585658778,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039912672981723006,
      "loss": 5.2184,
      "mean_token_accuracy": 0.2175995573401451,
      "num_tokens": 67558384.0,
      "step": 36240
    },
    {
      "entropy": 5.9910365581512455,
      "epoch": 3.200724125750618,
      "grad_norm": 1.25,
      "learning_rate": 0.0003991004628086958,
      "loss": 5.3352,
      "mean_token_accuracy": 0.21024331152439119,
      "num_tokens": 67567399.0,
      "step": 36245
    },
    {
      "entropy": 5.980331325531006,
      "epoch": 3.2011656658424585,
      "grad_norm": 1.25,
      "learning_rate": 0.0003990741933691041,
      "loss": 5.2363,
      "mean_token_accuracy": 0.21904250979423523,
      "num_tokens": 67576763.0,
      "step": 36250
    },
    {
      "entropy": 6.015458059310913,
      "epoch": 3.201607205934299,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039904792149896945,
      "loss": 5.3124,
      "mean_token_accuracy": 0.21270940601825714,
      "num_tokens": 67586527.0,
      "step": 36255
    },
    {
      "entropy": 5.9184845924377445,
      "epoch": 3.202048746026139,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003990216471988066,
      "loss": 5.2769,
      "mean_token_accuracy": 0.22009015679359437,
      "num_tokens": 67595624.0,
      "step": 36260
    },
    {
      "entropy": 5.994451475143433,
      "epoch": 3.2024902861179796,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003989953704691302,
      "loss": 5.1764,
      "mean_token_accuracy": 0.22688985168933867,
      "num_tokens": 67604194.0,
      "step": 36265
    },
    {
      "entropy": 6.012193822860718,
      "epoch": 3.2029318262098196,
      "grad_norm": 1.1875,
      "learning_rate": 0.000398969091310455,
      "loss": 5.2816,
      "mean_token_accuracy": 0.21460311710834504,
      "num_tokens": 67614262.0,
      "step": 36270
    },
    {
      "entropy": 6.077952146530151,
      "epoch": 3.20337336630166,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039894280972329585,
      "loss": 5.3762,
      "mean_token_accuracy": 0.2014770731329918,
      "num_tokens": 67622891.0,
      "step": 36275
    },
    {
      "entropy": 5.9244279861450195,
      "epoch": 3.2038149063935006,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039891652570816746,
      "loss": 5.1608,
      "mean_token_accuracy": 0.22275918275117873,
      "num_tokens": 67631756.0,
      "step": 36280
    },
    {
      "entropy": 5.937618017196655,
      "epoch": 3.2042564464853407,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003988902392655849,
      "loss": 5.2916,
      "mean_token_accuracy": 0.218116095662117,
      "num_tokens": 67639988.0,
      "step": 36285
    },
    {
      "entropy": 5.921435880661011,
      "epoch": 3.204697986577181,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039886395039606296,
      "loss": 5.24,
      "mean_token_accuracy": 0.21244366765022277,
      "num_tokens": 67648523.0,
      "step": 36290
    },
    {
      "entropy": 5.95641450881958,
      "epoch": 3.2051395266690217,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003988376591001167,
      "loss": 5.2036,
      "mean_token_accuracy": 0.21936520785093308,
      "num_tokens": 67658028.0,
      "step": 36295
    },
    {
      "entropy": 5.997090864181518,
      "epoch": 3.2055810667608617,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003988113653782611,
      "loss": 5.3136,
      "mean_token_accuracy": 0.21107933819293975,
      "num_tokens": 67667015.0,
      "step": 36300
    },
    {
      "entropy": 5.994421768188476,
      "epoch": 3.2060226068527022,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003987850692310113,
      "loss": 5.2353,
      "mean_token_accuracy": 0.21960291564464568,
      "num_tokens": 67675168.0,
      "step": 36305
    },
    {
      "entropy": 5.992223787307739,
      "epoch": 3.2064641469445427,
      "grad_norm": 1.390625,
      "learning_rate": 0.00039875877065888234,
      "loss": 5.2337,
      "mean_token_accuracy": 0.2171051248908043,
      "num_tokens": 67683440.0,
      "step": 36310
    },
    {
      "entropy": 5.920444202423096,
      "epoch": 3.206905687036383,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003987324696623895,
      "loss": 5.215,
      "mean_token_accuracy": 0.2259303778409958,
      "num_tokens": 67692730.0,
      "step": 36315
    },
    {
      "entropy": 5.9803447246551515,
      "epoch": 3.2073472271282233,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039870616624204804,
      "loss": 5.3281,
      "mean_token_accuracy": 0.2074700713157654,
      "num_tokens": 67703966.0,
      "step": 36320
    },
    {
      "entropy": 6.000200176239014,
      "epoch": 3.207788767220064,
      "grad_norm": 1.5703125,
      "learning_rate": 0.000398679860398373,
      "loss": 5.2381,
      "mean_token_accuracy": 0.21332917660474776,
      "num_tokens": 67712895.0,
      "step": 36325
    },
    {
      "entropy": 6.019316005706787,
      "epoch": 3.208230307311904,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003986535521318799,
      "loss": 5.2644,
      "mean_token_accuracy": 0.21804881244897842,
      "num_tokens": 67722152.0,
      "step": 36330
    },
    {
      "entropy": 5.9773167133331295,
      "epoch": 3.2086718474037443,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003986272414430839,
      "loss": 5.3401,
      "mean_token_accuracy": 0.2126815363764763,
      "num_tokens": 67732044.0,
      "step": 36335
    },
    {
      "entropy": 6.084668350219727,
      "epoch": 3.2091133874955844,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039860092833250075,
      "loss": 5.3199,
      "mean_token_accuracy": 0.2080204427242279,
      "num_tokens": 67741480.0,
      "step": 36340
    },
    {
      "entropy": 5.96791033744812,
      "epoch": 3.209554927587425,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039857461280064563,
      "loss": 5.2321,
      "mean_token_accuracy": 0.22001474499702453,
      "num_tokens": 67751928.0,
      "step": 36345
    },
    {
      "entropy": 5.948920726776123,
      "epoch": 3.2099964676792654,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039854829484803404,
      "loss": 5.221,
      "mean_token_accuracy": 0.22763042002916337,
      "num_tokens": 67760199.0,
      "step": 36350
    },
    {
      "entropy": 5.963639783859253,
      "epoch": 3.2104380077711054,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039852197447518174,
      "loss": 5.2026,
      "mean_token_accuracy": 0.22300264984369278,
      "num_tokens": 67770254.0,
      "step": 36355
    },
    {
      "entropy": 6.019513750076294,
      "epoch": 3.210879547862946,
      "grad_norm": 1.25,
      "learning_rate": 0.00039849565168260416,
      "loss": 5.2944,
      "mean_token_accuracy": 0.21785151958465576,
      "num_tokens": 67779703.0,
      "step": 36360
    },
    {
      "entropy": 5.880721473693848,
      "epoch": 3.2113210879547864,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000398469326470817,
      "loss": 5.2056,
      "mean_token_accuracy": 0.22101210802793503,
      "num_tokens": 67788850.0,
      "step": 36365
    },
    {
      "entropy": 5.969968557357788,
      "epoch": 3.2117626280466265,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039844299884033595,
      "loss": 5.2167,
      "mean_token_accuracy": 0.21787165254354476,
      "num_tokens": 67798072.0,
      "step": 36370
    },
    {
      "entropy": 5.96970944404602,
      "epoch": 3.212204168138467,
      "grad_norm": 1.25,
      "learning_rate": 0.00039841666879167684,
      "loss": 5.2389,
      "mean_token_accuracy": 0.2140273258090019,
      "num_tokens": 67806522.0,
      "step": 36375
    },
    {
      "entropy": 5.902454280853272,
      "epoch": 3.2126457082303075,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039839033632535525,
      "loss": 5.2285,
      "mean_token_accuracy": 0.2208619847893715,
      "num_tokens": 67815678.0,
      "step": 36380
    },
    {
      "entropy": 5.948849201202393,
      "epoch": 3.2130872483221475,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003983640014418872,
      "loss": 5.2223,
      "mean_token_accuracy": 0.22252450585365297,
      "num_tokens": 67825013.0,
      "step": 36385
    },
    {
      "entropy": 5.915253448486328,
      "epoch": 3.213528788413988,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003983376641417885,
      "loss": 5.1815,
      "mean_token_accuracy": 0.23232420086860656,
      "num_tokens": 67834062.0,
      "step": 36390
    },
    {
      "entropy": 6.015630578994751,
      "epoch": 3.2139703285058285,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003983113244255752,
      "loss": 5.3955,
      "mean_token_accuracy": 0.2037195608019829,
      "num_tokens": 67843361.0,
      "step": 36395
    },
    {
      "entropy": 5.9913270473480225,
      "epoch": 3.2144118685976686,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039828498229376315,
      "loss": 5.2895,
      "mean_token_accuracy": 0.21952436864376068,
      "num_tokens": 67852169.0,
      "step": 36400
    },
    {
      "entropy": 6.0551807403564455,
      "epoch": 3.214853408689509,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039825863774686836,
      "loss": 5.3902,
      "mean_token_accuracy": 0.20790864676237106,
      "num_tokens": 67862303.0,
      "step": 36405
    },
    {
      "entropy": 5.974743318557739,
      "epoch": 3.2152949487813496,
      "grad_norm": 1.28125,
      "learning_rate": 0.000398232290785407,
      "loss": 5.249,
      "mean_token_accuracy": 0.22085909992456437,
      "num_tokens": 67870879.0,
      "step": 36410
    },
    {
      "entropy": 5.954430389404297,
      "epoch": 3.2157364888731896,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039820594140989513,
      "loss": 5.2993,
      "mean_token_accuracy": 0.21206142753362656,
      "num_tokens": 67880613.0,
      "step": 36415
    },
    {
      "entropy": 5.945649766921997,
      "epoch": 3.21617802896503,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039817958962084886,
      "loss": 5.2219,
      "mean_token_accuracy": 0.21952070146799088,
      "num_tokens": 67890338.0,
      "step": 36420
    },
    {
      "entropy": 6.00754680633545,
      "epoch": 3.21661956905687,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003981532354187846,
      "loss": 5.2959,
      "mean_token_accuracy": 0.21111118644475937,
      "num_tokens": 67900620.0,
      "step": 36425
    },
    {
      "entropy": 6.005829334259033,
      "epoch": 3.2170611091487107,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039812687880421847,
      "loss": 5.3321,
      "mean_token_accuracy": 0.21910441815853118,
      "num_tokens": 67909823.0,
      "step": 36430
    },
    {
      "entropy": 6.012344121932983,
      "epoch": 3.217502649240551,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003981005197776668,
      "loss": 5.3599,
      "mean_token_accuracy": 0.21198814660310744,
      "num_tokens": 67919704.0,
      "step": 36435
    },
    {
      "entropy": 6.043837928771973,
      "epoch": 3.2179441893323912,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003980741583396461,
      "loss": 5.3075,
      "mean_token_accuracy": 0.2185260072350502,
      "num_tokens": 67929190.0,
      "step": 36440
    },
    {
      "entropy": 5.997927904129028,
      "epoch": 3.2183857294242317,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003980477944906724,
      "loss": 5.2697,
      "mean_token_accuracy": 0.2182636320590973,
      "num_tokens": 67938738.0,
      "step": 36445
    },
    {
      "entropy": 5.954585695266724,
      "epoch": 3.2188272695160722,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003980214282312625,
      "loss": 5.2698,
      "mean_token_accuracy": 0.21649525463581085,
      "num_tokens": 67947701.0,
      "step": 36450
    },
    {
      "entropy": 5.916094446182251,
      "epoch": 3.2192688096079123,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039799505956193275,
      "loss": 5.1782,
      "mean_token_accuracy": 0.2219523400068283,
      "num_tokens": 67956676.0,
      "step": 36455
    },
    {
      "entropy": 5.982527017593384,
      "epoch": 3.219710349699753,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039796868848319977,
      "loss": 5.2389,
      "mean_token_accuracy": 0.21813557744026185,
      "num_tokens": 67966832.0,
      "step": 36460
    },
    {
      "entropy": 5.946312761306762,
      "epoch": 3.2201518897915933,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003979423149955801,
      "loss": 5.2365,
      "mean_token_accuracy": 0.22046839594841003,
      "num_tokens": 67976305.0,
      "step": 36465
    },
    {
      "entropy": 5.905128383636475,
      "epoch": 3.2205934298834333,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003979159390995904,
      "loss": 5.1968,
      "mean_token_accuracy": 0.22174791991710663,
      "num_tokens": 67985153.0,
      "step": 36470
    },
    {
      "entropy": 5.979664754867554,
      "epoch": 3.221034969975274,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039788956079574736,
      "loss": 5.2893,
      "mean_token_accuracy": 0.21188875436782836,
      "num_tokens": 67994434.0,
      "step": 36475
    },
    {
      "entropy": 5.960645866394043,
      "epoch": 3.221476510067114,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039786318008456774,
      "loss": 5.369,
      "mean_token_accuracy": 0.2058834046125412,
      "num_tokens": 68004895.0,
      "step": 36480
    },
    {
      "entropy": 6.021360731124878,
      "epoch": 3.2219180501589544,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003978367969665682,
      "loss": 5.2676,
      "mean_token_accuracy": 0.21014028191566467,
      "num_tokens": 68014549.0,
      "step": 36485
    },
    {
      "entropy": 6.054633045196534,
      "epoch": 3.222359590250795,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003978104114422658,
      "loss": 5.2505,
      "mean_token_accuracy": 0.22020305842161178,
      "num_tokens": 68023586.0,
      "step": 36490
    },
    {
      "entropy": 5.997241878509522,
      "epoch": 3.222801130342635,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003977840235121772,
      "loss": 5.212,
      "mean_token_accuracy": 0.22043734043836594,
      "num_tokens": 68032695.0,
      "step": 36495
    },
    {
      "entropy": 6.035923814773559,
      "epoch": 3.2232426704344754,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039775763317681945,
      "loss": 5.2471,
      "mean_token_accuracy": 0.21311259418725967,
      "num_tokens": 68041093.0,
      "step": 36500
    },
    {
      "entropy": 6.042803525924683,
      "epoch": 3.223684210526316,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039773124043670957,
      "loss": 5.3189,
      "mean_token_accuracy": 0.2119938239455223,
      "num_tokens": 68051084.0,
      "step": 36505
    },
    {
      "entropy": 5.804147958755493,
      "epoch": 3.224125750618156,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039770484529236434,
      "loss": 5.1586,
      "mean_token_accuracy": 0.22428612858057023,
      "num_tokens": 68060595.0,
      "step": 36510
    },
    {
      "entropy": 5.9674952030181885,
      "epoch": 3.2245672907099965,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000397678447744301,
      "loss": 5.2117,
      "mean_token_accuracy": 0.22348068356513978,
      "num_tokens": 68069227.0,
      "step": 36515
    },
    {
      "entropy": 5.971409559249878,
      "epoch": 3.225008830801837,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003976520477930367,
      "loss": 5.2223,
      "mean_token_accuracy": 0.21774462908506392,
      "num_tokens": 68077790.0,
      "step": 36520
    },
    {
      "entropy": 5.880855369567871,
      "epoch": 3.225450370893677,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039762564543908854,
      "loss": 5.198,
      "mean_token_accuracy": 0.22641972601413726,
      "num_tokens": 68088004.0,
      "step": 36525
    },
    {
      "entropy": 6.0580305576324465,
      "epoch": 3.2258919109855175,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039759924068297366,
      "loss": 5.3098,
      "mean_token_accuracy": 0.21169273555278778,
      "num_tokens": 68098093.0,
      "step": 36530
    },
    {
      "entropy": 6.014315128326416,
      "epoch": 3.226333451077358,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003975728335252095,
      "loss": 5.2989,
      "mean_token_accuracy": 0.21179130375385286,
      "num_tokens": 68107605.0,
      "step": 36535
    },
    {
      "entropy": 5.962546682357788,
      "epoch": 3.226774991169198,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003975464239663131,
      "loss": 5.2649,
      "mean_token_accuracy": 0.21841962039470672,
      "num_tokens": 68116065.0,
      "step": 36540
    },
    {
      "entropy": 5.968414831161499,
      "epoch": 3.2272165312610386,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039752001200680197,
      "loss": 5.3107,
      "mean_token_accuracy": 0.2114235356450081,
      "num_tokens": 68126140.0,
      "step": 36545
    },
    {
      "entropy": 6.027139520645141,
      "epoch": 3.227658071352879,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039749359764719357,
      "loss": 5.2937,
      "mean_token_accuracy": 0.21271097660064697,
      "num_tokens": 68135880.0,
      "step": 36550
    },
    {
      "entropy": 6.02165584564209,
      "epoch": 3.228099611444719,
      "grad_norm": 1.25,
      "learning_rate": 0.0003974671808880052,
      "loss": 5.3229,
      "mean_token_accuracy": 0.20908619314432145,
      "num_tokens": 68145560.0,
      "step": 36555
    },
    {
      "entropy": 5.931200885772705,
      "epoch": 3.2285411515365596,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039744076172975446,
      "loss": 5.1694,
      "mean_token_accuracy": 0.22665969729423524,
      "num_tokens": 68154779.0,
      "step": 36560
    },
    {
      "entropy": 5.935416316986084,
      "epoch": 3.2289826916283997,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003974143401729587,
      "loss": 5.1908,
      "mean_token_accuracy": 0.22863437980413437,
      "num_tokens": 68164520.0,
      "step": 36565
    },
    {
      "entropy": 6.0169349193573,
      "epoch": 3.22942423172024,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003973879162181358,
      "loss": 5.2661,
      "mean_token_accuracy": 0.2169985666871071,
      "num_tokens": 68173055.0,
      "step": 36570
    },
    {
      "entropy": 6.018259143829345,
      "epoch": 3.2298657718120807,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039736148986580313,
      "loss": 5.3597,
      "mean_token_accuracy": 0.2131840854883194,
      "num_tokens": 68182885.0,
      "step": 36575
    },
    {
      "entropy": 6.019797229766846,
      "epoch": 3.2303073119039207,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003973350611164785,
      "loss": 5.304,
      "mean_token_accuracy": 0.2174667090177536,
      "num_tokens": 68191836.0,
      "step": 36580
    },
    {
      "entropy": 6.03419680595398,
      "epoch": 3.2307488519957612,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039730862997067965,
      "loss": 5.4254,
      "mean_token_accuracy": 0.20800370573997498,
      "num_tokens": 68201454.0,
      "step": 36585
    },
    {
      "entropy": 6.063929843902588,
      "epoch": 3.2311903920876017,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039728219642892423,
      "loss": 5.33,
      "mean_token_accuracy": 0.21359096020460128,
      "num_tokens": 68211581.0,
      "step": 36590
    },
    {
      "entropy": 6.010131740570069,
      "epoch": 3.231631932179442,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039725576049173014,
      "loss": 5.2572,
      "mean_token_accuracy": 0.2198995366692543,
      "num_tokens": 68220723.0,
      "step": 36595
    },
    {
      "entropy": 6.097632932662964,
      "epoch": 3.2320734722712823,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039722932215961515,
      "loss": 5.3311,
      "mean_token_accuracy": 0.209345506131649,
      "num_tokens": 68230017.0,
      "step": 36600
    },
    {
      "entropy": 6.011978530883789,
      "epoch": 3.2325150123631228,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039720288143309734,
      "loss": 5.1843,
      "mean_token_accuracy": 0.22663077861070632,
      "num_tokens": 68238220.0,
      "step": 36605
    },
    {
      "entropy": 5.983210706710816,
      "epoch": 3.232956552454963,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039717643831269463,
      "loss": 5.3336,
      "mean_token_accuracy": 0.20555115342140198,
      "num_tokens": 68248074.0,
      "step": 36610
    },
    {
      "entropy": 6.014733695983887,
      "epoch": 3.2333980925468033,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039714999279892495,
      "loss": 5.3136,
      "mean_token_accuracy": 0.210426364839077,
      "num_tokens": 68257520.0,
      "step": 36615
    },
    {
      "entropy": 6.020621204376221,
      "epoch": 3.2338396326386434,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003971235448923064,
      "loss": 5.2393,
      "mean_token_accuracy": 0.21546000689268113,
      "num_tokens": 68267055.0,
      "step": 36620
    },
    {
      "entropy": 6.017213916778564,
      "epoch": 3.234281172730484,
      "grad_norm": 1.234375,
      "learning_rate": 0.000397097094593357,
      "loss": 5.327,
      "mean_token_accuracy": 0.21233258843421937,
      "num_tokens": 68276944.0,
      "step": 36625
    },
    {
      "entropy": 5.9833015441894535,
      "epoch": 3.2347227128223244,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039707064190259495,
      "loss": 5.2633,
      "mean_token_accuracy": 0.21549321115016937,
      "num_tokens": 68285300.0,
      "step": 36630
    },
    {
      "entropy": 5.99281268119812,
      "epoch": 3.2351642529141644,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039704418682053847,
      "loss": 5.2786,
      "mean_token_accuracy": 0.20864466875791549,
      "num_tokens": 68294797.0,
      "step": 36635
    },
    {
      "entropy": 6.0106734275817875,
      "epoch": 3.235605793006005,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003970177293477058,
      "loss": 5.2359,
      "mean_token_accuracy": 0.22095174640417098,
      "num_tokens": 68303618.0,
      "step": 36640
    },
    {
      "entropy": 5.988797378540039,
      "epoch": 3.2360473330978454,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003969912694846152,
      "loss": 5.3204,
      "mean_token_accuracy": 0.22160472422838212,
      "num_tokens": 68313313.0,
      "step": 36645
    },
    {
      "entropy": 5.981370735168457,
      "epoch": 3.2364888731896855,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039696480723178497,
      "loss": 5.1946,
      "mean_token_accuracy": 0.22921810001134874,
      "num_tokens": 68321384.0,
      "step": 36650
    },
    {
      "entropy": 5.913290405273438,
      "epoch": 3.236930413281526,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003969383425897336,
      "loss": 5.2213,
      "mean_token_accuracy": 0.21978979855775832,
      "num_tokens": 68330427.0,
      "step": 36655
    },
    {
      "entropy": 5.977771949768067,
      "epoch": 3.2373719533733665,
      "grad_norm": 1.390625,
      "learning_rate": 0.00039691187555897944,
      "loss": 5.2387,
      "mean_token_accuracy": 0.22066641002893447,
      "num_tokens": 68339097.0,
      "step": 36660
    },
    {
      "entropy": 5.949122142791748,
      "epoch": 3.2378134934652065,
      "grad_norm": 1.25,
      "learning_rate": 0.0003968854061400409,
      "loss": 5.2255,
      "mean_token_accuracy": 0.22163910418748856,
      "num_tokens": 68349123.0,
      "step": 36665
    },
    {
      "entropy": 5.966205692291259,
      "epoch": 3.238255033557047,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003968589343334366,
      "loss": 5.2055,
      "mean_token_accuracy": 0.2160753756761551,
      "num_tokens": 68359043.0,
      "step": 36670
    },
    {
      "entropy": 6.032446384429932,
      "epoch": 3.2386965736488875,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039683246013968516,
      "loss": 5.3265,
      "mean_token_accuracy": 0.21044713258743286,
      "num_tokens": 68369103.0,
      "step": 36675
    },
    {
      "entropy": 6.064058399200439,
      "epoch": 3.2391381137407276,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003968059835593051,
      "loss": 5.4192,
      "mean_token_accuracy": 0.2006993517279625,
      "num_tokens": 68379010.0,
      "step": 36680
    },
    {
      "entropy": 6.020754814147949,
      "epoch": 3.239579653832568,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039677950459281495,
      "loss": 5.2774,
      "mean_token_accuracy": 0.20778267383575438,
      "num_tokens": 68388920.0,
      "step": 36685
    },
    {
      "entropy": 6.037098932266235,
      "epoch": 3.2400211939244086,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039675302324073373,
      "loss": 5.2799,
      "mean_token_accuracy": 0.21399130076169967,
      "num_tokens": 68397870.0,
      "step": 36690
    },
    {
      "entropy": 5.964752149581909,
      "epoch": 3.2404627340162486,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039672653950358,
      "loss": 5.2636,
      "mean_token_accuracy": 0.21833497285842896,
      "num_tokens": 68407948.0,
      "step": 36695
    },
    {
      "entropy": 5.975012540817261,
      "epoch": 3.240904274108089,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039670005338187254,
      "loss": 5.338,
      "mean_token_accuracy": 0.20829940736293792,
      "num_tokens": 68416863.0,
      "step": 36700
    },
    {
      "entropy": 6.013595676422119,
      "epoch": 3.241345814199929,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039667356487613025,
      "loss": 5.2953,
      "mean_token_accuracy": 0.21722155660390854,
      "num_tokens": 68425202.0,
      "step": 36705
    },
    {
      "entropy": 6.1152870655059814,
      "epoch": 3.2417873542917697,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003966470739868721,
      "loss": 5.3544,
      "mean_token_accuracy": 0.21192027926445006,
      "num_tokens": 68434693.0,
      "step": 36710
    },
    {
      "entropy": 6.007962322235107,
      "epoch": 3.24222889438361,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039662058071461684,
      "loss": 5.2583,
      "mean_token_accuracy": 0.22083498239517213,
      "num_tokens": 68444740.0,
      "step": 36715
    },
    {
      "entropy": 5.8963768005371096,
      "epoch": 3.2426704344754502,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003965940850598836,
      "loss": 5.1755,
      "mean_token_accuracy": 0.22560936063528061,
      "num_tokens": 68453411.0,
      "step": 36720
    },
    {
      "entropy": 5.990042877197266,
      "epoch": 3.2431119745672907,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003965675870231915,
      "loss": 5.3025,
      "mean_token_accuracy": 0.21469911187887192,
      "num_tokens": 68462295.0,
      "step": 36725
    },
    {
      "entropy": 5.979489660263061,
      "epoch": 3.243553514659131,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039654108660505937,
      "loss": 5.2865,
      "mean_token_accuracy": 0.21132988333702088,
      "num_tokens": 68472031.0,
      "step": 36730
    },
    {
      "entropy": 5.916295719146729,
      "epoch": 3.2439950547509713,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003965145838060066,
      "loss": 5.1451,
      "mean_token_accuracy": 0.2267352283000946,
      "num_tokens": 68481683.0,
      "step": 36735
    },
    {
      "entropy": 5.958977174758911,
      "epoch": 3.2444365948428118,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003964880786265522,
      "loss": 5.2436,
      "mean_token_accuracy": 0.21035130620002745,
      "num_tokens": 68491207.0,
      "step": 36740
    },
    {
      "entropy": 6.032250022888183,
      "epoch": 3.2448781349346523,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039646157106721535,
      "loss": 5.2745,
      "mean_token_accuracy": 0.21385041624307632,
      "num_tokens": 68501462.0,
      "step": 36745
    },
    {
      "entropy": 5.939788484573365,
      "epoch": 3.2453196750264923,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039643506112851544,
      "loss": 5.1015,
      "mean_token_accuracy": 0.22676249593496323,
      "num_tokens": 68509922.0,
      "step": 36750
    },
    {
      "entropy": 5.866763401031494,
      "epoch": 3.245761215118333,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003964085488109718,
      "loss": 5.1345,
      "mean_token_accuracy": 0.23278581351041794,
      "num_tokens": 68518839.0,
      "step": 36755
    },
    {
      "entropy": 5.942041683197021,
      "epoch": 3.246202755210173,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039638203411510373,
      "loss": 5.239,
      "mean_token_accuracy": 0.21989279091358185,
      "num_tokens": 68528298.0,
      "step": 36760
    },
    {
      "entropy": 5.961031818389893,
      "epoch": 3.2466442953020134,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00039635551704143065,
      "loss": 5.1571,
      "mean_token_accuracy": 0.22617117911577225,
      "num_tokens": 68536638.0,
      "step": 36765
    },
    {
      "entropy": 5.955879783630371,
      "epoch": 3.247085835393854,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003963289975904719,
      "loss": 5.2845,
      "mean_token_accuracy": 0.20861005187034606,
      "num_tokens": 68545700.0,
      "step": 36770
    },
    {
      "entropy": 5.89948148727417,
      "epoch": 3.247527375485694,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003963024757627472,
      "loss": 5.2057,
      "mean_token_accuracy": 0.2177538439631462,
      "num_tokens": 68555312.0,
      "step": 36775
    },
    {
      "entropy": 6.04776349067688,
      "epoch": 3.2479689155775344,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039627595155877597,
      "loss": 5.324,
      "mean_token_accuracy": 0.20562739074230194,
      "num_tokens": 68563543.0,
      "step": 36780
    },
    {
      "entropy": 6.008139753341675,
      "epoch": 3.248410455669375,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003962494249790779,
      "loss": 5.3153,
      "mean_token_accuracy": 0.2140966236591339,
      "num_tokens": 68573111.0,
      "step": 36785
    },
    {
      "entropy": 5.933464002609253,
      "epoch": 3.248851995761215,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003962228960241724,
      "loss": 5.2206,
      "mean_token_accuracy": 0.21797924041748046,
      "num_tokens": 68582106.0,
      "step": 36790
    },
    {
      "entropy": 6.064051866531372,
      "epoch": 3.2492935358530555,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003961963646945794,
      "loss": 5.3783,
      "mean_token_accuracy": 0.20687713772058486,
      "num_tokens": 68592054.0,
      "step": 36795
    },
    {
      "entropy": 6.058645582199096,
      "epoch": 3.249735075944896,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039616983099081856,
      "loss": 5.3084,
      "mean_token_accuracy": 0.2211517333984375,
      "num_tokens": 68600745.0,
      "step": 36800
    },
    {
      "entropy": 5.937452602386474,
      "epoch": 3.250176616036736,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039614329491340956,
      "loss": 5.1346,
      "mean_token_accuracy": 0.2297213450074196,
      "num_tokens": 68610174.0,
      "step": 36805
    },
    {
      "entropy": 5.898287057876587,
      "epoch": 3.2506181561285765,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039611675646287234,
      "loss": 5.2483,
      "mean_token_accuracy": 0.22211646735668183,
      "num_tokens": 68619945.0,
      "step": 36810
    },
    {
      "entropy": 5.942168045043945,
      "epoch": 3.251059696220417,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003960902156397268,
      "loss": 5.1996,
      "mean_token_accuracy": 0.22048505693674086,
      "num_tokens": 68629095.0,
      "step": 36815
    },
    {
      "entropy": 5.882517671585083,
      "epoch": 3.251501236312257,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039606367244449277,
      "loss": 5.1611,
      "mean_token_accuracy": 0.23124985992908478,
      "num_tokens": 68638866.0,
      "step": 36820
    },
    {
      "entropy": 5.9153917789459225,
      "epoch": 3.2519427764040976,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039603712687769023,
      "loss": 5.2232,
      "mean_token_accuracy": 0.2216994047164917,
      "num_tokens": 68648366.0,
      "step": 36825
    },
    {
      "entropy": 6.018566036224366,
      "epoch": 3.252384316495938,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039601057893983924,
      "loss": 5.2648,
      "mean_token_accuracy": 0.2244268223643303,
      "num_tokens": 68657000.0,
      "step": 36830
    },
    {
      "entropy": 6.012121677398682,
      "epoch": 3.252825856587778,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003959840286314598,
      "loss": 5.2607,
      "mean_token_accuracy": 0.21501076221466064,
      "num_tokens": 68666518.0,
      "step": 36835
    },
    {
      "entropy": 5.9626220703125,
      "epoch": 3.2532673966796186,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003959574759530721,
      "loss": 5.2579,
      "mean_token_accuracy": 0.2111723691225052,
      "num_tokens": 68675361.0,
      "step": 36840
    },
    {
      "entropy": 5.9566868305206295,
      "epoch": 3.2537089367714587,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003959309209051962,
      "loss": 5.2512,
      "mean_token_accuracy": 0.220243039727211,
      "num_tokens": 68684363.0,
      "step": 36845
    },
    {
      "entropy": 6.009092092514038,
      "epoch": 3.254150476863299,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039590436348835236,
      "loss": 5.2864,
      "mean_token_accuracy": 0.21560091525316238,
      "num_tokens": 68692856.0,
      "step": 36850
    },
    {
      "entropy": 6.034383153915405,
      "epoch": 3.2545920169551397,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039587780370306075,
      "loss": 5.2409,
      "mean_token_accuracy": 0.215257728099823,
      "num_tokens": 68701736.0,
      "step": 36855
    },
    {
      "entropy": 5.959290266036987,
      "epoch": 3.2550335570469797,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003958512415498417,
      "loss": 5.3028,
      "mean_token_accuracy": 0.214278145134449,
      "num_tokens": 68710936.0,
      "step": 36860
    },
    {
      "entropy": 6.007812213897705,
      "epoch": 3.25547509713882,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003958246770292156,
      "loss": 5.3194,
      "mean_token_accuracy": 0.21177174299955367,
      "num_tokens": 68720189.0,
      "step": 36865
    },
    {
      "entropy": 5.949674463272094,
      "epoch": 3.2559166372306607,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003957981101417028,
      "loss": 5.2623,
      "mean_token_accuracy": 0.2112271413207054,
      "num_tokens": 68729916.0,
      "step": 36870
    },
    {
      "entropy": 6.007785844802856,
      "epoch": 3.2563581773225008,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003957715408878237,
      "loss": 5.3046,
      "mean_token_accuracy": 0.20754634588956833,
      "num_tokens": 68739082.0,
      "step": 36875
    },
    {
      "entropy": 5.989293813705444,
      "epoch": 3.2567997174143413,
      "grad_norm": 1.25,
      "learning_rate": 0.00039574496926809885,
      "loss": 5.3319,
      "mean_token_accuracy": 0.2086064413189888,
      "num_tokens": 68748529.0,
      "step": 36880
    },
    {
      "entropy": 5.874357557296753,
      "epoch": 3.2572412575061813,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003957183952830487,
      "loss": 5.0809,
      "mean_token_accuracy": 0.24046206325292588,
      "num_tokens": 68757757.0,
      "step": 36885
    },
    {
      "entropy": 5.898354530334473,
      "epoch": 3.257682797598022,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039569181893319376,
      "loss": 5.1139,
      "mean_token_accuracy": 0.23264981657266617,
      "num_tokens": 68766825.0,
      "step": 36890
    },
    {
      "entropy": 5.967986249923706,
      "epoch": 3.2581243376898623,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003956652402190549,
      "loss": 5.2636,
      "mean_token_accuracy": 0.21830902099609376,
      "num_tokens": 68775683.0,
      "step": 36895
    },
    {
      "entropy": 5.969089508056641,
      "epoch": 3.2585658777817024,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003956386591411525,
      "loss": 5.2351,
      "mean_token_accuracy": 0.21438551545143128,
      "num_tokens": 68784567.0,
      "step": 36900
    },
    {
      "entropy": 5.961730098724365,
      "epoch": 3.259007417873543,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039561207570000734,
      "loss": 5.329,
      "mean_token_accuracy": 0.2120150864124298,
      "num_tokens": 68793971.0,
      "step": 36905
    },
    {
      "entropy": 6.013757801055908,
      "epoch": 3.2594489579653834,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003955854898961403,
      "loss": 5.3503,
      "mean_token_accuracy": 0.21001618057489396,
      "num_tokens": 68803354.0,
      "step": 36910
    },
    {
      "entropy": 5.989643812179565,
      "epoch": 3.2598904980572234,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000395558901730072,
      "loss": 5.2248,
      "mean_token_accuracy": 0.21554286181926727,
      "num_tokens": 68812169.0,
      "step": 36915
    },
    {
      "entropy": 5.909684324264527,
      "epoch": 3.260332038149064,
      "grad_norm": 1.25,
      "learning_rate": 0.00039553231120232355,
      "loss": 5.1541,
      "mean_token_accuracy": 0.22430993765592575,
      "num_tokens": 68821242.0,
      "step": 36920
    },
    {
      "entropy": 6.006958627700806,
      "epoch": 3.2607735782409044,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003955057183134155,
      "loss": 5.2795,
      "mean_token_accuracy": 0.21418847590684892,
      "num_tokens": 68831841.0,
      "step": 36925
    },
    {
      "entropy": 6.0323080062866214,
      "epoch": 3.2612151183327445,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000395479123063869,
      "loss": 5.338,
      "mean_token_accuracy": 0.2072613298892975,
      "num_tokens": 68841439.0,
      "step": 36930
    },
    {
      "entropy": 5.966203737258911,
      "epoch": 3.261656658424585,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039545252545420507,
      "loss": 5.2071,
      "mean_token_accuracy": 0.21929606348276137,
      "num_tokens": 68851619.0,
      "step": 36935
    },
    {
      "entropy": 5.9751029968261715,
      "epoch": 3.2620981985164255,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003954259254849447,
      "loss": 5.1864,
      "mean_token_accuracy": 0.21969714313745498,
      "num_tokens": 68859955.0,
      "step": 36940
    },
    {
      "entropy": 5.998754501342773,
      "epoch": 3.2625397386082655,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003953993231566088,
      "loss": 5.2624,
      "mean_token_accuracy": 0.218265999853611,
      "num_tokens": 68868941.0,
      "step": 36945
    },
    {
      "entropy": 6.016767454147339,
      "epoch": 3.262981278700106,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003953727184697187,
      "loss": 5.3394,
      "mean_token_accuracy": 0.2103082373738289,
      "num_tokens": 68877096.0,
      "step": 36950
    },
    {
      "entropy": 5.969484663009643,
      "epoch": 3.2634228187919465,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003953461114247954,
      "loss": 5.2782,
      "mean_token_accuracy": 0.21296564638614654,
      "num_tokens": 68886791.0,
      "step": 36955
    },
    {
      "entropy": 5.9946129322052,
      "epoch": 3.2638643588837866,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039531950202236046,
      "loss": 5.2299,
      "mean_token_accuracy": 0.22257154881954194,
      "num_tokens": 68896953.0,
      "step": 36960
    },
    {
      "entropy": 6.003227376937867,
      "epoch": 3.264305898975627,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003952928902629347,
      "loss": 5.2546,
      "mean_token_accuracy": 0.2099197432398796,
      "num_tokens": 68905221.0,
      "step": 36965
    },
    {
      "entropy": 6.001596689224243,
      "epoch": 3.2647474390674676,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039526627614703977,
      "loss": 5.3094,
      "mean_token_accuracy": 0.20775667876005172,
      "num_tokens": 68915880.0,
      "step": 36970
    },
    {
      "entropy": 5.936873245239258,
      "epoch": 3.2651889791593076,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003952396596751968,
      "loss": 5.0989,
      "mean_token_accuracy": 0.2304267555475235,
      "num_tokens": 68924821.0,
      "step": 36975
    },
    {
      "entropy": 6.008468818664551,
      "epoch": 3.265630519251148,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003952130408479273,
      "loss": 5.3014,
      "mean_token_accuracy": 0.2096147805452347,
      "num_tokens": 68934409.0,
      "step": 36980
    },
    {
      "entropy": 5.865411806106567,
      "epoch": 3.266072059342988,
      "grad_norm": 1.25,
      "learning_rate": 0.00039518641966575273,
      "loss": 5.2016,
      "mean_token_accuracy": 0.226019486784935,
      "num_tokens": 68943165.0,
      "step": 36985
    },
    {
      "entropy": 6.031565761566162,
      "epoch": 3.2665135994348287,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003951597961291945,
      "loss": 5.3647,
      "mean_token_accuracy": 0.2036270394921303,
      "num_tokens": 68954300.0,
      "step": 36990
    },
    {
      "entropy": 6.028769207000733,
      "epoch": 3.266955139526669,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039513317023877424,
      "loss": 5.3408,
      "mean_token_accuracy": 0.21009333580732345,
      "num_tokens": 68963658.0,
      "step": 36995
    },
    {
      "entropy": 6.140779447555542,
      "epoch": 3.267396679618509,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003951065419950134,
      "loss": 5.4379,
      "mean_token_accuracy": 0.20502290278673171,
      "num_tokens": 68974478.0,
      "step": 37000
    },
    {
      "entropy": 6.001426362991333,
      "epoch": 3.2678382197103497,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039507991139843383,
      "loss": 5.3,
      "mean_token_accuracy": 0.21667294800281525,
      "num_tokens": 68984612.0,
      "step": 37005
    },
    {
      "entropy": 6.073200225830078,
      "epoch": 3.26827975980219,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039505327844955706,
      "loss": 5.341,
      "mean_token_accuracy": 0.20606469064950944,
      "num_tokens": 68994434.0,
      "step": 37010
    },
    {
      "entropy": 5.958566665649414,
      "epoch": 3.2687212998940303,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003950266431489049,
      "loss": 5.2299,
      "mean_token_accuracy": 0.22427428364753724,
      "num_tokens": 69003691.0,
      "step": 37015
    },
    {
      "entropy": 5.980310487747192,
      "epoch": 3.2691628399858708,
      "grad_norm": 1.25,
      "learning_rate": 0.0003950000054969989,
      "loss": 5.3008,
      "mean_token_accuracy": 0.21981224417686462,
      "num_tokens": 69012814.0,
      "step": 37020
    },
    {
      "entropy": 5.974028205871582,
      "epoch": 3.269604380077711,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039497336549436124,
      "loss": 5.2527,
      "mean_token_accuracy": 0.22313983887434005,
      "num_tokens": 69021971.0,
      "step": 37025
    },
    {
      "entropy": 6.062710285186768,
      "epoch": 3.2700459201695513,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003949467231415135,
      "loss": 5.3145,
      "mean_token_accuracy": 0.2118724465370178,
      "num_tokens": 69031760.0,
      "step": 37030
    },
    {
      "entropy": 6.036439704895019,
      "epoch": 3.270487460261392,
      "grad_norm": 1.25,
      "learning_rate": 0.00039492007843897764,
      "loss": 5.337,
      "mean_token_accuracy": 0.20822500586509704,
      "num_tokens": 69040919.0,
      "step": 37035
    },
    {
      "entropy": 5.904181385040284,
      "epoch": 3.270929000353232,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003948934313872757,
      "loss": 5.248,
      "mean_token_accuracy": 0.22527816891670227,
      "num_tokens": 69049735.0,
      "step": 37040
    },
    {
      "entropy": 6.057449865341186,
      "epoch": 3.2713705404450724,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039486678198692954,
      "loss": 5.3776,
      "mean_token_accuracy": 0.20584150403738022,
      "num_tokens": 69059063.0,
      "step": 37045
    },
    {
      "entropy": 6.016906833648681,
      "epoch": 3.271812080536913,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003948401302384614,
      "loss": 5.3011,
      "mean_token_accuracy": 0.21573506146669388,
      "num_tokens": 69068851.0,
      "step": 37050
    },
    {
      "entropy": 6.009575271606446,
      "epoch": 3.272253620628753,
      "grad_norm": 1.5625,
      "learning_rate": 0.00039481347614239313,
      "loss": 5.2068,
      "mean_token_accuracy": 0.22009219527244567,
      "num_tokens": 69078646.0,
      "step": 37055
    },
    {
      "entropy": 5.924802255630493,
      "epoch": 3.2726951607205934,
      "grad_norm": 1.25,
      "learning_rate": 0.0003947868196992471,
      "loss": 5.2352,
      "mean_token_accuracy": 0.21860019862651825,
      "num_tokens": 69087166.0,
      "step": 37060
    },
    {
      "entropy": 5.927752351760864,
      "epoch": 3.273136700812434,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003947601609095454,
      "loss": 5.2846,
      "mean_token_accuracy": 0.22102758437395095,
      "num_tokens": 69096975.0,
      "step": 37065
    },
    {
      "entropy": 5.97263994216919,
      "epoch": 3.273578240904274,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003947334997738103,
      "loss": 5.2483,
      "mean_token_accuracy": 0.21940518468618392,
      "num_tokens": 69106369.0,
      "step": 37070
    },
    {
      "entropy": 5.9098639488220215,
      "epoch": 3.2740197809961145,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003947068362925639,
      "loss": 5.1989,
      "mean_token_accuracy": 0.22746925204992294,
      "num_tokens": 69115807.0,
      "step": 37075
    },
    {
      "entropy": 6.012418413162232,
      "epoch": 3.274461321087955,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003946801704663288,
      "loss": 5.3511,
      "mean_token_accuracy": 0.21316528022289277,
      "num_tokens": 69125552.0,
      "step": 37080
    },
    {
      "entropy": 5.985368013381958,
      "epoch": 3.274902861179795,
      "grad_norm": 1.375,
      "learning_rate": 0.0003946535022956271,
      "loss": 5.2916,
      "mean_token_accuracy": 0.21559430211782454,
      "num_tokens": 69134792.0,
      "step": 37085
    },
    {
      "entropy": 6.089306306838989,
      "epoch": 3.2753444012716355,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003946268317809814,
      "loss": 5.3078,
      "mean_token_accuracy": 0.2088360905647278,
      "num_tokens": 69144074.0,
      "step": 37090
    },
    {
      "entropy": 5.982033014297485,
      "epoch": 3.275785941363476,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003946001589229141,
      "loss": 5.2211,
      "mean_token_accuracy": 0.224603071808815,
      "num_tokens": 69153448.0,
      "step": 37095
    },
    {
      "entropy": 5.886073923110962,
      "epoch": 3.276227481455316,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039457348372194776,
      "loss": 5.2165,
      "mean_token_accuracy": 0.22356079369783402,
      "num_tokens": 69162915.0,
      "step": 37100
    },
    {
      "entropy": 5.932465934753418,
      "epoch": 3.2766690215471566,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003945468061786048,
      "loss": 5.142,
      "mean_token_accuracy": 0.22639501392841338,
      "num_tokens": 69172611.0,
      "step": 37105
    },
    {
      "entropy": 5.992615509033203,
      "epoch": 3.277110561638997,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003945201262934079,
      "loss": 5.3357,
      "mean_token_accuracy": 0.21019552201032637,
      "num_tokens": 69182347.0,
      "step": 37110
    },
    {
      "entropy": 5.923898935317993,
      "epoch": 3.277552101730837,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003944934440668797,
      "loss": 5.2218,
      "mean_token_accuracy": 0.22598378658294677,
      "num_tokens": 69191223.0,
      "step": 37115
    },
    {
      "entropy": 6.045789289474487,
      "epoch": 3.2779936418226776,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003944667594995429,
      "loss": 5.3869,
      "mean_token_accuracy": 0.2047243371605873,
      "num_tokens": 69201037.0,
      "step": 37120
    },
    {
      "entropy": 5.954840421676636,
      "epoch": 3.2784351819145177,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003944400725919203,
      "loss": 5.1382,
      "mean_token_accuracy": 0.22598475366830825,
      "num_tokens": 69210846.0,
      "step": 37125
    },
    {
      "entropy": 5.917153692245483,
      "epoch": 3.278876722006358,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003944133833445345,
      "loss": 5.2675,
      "mean_token_accuracy": 0.2198271483182907,
      "num_tokens": 69219696.0,
      "step": 37130
    },
    {
      "entropy": 6.010452651977539,
      "epoch": 3.2793182620981987,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039438669175790854,
      "loss": 5.2593,
      "mean_token_accuracy": 0.2072984278202057,
      "num_tokens": 69228715.0,
      "step": 37135
    },
    {
      "entropy": 6.025296068191528,
      "epoch": 3.2797598021900387,
      "grad_norm": 1.125,
      "learning_rate": 0.00039435999783256505,
      "loss": 5.2363,
      "mean_token_accuracy": 0.2089283734560013,
      "num_tokens": 69238501.0,
      "step": 37140
    },
    {
      "entropy": 5.9550614833831785,
      "epoch": 3.280201342281879,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003943333015690272,
      "loss": 5.2601,
      "mean_token_accuracy": 0.2127617433667183,
      "num_tokens": 69247685.0,
      "step": 37145
    },
    {
      "entropy": 5.9405608654022215,
      "epoch": 3.2806428823737197,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003943066029678178,
      "loss": 5.2682,
      "mean_token_accuracy": 0.21274785846471786,
      "num_tokens": 69256556.0,
      "step": 37150
    },
    {
      "entropy": 5.883362436294556,
      "epoch": 3.2810844224655598,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003942799020294599,
      "loss": 5.1776,
      "mean_token_accuracy": 0.22034913003444673,
      "num_tokens": 69264879.0,
      "step": 37155
    },
    {
      "entropy": 5.9848345756530765,
      "epoch": 3.2815259625574003,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039425319875447664,
      "loss": 5.3146,
      "mean_token_accuracy": 0.20870003551244737,
      "num_tokens": 69273462.0,
      "step": 37160
    },
    {
      "entropy": 6.042672204971313,
      "epoch": 3.2819675026492403,
      "grad_norm": 1.359375,
      "learning_rate": 0.00039422649314339094,
      "loss": 5.306,
      "mean_token_accuracy": 0.21796911656856538,
      "num_tokens": 69281901.0,
      "step": 37165
    },
    {
      "entropy": 5.992718029022217,
      "epoch": 3.282409042741081,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003941997851967261,
      "loss": 5.3109,
      "mean_token_accuracy": 0.21354688704013824,
      "num_tokens": 69291075.0,
      "step": 37170
    },
    {
      "entropy": 5.985522985458374,
      "epoch": 3.2828505828329213,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039417307491500513,
      "loss": 5.2616,
      "mean_token_accuracy": 0.2204861968755722,
      "num_tokens": 69299475.0,
      "step": 37175
    },
    {
      "entropy": 5.925528192520142,
      "epoch": 3.2832921229247614,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039414636229875155,
      "loss": 5.1962,
      "mean_token_accuracy": 0.2225355699658394,
      "num_tokens": 69308208.0,
      "step": 37180
    },
    {
      "entropy": 5.985436868667603,
      "epoch": 3.283733663016602,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003941196473484884,
      "loss": 5.3601,
      "mean_token_accuracy": 0.21173853874206544,
      "num_tokens": 69317642.0,
      "step": 37185
    },
    {
      "entropy": 5.957544660568237,
      "epoch": 3.2841752031084424,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039409293006473917,
      "loss": 5.1678,
      "mean_token_accuracy": 0.21648416966199874,
      "num_tokens": 69326940.0,
      "step": 37190
    },
    {
      "entropy": 5.996731424331665,
      "epoch": 3.2846167432002824,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039406621044802716,
      "loss": 5.3281,
      "mean_token_accuracy": 0.21176071912050248,
      "num_tokens": 69337356.0,
      "step": 37195
    },
    {
      "entropy": 5.905270624160766,
      "epoch": 3.285058283292123,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039403948849887584,
      "loss": 5.1924,
      "mean_token_accuracy": 0.2237560346722603,
      "num_tokens": 69346484.0,
      "step": 37200
    },
    {
      "entropy": 5.983669805526733,
      "epoch": 3.2854998233839634,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003940127642178085,
      "loss": 5.3069,
      "mean_token_accuracy": 0.21381033807992936,
      "num_tokens": 69355964.0,
      "step": 37205
    },
    {
      "entropy": 6.013056850433349,
      "epoch": 3.2859413634758035,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039398603760534893,
      "loss": 5.3207,
      "mean_token_accuracy": 0.21312824189662932,
      "num_tokens": 69365270.0,
      "step": 37210
    },
    {
      "entropy": 5.92778468132019,
      "epoch": 3.286382903567644,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003939593086620206,
      "loss": 5.1149,
      "mean_token_accuracy": 0.2342902421951294,
      "num_tokens": 69374514.0,
      "step": 37215
    },
    {
      "entropy": 5.974517393112182,
      "epoch": 3.2868244436594845,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039393257738834693,
      "loss": 5.2821,
      "mean_token_accuracy": 0.21405384838581085,
      "num_tokens": 69384821.0,
      "step": 37220
    },
    {
      "entropy": 6.007110500335694,
      "epoch": 3.2872659837513245,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003939058437848518,
      "loss": 5.3622,
      "mean_token_accuracy": 0.2118734285235405,
      "num_tokens": 69394835.0,
      "step": 37225
    },
    {
      "entropy": 6.0043370723724365,
      "epoch": 3.287707523843165,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039387910785205875,
      "loss": 5.3479,
      "mean_token_accuracy": 0.21523149758577348,
      "num_tokens": 69405525.0,
      "step": 37230
    },
    {
      "entropy": 5.932633018493652,
      "epoch": 3.2881490639350055,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039385236959049166,
      "loss": 5.1629,
      "mean_token_accuracy": 0.22342934012413024,
      "num_tokens": 69414512.0,
      "step": 37235
    },
    {
      "entropy": 6.044431591033936,
      "epoch": 3.2885906040268456,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039382562900067413,
      "loss": 5.2746,
      "mean_token_accuracy": 0.21099649518728256,
      "num_tokens": 69424077.0,
      "step": 37240
    },
    {
      "entropy": 6.049879503250122,
      "epoch": 3.289032144118686,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039379888608313026,
      "loss": 5.2774,
      "mean_token_accuracy": 0.21768261045217513,
      "num_tokens": 69433678.0,
      "step": 37245
    },
    {
      "entropy": 5.992488384246826,
      "epoch": 3.2894736842105265,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003937721408383837,
      "loss": 5.276,
      "mean_token_accuracy": 0.21555041670799255,
      "num_tokens": 69443666.0,
      "step": 37250
    },
    {
      "entropy": 5.9817784309387205,
      "epoch": 3.2899152243023666,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003937453932669584,
      "loss": 5.2013,
      "mean_token_accuracy": 0.21830229908227922,
      "num_tokens": 69452430.0,
      "step": 37255
    },
    {
      "entropy": 5.9433879375457765,
      "epoch": 3.290356764394207,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003937186433693785,
      "loss": 5.2751,
      "mean_token_accuracy": 0.21650762856006622,
      "num_tokens": 69462223.0,
      "step": 37260
    },
    {
      "entropy": 5.968553113937378,
      "epoch": 3.290798304486047,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039369189114616776,
      "loss": 5.2696,
      "mean_token_accuracy": 0.2136049047112465,
      "num_tokens": 69471838.0,
      "step": 37265
    },
    {
      "entropy": 6.039465713500976,
      "epoch": 3.2912398445778877,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039366513659785046,
      "loss": 5.2378,
      "mean_token_accuracy": 0.21809312403202058,
      "num_tokens": 69481691.0,
      "step": 37270
    },
    {
      "entropy": 5.980565214157105,
      "epoch": 3.291681384669728,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039363837972495057,
      "loss": 5.32,
      "mean_token_accuracy": 0.20957124829292298,
      "num_tokens": 69490687.0,
      "step": 37275
    },
    {
      "entropy": 5.960462856292724,
      "epoch": 3.292122924761568,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039361162052799236,
      "loss": 5.2728,
      "mean_token_accuracy": 0.21996238082647324,
      "num_tokens": 69500622.0,
      "step": 37280
    },
    {
      "entropy": 6.069161701202392,
      "epoch": 3.2925644648534087,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003935848590074999,
      "loss": 5.3728,
      "mean_token_accuracy": 0.20844705998897553,
      "num_tokens": 69510262.0,
      "step": 37285
    },
    {
      "entropy": 6.084574460983276,
      "epoch": 3.293006004945249,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003935580951639975,
      "loss": 5.3257,
      "mean_token_accuracy": 0.21718741357326507,
      "num_tokens": 69519536.0,
      "step": 37290
    },
    {
      "entropy": 5.9613879203796385,
      "epoch": 3.2934475450370893,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039353132899800947,
      "loss": 5.2188,
      "mean_token_accuracy": 0.21787037551403046,
      "num_tokens": 69529038.0,
      "step": 37295
    },
    {
      "entropy": 5.933922624588012,
      "epoch": 3.2938890851289298,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00039350456051006016,
      "loss": 5.2711,
      "mean_token_accuracy": 0.21532929539680482,
      "num_tokens": 69537411.0,
      "step": 37300
    },
    {
      "entropy": 5.950752019882202,
      "epoch": 3.29433062522077,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003934777897006738,
      "loss": 5.2657,
      "mean_token_accuracy": 0.20935559570789336,
      "num_tokens": 69546403.0,
      "step": 37305
    },
    {
      "entropy": 5.952844858169556,
      "epoch": 3.2947721653126103,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003934510165703751,
      "loss": 5.197,
      "mean_token_accuracy": 0.22282249480485916,
      "num_tokens": 69555676.0,
      "step": 37310
    },
    {
      "entropy": 6.023096084594727,
      "epoch": 3.295213705404451,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039342424111968815,
      "loss": 5.331,
      "mean_token_accuracy": 0.21563151627779006,
      "num_tokens": 69564939.0,
      "step": 37315
    },
    {
      "entropy": 5.972294759750366,
      "epoch": 3.295655245496291,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039339746334913773,
      "loss": 5.1922,
      "mean_token_accuracy": 0.22106217294931413,
      "num_tokens": 69573249.0,
      "step": 37320
    },
    {
      "entropy": 6.019758844375611,
      "epoch": 3.2960967855881314,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039337068325924837,
      "loss": 5.3773,
      "mean_token_accuracy": 0.2077336311340332,
      "num_tokens": 69583702.0,
      "step": 37325
    },
    {
      "entropy": 5.887254905700684,
      "epoch": 3.296538325679972,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039334390085054465,
      "loss": 5.1156,
      "mean_token_accuracy": 0.23763141483068467,
      "num_tokens": 69592726.0,
      "step": 37330
    },
    {
      "entropy": 5.988497543334961,
      "epoch": 3.296979865771812,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003933171161235512,
      "loss": 5.3543,
      "mean_token_accuracy": 0.20368548780679702,
      "num_tokens": 69602043.0,
      "step": 37335
    },
    {
      "entropy": 6.065269899368286,
      "epoch": 3.2974214058636524,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039329032907879283,
      "loss": 5.37,
      "mean_token_accuracy": 0.2126534715294838,
      "num_tokens": 69612551.0,
      "step": 37340
    },
    {
      "entropy": 6.035435628890991,
      "epoch": 3.297862945955493,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003932635397167941,
      "loss": 5.2454,
      "mean_token_accuracy": 0.2259484738111496,
      "num_tokens": 69621628.0,
      "step": 37345
    },
    {
      "entropy": 5.901703500747681,
      "epoch": 3.298304486047333,
      "grad_norm": 1.25,
      "learning_rate": 0.00039323674803807987,
      "loss": 5.1337,
      "mean_token_accuracy": 0.22718448638916017,
      "num_tokens": 69630998.0,
      "step": 37350
    },
    {
      "entropy": 5.980437278747559,
      "epoch": 3.2987460261391734,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000393209954043175,
      "loss": 5.3321,
      "mean_token_accuracy": 0.20545465648174285,
      "num_tokens": 69640528.0,
      "step": 37355
    },
    {
      "entropy": 6.012602376937866,
      "epoch": 3.299187566231014,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039318315773260434,
      "loss": 5.223,
      "mean_token_accuracy": 0.22245216071605683,
      "num_tokens": 69649175.0,
      "step": 37360
    },
    {
      "entropy": 6.049613904953003,
      "epoch": 3.299629106322854,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003931563591068929,
      "loss": 5.2543,
      "mean_token_accuracy": 0.21197005808353425,
      "num_tokens": 69657652.0,
      "step": 37365
    },
    {
      "entropy": 5.905295705795288,
      "epoch": 3.3000706464146945,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039312955816656547,
      "loss": 5.2606,
      "mean_token_accuracy": 0.21827309280633928,
      "num_tokens": 69667796.0,
      "step": 37370
    },
    {
      "entropy": 5.927179145812988,
      "epoch": 3.300512186506535,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003931027549121473,
      "loss": 5.1987,
      "mean_token_accuracy": 0.2234913170337677,
      "num_tokens": 69676785.0,
      "step": 37375
    },
    {
      "entropy": 6.07060980796814,
      "epoch": 3.300953726598375,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003930759493441633,
      "loss": 5.2958,
      "mean_token_accuracy": 0.2072490006685257,
      "num_tokens": 69685895.0,
      "step": 37380
    },
    {
      "entropy": 5.992298173904419,
      "epoch": 3.3013952666902155,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003930491414631386,
      "loss": 5.2086,
      "mean_token_accuracy": 0.2230420306324959,
      "num_tokens": 69694712.0,
      "step": 37385
    },
    {
      "entropy": 6.052926445007325,
      "epoch": 3.301836806782056,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039302233126959827,
      "loss": 5.3465,
      "mean_token_accuracy": 0.2007141724228859,
      "num_tokens": 69704380.0,
      "step": 37390
    },
    {
      "entropy": 5.883172702789307,
      "epoch": 3.302278346873896,
      "grad_norm": 1.25,
      "learning_rate": 0.00039299551876406777,
      "loss": 5.1735,
      "mean_token_accuracy": 0.2243261843919754,
      "num_tokens": 69714431.0,
      "step": 37395
    },
    {
      "entropy": 5.919784164428711,
      "epoch": 3.3027198869657366,
      "grad_norm": 1.15625,
      "learning_rate": 0.000392968703947072,
      "loss": 5.2092,
      "mean_token_accuracy": 0.2194569781422615,
      "num_tokens": 69722970.0,
      "step": 37400
    },
    {
      "entropy": 5.976489686965943,
      "epoch": 3.3031614270575766,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003929418868191365,
      "loss": 5.2844,
      "mean_token_accuracy": 0.2169706791639328,
      "num_tokens": 69732308.0,
      "step": 37405
    },
    {
      "entropy": 5.981862783432007,
      "epoch": 3.303602967149417,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003929150673807865,
      "loss": 5.2059,
      "mean_token_accuracy": 0.21742572039365768,
      "num_tokens": 69741290.0,
      "step": 37410
    },
    {
      "entropy": 5.98514723777771,
      "epoch": 3.3040445072412576,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003928882456325474,
      "loss": 5.2312,
      "mean_token_accuracy": 0.22098461389541627,
      "num_tokens": 69750658.0,
      "step": 37415
    },
    {
      "entropy": 5.944821786880493,
      "epoch": 3.3044860473330977,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003928614215749446,
      "loss": 5.3607,
      "mean_token_accuracy": 0.21173785477876664,
      "num_tokens": 69760857.0,
      "step": 37420
    },
    {
      "entropy": 5.965083980560303,
      "epoch": 3.304927587424938,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003928345952085036,
      "loss": 5.2841,
      "mean_token_accuracy": 0.2211543560028076,
      "num_tokens": 69770217.0,
      "step": 37425
    },
    {
      "entropy": 5.998415946960449,
      "epoch": 3.3053691275167787,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003928077665337499,
      "loss": 5.3069,
      "mean_token_accuracy": 0.213831527531147,
      "num_tokens": 69779864.0,
      "step": 37430
    },
    {
      "entropy": 5.929786014556885,
      "epoch": 3.3058106676086187,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039278093555120895,
      "loss": 5.2214,
      "mean_token_accuracy": 0.21956982761621474,
      "num_tokens": 69789337.0,
      "step": 37435
    },
    {
      "entropy": 6.06337080001831,
      "epoch": 3.3062522077004592,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003927541022614066,
      "loss": 5.3356,
      "mean_token_accuracy": 0.2071218192577362,
      "num_tokens": 69799230.0,
      "step": 37440
    },
    {
      "entropy": 5.96337194442749,
      "epoch": 3.3066937477922993,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039272726666486824,
      "loss": 5.2029,
      "mean_token_accuracy": 0.22627685368061065,
      "num_tokens": 69807741.0,
      "step": 37445
    },
    {
      "entropy": 5.9499200820922855,
      "epoch": 3.30713528788414,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003927004287621198,
      "loss": 5.2841,
      "mean_token_accuracy": 0.21028792709112168,
      "num_tokens": 69818007.0,
      "step": 37450
    },
    {
      "entropy": 5.936650037765503,
      "epoch": 3.3075768279759803,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003926735885536868,
      "loss": 5.1852,
      "mean_token_accuracy": 0.21990627497434617,
      "num_tokens": 69827225.0,
      "step": 37455
    },
    {
      "entropy": 5.951449251174926,
      "epoch": 3.3080183680678203,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003926467460400952,
      "loss": 5.2376,
      "mean_token_accuracy": 0.22082276195287703,
      "num_tokens": 69836734.0,
      "step": 37460
    },
    {
      "entropy": 6.095274496078491,
      "epoch": 3.308459908159661,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003926199012218707,
      "loss": 5.3051,
      "mean_token_accuracy": 0.2152782306075096,
      "num_tokens": 69846064.0,
      "step": 37465
    },
    {
      "entropy": 5.951054048538208,
      "epoch": 3.3089014482515013,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039259305409953925,
      "loss": 5.2244,
      "mean_token_accuracy": 0.21960707008838654,
      "num_tokens": 69855352.0,
      "step": 37470
    },
    {
      "entropy": 5.872906446456909,
      "epoch": 3.3093429883433414,
      "grad_norm": 1.328125,
      "learning_rate": 0.00039256620467362674,
      "loss": 5.2068,
      "mean_token_accuracy": 0.21518323868513106,
      "num_tokens": 69863070.0,
      "step": 37475
    },
    {
      "entropy": 5.8781615734100345,
      "epoch": 3.309784528435182,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039253935294465914,
      "loss": 5.2295,
      "mean_token_accuracy": 0.2221367195248604,
      "num_tokens": 69872056.0,
      "step": 37480
    },
    {
      "entropy": 6.008704280853271,
      "epoch": 3.3102260685270224,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003925124989131625,
      "loss": 5.2215,
      "mean_token_accuracy": 0.2186839297413826,
      "num_tokens": 69882125.0,
      "step": 37485
    },
    {
      "entropy": 6.0158473491668705,
      "epoch": 3.3106676086188624,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003924856425796628,
      "loss": 5.2785,
      "mean_token_accuracy": 0.21347853988409043,
      "num_tokens": 69891378.0,
      "step": 37490
    },
    {
      "entropy": 5.974108409881592,
      "epoch": 3.311109148710703,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039245878394468623,
      "loss": 5.1903,
      "mean_token_accuracy": 0.22271306216716766,
      "num_tokens": 69899545.0,
      "step": 37495
    },
    {
      "entropy": 6.008350133895874,
      "epoch": 3.3115506888025434,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003924319230087588,
      "loss": 5.2371,
      "mean_token_accuracy": 0.21929647922515869,
      "num_tokens": 69908647.0,
      "step": 37500
    },
    {
      "entropy": 5.964610195159912,
      "epoch": 3.3119922288943835,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003924050597724068,
      "loss": 5.2105,
      "mean_token_accuracy": 0.22464802265167236,
      "num_tokens": 69916934.0,
      "step": 37505
    },
    {
      "entropy": 5.88093318939209,
      "epoch": 3.312433768986224,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039237819423615654,
      "loss": 5.2177,
      "mean_token_accuracy": 0.2153438597917557,
      "num_tokens": 69926420.0,
      "step": 37510
    },
    {
      "entropy": 5.9142327308654785,
      "epoch": 3.3128753090780645,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039235132640053414,
      "loss": 5.201,
      "mean_token_accuracy": 0.22009824365377426,
      "num_tokens": 69935021.0,
      "step": 37515
    },
    {
      "entropy": 6.092583036422729,
      "epoch": 3.3133168491699045,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039232445626606605,
      "loss": 5.3099,
      "mean_token_accuracy": 0.20594043731689454,
      "num_tokens": 69944570.0,
      "step": 37520
    },
    {
      "entropy": 5.958012294769287,
      "epoch": 3.313758389261745,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039229758383327844,
      "loss": 5.2702,
      "mean_token_accuracy": 0.2232776790857315,
      "num_tokens": 69953713.0,
      "step": 37525
    },
    {
      "entropy": 5.9233471870422365,
      "epoch": 3.3141999293535855,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039227070910269807,
      "loss": 5.228,
      "mean_token_accuracy": 0.22861203998327256,
      "num_tokens": 69962889.0,
      "step": 37530
    },
    {
      "entropy": 6.000878620147705,
      "epoch": 3.3146414694454256,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039224383207485115,
      "loss": 5.2807,
      "mean_token_accuracy": 0.2208830863237381,
      "num_tokens": 69971677.0,
      "step": 37535
    },
    {
      "entropy": 5.981773567199707,
      "epoch": 3.315083009537266,
      "grad_norm": 1.125,
      "learning_rate": 0.0003922169527502642,
      "loss": 5.3004,
      "mean_token_accuracy": 0.21024901866912843,
      "num_tokens": 69980674.0,
      "step": 37540
    },
    {
      "entropy": 6.036190748214722,
      "epoch": 3.315524549629106,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039219007112946375,
      "loss": 5.2751,
      "mean_token_accuracy": 0.2094022139906883,
      "num_tokens": 69990377.0,
      "step": 37545
    },
    {
      "entropy": 5.910449886322022,
      "epoch": 3.3159660897209466,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003921631872129766,
      "loss": 5.1912,
      "mean_token_accuracy": 0.2195703938603401,
      "num_tokens": 69999540.0,
      "step": 37550
    },
    {
      "entropy": 6.028537654876709,
      "epoch": 3.316407629812787,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003921363010013292,
      "loss": 5.3105,
      "mean_token_accuracy": 0.21327610462903976,
      "num_tokens": 70008543.0,
      "step": 37555
    },
    {
      "entropy": 6.021065711975098,
      "epoch": 3.316849169904627,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039210941249504823,
      "loss": 5.286,
      "mean_token_accuracy": 0.21157125234603882,
      "num_tokens": 70018616.0,
      "step": 37560
    },
    {
      "entropy": 5.996615171432495,
      "epoch": 3.3172907099964677,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003920825216946605,
      "loss": 5.2221,
      "mean_token_accuracy": 0.22198625952005385,
      "num_tokens": 70027678.0,
      "step": 37565
    },
    {
      "entropy": 5.893824529647827,
      "epoch": 3.317732250088308,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003920556286006927,
      "loss": 5.2376,
      "mean_token_accuracy": 0.2129702538251877,
      "num_tokens": 70038007.0,
      "step": 37570
    },
    {
      "entropy": 5.957266330718994,
      "epoch": 3.3181737901801482,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003920287332136718,
      "loss": 5.2638,
      "mean_token_accuracy": 0.22110532224178314,
      "num_tokens": 70046455.0,
      "step": 37575
    },
    {
      "entropy": 5.971049737930298,
      "epoch": 3.3186153302719887,
      "grad_norm": 1.25,
      "learning_rate": 0.0003920018355341245,
      "loss": 5.2275,
      "mean_token_accuracy": 0.2257399931550026,
      "num_tokens": 70054852.0,
      "step": 37580
    },
    {
      "entropy": 6.006304121017456,
      "epoch": 3.319056870363829,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039197493556257783,
      "loss": 5.3376,
      "mean_token_accuracy": 0.21273150891065598,
      "num_tokens": 70064634.0,
      "step": 37585
    },
    {
      "entropy": 6.012460136413575,
      "epoch": 3.3194984104556693,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003919480332995587,
      "loss": 5.2814,
      "mean_token_accuracy": 0.20310094207525253,
      "num_tokens": 70074079.0,
      "step": 37590
    },
    {
      "entropy": 5.970696210861206,
      "epoch": 3.31993995054751,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003919211287455941,
      "loss": 5.2275,
      "mean_token_accuracy": 0.21499452143907546,
      "num_tokens": 70082018.0,
      "step": 37595
    },
    {
      "entropy": 6.021362543106079,
      "epoch": 3.32038149063935,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003918942219012111,
      "loss": 5.3504,
      "mean_token_accuracy": 0.21392042338848113,
      "num_tokens": 70091506.0,
      "step": 37600
    },
    {
      "entropy": 6.0029912948608395,
      "epoch": 3.3208230307311903,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039186731276693676,
      "loss": 5.2582,
      "mean_token_accuracy": 0.21559048742055892,
      "num_tokens": 70100435.0,
      "step": 37605
    },
    {
      "entropy": 5.927324485778809,
      "epoch": 3.321264570823031,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039184040134329824,
      "loss": 5.1528,
      "mean_token_accuracy": 0.21877959221601487,
      "num_tokens": 70109395.0,
      "step": 37610
    },
    {
      "entropy": 5.971081066131592,
      "epoch": 3.321706110914871,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003918134876308227,
      "loss": 5.3164,
      "mean_token_accuracy": 0.2111333817243576,
      "num_tokens": 70119412.0,
      "step": 37615
    },
    {
      "entropy": 5.999567699432373,
      "epoch": 3.3221476510067114,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00039178657163003734,
      "loss": 5.2818,
      "mean_token_accuracy": 0.215618397295475,
      "num_tokens": 70128768.0,
      "step": 37620
    },
    {
      "entropy": 5.857863807678223,
      "epoch": 3.322589191098552,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039175965334146956,
      "loss": 5.1169,
      "mean_token_accuracy": 0.22888162583112717,
      "num_tokens": 70138396.0,
      "step": 37625
    },
    {
      "entropy": 6.0752747535705565,
      "epoch": 3.323030731190392,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003917327327656465,
      "loss": 5.3607,
      "mean_token_accuracy": 0.20414792448282243,
      "num_tokens": 70147661.0,
      "step": 37630
    },
    {
      "entropy": 5.989727735519409,
      "epoch": 3.3234722712822324,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039170580990309555,
      "loss": 5.2405,
      "mean_token_accuracy": 0.22364400774240495,
      "num_tokens": 70156355.0,
      "step": 37635
    },
    {
      "entropy": 5.985971450805664,
      "epoch": 3.323913811374073,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039167888475434424,
      "loss": 5.2859,
      "mean_token_accuracy": 0.214864781498909,
      "num_tokens": 70166042.0,
      "step": 37640
    },
    {
      "entropy": 6.087683439254761,
      "epoch": 3.324355351465913,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039165195731991986,
      "loss": 5.3491,
      "mean_token_accuracy": 0.20834795385599136,
      "num_tokens": 70175644.0,
      "step": 37645
    },
    {
      "entropy": 6.013585901260376,
      "epoch": 3.3247968915577535,
      "grad_norm": 1.25,
      "learning_rate": 0.00039162502760035006,
      "loss": 5.2041,
      "mean_token_accuracy": 0.21962591409683227,
      "num_tokens": 70185002.0,
      "step": 37650
    },
    {
      "entropy": 5.978408145904541,
      "epoch": 3.325238431649594,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003915980955961623,
      "loss": 5.2547,
      "mean_token_accuracy": 0.21366490423679352,
      "num_tokens": 70194585.0,
      "step": 37655
    },
    {
      "entropy": 5.997512769699097,
      "epoch": 3.325679971741434,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039157116130788416,
      "loss": 5.3005,
      "mean_token_accuracy": 0.21237581521272658,
      "num_tokens": 70203642.0,
      "step": 37660
    },
    {
      "entropy": 5.990277290344238,
      "epoch": 3.3261215118332745,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039154422473604316,
      "loss": 5.1787,
      "mean_token_accuracy": 0.23106301724910736,
      "num_tokens": 70212554.0,
      "step": 37665
    },
    {
      "entropy": 5.904118013381958,
      "epoch": 3.326563051925115,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003915172858811672,
      "loss": 5.2437,
      "mean_token_accuracy": 0.21608608812093735,
      "num_tokens": 70222142.0,
      "step": 37670
    },
    {
      "entropy": 5.982882070541382,
      "epoch": 3.327004592016955,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003914903447437838,
      "loss": 5.2719,
      "mean_token_accuracy": 0.21781385093927383,
      "num_tokens": 70231806.0,
      "step": 37675
    },
    {
      "entropy": 6.062178897857666,
      "epoch": 3.3274461321087956,
      "grad_norm": 1.25,
      "learning_rate": 0.00039146340132442084,
      "loss": 5.315,
      "mean_token_accuracy": 0.21270308643579483,
      "num_tokens": 70241344.0,
      "step": 37680
    },
    {
      "entropy": 6.101764774322509,
      "epoch": 3.3278876722006356,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039143645562360606,
      "loss": 5.3641,
      "mean_token_accuracy": 0.20285885781049728,
      "num_tokens": 70250620.0,
      "step": 37685
    },
    {
      "entropy": 5.9389783382415775,
      "epoch": 3.328329212292476,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003914095076418674,
      "loss": 5.2723,
      "mean_token_accuracy": 0.21841603070497512,
      "num_tokens": 70259736.0,
      "step": 37690
    },
    {
      "entropy": 5.905954313278198,
      "epoch": 3.3287707523843166,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039138255737973255,
      "loss": 5.2927,
      "mean_token_accuracy": 0.20527712404727935,
      "num_tokens": 70270070.0,
      "step": 37695
    },
    {
      "entropy": 6.0574134349822994,
      "epoch": 3.3292122924761567,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003913556048377297,
      "loss": 5.2655,
      "mean_token_accuracy": 0.2182820752263069,
      "num_tokens": 70279720.0,
      "step": 37700
    },
    {
      "entropy": 6.00329098701477,
      "epoch": 3.329653832567997,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003913286500163867,
      "loss": 5.2724,
      "mean_token_accuracy": 0.21788561344146729,
      "num_tokens": 70288717.0,
      "step": 37705
    },
    {
      "entropy": 6.0121983051300045,
      "epoch": 3.3300953726598377,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039130169291623165,
      "loss": 5.2477,
      "mean_token_accuracy": 0.2174131229519844,
      "num_tokens": 70298016.0,
      "step": 37710
    },
    {
      "entropy": 6.0367895603179935,
      "epoch": 3.3305369127516777,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039127473353779257,
      "loss": 5.4118,
      "mean_token_accuracy": 0.20297054648399354,
      "num_tokens": 70308535.0,
      "step": 37715
    },
    {
      "entropy": 6.022296667098999,
      "epoch": 3.3309784528435182,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003912477718815975,
      "loss": 5.2991,
      "mean_token_accuracy": 0.21490044742822648,
      "num_tokens": 70317713.0,
      "step": 37720
    },
    {
      "entropy": 5.978928470611573,
      "epoch": 3.3314199929353583,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003912208079481748,
      "loss": 5.3311,
      "mean_token_accuracy": 0.2074248507618904,
      "num_tokens": 70326953.0,
      "step": 37725
    },
    {
      "entropy": 5.94328088760376,
      "epoch": 3.331861533027199,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003911938417380525,
      "loss": 5.2334,
      "mean_token_accuracy": 0.22361615747213365,
      "num_tokens": 70336138.0,
      "step": 37730
    },
    {
      "entropy": 5.981195735931396,
      "epoch": 3.3323030731190393,
      "grad_norm": 1.25,
      "learning_rate": 0.0003911668732517589,
      "loss": 5.2312,
      "mean_token_accuracy": 0.22180334329605103,
      "num_tokens": 70345373.0,
      "step": 37735
    },
    {
      "entropy": 5.9639808177948,
      "epoch": 3.3327446132108793,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003911399024898224,
      "loss": 5.1851,
      "mean_token_accuracy": 0.22004415094852448,
      "num_tokens": 70355435.0,
      "step": 37740
    },
    {
      "entropy": 5.968702936172486,
      "epoch": 3.33318615330272,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003911129294527712,
      "loss": 5.2105,
      "mean_token_accuracy": 0.21345116347074508,
      "num_tokens": 70365140.0,
      "step": 37745
    },
    {
      "entropy": 5.997754192352295,
      "epoch": 3.3336276933945603,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003910859541411338,
      "loss": 5.2485,
      "mean_token_accuracy": 0.21144194155931473,
      "num_tokens": 70373017.0,
      "step": 37750
    },
    {
      "entropy": 5.938621568679809,
      "epoch": 3.3340692334864004,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003910589765554386,
      "loss": 5.1819,
      "mean_token_accuracy": 0.2217613860964775,
      "num_tokens": 70382651.0,
      "step": 37755
    },
    {
      "entropy": 5.956889820098877,
      "epoch": 3.334510773578241,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000391031996696214,
      "loss": 5.1951,
      "mean_token_accuracy": 0.2156039744615555,
      "num_tokens": 70391168.0,
      "step": 37760
    },
    {
      "entropy": 5.9626165390014645,
      "epoch": 3.3349523136700814,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039100501456398857,
      "loss": 5.262,
      "mean_token_accuracy": 0.21243127882480622,
      "num_tokens": 70400180.0,
      "step": 37765
    },
    {
      "entropy": 5.963349866867065,
      "epoch": 3.3353938537619214,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003909780301592909,
      "loss": 5.2429,
      "mean_token_accuracy": 0.2267365872859955,
      "num_tokens": 70410554.0,
      "step": 37770
    },
    {
      "entropy": 6.034364414215088,
      "epoch": 3.335835393853762,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00039095104348264956,
      "loss": 5.2246,
      "mean_token_accuracy": 0.21812936663627625,
      "num_tokens": 70418991.0,
      "step": 37775
    },
    {
      "entropy": 5.9807226181030275,
      "epoch": 3.3362769339456024,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039092405453459327,
      "loss": 5.2756,
      "mean_token_accuracy": 0.2184170663356781,
      "num_tokens": 70428055.0,
      "step": 37780
    },
    {
      "entropy": 5.931095743179322,
      "epoch": 3.3367184740374425,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003908970633156506,
      "loss": 5.2092,
      "mean_token_accuracy": 0.22584544718265534,
      "num_tokens": 70437147.0,
      "step": 37785
    },
    {
      "entropy": 6.061899137496948,
      "epoch": 3.337160014129283,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039087006982635046,
      "loss": 5.2845,
      "mean_token_accuracy": 0.21069404482841492,
      "num_tokens": 70447160.0,
      "step": 37790
    },
    {
      "entropy": 5.9944133281707765,
      "epoch": 3.3376015542211235,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003908430740672215,
      "loss": 5.182,
      "mean_token_accuracy": 0.22129735797643663,
      "num_tokens": 70455309.0,
      "step": 37795
    },
    {
      "entropy": 5.9759192943573,
      "epoch": 3.3380430943129635,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00039081607603879266,
      "loss": 5.2839,
      "mean_token_accuracy": 0.21438368409872055,
      "num_tokens": 70464137.0,
      "step": 37800
    },
    {
      "entropy": 5.958385419845581,
      "epoch": 3.338484634404804,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039078907574159274,
      "loss": 5.266,
      "mean_token_accuracy": 0.21923424005508424,
      "num_tokens": 70473441.0,
      "step": 37805
    },
    {
      "entropy": 6.044699382781983,
      "epoch": 3.3389261744966445,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039076207317615064,
      "loss": 5.3588,
      "mean_token_accuracy": 0.2059476390480995,
      "num_tokens": 70482572.0,
      "step": 37810
    },
    {
      "entropy": 6.000653648376465,
      "epoch": 3.3393677145884846,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003907350683429954,
      "loss": 5.2659,
      "mean_token_accuracy": 0.21842039823532106,
      "num_tokens": 70491928.0,
      "step": 37815
    },
    {
      "entropy": 5.979468488693238,
      "epoch": 3.339809254680325,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000390708061242656,
      "loss": 5.2489,
      "mean_token_accuracy": 0.22641881108283995,
      "num_tokens": 70501340.0,
      "step": 37820
    },
    {
      "entropy": 6.023280954360962,
      "epoch": 3.340250794772165,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039068105187566146,
      "loss": 5.2434,
      "mean_token_accuracy": 0.21668004542589187,
      "num_tokens": 70510344.0,
      "step": 37825
    },
    {
      "entropy": 5.915645503997803,
      "epoch": 3.3406923348640056,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039065404024254084,
      "loss": 5.1279,
      "mean_token_accuracy": 0.22845466583967208,
      "num_tokens": 70519407.0,
      "step": 37830
    },
    {
      "entropy": 5.943674993515015,
      "epoch": 3.341133874955846,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003906270263438234,
      "loss": 5.2445,
      "mean_token_accuracy": 0.2159701555967331,
      "num_tokens": 70529227.0,
      "step": 37835
    },
    {
      "entropy": 6.09865550994873,
      "epoch": 3.341575415047686,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003906000101800383,
      "loss": 5.4351,
      "mean_token_accuracy": 0.19892356395721436,
      "num_tokens": 70539399.0,
      "step": 37840
    },
    {
      "entropy": 6.088089036941528,
      "epoch": 3.3420169551395267,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039057299175171477,
      "loss": 5.3151,
      "mean_token_accuracy": 0.21134002655744552,
      "num_tokens": 70548419.0,
      "step": 37845
    },
    {
      "entropy": 5.991208219528199,
      "epoch": 3.342458495231367,
      "grad_norm": 1.25,
      "learning_rate": 0.000390545971059382,
      "loss": 5.2699,
      "mean_token_accuracy": 0.21565017253160476,
      "num_tokens": 70556847.0,
      "step": 37850
    },
    {
      "entropy": 5.9268700122833256,
      "epoch": 3.3429000353232072,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039051894810356947,
      "loss": 5.2023,
      "mean_token_accuracy": 0.22536755055189134,
      "num_tokens": 70565639.0,
      "step": 37855
    },
    {
      "entropy": 5.971040916442871,
      "epoch": 3.3433415754150477,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003904919228848063,
      "loss": 5.2559,
      "mean_token_accuracy": 0.21318635046482087,
      "num_tokens": 70575433.0,
      "step": 37860
    },
    {
      "entropy": 5.986291885375977,
      "epoch": 3.343783115506888,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039046489540362216,
      "loss": 5.2365,
      "mean_token_accuracy": 0.21883244663476945,
      "num_tokens": 70584959.0,
      "step": 37865
    },
    {
      "entropy": 6.012225484848022,
      "epoch": 3.3442246555987283,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039043786566054627,
      "loss": 5.2809,
      "mean_token_accuracy": 0.2210377648472786,
      "num_tokens": 70594448.0,
      "step": 37870
    },
    {
      "entropy": 6.010999155044556,
      "epoch": 3.3446661956905688,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003904108336561084,
      "loss": 5.3379,
      "mean_token_accuracy": 0.20596355050802231,
      "num_tokens": 70604305.0,
      "step": 37875
    },
    {
      "entropy": 5.979026985168457,
      "epoch": 3.345107735782409,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003903837993908379,
      "loss": 5.1905,
      "mean_token_accuracy": 0.22509963065385818,
      "num_tokens": 70613344.0,
      "step": 37880
    },
    {
      "entropy": 5.9429192543029785,
      "epoch": 3.3455492758742493,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003903567628652643,
      "loss": 5.2144,
      "mean_token_accuracy": 0.2237329974770546,
      "num_tokens": 70622237.0,
      "step": 37885
    },
    {
      "entropy": 5.974455833435059,
      "epoch": 3.34599081596609,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003903297240799174,
      "loss": 5.2005,
      "mean_token_accuracy": 0.21911902874708175,
      "num_tokens": 70631775.0,
      "step": 37890
    },
    {
      "entropy": 5.999296617507935,
      "epoch": 3.34643235605793,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039030268303532674,
      "loss": 5.2207,
      "mean_token_accuracy": 0.21710936725139618,
      "num_tokens": 70640505.0,
      "step": 37895
    },
    {
      "entropy": 5.906876039505005,
      "epoch": 3.3468738961497704,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039027563973202217,
      "loss": 5.099,
      "mean_token_accuracy": 0.23085024058818818,
      "num_tokens": 70650370.0,
      "step": 37900
    },
    {
      "entropy": 6.054568099975586,
      "epoch": 3.347315436241611,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003902485941705333,
      "loss": 5.3326,
      "mean_token_accuracy": 0.20854728072881698,
      "num_tokens": 70660633.0,
      "step": 37905
    },
    {
      "entropy": 6.0466461181640625,
      "epoch": 3.347756976333451,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003902215463513901,
      "loss": 5.3741,
      "mean_token_accuracy": 0.20846243500709533,
      "num_tokens": 70670032.0,
      "step": 37910
    },
    {
      "entropy": 5.933095502853393,
      "epoch": 3.3481985164252914,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003901944962751222,
      "loss": 5.1926,
      "mean_token_accuracy": 0.2165850967168808,
      "num_tokens": 70679839.0,
      "step": 37915
    },
    {
      "entropy": 6.016415309906006,
      "epoch": 3.348640056517132,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003901674439422597,
      "loss": 5.2452,
      "mean_token_accuracy": 0.21590664237737656,
      "num_tokens": 70688502.0,
      "step": 37920
    },
    {
      "entropy": 6.030062675476074,
      "epoch": 3.349081596608972,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039014038935333254,
      "loss": 5.2828,
      "mean_token_accuracy": 0.20847204327583313,
      "num_tokens": 70697665.0,
      "step": 37925
    },
    {
      "entropy": 5.9861620426177975,
      "epoch": 3.3495231367008125,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003901133325088706,
      "loss": 5.2768,
      "mean_token_accuracy": 0.215015709400177,
      "num_tokens": 70707006.0,
      "step": 37930
    },
    {
      "entropy": 5.848319053649902,
      "epoch": 3.349964676792653,
      "grad_norm": 1.21875,
      "learning_rate": 0.000390086273409404,
      "loss": 5.1489,
      "mean_token_accuracy": 0.223651222884655,
      "num_tokens": 70716296.0,
      "step": 37935
    },
    {
      "entropy": 6.051487588882447,
      "epoch": 3.350406216884493,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039005921205546265,
      "loss": 5.2873,
      "mean_token_accuracy": 0.21280343532562257,
      "num_tokens": 70725952.0,
      "step": 37940
    },
    {
      "entropy": 6.0055008888244625,
      "epoch": 3.3508477569763335,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003900321484475767,
      "loss": 5.2065,
      "mean_token_accuracy": 0.21810637414455414,
      "num_tokens": 70734965.0,
      "step": 37945
    },
    {
      "entropy": 5.951455783843994,
      "epoch": 3.351289297068174,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039000508258627654,
      "loss": 5.2365,
      "mean_token_accuracy": 0.2221560388803482,
      "num_tokens": 70743876.0,
      "step": 37950
    },
    {
      "entropy": 5.911768102645874,
      "epoch": 3.351730837160014,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038997801447209217,
      "loss": 5.1897,
      "mean_token_accuracy": 0.21699351370334624,
      "num_tokens": 70752696.0,
      "step": 37955
    },
    {
      "entropy": 6.005338287353515,
      "epoch": 3.3521723772518546,
      "grad_norm": 1.25,
      "learning_rate": 0.00038995094410555395,
      "loss": 5.3359,
      "mean_token_accuracy": 0.2076657846570015,
      "num_tokens": 70761733.0,
      "step": 37960
    },
    {
      "entropy": 6.0486548900604244,
      "epoch": 3.3526139173436946,
      "grad_norm": 1.171875,
      "learning_rate": 0.000389923871487192,
      "loss": 5.2865,
      "mean_token_accuracy": 0.21885304898023605,
      "num_tokens": 70770939.0,
      "step": 37965
    },
    {
      "entropy": 6.036975145339966,
      "epoch": 3.353055457435535,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038989679661753684,
      "loss": 5.2153,
      "mean_token_accuracy": 0.21381874084472657,
      "num_tokens": 70780864.0,
      "step": 37970
    },
    {
      "entropy": 5.9559835433959964,
      "epoch": 3.3534969975273756,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003898697194971187,
      "loss": 5.2877,
      "mean_token_accuracy": 0.21003559678792955,
      "num_tokens": 70789907.0,
      "step": 37975
    },
    {
      "entropy": 6.020008373260498,
      "epoch": 3.3539385376192157,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003898426401264682,
      "loss": 5.2702,
      "mean_token_accuracy": 0.21211745142936705,
      "num_tokens": 70799238.0,
      "step": 37980
    },
    {
      "entropy": 6.039739894866943,
      "epoch": 3.354380077711056,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038981555850611566,
      "loss": 5.3226,
      "mean_token_accuracy": 0.21304027438163758,
      "num_tokens": 70808295.0,
      "step": 37985
    },
    {
      "entropy": 6.045153856277466,
      "epoch": 3.3548216178028967,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038978847463659155,
      "loss": 5.2622,
      "mean_token_accuracy": 0.2130185917019844,
      "num_tokens": 70817277.0,
      "step": 37990
    },
    {
      "entropy": 5.990949535369873,
      "epoch": 3.3552631578947367,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003897613885184267,
      "loss": 5.2519,
      "mean_token_accuracy": 0.21387745290994645,
      "num_tokens": 70827307.0,
      "step": 37995
    },
    {
      "entropy": 6.00031476020813,
      "epoch": 3.3557046979865772,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003897343001521513,
      "loss": 5.1724,
      "mean_token_accuracy": 0.21777165532112122,
      "num_tokens": 70836212.0,
      "step": 38000
    },
    {
      "entropy": 6.0080126285552975,
      "epoch": 3.3561462380784173,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003897072095382963,
      "loss": 5.2631,
      "mean_token_accuracy": 0.21201852709054947,
      "num_tokens": 70844719.0,
      "step": 38005
    },
    {
      "entropy": 6.013821077346802,
      "epoch": 3.3565877781702578,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003896801166773923,
      "loss": 5.3095,
      "mean_token_accuracy": 0.20902844071388244,
      "num_tokens": 70853664.0,
      "step": 38010
    },
    {
      "entropy": 6.0519853115081785,
      "epoch": 3.3570293182620983,
      "grad_norm": 1.46875,
      "learning_rate": 0.00038965302156997,
      "loss": 5.3551,
      "mean_token_accuracy": 0.20753247737884523,
      "num_tokens": 70863881.0,
      "step": 38015
    },
    {
      "entropy": 5.952532052993774,
      "epoch": 3.3574708583539383,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038962592421656027,
      "loss": 5.1981,
      "mean_token_accuracy": 0.2222941756248474,
      "num_tokens": 70873298.0,
      "step": 38020
    },
    {
      "entropy": 6.006985855102539,
      "epoch": 3.357912398445779,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038959882461769385,
      "loss": 5.2847,
      "mean_token_accuracy": 0.21703377217054368,
      "num_tokens": 70882345.0,
      "step": 38025
    },
    {
      "entropy": 6.047765302658081,
      "epoch": 3.3583539385376193,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038957172277390167,
      "loss": 5.2622,
      "mean_token_accuracy": 0.21113474369049073,
      "num_tokens": 70891239.0,
      "step": 38030
    },
    {
      "entropy": 5.9025633335113525,
      "epoch": 3.3587954786294594,
      "grad_norm": 1.25,
      "learning_rate": 0.00038954461868571455,
      "loss": 5.1675,
      "mean_token_accuracy": 0.22013039290904998,
      "num_tokens": 70900652.0,
      "step": 38035
    },
    {
      "entropy": 5.954915857315063,
      "epoch": 3.3592370187213,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003895175123536635,
      "loss": 5.2723,
      "mean_token_accuracy": 0.2156171053647995,
      "num_tokens": 70909757.0,
      "step": 38040
    },
    {
      "entropy": 6.025122022628784,
      "epoch": 3.3596785588131404,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003894904037782795,
      "loss": 5.3473,
      "mean_token_accuracy": 0.2097575142979622,
      "num_tokens": 70918666.0,
      "step": 38045
    },
    {
      "entropy": 5.963583517074585,
      "epoch": 3.3601200989049804,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003894632929600936,
      "loss": 5.1763,
      "mean_token_accuracy": 0.22681639194488526,
      "num_tokens": 70928308.0,
      "step": 38050
    },
    {
      "entropy": 6.00816764831543,
      "epoch": 3.360561638996821,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038943617989963697,
      "loss": 5.3305,
      "mean_token_accuracy": 0.21161584109067916,
      "num_tokens": 70938305.0,
      "step": 38055
    },
    {
      "entropy": 5.9869779586792,
      "epoch": 3.3610031790886614,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00038940906459744064,
      "loss": 5.2805,
      "mean_token_accuracy": 0.2113413468003273,
      "num_tokens": 70947791.0,
      "step": 38060
    },
    {
      "entropy": 5.968735122680664,
      "epoch": 3.3614447191805015,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003893819470540358,
      "loss": 5.2452,
      "mean_token_accuracy": 0.2185683012008667,
      "num_tokens": 70956304.0,
      "step": 38065
    },
    {
      "entropy": 6.016028308868409,
      "epoch": 3.361886259272342,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038935482726995356,
      "loss": 5.2911,
      "mean_token_accuracy": 0.21478083580732346,
      "num_tokens": 70965897.0,
      "step": 38070
    },
    {
      "entropy": 5.993871688842773,
      "epoch": 3.3623277993641825,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003893277052457254,
      "loss": 5.2734,
      "mean_token_accuracy": 0.21575081646442412,
      "num_tokens": 70976056.0,
      "step": 38075
    },
    {
      "entropy": 5.9698821067810055,
      "epoch": 3.3627693394560225,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003893005809818825,
      "loss": 5.2557,
      "mean_token_accuracy": 0.21777358651161194,
      "num_tokens": 70985275.0,
      "step": 38080
    },
    {
      "entropy": 6.0400598526000975,
      "epoch": 3.363210879547863,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003892734544789562,
      "loss": 5.374,
      "mean_token_accuracy": 0.20553468018770218,
      "num_tokens": 70995419.0,
      "step": 38085
    },
    {
      "entropy": 5.998588228225708,
      "epoch": 3.3636524196397035,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038924632573747795,
      "loss": 5.263,
      "mean_token_accuracy": 0.21711350977420807,
      "num_tokens": 71004015.0,
      "step": 38090
    },
    {
      "entropy": 6.020102691650391,
      "epoch": 3.3640939597315436,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003892191947579791,
      "loss": 5.3092,
      "mean_token_accuracy": 0.21791349202394486,
      "num_tokens": 71013509.0,
      "step": 38095
    },
    {
      "entropy": 5.994280910491943,
      "epoch": 3.364535499823384,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038919206154099117,
      "loss": 5.3537,
      "mean_token_accuracy": 0.20823130905628204,
      "num_tokens": 71023949.0,
      "step": 38100
    },
    {
      "entropy": 5.977526092529297,
      "epoch": 3.364977039915224,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003891649260870458,
      "loss": 5.2453,
      "mean_token_accuracy": 0.2182032734155655,
      "num_tokens": 71033836.0,
      "step": 38105
    },
    {
      "entropy": 5.994592761993408,
      "epoch": 3.3654185800070646,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003891377883966744,
      "loss": 5.2313,
      "mean_token_accuracy": 0.22235547602176667,
      "num_tokens": 71043067.0,
      "step": 38110
    },
    {
      "entropy": 5.902916812896729,
      "epoch": 3.365860120098905,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038911064847040863,
      "loss": 5.2711,
      "mean_token_accuracy": 0.21799775660037995,
      "num_tokens": 71052370.0,
      "step": 38115
    },
    {
      "entropy": 5.975499534606934,
      "epoch": 3.366301660190745,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038908350630878025,
      "loss": 5.2873,
      "mean_token_accuracy": 0.21899012178182603,
      "num_tokens": 71061258.0,
      "step": 38120
    },
    {
      "entropy": 6.004244947433472,
      "epoch": 3.3667432002825857,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038905636191232073,
      "loss": 5.2024,
      "mean_token_accuracy": 0.22524917423725127,
      "num_tokens": 71070395.0,
      "step": 38125
    },
    {
      "entropy": 5.927587366104126,
      "epoch": 3.367184740374426,
      "grad_norm": 1.125,
      "learning_rate": 0.000389029215281562,
      "loss": 5.1621,
      "mean_token_accuracy": 0.23154226541519166,
      "num_tokens": 71079365.0,
      "step": 38130
    },
    {
      "entropy": 5.989171028137207,
      "epoch": 3.367626280466266,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003890020664170358,
      "loss": 5.2838,
      "mean_token_accuracy": 0.21475808769464494,
      "num_tokens": 71088163.0,
      "step": 38135
    },
    {
      "entropy": 6.0078911781311035,
      "epoch": 3.3680678205581067,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038897491531927393,
      "loss": 5.2407,
      "mean_token_accuracy": 0.2203908696770668,
      "num_tokens": 71097432.0,
      "step": 38140
    },
    {
      "entropy": 6.019986200332641,
      "epoch": 3.3685093606499468,
      "grad_norm": 1.375,
      "learning_rate": 0.00038894776198880826,
      "loss": 5.2703,
      "mean_token_accuracy": 0.2117152437567711,
      "num_tokens": 71105411.0,
      "step": 38145
    },
    {
      "entropy": 6.013103723526001,
      "epoch": 3.3689509007417873,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038892060642617085,
      "loss": 5.3694,
      "mean_token_accuracy": 0.20626052916049958,
      "num_tokens": 71114357.0,
      "step": 38150
    },
    {
      "entropy": 5.870814800262451,
      "epoch": 3.3693924408336278,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003888934486318935,
      "loss": 5.1651,
      "mean_token_accuracy": 0.23229147642850875,
      "num_tokens": 71123089.0,
      "step": 38155
    },
    {
      "entropy": 5.958753442764282,
      "epoch": 3.369833980925468,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038886628860650824,
      "loss": 5.2681,
      "mean_token_accuracy": 0.2206153929233551,
      "num_tokens": 71133059.0,
      "step": 38160
    },
    {
      "entropy": 6.114358711242676,
      "epoch": 3.3702755210173083,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038883912635054714,
      "loss": 5.3259,
      "mean_token_accuracy": 0.20916418433189393,
      "num_tokens": 71142230.0,
      "step": 38165
    },
    {
      "entropy": 5.9347044944763185,
      "epoch": 3.370717061109149,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003888119618645423,
      "loss": 5.1852,
      "mean_token_accuracy": 0.22622699588537215,
      "num_tokens": 71151215.0,
      "step": 38170
    },
    {
      "entropy": 6.005006265640259,
      "epoch": 3.371158601200989,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003887847951490259,
      "loss": 5.3093,
      "mean_token_accuracy": 0.21715658903121948,
      "num_tokens": 71161169.0,
      "step": 38175
    },
    {
      "entropy": 5.998594188690186,
      "epoch": 3.3716001412928294,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003887576262045301,
      "loss": 5.3176,
      "mean_token_accuracy": 0.20854142010211946,
      "num_tokens": 71171505.0,
      "step": 38180
    },
    {
      "entropy": 6.0004924774169925,
      "epoch": 3.37204168138467,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038873045503158706,
      "loss": 5.3105,
      "mean_token_accuracy": 0.2146690621972084,
      "num_tokens": 71180666.0,
      "step": 38185
    },
    {
      "entropy": 5.993875551223755,
      "epoch": 3.37248322147651,
      "grad_norm": 2.390625,
      "learning_rate": 0.00038870328163072905,
      "loss": 5.148,
      "mean_token_accuracy": 0.22120733708143234,
      "num_tokens": 71189225.0,
      "step": 38190
    },
    {
      "entropy": 5.983652973175049,
      "epoch": 3.3729247615683504,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038867610600248843,
      "loss": 5.2487,
      "mean_token_accuracy": 0.21734318733215333,
      "num_tokens": 71198601.0,
      "step": 38195
    },
    {
      "entropy": 5.936203098297119,
      "epoch": 3.373366301660191,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038864892814739754,
      "loss": 5.2882,
      "mean_token_accuracy": 0.21447401791810988,
      "num_tokens": 71208437.0,
      "step": 38200
    },
    {
      "entropy": 6.03398551940918,
      "epoch": 3.373807841752031,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038862174806598884,
      "loss": 5.3339,
      "mean_token_accuracy": 0.21203531920909882,
      "num_tokens": 71218992.0,
      "step": 38205
    },
    {
      "entropy": 5.979867696762085,
      "epoch": 3.3742493818438715,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003885945657587946,
      "loss": 5.2248,
      "mean_token_accuracy": 0.22125543355941774,
      "num_tokens": 71228479.0,
      "step": 38210
    },
    {
      "entropy": 6.029947805404663,
      "epoch": 3.374690921935712,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038856738122634756,
      "loss": 5.2647,
      "mean_token_accuracy": 0.21593856811523438,
      "num_tokens": 71238060.0,
      "step": 38215
    },
    {
      "entropy": 5.861026048660278,
      "epoch": 3.375132462027552,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038854019446918006,
      "loss": 5.0908,
      "mean_token_accuracy": 0.22700021862983705,
      "num_tokens": 71247525.0,
      "step": 38220
    },
    {
      "entropy": 5.925587129592896,
      "epoch": 3.3755740021193925,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003885130054878247,
      "loss": 5.2814,
      "mean_token_accuracy": 0.21511693745851518,
      "num_tokens": 71257157.0,
      "step": 38225
    },
    {
      "entropy": 5.956732177734375,
      "epoch": 3.376015542211233,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038848581428281413,
      "loss": 5.1977,
      "mean_token_accuracy": 0.21725669056177138,
      "num_tokens": 71267026.0,
      "step": 38230
    },
    {
      "entropy": 5.945455837249756,
      "epoch": 3.376457082303073,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038845862085468103,
      "loss": 5.1391,
      "mean_token_accuracy": 0.23540774881839752,
      "num_tokens": 71276303.0,
      "step": 38235
    },
    {
      "entropy": 5.993433094024658,
      "epoch": 3.3768986223949136,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003884314252039581,
      "loss": 5.3224,
      "mean_token_accuracy": 0.21369817852973938,
      "num_tokens": 71285546.0,
      "step": 38240
    },
    {
      "entropy": 5.990798997879028,
      "epoch": 3.3773401624867536,
      "grad_norm": 1.328125,
      "learning_rate": 0.000388404227331178,
      "loss": 5.2635,
      "mean_token_accuracy": 0.21826103925704957,
      "num_tokens": 71293675.0,
      "step": 38245
    },
    {
      "entropy": 5.941787672042847,
      "epoch": 3.377781702578594,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038837702723687364,
      "loss": 5.2706,
      "mean_token_accuracy": 0.21853090822696686,
      "num_tokens": 71303259.0,
      "step": 38250
    },
    {
      "entropy": 5.97865309715271,
      "epoch": 3.3782232426704346,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003883498249215777,
      "loss": 5.2853,
      "mean_token_accuracy": 0.216414238512516,
      "num_tokens": 71311856.0,
      "step": 38255
    },
    {
      "entropy": 6.020687913894653,
      "epoch": 3.3786647827622747,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003883226203858232,
      "loss": 5.2714,
      "mean_token_accuracy": 0.21585872024297714,
      "num_tokens": 71321187.0,
      "step": 38260
    },
    {
      "entropy": 6.070421648025513,
      "epoch": 3.379106322854115,
      "grad_norm": 1.328125,
      "learning_rate": 0.000388295413630143,
      "loss": 5.3162,
      "mean_token_accuracy": 0.2098655566573143,
      "num_tokens": 71331214.0,
      "step": 38265
    },
    {
      "entropy": 6.045197010040283,
      "epoch": 3.3795478629459557,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003882682046550701,
      "loss": 5.3721,
      "mean_token_accuracy": 0.2068891555070877,
      "num_tokens": 71341425.0,
      "step": 38270
    },
    {
      "entropy": 6.010863590240478,
      "epoch": 3.3799894030377957,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003882409934611375,
      "loss": 5.3166,
      "mean_token_accuracy": 0.2175229609012604,
      "num_tokens": 71350434.0,
      "step": 38275
    },
    {
      "entropy": 6.007431221008301,
      "epoch": 3.380430943129636,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038821378004887824,
      "loss": 5.3065,
      "mean_token_accuracy": 0.20373265147209169,
      "num_tokens": 71359701.0,
      "step": 38280
    },
    {
      "entropy": 5.9970320701599125,
      "epoch": 3.3808724832214763,
      "grad_norm": 1.125,
      "learning_rate": 0.00038818656441882537,
      "loss": 5.283,
      "mean_token_accuracy": 0.2197606936097145,
      "num_tokens": 71369396.0,
      "step": 38285
    },
    {
      "entropy": 6.0015904903411865,
      "epoch": 3.3813140233133168,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003881593465715122,
      "loss": 5.2796,
      "mean_token_accuracy": 0.21194355338811874,
      "num_tokens": 71379057.0,
      "step": 38290
    },
    {
      "entropy": 6.021350431442261,
      "epoch": 3.3817555634051573,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003881321265074718,
      "loss": 5.2778,
      "mean_token_accuracy": 0.213846792280674,
      "num_tokens": 71388794.0,
      "step": 38295
    },
    {
      "entropy": 6.054381322860718,
      "epoch": 3.3821971034969973,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003881049042272372,
      "loss": 5.2498,
      "mean_token_accuracy": 0.22152558267116546,
      "num_tokens": 71397978.0,
      "step": 38300
    },
    {
      "entropy": 5.953231954574585,
      "epoch": 3.382638643588838,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003880776797313419,
      "loss": 5.2426,
      "mean_token_accuracy": 0.2176307886838913,
      "num_tokens": 71406817.0,
      "step": 38305
    },
    {
      "entropy": 5.952364540100097,
      "epoch": 3.3830801836806783,
      "grad_norm": 1.34375,
      "learning_rate": 0.00038805045302031926,
      "loss": 5.268,
      "mean_token_accuracy": 0.21371648609638214,
      "num_tokens": 71415433.0,
      "step": 38310
    },
    {
      "entropy": 6.051973152160644,
      "epoch": 3.3835217237725184,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003880232240947025,
      "loss": 5.3166,
      "mean_token_accuracy": 0.2140570431947708,
      "num_tokens": 71424337.0,
      "step": 38315
    },
    {
      "entropy": 5.987165021896362,
      "epoch": 3.383963263864359,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003879959929550251,
      "loss": 5.1631,
      "mean_token_accuracy": 0.22585549652576448,
      "num_tokens": 71433294.0,
      "step": 38320
    },
    {
      "entropy": 6.019654083251953,
      "epoch": 3.3844048039561994,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003879687596018204,
      "loss": 5.2014,
      "mean_token_accuracy": 0.22521477639675141,
      "num_tokens": 71442613.0,
      "step": 38325
    },
    {
      "entropy": 5.9229823589324955,
      "epoch": 3.3848463440480394,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003879415240356219,
      "loss": 5.2451,
      "mean_token_accuracy": 0.2176499053835869,
      "num_tokens": 71451806.0,
      "step": 38330
    },
    {
      "entropy": 5.9520162582397464,
      "epoch": 3.38528788413988,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00038791428625696323,
      "loss": 5.189,
      "mean_token_accuracy": 0.22187260389328003,
      "num_tokens": 71461175.0,
      "step": 38335
    },
    {
      "entropy": 6.042097759246826,
      "epoch": 3.3857294242317204,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003878870462663779,
      "loss": 5.3306,
      "mean_token_accuracy": 0.20736541152000426,
      "num_tokens": 71471370.0,
      "step": 38340
    },
    {
      "entropy": 6.071519088745117,
      "epoch": 3.3861709643235605,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038785980406439964,
      "loss": 5.2777,
      "mean_token_accuracy": 0.21257498115301132,
      "num_tokens": 71480466.0,
      "step": 38345
    },
    {
      "entropy": 6.012036943435669,
      "epoch": 3.386612504415401,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038783255965156187,
      "loss": 5.2224,
      "mean_token_accuracy": 0.22095677703619004,
      "num_tokens": 71490326.0,
      "step": 38350
    },
    {
      "entropy": 5.986230802536011,
      "epoch": 3.3870540445072415,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003878053130283985,
      "loss": 5.2687,
      "mean_token_accuracy": 0.21011969298124314,
      "num_tokens": 71500153.0,
      "step": 38355
    },
    {
      "entropy": 6.035343933105469,
      "epoch": 3.3874955845990815,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003877780641954432,
      "loss": 5.3398,
      "mean_token_accuracy": 0.20762912184000015,
      "num_tokens": 71509520.0,
      "step": 38360
    },
    {
      "entropy": 6.016529703140259,
      "epoch": 3.387937124690922,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038775081315322967,
      "loss": 5.3125,
      "mean_token_accuracy": 0.21046209186315537,
      "num_tokens": 71519065.0,
      "step": 38365
    },
    {
      "entropy": 5.956568908691406,
      "epoch": 3.3883786647827625,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038772355990229204,
      "loss": 5.155,
      "mean_token_accuracy": 0.22655546814203262,
      "num_tokens": 71527789.0,
      "step": 38370
    },
    {
      "entropy": 6.054072618484497,
      "epoch": 3.3888202048746026,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003876963044431638,
      "loss": 5.2806,
      "mean_token_accuracy": 0.21510057747364045,
      "num_tokens": 71536721.0,
      "step": 38375
    },
    {
      "entropy": 6.051922369003296,
      "epoch": 3.389261744966443,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003876690467763791,
      "loss": 5.31,
      "mean_token_accuracy": 0.21769679188728333,
      "num_tokens": 71545358.0,
      "step": 38380
    },
    {
      "entropy": 5.962173891067505,
      "epoch": 3.389703285058283,
      "grad_norm": 1.125,
      "learning_rate": 0.0003876417869024718,
      "loss": 5.294,
      "mean_token_accuracy": 0.21101083606481552,
      "num_tokens": 71554550.0,
      "step": 38385
    },
    {
      "entropy": 5.9061394214630125,
      "epoch": 3.3901448251501236,
      "grad_norm": 1.1875,
      "learning_rate": 0.000387614524821976,
      "loss": 5.1844,
      "mean_token_accuracy": 0.2294047310948372,
      "num_tokens": 71562898.0,
      "step": 38390
    },
    {
      "entropy": 6.001932382583618,
      "epoch": 3.390586365241964,
      "grad_norm": 1.25,
      "learning_rate": 0.00038758726053542575,
      "loss": 5.2256,
      "mean_token_accuracy": 0.2130008116364479,
      "num_tokens": 71571807.0,
      "step": 38395
    },
    {
      "entropy": 6.054019546508789,
      "epoch": 3.391027905333804,
      "grad_norm": 1.234375,
      "learning_rate": 0.000387559994043355,
      "loss": 5.2994,
      "mean_token_accuracy": 0.20502106845378876,
      "num_tokens": 71581311.0,
      "step": 38400
    },
    {
      "entropy": 5.939312744140625,
      "epoch": 3.3914694454256447,
      "grad_norm": 1.265625,
      "learning_rate": 0.000387532725346298,
      "loss": 5.2356,
      "mean_token_accuracy": 0.22570940256118774,
      "num_tokens": 71591316.0,
      "step": 38405
    },
    {
      "entropy": 5.972229480743408,
      "epoch": 3.391910985517485,
      "grad_norm": 1.3125,
      "learning_rate": 0.000387505454444789,
      "loss": 5.2738,
      "mean_token_accuracy": 0.21405325531959535,
      "num_tokens": 71600750.0,
      "step": 38410
    },
    {
      "entropy": 6.024455642700195,
      "epoch": 3.392352525609325,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003874781813393621,
      "loss": 5.2861,
      "mean_token_accuracy": 0.21291105151176454,
      "num_tokens": 71609886.0,
      "step": 38415
    },
    {
      "entropy": 6.028834438323974,
      "epoch": 3.3927940657011657,
      "grad_norm": 1.125,
      "learning_rate": 0.00038745090603055165,
      "loss": 5.3675,
      "mean_token_accuracy": 0.21692642271518708,
      "num_tokens": 71620062.0,
      "step": 38420
    },
    {
      "entropy": 6.09269528388977,
      "epoch": 3.3932356057930058,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00038742362851889185,
      "loss": 5.2179,
      "mean_token_accuracy": 0.2250094324350357,
      "num_tokens": 71629302.0,
      "step": 38425
    },
    {
      "entropy": 6.0477707862854,
      "epoch": 3.3936771458848463,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003873963488049171,
      "loss": 5.3321,
      "mean_token_accuracy": 0.2110600546002388,
      "num_tokens": 71637900.0,
      "step": 38430
    },
    {
      "entropy": 5.911991786956787,
      "epoch": 3.3941186859766868,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003873690668891618,
      "loss": 5.1847,
      "mean_token_accuracy": 0.22871804386377334,
      "num_tokens": 71646878.0,
      "step": 38435
    },
    {
      "entropy": 5.932006692886352,
      "epoch": 3.394560226068527,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003873417827721604,
      "loss": 5.2038,
      "mean_token_accuracy": 0.22075522541999817,
      "num_tokens": 71655899.0,
      "step": 38440
    },
    {
      "entropy": 5.980220937728882,
      "epoch": 3.3950017661603673,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003873144964544473,
      "loss": 5.2125,
      "mean_token_accuracy": 0.22339365929365157,
      "num_tokens": 71664369.0,
      "step": 38445
    },
    {
      "entropy": 6.0673057556152346,
      "epoch": 3.395443306252208,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003872872079365572,
      "loss": 5.2816,
      "mean_token_accuracy": 0.21374195367097854,
      "num_tokens": 71674131.0,
      "step": 38450
    },
    {
      "entropy": 5.97225604057312,
      "epoch": 3.395884846344048,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038725991721902446,
      "loss": 5.2095,
      "mean_token_accuracy": 0.21933724135160446,
      "num_tokens": 71683994.0,
      "step": 38455
    },
    {
      "entropy": 5.912764978408814,
      "epoch": 3.3963263864358884,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003872326243023838,
      "loss": 5.1996,
      "mean_token_accuracy": 0.22173389047384262,
      "num_tokens": 71693901.0,
      "step": 38460
    },
    {
      "entropy": 5.958762168884277,
      "epoch": 3.396767926527729,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038720532918716993,
      "loss": 5.246,
      "mean_token_accuracy": 0.2190515622496605,
      "num_tokens": 71703104.0,
      "step": 38465
    },
    {
      "entropy": 5.975255870819092,
      "epoch": 3.397209466619569,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003871780318739174,
      "loss": 5.1479,
      "mean_token_accuracy": 0.22877007424831391,
      "num_tokens": 71711619.0,
      "step": 38470
    },
    {
      "entropy": 5.992406749725342,
      "epoch": 3.3976510067114094,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003871507323631611,
      "loss": 5.2345,
      "mean_token_accuracy": 0.2233680874109268,
      "num_tokens": 71720728.0,
      "step": 38475
    },
    {
      "entropy": 6.017480325698853,
      "epoch": 3.39809254680325,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038712343065543575,
      "loss": 5.2689,
      "mean_token_accuracy": 0.2171694129705429,
      "num_tokens": 71729444.0,
      "step": 38480
    },
    {
      "entropy": 6.014288282394409,
      "epoch": 3.39853408689509,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038709612675127606,
      "loss": 5.2674,
      "mean_token_accuracy": 0.21227735728025438,
      "num_tokens": 71738953.0,
      "step": 38485
    },
    {
      "entropy": 5.99949631690979,
      "epoch": 3.3989756269869305,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003870688206512171,
      "loss": 5.267,
      "mean_token_accuracy": 0.2187135025858879,
      "num_tokens": 71748150.0,
      "step": 38490
    },
    {
      "entropy": 5.992556810379028,
      "epoch": 3.399417167078771,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003870415123557936,
      "loss": 5.2714,
      "mean_token_accuracy": 0.20720428228378296,
      "num_tokens": 71758387.0,
      "step": 38495
    },
    {
      "entropy": 6.066421461105347,
      "epoch": 3.399858707170611,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003870142018655406,
      "loss": 5.375,
      "mean_token_accuracy": 0.21013652831315993,
      "num_tokens": 71768539.0,
      "step": 38500
    },
    {
      "entropy": 6.016310119628907,
      "epoch": 3.4003002472624515,
      "grad_norm": 1.125,
      "learning_rate": 0.00038698688918099313,
      "loss": 5.2828,
      "mean_token_accuracy": 0.21977334916591645,
      "num_tokens": 71777976.0,
      "step": 38505
    },
    {
      "entropy": 6.103932952880859,
      "epoch": 3.400741787354292,
      "grad_norm": 1.328125,
      "learning_rate": 0.00038695957430268617,
      "loss": 5.3326,
      "mean_token_accuracy": 0.21166000068187713,
      "num_tokens": 71786501.0,
      "step": 38510
    },
    {
      "entropy": 6.051879978179931,
      "epoch": 3.401183327446132,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003869322572311549,
      "loss": 5.2964,
      "mean_token_accuracy": 0.21456958055496217,
      "num_tokens": 71795956.0,
      "step": 38515
    },
    {
      "entropy": 5.994974136352539,
      "epoch": 3.4016248675379726,
      "grad_norm": 1.375,
      "learning_rate": 0.0003869049379669342,
      "loss": 5.2077,
      "mean_token_accuracy": 0.21910603791475297,
      "num_tokens": 71805306.0,
      "step": 38520
    },
    {
      "entropy": 5.941828918457031,
      "epoch": 3.4020664076298126,
      "grad_norm": 1.390625,
      "learning_rate": 0.00038687761651055955,
      "loss": 5.1802,
      "mean_token_accuracy": 0.21986400932073594,
      "num_tokens": 71813452.0,
      "step": 38525
    },
    {
      "entropy": 5.987105798721314,
      "epoch": 3.402507947721653,
      "grad_norm": 1.328125,
      "learning_rate": 0.00038685029286256596,
      "loss": 5.2254,
      "mean_token_accuracy": 0.21986689269542695,
      "num_tokens": 71822805.0,
      "step": 38530
    },
    {
      "entropy": 6.0323646545410154,
      "epoch": 3.4029494878134936,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038682296702348873,
      "loss": 5.3181,
      "mean_token_accuracy": 0.21019479036331176,
      "num_tokens": 71831366.0,
      "step": 38535
    },
    {
      "entropy": 5.960687255859375,
      "epoch": 3.4033910279053337,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003867956389938633,
      "loss": 5.2114,
      "mean_token_accuracy": 0.21913733929395676,
      "num_tokens": 71841079.0,
      "step": 38540
    },
    {
      "entropy": 5.953130722045898,
      "epoch": 3.403832567997174,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038676830877422473,
      "loss": 5.2105,
      "mean_token_accuracy": 0.22349559068679808,
      "num_tokens": 71850390.0,
      "step": 38545
    },
    {
      "entropy": 6.018988609313965,
      "epoch": 3.4042741080890146,
      "grad_norm": 1.375,
      "learning_rate": 0.0003867409763651086,
      "loss": 5.2977,
      "mean_token_accuracy": 0.21924371719360353,
      "num_tokens": 71859624.0,
      "step": 38550
    },
    {
      "entropy": 6.059893655776977,
      "epoch": 3.4047156481808547,
      "grad_norm": 1.125,
      "learning_rate": 0.00038671364176705037,
      "loss": 5.3632,
      "mean_token_accuracy": 0.21189618557691575,
      "num_tokens": 71869491.0,
      "step": 38555
    },
    {
      "entropy": 5.963692331314087,
      "epoch": 3.405157188272695,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003866863049805854,
      "loss": 5.2335,
      "mean_token_accuracy": 0.208090902864933,
      "num_tokens": 71878572.0,
      "step": 38560
    },
    {
      "entropy": 5.979174184799194,
      "epoch": 3.4055987283645353,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003866589660062492,
      "loss": 5.2664,
      "mean_token_accuracy": 0.22312346547842027,
      "num_tokens": 71887082.0,
      "step": 38565
    },
    {
      "entropy": 5.933092069625855,
      "epoch": 3.4060402684563758,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003866316248445773,
      "loss": 5.2956,
      "mean_token_accuracy": 0.21180433332920073,
      "num_tokens": 71896396.0,
      "step": 38570
    },
    {
      "entropy": 6.02539029121399,
      "epoch": 3.4064818085482163,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003866042814961055,
      "loss": 5.2687,
      "mean_token_accuracy": 0.21809718310832976,
      "num_tokens": 71905796.0,
      "step": 38575
    },
    {
      "entropy": 6.053640270233155,
      "epoch": 3.4069233486400563,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038657693596136925,
      "loss": 5.2107,
      "mean_token_accuracy": 0.2251809984445572,
      "num_tokens": 71914680.0,
      "step": 38580
    },
    {
      "entropy": 5.852109670639038,
      "epoch": 3.407364888731897,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038654958824090415,
      "loss": 5.114,
      "mean_token_accuracy": 0.24525896310806275,
      "num_tokens": 71924315.0,
      "step": 38585
    },
    {
      "entropy": 5.986156415939331,
      "epoch": 3.4078064288237373,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003865222383352462,
      "loss": 5.2978,
      "mean_token_accuracy": 0.21113371849060059,
      "num_tokens": 71934729.0,
      "step": 38590
    },
    {
      "entropy": 5.9019522190094,
      "epoch": 3.4082479689155774,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000386494886244931,
      "loss": 5.1928,
      "mean_token_accuracy": 0.2242108702659607,
      "num_tokens": 71944550.0,
      "step": 38595
    },
    {
      "entropy": 6.038937377929687,
      "epoch": 3.408689509007418,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003864675319704944,
      "loss": 5.2958,
      "mean_token_accuracy": 0.21374621242284775,
      "num_tokens": 71953249.0,
      "step": 38600
    },
    {
      "entropy": 6.004788017272949,
      "epoch": 3.4091310490992583,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003864401755124722,
      "loss": 5.2312,
      "mean_token_accuracy": 0.21812148690223693,
      "num_tokens": 71961656.0,
      "step": 38605
    },
    {
      "entropy": 6.0151573657989506,
      "epoch": 3.4095725891910984,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003864128168714004,
      "loss": 5.3121,
      "mean_token_accuracy": 0.21275315135717393,
      "num_tokens": 71970767.0,
      "step": 38610
    },
    {
      "entropy": 5.960518646240234,
      "epoch": 3.410014129282939,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038638545604781485,
      "loss": 5.2782,
      "mean_token_accuracy": 0.21506432741880416,
      "num_tokens": 71980593.0,
      "step": 38615
    },
    {
      "entropy": 5.9823438167572025,
      "epoch": 3.4104556693747794,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038635809304225153,
      "loss": 5.1903,
      "mean_token_accuracy": 0.2218498855829239,
      "num_tokens": 71989877.0,
      "step": 38620
    },
    {
      "entropy": 5.97461051940918,
      "epoch": 3.4108972094666195,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038633072785524656,
      "loss": 5.202,
      "mean_token_accuracy": 0.22084764540195465,
      "num_tokens": 71998395.0,
      "step": 38625
    },
    {
      "entropy": 6.00566554069519,
      "epoch": 3.41133874955846,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038630336048733595,
      "loss": 5.2748,
      "mean_token_accuracy": 0.21229512691497804,
      "num_tokens": 72007457.0,
      "step": 38630
    },
    {
      "entropy": 6.064391851425171,
      "epoch": 3.4117802896503004,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038627599093905574,
      "loss": 5.3107,
      "mean_token_accuracy": 0.2121117353439331,
      "num_tokens": 72017000.0,
      "step": 38635
    },
    {
      "entropy": 6.05558967590332,
      "epoch": 3.4122218297421405,
      "grad_norm": 1.125,
      "learning_rate": 0.00038624861921094224,
      "loss": 5.3075,
      "mean_token_accuracy": 0.213684843480587,
      "num_tokens": 72027096.0,
      "step": 38640
    },
    {
      "entropy": 5.978060483932495,
      "epoch": 3.412663369833981,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003862212453035315,
      "loss": 5.303,
      "mean_token_accuracy": 0.21145523935556412,
      "num_tokens": 72035473.0,
      "step": 38645
    },
    {
      "entropy": 6.034113073348999,
      "epoch": 3.4131049099258215,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038619386921735977,
      "loss": 5.3004,
      "mean_token_accuracy": 0.21630522906780242,
      "num_tokens": 72044704.0,
      "step": 38650
    },
    {
      "entropy": 6.010805892944336,
      "epoch": 3.4135464500176615,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038616649095296347,
      "loss": 5.301,
      "mean_token_accuracy": 0.21349700391292573,
      "num_tokens": 72054201.0,
      "step": 38655
    },
    {
      "entropy": 6.022091674804687,
      "epoch": 3.413987990109502,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038613911051087885,
      "loss": 5.2003,
      "mean_token_accuracy": 0.22079873830080032,
      "num_tokens": 72063764.0,
      "step": 38660
    },
    {
      "entropy": 6.039940309524536,
      "epoch": 3.414429530201342,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003861117278916422,
      "loss": 5.3589,
      "mean_token_accuracy": 0.20658409595489502,
      "num_tokens": 72074311.0,
      "step": 38665
    },
    {
      "entropy": 6.087201976776123,
      "epoch": 3.4148710702931826,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038608434309579,
      "loss": 5.3697,
      "mean_token_accuracy": 0.2105634704232216,
      "num_tokens": 72084745.0,
      "step": 38670
    },
    {
      "entropy": 6.001833772659301,
      "epoch": 3.415312610385023,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003860569561238587,
      "loss": 5.2444,
      "mean_token_accuracy": 0.21669663041830062,
      "num_tokens": 72094609.0,
      "step": 38675
    },
    {
      "entropy": 6.035360717773438,
      "epoch": 3.415754150476863,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003860295669763848,
      "loss": 5.3054,
      "mean_token_accuracy": 0.20585043579339982,
      "num_tokens": 72104927.0,
      "step": 38680
    },
    {
      "entropy": 6.029027509689331,
      "epoch": 3.4161956905687036,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038600217565390487,
      "loss": 5.2342,
      "mean_token_accuracy": 0.21520543843507767,
      "num_tokens": 72114891.0,
      "step": 38685
    },
    {
      "entropy": 5.955369567871093,
      "epoch": 3.416637230660544,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00038597478215695544,
      "loss": 5.1577,
      "mean_token_accuracy": 0.22727712839841843,
      "num_tokens": 72123685.0,
      "step": 38690
    },
    {
      "entropy": 5.974270391464233,
      "epoch": 3.417078770752384,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038594738648607316,
      "loss": 5.2151,
      "mean_token_accuracy": 0.22041327506303787,
      "num_tokens": 72132828.0,
      "step": 38695
    },
    {
      "entropy": 5.967808675765991,
      "epoch": 3.4175203108442247,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038591998864179455,
      "loss": 5.2627,
      "mean_token_accuracy": 0.22109427601099013,
      "num_tokens": 72142508.0,
      "step": 38700
    },
    {
      "entropy": 5.966383600234986,
      "epoch": 3.4179618509360647,
      "grad_norm": 1.25,
      "learning_rate": 0.0003858925886246566,
      "loss": 5.2697,
      "mean_token_accuracy": 0.2188413456082344,
      "num_tokens": 72152638.0,
      "step": 38705
    },
    {
      "entropy": 6.051748609542846,
      "epoch": 3.4184033910279052,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003858651864351957,
      "loss": 5.3334,
      "mean_token_accuracy": 0.20832593441009523,
      "num_tokens": 72161875.0,
      "step": 38710
    },
    {
      "entropy": 5.9354846477508545,
      "epoch": 3.4188449311197457,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038583778207394904,
      "loss": 5.2237,
      "mean_token_accuracy": 0.2190447047352791,
      "num_tokens": 72171940.0,
      "step": 38715
    },
    {
      "entropy": 5.944736909866333,
      "epoch": 3.419286471211586,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003858103755414533,
      "loss": 5.2578,
      "mean_token_accuracy": 0.2167096272110939,
      "num_tokens": 72180997.0,
      "step": 38720
    },
    {
      "entropy": 6.022558546066284,
      "epoch": 3.4197280113034263,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038578296683824514,
      "loss": 5.2544,
      "mean_token_accuracy": 0.22365372627973557,
      "num_tokens": 72190753.0,
      "step": 38725
    },
    {
      "entropy": 5.931233882904053,
      "epoch": 3.420169551395267,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003857555559648618,
      "loss": 5.1064,
      "mean_token_accuracy": 0.22344397157430648,
      "num_tokens": 72200192.0,
      "step": 38730
    },
    {
      "entropy": 5.973051977157593,
      "epoch": 3.420611091487107,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00038572814292184006,
      "loss": 5.2296,
      "mean_token_accuracy": 0.22217322885990143,
      "num_tokens": 72208562.0,
      "step": 38735
    },
    {
      "entropy": 5.948775339126587,
      "epoch": 3.4210526315789473,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038570072770971697,
      "loss": 5.2656,
      "mean_token_accuracy": 0.21426888555288315,
      "num_tokens": 72219275.0,
      "step": 38740
    },
    {
      "entropy": 6.0971489429473875,
      "epoch": 3.421494171670788,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038567331032902957,
      "loss": 5.3571,
      "mean_token_accuracy": 0.20384574085474014,
      "num_tokens": 72229115.0,
      "step": 38745
    },
    {
      "entropy": 6.018280696868897,
      "epoch": 3.421935711762628,
      "grad_norm": 1.21875,
      "learning_rate": 0.000385645890780315,
      "loss": 5.3019,
      "mean_token_accuracy": 0.21395204216241837,
      "num_tokens": 72238480.0,
      "step": 38750
    },
    {
      "entropy": 6.007220220565796,
      "epoch": 3.4223772518544684,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003856184690641104,
      "loss": 5.2329,
      "mean_token_accuracy": 0.21768446266651154,
      "num_tokens": 72248594.0,
      "step": 38755
    },
    {
      "entropy": 6.010255289077759,
      "epoch": 3.422818791946309,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003855910451809529,
      "loss": 5.2688,
      "mean_token_accuracy": 0.21821995526552201,
      "num_tokens": 72257627.0,
      "step": 38760
    },
    {
      "entropy": 6.0127626895904545,
      "epoch": 3.423260332038149,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003855636191313796,
      "loss": 5.2823,
      "mean_token_accuracy": 0.2170691579580307,
      "num_tokens": 72265973.0,
      "step": 38765
    },
    {
      "entropy": 5.9710687637329105,
      "epoch": 3.4237018721299894,
      "grad_norm": 1.203125,
      "learning_rate": 0.000385536190915928,
      "loss": 5.2539,
      "mean_token_accuracy": 0.22078134417533873,
      "num_tokens": 72274669.0,
      "step": 38770
    },
    {
      "entropy": 5.990645980834961,
      "epoch": 3.42414341222183,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003855087605351352,
      "loss": 5.3003,
      "mean_token_accuracy": 0.21565737277269365,
      "num_tokens": 72283644.0,
      "step": 38775
    },
    {
      "entropy": 6.1021829605102536,
      "epoch": 3.42458495231367,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003854813279895387,
      "loss": 5.2776,
      "mean_token_accuracy": 0.21409431248903274,
      "num_tokens": 72292534.0,
      "step": 38780
    },
    {
      "entropy": 6.002764654159546,
      "epoch": 3.4250264924055105,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003854538932796758,
      "loss": 5.2775,
      "mean_token_accuracy": 0.21559703052043916,
      "num_tokens": 72301790.0,
      "step": 38785
    },
    {
      "entropy": 5.946360731124878,
      "epoch": 3.425468032497351,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003854264564060839,
      "loss": 5.1892,
      "mean_token_accuracy": 0.2178877428174019,
      "num_tokens": 72309674.0,
      "step": 38790
    },
    {
      "entropy": 5.942417478561401,
      "epoch": 3.425909572589191,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003853990173693006,
      "loss": 5.2231,
      "mean_token_accuracy": 0.21892310976982116,
      "num_tokens": 72319332.0,
      "step": 38795
    },
    {
      "entropy": 6.0720172882080075,
      "epoch": 3.4263511126810315,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003853715761698633,
      "loss": 5.3063,
      "mean_token_accuracy": 0.20556581616401673,
      "num_tokens": 72328553.0,
      "step": 38800
    },
    {
      "entropy": 6.037856388092041,
      "epoch": 3.4267926527728716,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003853441328083096,
      "loss": 5.2722,
      "mean_token_accuracy": 0.20976270586252213,
      "num_tokens": 72337429.0,
      "step": 38805
    },
    {
      "entropy": 6.080655145645141,
      "epoch": 3.427234192864712,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038531668728517703,
      "loss": 5.2997,
      "mean_token_accuracy": 0.2096072867512703,
      "num_tokens": 72346638.0,
      "step": 38810
    },
    {
      "entropy": 5.985206937789917,
      "epoch": 3.4276757329565526,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038528923960100335,
      "loss": 5.2124,
      "mean_token_accuracy": 0.22305674552917482,
      "num_tokens": 72356157.0,
      "step": 38815
    },
    {
      "entropy": 5.993476724624633,
      "epoch": 3.4281172730483926,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003852617897563263,
      "loss": 5.2791,
      "mean_token_accuracy": 0.2127846136689186,
      "num_tokens": 72366323.0,
      "step": 38820
    },
    {
      "entropy": 5.958876228332519,
      "epoch": 3.428558813140233,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038523433775168337,
      "loss": 5.2418,
      "mean_token_accuracy": 0.21875501275062562,
      "num_tokens": 72375167.0,
      "step": 38825
    },
    {
      "entropy": 5.928659248352051,
      "epoch": 3.4290003532320736,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038520688358761244,
      "loss": 5.21,
      "mean_token_accuracy": 0.22340206652879716,
      "num_tokens": 72384292.0,
      "step": 38830
    },
    {
      "entropy": 5.948602247238159,
      "epoch": 3.4294418933239137,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038517942726465137,
      "loss": 5.2191,
      "mean_token_accuracy": 0.21714774817228316,
      "num_tokens": 72394008.0,
      "step": 38835
    },
    {
      "entropy": 6.013561630249024,
      "epoch": 3.429883433415754,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038515196878333803,
      "loss": 5.2523,
      "mean_token_accuracy": 0.21744268387556076,
      "num_tokens": 72403074.0,
      "step": 38840
    },
    {
      "entropy": 6.017801284790039,
      "epoch": 3.4303249735075942,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003851245081442102,
      "loss": 5.3419,
      "mean_token_accuracy": 0.21118799597024918,
      "num_tokens": 72412788.0,
      "step": 38845
    },
    {
      "entropy": 5.982899618148804,
      "epoch": 3.4307665135994347,
      "grad_norm": 1.125,
      "learning_rate": 0.0003850970453478059,
      "loss": 5.2764,
      "mean_token_accuracy": 0.215990848839283,
      "num_tokens": 72421923.0,
      "step": 38850
    },
    {
      "entropy": 6.020734405517578,
      "epoch": 3.4312080536912752,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003850695803946631,
      "loss": 5.2791,
      "mean_token_accuracy": 0.21497651934623718,
      "num_tokens": 72431696.0,
      "step": 38855
    },
    {
      "entropy": 6.055990743637085,
      "epoch": 3.4316495937831153,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038504211328531986,
      "loss": 5.3114,
      "mean_token_accuracy": 0.21458837389945984,
      "num_tokens": 72441368.0,
      "step": 38860
    },
    {
      "entropy": 6.037660217285156,
      "epoch": 3.432091133874956,
      "grad_norm": 1.25,
      "learning_rate": 0.00038501464402031417,
      "loss": 5.2655,
      "mean_token_accuracy": 0.22582121044397355,
      "num_tokens": 72449915.0,
      "step": 38865
    },
    {
      "entropy": 6.0332764148712155,
      "epoch": 3.4325326739667963,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038498717260018425,
      "loss": 5.2389,
      "mean_token_accuracy": 0.2175449699163437,
      "num_tokens": 72459549.0,
      "step": 38870
    },
    {
      "entropy": 5.990675640106201,
      "epoch": 3.4329742140586363,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003849596990254681,
      "loss": 5.2952,
      "mean_token_accuracy": 0.2152669370174408,
      "num_tokens": 72469710.0,
      "step": 38875
    },
    {
      "entropy": 5.9523194313049315,
      "epoch": 3.433415754150477,
      "grad_norm": 1.203125,
      "learning_rate": 0.000384932223296704,
      "loss": 5.2986,
      "mean_token_accuracy": 0.2121440589427948,
      "num_tokens": 72478357.0,
      "step": 38880
    },
    {
      "entropy": 5.909491968154907,
      "epoch": 3.4338572942423173,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038490474541443004,
      "loss": 5.2063,
      "mean_token_accuracy": 0.22501663565635682,
      "num_tokens": 72487950.0,
      "step": 38885
    },
    {
      "entropy": 6.04036955833435,
      "epoch": 3.4342988343341574,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003848772653791848,
      "loss": 5.2716,
      "mean_token_accuracy": 0.21551418751478196,
      "num_tokens": 72496188.0,
      "step": 38890
    },
    {
      "entropy": 6.010817909240723,
      "epoch": 3.434740374425998,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003848497831915064,
      "loss": 5.2339,
      "mean_token_accuracy": 0.22527382522821426,
      "num_tokens": 72505361.0,
      "step": 38895
    },
    {
      "entropy": 5.97498459815979,
      "epoch": 3.4351819145178384,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003848222988519331,
      "loss": 5.1999,
      "mean_token_accuracy": 0.23006894141435624,
      "num_tokens": 72515254.0,
      "step": 38900
    },
    {
      "entropy": 6.035889053344727,
      "epoch": 3.4356234546096784,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00038479481236100354,
      "loss": 5.2627,
      "mean_token_accuracy": 0.21516606956720352,
      "num_tokens": 72523971.0,
      "step": 38905
    },
    {
      "entropy": 5.994227361679077,
      "epoch": 3.436064994701519,
      "grad_norm": 1.171875,
      "learning_rate": 0.000384767323719256,
      "loss": 5.2774,
      "mean_token_accuracy": 0.2166067913174629,
      "num_tokens": 72534130.0,
      "step": 38910
    },
    {
      "entropy": 6.072881603240967,
      "epoch": 3.4365065347933594,
      "grad_norm": 1.234375,
      "learning_rate": 0.000384739832927229,
      "loss": 5.3458,
      "mean_token_accuracy": 0.19950393289327623,
      "num_tokens": 72544665.0,
      "step": 38915
    },
    {
      "entropy": 6.029995441436768,
      "epoch": 3.4369480748851995,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038471233998546117,
      "loss": 5.1338,
      "mean_token_accuracy": 0.23157313764095305,
      "num_tokens": 72553001.0,
      "step": 38920
    },
    {
      "entropy": 6.105037689208984,
      "epoch": 3.43738961497704,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038468484489449095,
      "loss": 5.4009,
      "mean_token_accuracy": 0.2035009667277336,
      "num_tokens": 72562363.0,
      "step": 38925
    },
    {
      "entropy": 6.0384681224823,
      "epoch": 3.4378311550688805,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000384657347654857,
      "loss": 5.3019,
      "mean_token_accuracy": 0.20676166862249373,
      "num_tokens": 72570969.0,
      "step": 38930
    },
    {
      "entropy": 6.021579360961914,
      "epoch": 3.4382726951607205,
      "grad_norm": 1.25,
      "learning_rate": 0.00038462984826709794,
      "loss": 5.207,
      "mean_token_accuracy": 0.21583531349897384,
      "num_tokens": 72580288.0,
      "step": 38935
    },
    {
      "entropy": 6.045266485214233,
      "epoch": 3.438714235252561,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003846023467317525,
      "loss": 5.2759,
      "mean_token_accuracy": 0.20860904455184937,
      "num_tokens": 72588956.0,
      "step": 38940
    },
    {
      "entropy": 5.984873104095459,
      "epoch": 3.439155775344401,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003845748430493595,
      "loss": 5.2515,
      "mean_token_accuracy": 0.2231421411037445,
      "num_tokens": 72597456.0,
      "step": 38945
    },
    {
      "entropy": 6.07704529762268,
      "epoch": 3.4395973154362416,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003845473372204576,
      "loss": 5.3512,
      "mean_token_accuracy": 0.21037630289793013,
      "num_tokens": 72607805.0,
      "step": 38950
    },
    {
      "entropy": 6.017535209655762,
      "epoch": 3.440038855528082,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038451982924558566,
      "loss": 5.2324,
      "mean_token_accuracy": 0.21337262094020842,
      "num_tokens": 72617767.0,
      "step": 38955
    },
    {
      "entropy": 5.948566007614136,
      "epoch": 3.440480395619922,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003844923191252825,
      "loss": 5.1322,
      "mean_token_accuracy": 0.22988792657852172,
      "num_tokens": 72626918.0,
      "step": 38960
    },
    {
      "entropy": 5.986517381668091,
      "epoch": 3.4409219357117626,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038446480686008715,
      "loss": 5.2884,
      "mean_token_accuracy": 0.2141187980771065,
      "num_tokens": 72637486.0,
      "step": 38965
    },
    {
      "entropy": 5.990935468673706,
      "epoch": 3.441363475803603,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003844372924505385,
      "loss": 5.175,
      "mean_token_accuracy": 0.22464661598205565,
      "num_tokens": 72646168.0,
      "step": 38970
    },
    {
      "entropy": 6.043038082122803,
      "epoch": 3.441805015895443,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003844097758971755,
      "loss": 5.3998,
      "mean_token_accuracy": 0.20292637944221498,
      "num_tokens": 72656970.0,
      "step": 38975
    },
    {
      "entropy": 6.0416522979736325,
      "epoch": 3.4422465559872837,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003843822572005372,
      "loss": 5.2803,
      "mean_token_accuracy": 0.21840384006500244,
      "num_tokens": 72666687.0,
      "step": 38980
    },
    {
      "entropy": 5.984784841537476,
      "epoch": 3.4426880960791237,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038435473636116275,
      "loss": 5.2551,
      "mean_token_accuracy": 0.21863355487585068,
      "num_tokens": 72675990.0,
      "step": 38985
    },
    {
      "entropy": 5.965013313293457,
      "epoch": 3.4431296361709642,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038432721337959113,
      "loss": 5.2274,
      "mean_token_accuracy": 0.21659006625413896,
      "num_tokens": 72684946.0,
      "step": 38990
    },
    {
      "entropy": 6.0062566757202145,
      "epoch": 3.4435711762628047,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003842996882563616,
      "loss": 5.3759,
      "mean_token_accuracy": 0.21072787642478943,
      "num_tokens": 72694836.0,
      "step": 38995
    },
    {
      "entropy": 6.056128358840942,
      "epoch": 3.444012716354645,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003842721609920134,
      "loss": 5.2176,
      "mean_token_accuracy": 0.22629913240671157,
      "num_tokens": 72703940.0,
      "step": 39000
    },
    {
      "epoch": 3.444012716354645,
      "eval_entropy": 5.768855893058388,
      "eval_loss": 5.33239221572876,
      "eval_mean_token_accuracy": 0.22184960317301183,
      "eval_num_tokens": 72703940.0,
      "eval_runtime": 20.2326,
      "eval_samples_per_second": 1745.454,
      "eval_steps_per_second": 218.213,
      "step": 39000
    },
    {
      "entropy": 5.989181280136108,
      "epoch": 3.4444542564464853,
      "grad_norm": 1.25,
      "learning_rate": 0.0003842446315870857,
      "loss": 5.2181,
      "mean_token_accuracy": 0.21702299416065216,
      "num_tokens": 72712630.0,
      "step": 39005
    },
    {
      "entropy": 5.9531824588775635,
      "epoch": 3.444895796538326,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003842171000421178,
      "loss": 5.2988,
      "mean_token_accuracy": 0.21340600550174713,
      "num_tokens": 72722263.0,
      "step": 39010
    },
    {
      "entropy": 6.04642333984375,
      "epoch": 3.445337336630166,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003841895663576489,
      "loss": 5.3115,
      "mean_token_accuracy": 0.21188612878322602,
      "num_tokens": 72731731.0,
      "step": 39015
    },
    {
      "entropy": 6.013098478317261,
      "epoch": 3.4457788767220063,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003841620305342186,
      "loss": 5.3438,
      "mean_token_accuracy": 0.2081746131181717,
      "num_tokens": 72741633.0,
      "step": 39020
    },
    {
      "entropy": 6.110986137390137,
      "epoch": 3.446220416813847,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003841344925723662,
      "loss": 5.3788,
      "mean_token_accuracy": 0.20659889578819274,
      "num_tokens": 72751713.0,
      "step": 39025
    },
    {
      "entropy": 6.072715997695923,
      "epoch": 3.446661956905687,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003841069524726311,
      "loss": 5.3249,
      "mean_token_accuracy": 0.21469690054655075,
      "num_tokens": 72761567.0,
      "step": 39030
    },
    {
      "entropy": 5.983273124694824,
      "epoch": 3.4471034969975274,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038407941023555295,
      "loss": 5.2378,
      "mean_token_accuracy": 0.22081244736909866,
      "num_tokens": 72770443.0,
      "step": 39035
    },
    {
      "entropy": 6.020238018035888,
      "epoch": 3.447545037089368,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038405186586167115,
      "loss": 5.29,
      "mean_token_accuracy": 0.21243415176868438,
      "num_tokens": 72780104.0,
      "step": 39040
    },
    {
      "entropy": 6.012662792205811,
      "epoch": 3.447986577181208,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003840243193515253,
      "loss": 5.2632,
      "mean_token_accuracy": 0.21788278222084045,
      "num_tokens": 72788478.0,
      "step": 39045
    },
    {
      "entropy": 6.037291431427002,
      "epoch": 3.4484281172730484,
      "grad_norm": 1.140625,
      "learning_rate": 0.000383996770705655,
      "loss": 5.3353,
      "mean_token_accuracy": 0.2054491177201271,
      "num_tokens": 72798461.0,
      "step": 39050
    },
    {
      "entropy": 6.0089503765106205,
      "epoch": 3.448869657364889,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003839692199246001,
      "loss": 5.3134,
      "mean_token_accuracy": 0.2168915331363678,
      "num_tokens": 72809085.0,
      "step": 39055
    },
    {
      "entropy": 6.079589748382569,
      "epoch": 3.449311197456729,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038394166700890005,
      "loss": 5.354,
      "mean_token_accuracy": 0.21140784323215484,
      "num_tokens": 72819518.0,
      "step": 39060
    },
    {
      "entropy": 5.983468341827392,
      "epoch": 3.4497527375485695,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038391411195909473,
      "loss": 5.1555,
      "mean_token_accuracy": 0.22240890562534332,
      "num_tokens": 72828629.0,
      "step": 39065
    },
    {
      "entropy": 6.025777530670166,
      "epoch": 3.45019427764041,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003838865547757238,
      "loss": 5.2286,
      "mean_token_accuracy": 0.20994078516960143,
      "num_tokens": 72838740.0,
      "step": 39070
    },
    {
      "entropy": 5.966018342971802,
      "epoch": 3.45063581773225,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003838589954593273,
      "loss": 5.1727,
      "mean_token_accuracy": 0.22675064504146575,
      "num_tokens": 72847563.0,
      "step": 39075
    },
    {
      "entropy": 5.965786266326904,
      "epoch": 3.4510773578240905,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000383831434010445,
      "loss": 5.1857,
      "mean_token_accuracy": 0.21821920722723007,
      "num_tokens": 72856360.0,
      "step": 39080
    },
    {
      "entropy": 5.936714029312133,
      "epoch": 3.4515188979159306,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038380387042961673,
      "loss": 5.158,
      "mean_token_accuracy": 0.22447510659694672,
      "num_tokens": 72864428.0,
      "step": 39085
    },
    {
      "entropy": 5.960104274749756,
      "epoch": 3.451960438007771,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038377630471738257,
      "loss": 5.2885,
      "mean_token_accuracy": 0.21607982069253923,
      "num_tokens": 72874341.0,
      "step": 39090
    },
    {
      "entropy": 5.997353410720825,
      "epoch": 3.4524019780996116,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003837487368742824,
      "loss": 5.2678,
      "mean_token_accuracy": 0.2127060428261757,
      "num_tokens": 72884292.0,
      "step": 39095
    },
    {
      "entropy": 6.066124153137207,
      "epoch": 3.4528435181914516,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003837211669008564,
      "loss": 5.3156,
      "mean_token_accuracy": 0.20617655366659166,
      "num_tokens": 72893771.0,
      "step": 39100
    },
    {
      "entropy": 5.977844715118408,
      "epoch": 3.453285058283292,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003836935947976445,
      "loss": 5.2309,
      "mean_token_accuracy": 0.22768858224153518,
      "num_tokens": 72903649.0,
      "step": 39105
    },
    {
      "entropy": 6.012694644927978,
      "epoch": 3.4537265983751326,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000383666020565187,
      "loss": 5.2792,
      "mean_token_accuracy": 0.21735939681529998,
      "num_tokens": 72913471.0,
      "step": 39110
    },
    {
      "entropy": 6.057227897644043,
      "epoch": 3.4541681384669727,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003836384442040238,
      "loss": 5.3051,
      "mean_token_accuracy": 0.210194830596447,
      "num_tokens": 72924012.0,
      "step": 39115
    },
    {
      "entropy": 6.050476932525635,
      "epoch": 3.454609678558813,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00038361086571469545,
      "loss": 5.268,
      "mean_token_accuracy": 0.222910313308239,
      "num_tokens": 72933055.0,
      "step": 39120
    },
    {
      "entropy": 5.961115217208862,
      "epoch": 3.4550512186506532,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003835832850977419,
      "loss": 5.2957,
      "mean_token_accuracy": 0.2092966079711914,
      "num_tokens": 72942575.0,
      "step": 39125
    },
    {
      "entropy": 5.984367799758911,
      "epoch": 3.4554927587424937,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003835557023537036,
      "loss": 5.2173,
      "mean_token_accuracy": 0.21515037566423417,
      "num_tokens": 72951106.0,
      "step": 39130
    },
    {
      "entropy": 5.976364850997925,
      "epoch": 3.4559342988343342,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003835281174831208,
      "loss": 5.3289,
      "mean_token_accuracy": 0.20996688306331635,
      "num_tokens": 72960667.0,
      "step": 39135
    },
    {
      "entropy": 6.011661195755005,
      "epoch": 3.4563758389261743,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038350053048653393,
      "loss": 5.1759,
      "mean_token_accuracy": 0.22326710373163222,
      "num_tokens": 72969606.0,
      "step": 39140
    },
    {
      "entropy": 6.104644393920898,
      "epoch": 3.456817379018015,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038347294136448337,
      "loss": 5.4425,
      "mean_token_accuracy": 0.19850173443555832,
      "num_tokens": 72978953.0,
      "step": 39145
    },
    {
      "entropy": 5.935787343978882,
      "epoch": 3.4572589191098553,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038344535011750953,
      "loss": 5.155,
      "mean_token_accuracy": 0.22917643785476685,
      "num_tokens": 72987398.0,
      "step": 39150
    },
    {
      "entropy": 6.040307235717774,
      "epoch": 3.4577004592016953,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000383417756746153,
      "loss": 5.2499,
      "mean_token_accuracy": 0.2181839868426323,
      "num_tokens": 72996340.0,
      "step": 39155
    },
    {
      "entropy": 6.000356388092041,
      "epoch": 3.458141999293536,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003833901612509542,
      "loss": 5.2634,
      "mean_token_accuracy": 0.22097819447517394,
      "num_tokens": 73006296.0,
      "step": 39160
    },
    {
      "entropy": 5.98387188911438,
      "epoch": 3.4585835393853763,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038336256363245385,
      "loss": 5.2919,
      "mean_token_accuracy": 0.22157441079616547,
      "num_tokens": 73016377.0,
      "step": 39165
    },
    {
      "entropy": 6.00056209564209,
      "epoch": 3.4590250794772164,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003833349638911925,
      "loss": 5.2417,
      "mean_token_accuracy": 0.21006400734186173,
      "num_tokens": 73025876.0,
      "step": 39170
    },
    {
      "entropy": 6.062374877929687,
      "epoch": 3.459466619569057,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038330736202771086,
      "loss": 5.3147,
      "mean_token_accuracy": 0.21338757127523422,
      "num_tokens": 73034676.0,
      "step": 39175
    },
    {
      "entropy": 6.064734649658203,
      "epoch": 3.4599081596608974,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038327975804254956,
      "loss": 5.2767,
      "mean_token_accuracy": 0.2127540960907936,
      "num_tokens": 73044490.0,
      "step": 39180
    },
    {
      "entropy": 5.953133010864258,
      "epoch": 3.4603496997527374,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003832521519362494,
      "loss": 5.267,
      "mean_token_accuracy": 0.2199898600578308,
      "num_tokens": 73054566.0,
      "step": 39185
    },
    {
      "entropy": 5.931606197357178,
      "epoch": 3.460791239844578,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00038322454370935113,
      "loss": 5.2918,
      "mean_token_accuracy": 0.21461794525384903,
      "num_tokens": 73064352.0,
      "step": 39190
    },
    {
      "entropy": 6.0156491756439205,
      "epoch": 3.4612327799364184,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003831969333623956,
      "loss": 5.2141,
      "mean_token_accuracy": 0.22218268513679504,
      "num_tokens": 73073736.0,
      "step": 39195
    },
    {
      "entropy": 6.033906412124634,
      "epoch": 3.4616743200282585,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038316932089592376,
      "loss": 5.3425,
      "mean_token_accuracy": 0.20722867399454117,
      "num_tokens": 73083138.0,
      "step": 39200
    },
    {
      "entropy": 6.0255859375,
      "epoch": 3.462115860120099,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038314170631047634,
      "loss": 5.3261,
      "mean_token_accuracy": 0.21655897945165634,
      "num_tokens": 73091262.0,
      "step": 39205
    },
    {
      "entropy": 5.996055698394775,
      "epoch": 3.4625574002119395,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003831140896065944,
      "loss": 5.2652,
      "mean_token_accuracy": 0.21412338465452194,
      "num_tokens": 73099827.0,
      "step": 39210
    },
    {
      "entropy": 5.971337127685547,
      "epoch": 3.4629989403037795,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038308647078481894,
      "loss": 5.2285,
      "mean_token_accuracy": 0.21575799882411956,
      "num_tokens": 73109231.0,
      "step": 39215
    },
    {
      "entropy": 5.919323873519898,
      "epoch": 3.46344048039562,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038305884984569103,
      "loss": 5.1929,
      "mean_token_accuracy": 0.22203171551227568,
      "num_tokens": 73117822.0,
      "step": 39220
    },
    {
      "entropy": 5.949937629699707,
      "epoch": 3.46388202048746,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003830312267897516,
      "loss": 5.2285,
      "mean_token_accuracy": 0.22175202071666716,
      "num_tokens": 73128004.0,
      "step": 39225
    },
    {
      "entropy": 6.039695692062378,
      "epoch": 3.4643235605793006,
      "grad_norm": 1.3125,
      "learning_rate": 0.000383003601617542,
      "loss": 5.228,
      "mean_token_accuracy": 0.21485500484704972,
      "num_tokens": 73136888.0,
      "step": 39230
    },
    {
      "entropy": 6.056411981582642,
      "epoch": 3.464765100671141,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038297597432960324,
      "loss": 5.3344,
      "mean_token_accuracy": 0.212411929666996,
      "num_tokens": 73146504.0,
      "step": 39235
    },
    {
      "entropy": 6.020932292938232,
      "epoch": 3.465206640762981,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003829483449264766,
      "loss": 5.2942,
      "mean_token_accuracy": 0.213856340944767,
      "num_tokens": 73155620.0,
      "step": 39240
    },
    {
      "entropy": 5.906854867935181,
      "epoch": 3.4656481808548216,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003829207134087032,
      "loss": 5.183,
      "mean_token_accuracy": 0.2206580474972725,
      "num_tokens": 73164235.0,
      "step": 39245
    },
    {
      "entropy": 5.97655234336853,
      "epoch": 3.466089720946662,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003828930797768244,
      "loss": 5.3172,
      "mean_token_accuracy": 0.21156201511621475,
      "num_tokens": 73173583.0,
      "step": 39250
    },
    {
      "entropy": 6.05130295753479,
      "epoch": 3.466531261038502,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003828654440313815,
      "loss": 5.2486,
      "mean_token_accuracy": 0.2195331871509552,
      "num_tokens": 73182244.0,
      "step": 39255
    },
    {
      "entropy": 5.968510627746582,
      "epoch": 3.4669728011303427,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038283780617291604,
      "loss": 5.1853,
      "mean_token_accuracy": 0.22563152760267258,
      "num_tokens": 73190984.0,
      "step": 39260
    },
    {
      "entropy": 5.994404411315918,
      "epoch": 3.4674143412221827,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038281016620196915,
      "loss": 5.2776,
      "mean_token_accuracy": 0.21482659876346588,
      "num_tokens": 73201196.0,
      "step": 39265
    },
    {
      "entropy": 5.90607271194458,
      "epoch": 3.4678558813140232,
      "grad_norm": 1.25,
      "learning_rate": 0.00038278252411908247,
      "loss": 5.1521,
      "mean_token_accuracy": 0.22878921777009964,
      "num_tokens": 73210054.0,
      "step": 39270
    },
    {
      "entropy": 5.961556816101075,
      "epoch": 3.4682974214058637,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003827548799247975,
      "loss": 5.213,
      "mean_token_accuracy": 0.22009462863206863,
      "num_tokens": 73219621.0,
      "step": 39275
    },
    {
      "entropy": 6.003859519958496,
      "epoch": 3.4687389614977038,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003827272336196556,
      "loss": 5.2183,
      "mean_token_accuracy": 0.219416780769825,
      "num_tokens": 73227838.0,
      "step": 39280
    },
    {
      "entropy": 5.96722903251648,
      "epoch": 3.4691805015895443,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003826995852041985,
      "loss": 5.2565,
      "mean_token_accuracy": 0.2171940252184868,
      "num_tokens": 73237591.0,
      "step": 39285
    },
    {
      "entropy": 5.996922874450684,
      "epoch": 3.4696220416813848,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00038267193467896775,
      "loss": 5.2922,
      "mean_token_accuracy": 0.21561949998140334,
      "num_tokens": 73246254.0,
      "step": 39290
    },
    {
      "entropy": 5.999179649353027,
      "epoch": 3.470063581773225,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003826442820445051,
      "loss": 5.3135,
      "mean_token_accuracy": 0.20819493681192397,
      "num_tokens": 73255941.0,
      "step": 39295
    },
    {
      "entropy": 6.050292873382569,
      "epoch": 3.4705051218650653,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038261662730135215,
      "loss": 5.3715,
      "mean_token_accuracy": 0.2096080780029297,
      "num_tokens": 73266543.0,
      "step": 39300
    },
    {
      "entropy": 5.995219421386719,
      "epoch": 3.470946661956906,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003825889704500507,
      "loss": 5.1879,
      "mean_token_accuracy": 0.2301987960934639,
      "num_tokens": 73274894.0,
      "step": 39305
    },
    {
      "entropy": 6.03801736831665,
      "epoch": 3.471388202048746,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038256131149114246,
      "loss": 5.3269,
      "mean_token_accuracy": 0.21139594465494155,
      "num_tokens": 73284126.0,
      "step": 39310
    },
    {
      "entropy": 5.967554664611816,
      "epoch": 3.4718297421405864,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003825336504251694,
      "loss": 5.1713,
      "mean_token_accuracy": 0.22161697447299958,
      "num_tokens": 73293300.0,
      "step": 39315
    },
    {
      "entropy": 6.019478750228882,
      "epoch": 3.472271282232427,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003825059872526733,
      "loss": 5.3228,
      "mean_token_accuracy": 0.2187269538640976,
      "num_tokens": 73302032.0,
      "step": 39320
    },
    {
      "entropy": 6.047161006927491,
      "epoch": 3.472712822324267,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003824783219741959,
      "loss": 5.2502,
      "mean_token_accuracy": 0.2143105685710907,
      "num_tokens": 73310187.0,
      "step": 39325
    },
    {
      "entropy": 5.912484455108642,
      "epoch": 3.4731543624161074,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038245065459027944,
      "loss": 5.1713,
      "mean_token_accuracy": 0.22393191158771514,
      "num_tokens": 73319667.0,
      "step": 39330
    },
    {
      "entropy": 5.953534317016602,
      "epoch": 3.473595902507948,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038242298510146567,
      "loss": 5.197,
      "mean_token_accuracy": 0.2247891381382942,
      "num_tokens": 73329217.0,
      "step": 39335
    },
    {
      "entropy": 6.034562301635742,
      "epoch": 3.474037442599788,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003823953135082967,
      "loss": 5.2376,
      "mean_token_accuracy": 0.2182897686958313,
      "num_tokens": 73337898.0,
      "step": 39340
    },
    {
      "entropy": 5.964702177047729,
      "epoch": 3.4744789826916285,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038236763981131465,
      "loss": 5.1815,
      "mean_token_accuracy": 0.22537890672683716,
      "num_tokens": 73346551.0,
      "step": 39345
    },
    {
      "entropy": 6.045445489883423,
      "epoch": 3.474920522783469,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003823399640110616,
      "loss": 5.3828,
      "mean_token_accuracy": 0.20673292130231857,
      "num_tokens": 73356821.0,
      "step": 39350
    },
    {
      "entropy": 6.049254369735718,
      "epoch": 3.475362062875309,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003823122861080797,
      "loss": 5.2772,
      "mean_token_accuracy": 0.2142346605658531,
      "num_tokens": 73366161.0,
      "step": 39355
    },
    {
      "entropy": 6.112970399856567,
      "epoch": 3.4758036029671495,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038228460610291123,
      "loss": 5.3338,
      "mean_token_accuracy": 0.20989432483911513,
      "num_tokens": 73375295.0,
      "step": 39360
    },
    {
      "entropy": 6.079036664962769,
      "epoch": 3.4762451430589896,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038225692399609826,
      "loss": 5.3193,
      "mean_token_accuracy": 0.2097224235534668,
      "num_tokens": 73386062.0,
      "step": 39365
    },
    {
      "entropy": 5.959751033782959,
      "epoch": 3.47668668315083,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003822292397881831,
      "loss": 5.2429,
      "mean_token_accuracy": 0.21946116536855698,
      "num_tokens": 73395206.0,
      "step": 39370
    },
    {
      "entropy": 6.040965175628662,
      "epoch": 3.4771282232426706,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038220155347970823,
      "loss": 5.3192,
      "mean_token_accuracy": 0.21142590939998626,
      "num_tokens": 73404971.0,
      "step": 39375
    },
    {
      "entropy": 6.127564907073975,
      "epoch": 3.4775697633345106,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038217386507121586,
      "loss": 5.4293,
      "mean_token_accuracy": 0.1993415892124176,
      "num_tokens": 73415120.0,
      "step": 39380
    },
    {
      "entropy": 6.048809814453125,
      "epoch": 3.478011303426351,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038214617456324844,
      "loss": 5.2951,
      "mean_token_accuracy": 0.21245165467262267,
      "num_tokens": 73424575.0,
      "step": 39385
    },
    {
      "entropy": 6.0043710231781,
      "epoch": 3.4784528435181916,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003821184819563484,
      "loss": 5.3158,
      "mean_token_accuracy": 0.211817429959774,
      "num_tokens": 73434539.0,
      "step": 39390
    },
    {
      "entropy": 5.926330089569092,
      "epoch": 3.4788943836100317,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038209078725105815,
      "loss": 5.1244,
      "mean_token_accuracy": 0.23256560713052749,
      "num_tokens": 73443134.0,
      "step": 39395
    },
    {
      "entropy": 6.020510530471801,
      "epoch": 3.479335923701872,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038206309044792034,
      "loss": 5.2327,
      "mean_token_accuracy": 0.220901021361351,
      "num_tokens": 73452332.0,
      "step": 39400
    },
    {
      "entropy": 5.931895399093628,
      "epoch": 3.4797774637937122,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038203539154747746,
      "loss": 5.2489,
      "mean_token_accuracy": 0.2196788877248764,
      "num_tokens": 73462066.0,
      "step": 39405
    },
    {
      "entropy": 5.990206527709961,
      "epoch": 3.4802190038855527,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003820076905502722,
      "loss": 5.2722,
      "mean_token_accuracy": 0.2120550751686096,
      "num_tokens": 73471676.0,
      "step": 39410
    },
    {
      "entropy": 6.075296783447266,
      "epoch": 3.480660543977393,
      "grad_norm": 1.1875,
      "learning_rate": 0.000381979987456847,
      "loss": 5.317,
      "mean_token_accuracy": 0.2136914238333702,
      "num_tokens": 73481834.0,
      "step": 39415
    },
    {
      "entropy": 5.931101131439209,
      "epoch": 3.4811020840692333,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00038195228226774473,
      "loss": 5.1594,
      "mean_token_accuracy": 0.22470639646053314,
      "num_tokens": 73491153.0,
      "step": 39420
    },
    {
      "entropy": 5.9991672992706295,
      "epoch": 3.4815436241610738,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038192457498350814,
      "loss": 5.1961,
      "mean_token_accuracy": 0.22118572741746903,
      "num_tokens": 73499961.0,
      "step": 39425
    },
    {
      "entropy": 6.043224859237671,
      "epoch": 3.4819851642529143,
      "grad_norm": 1.25,
      "learning_rate": 0.00038189686560467995,
      "loss": 5.3196,
      "mean_token_accuracy": 0.20921632200479506,
      "num_tokens": 73509016.0,
      "step": 39430
    },
    {
      "entropy": 5.9816460609436035,
      "epoch": 3.4824267043447543,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003818691541318029,
      "loss": 5.1834,
      "mean_token_accuracy": 0.22925851196050645,
      "num_tokens": 73517494.0,
      "step": 39435
    },
    {
      "entropy": 5.96790041923523,
      "epoch": 3.482868244436595,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038184144056542,
      "loss": 5.2429,
      "mean_token_accuracy": 0.22234198302030564,
      "num_tokens": 73527385.0,
      "step": 39440
    },
    {
      "entropy": 6.081214427947998,
      "epoch": 3.4833097845284353,
      "grad_norm": 1.375,
      "learning_rate": 0.0003818137249060739,
      "loss": 5.3129,
      "mean_token_accuracy": 0.21349373310804368,
      "num_tokens": 73535832.0,
      "step": 39445
    },
    {
      "entropy": 5.991124820709229,
      "epoch": 3.4837513246202754,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003817860071543078,
      "loss": 5.2338,
      "mean_token_accuracy": 0.2202399641275406,
      "num_tokens": 73545390.0,
      "step": 39450
    },
    {
      "entropy": 5.985174703598022,
      "epoch": 3.484192864712116,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00038175828731066446,
      "loss": 5.1637,
      "mean_token_accuracy": 0.23110085278749465,
      "num_tokens": 73554157.0,
      "step": 39455
    },
    {
      "entropy": 5.927543354034424,
      "epoch": 3.4846344048039564,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038173056537568707,
      "loss": 5.2322,
      "mean_token_accuracy": 0.22311265915632247,
      "num_tokens": 73563723.0,
      "step": 39460
    },
    {
      "entropy": 5.97398533821106,
      "epoch": 3.4850759448957964,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003817028413499185,
      "loss": 5.2168,
      "mean_token_accuracy": 0.2147776424884796,
      "num_tokens": 73572614.0,
      "step": 39465
    },
    {
      "entropy": 6.016584968566894,
      "epoch": 3.485517484987637,
      "grad_norm": 1.15625,
      "learning_rate": 0.000381675115233902,
      "loss": 5.3293,
      "mean_token_accuracy": 0.21292472630739212,
      "num_tokens": 73581446.0,
      "step": 39470
    },
    {
      "entropy": 6.027819871902466,
      "epoch": 3.4859590250794774,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038164738702818065,
      "loss": 5.2541,
      "mean_token_accuracy": 0.22050343602895736,
      "num_tokens": 73591426.0,
      "step": 39475
    },
    {
      "entropy": 5.963691711425781,
      "epoch": 3.4864005651713175,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038161965673329766,
      "loss": 5.1515,
      "mean_token_accuracy": 0.22831064611673355,
      "num_tokens": 73600262.0,
      "step": 39480
    },
    {
      "entropy": 5.955472946166992,
      "epoch": 3.486842105263158,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003815919243497963,
      "loss": 5.1833,
      "mean_token_accuracy": 0.2243058815598488,
      "num_tokens": 73609439.0,
      "step": 39485
    },
    {
      "entropy": 5.961241817474365,
      "epoch": 3.4872836453549985,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003815641898782197,
      "loss": 5.3617,
      "mean_token_accuracy": 0.21034234911203384,
      "num_tokens": 73619087.0,
      "step": 39490
    },
    {
      "entropy": 5.983580446243286,
      "epoch": 3.4877251854468385,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038153645331911117,
      "loss": 5.1175,
      "mean_token_accuracy": 0.2301471695303917,
      "num_tokens": 73627944.0,
      "step": 39495
    },
    {
      "entropy": 5.989296388626099,
      "epoch": 3.488166725538679,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003815087146730142,
      "loss": 5.262,
      "mean_token_accuracy": 0.21819087415933608,
      "num_tokens": 73636596.0,
      "step": 39500
    },
    {
      "entropy": 6.007274389266968,
      "epoch": 3.488608265630519,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003814809739404721,
      "loss": 5.2661,
      "mean_token_accuracy": 0.21433217823505402,
      "num_tokens": 73645012.0,
      "step": 39505
    },
    {
      "entropy": 5.996837711334228,
      "epoch": 3.4890498057223596,
      "grad_norm": 1.40625,
      "learning_rate": 0.00038145323112202817,
      "loss": 5.1881,
      "mean_token_accuracy": 0.21787006556987762,
      "num_tokens": 73653661.0,
      "step": 39510
    },
    {
      "entropy": 6.028957843780518,
      "epoch": 3.4894913458142,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038142548621822607,
      "loss": 5.3194,
      "mean_token_accuracy": 0.21274663507938385,
      "num_tokens": 73663302.0,
      "step": 39515
    },
    {
      "entropy": 5.9511829853057865,
      "epoch": 3.48993288590604,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038139773922960923,
      "loss": 5.1687,
      "mean_token_accuracy": 0.225909323990345,
      "num_tokens": 73672513.0,
      "step": 39520
    },
    {
      "entropy": 5.978934288024902,
      "epoch": 3.4903744259978806,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003813699901567212,
      "loss": 5.2655,
      "mean_token_accuracy": 0.21746921688318252,
      "num_tokens": 73681803.0,
      "step": 39525
    },
    {
      "entropy": 6.0368424415588375,
      "epoch": 3.490815966089721,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038134223900010547,
      "loss": 5.3627,
      "mean_token_accuracy": 0.21474342048168182,
      "num_tokens": 73692107.0,
      "step": 39530
    },
    {
      "entropy": 5.997737979888916,
      "epoch": 3.491257506181561,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038131448576030573,
      "loss": 5.1408,
      "mean_token_accuracy": 0.23078034073114395,
      "num_tokens": 73701371.0,
      "step": 39535
    },
    {
      "entropy": 6.046863555908203,
      "epoch": 3.4916990462734017,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003812867304378658,
      "loss": 5.3432,
      "mean_token_accuracy": 0.21298041194677353,
      "num_tokens": 73710647.0,
      "step": 39540
    },
    {
      "entropy": 6.103954696655274,
      "epoch": 3.4921405863652417,
      "grad_norm": 1.34375,
      "learning_rate": 0.00038125897303332924,
      "loss": 5.3471,
      "mean_token_accuracy": 0.2045069679617882,
      "num_tokens": 73720144.0,
      "step": 39545
    },
    {
      "entropy": 6.020205020904541,
      "epoch": 3.492582126457082,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003812312135472398,
      "loss": 5.2895,
      "mean_token_accuracy": 0.21409862190485002,
      "num_tokens": 73730149.0,
      "step": 39550
    },
    {
      "entropy": 5.934813404083252,
      "epoch": 3.4930236665489227,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038120345198014135,
      "loss": 5.1471,
      "mean_token_accuracy": 0.22133724689483641,
      "num_tokens": 73739424.0,
      "step": 39555
    },
    {
      "entropy": 6.003522253036499,
      "epoch": 3.4934652066407628,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038117568833257766,
      "loss": 5.2756,
      "mean_token_accuracy": 0.21643869578838348,
      "num_tokens": 73748747.0,
      "step": 39560
    },
    {
      "entropy": 5.971637630462647,
      "epoch": 3.4939067467326033,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038114792260509256,
      "loss": 5.2917,
      "mean_token_accuracy": 0.21960335075855256,
      "num_tokens": 73758400.0,
      "step": 39565
    },
    {
      "entropy": 6.042917013168335,
      "epoch": 3.4943482868244438,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003811201547982301,
      "loss": 5.3072,
      "mean_token_accuracy": 0.21143312007188797,
      "num_tokens": 73766521.0,
      "step": 39570
    },
    {
      "entropy": 6.044754123687744,
      "epoch": 3.494789826916284,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003810923849125341,
      "loss": 5.2596,
      "mean_token_accuracy": 0.22478238344192505,
      "num_tokens": 73776017.0,
      "step": 39575
    },
    {
      "entropy": 6.003633642196656,
      "epoch": 3.4952313670081243,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038106461294854866,
      "loss": 5.2435,
      "mean_token_accuracy": 0.2197526216506958,
      "num_tokens": 73785974.0,
      "step": 39580
    },
    {
      "entropy": 6.0101480960845945,
      "epoch": 3.495672907099965,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003810368389068178,
      "loss": 5.3515,
      "mean_token_accuracy": 0.2127833053469658,
      "num_tokens": 73795533.0,
      "step": 39585
    },
    {
      "entropy": 5.963533496856689,
      "epoch": 3.496114447191805,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00038100906278788554,
      "loss": 5.1873,
      "mean_token_accuracy": 0.2228034555912018,
      "num_tokens": 73804189.0,
      "step": 39590
    },
    {
      "entropy": 5.98062949180603,
      "epoch": 3.4965559872836454,
      "grad_norm": 1.34375,
      "learning_rate": 0.000380981284592296,
      "loss": 5.2323,
      "mean_token_accuracy": 0.22544790655374528,
      "num_tokens": 73812881.0,
      "step": 39595
    },
    {
      "entropy": 6.038416481018066,
      "epoch": 3.496997527375486,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003809535043205934,
      "loss": 5.4006,
      "mean_token_accuracy": 0.19732156693935393,
      "num_tokens": 73823209.0,
      "step": 39600
    },
    {
      "entropy": 6.0375806331634525,
      "epoch": 3.497439067467326,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003809257219733219,
      "loss": 5.2094,
      "mean_token_accuracy": 0.22163084894418716,
      "num_tokens": 73832054.0,
      "step": 39605
    },
    {
      "entropy": 6.0413689613342285,
      "epoch": 3.4978806075591664,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038089793755102574,
      "loss": 5.2609,
      "mean_token_accuracy": 0.21732541918754578,
      "num_tokens": 73840902.0,
      "step": 39610
    },
    {
      "entropy": 5.977435207366943,
      "epoch": 3.498322147651007,
      "grad_norm": 1.328125,
      "learning_rate": 0.00038087015105424925,
      "loss": 5.1667,
      "mean_token_accuracy": 0.21819917112588882,
      "num_tokens": 73850447.0,
      "step": 39615
    },
    {
      "entropy": 5.975948810577393,
      "epoch": 3.498763687742847,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038084236248353673,
      "loss": 5.2415,
      "mean_token_accuracy": 0.2204771190881729,
      "num_tokens": 73859918.0,
      "step": 39620
    },
    {
      "entropy": 5.961999702453613,
      "epoch": 3.4992052278346875,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038081457183943246,
      "loss": 5.2213,
      "mean_token_accuracy": 0.22773680388927459,
      "num_tokens": 73868579.0,
      "step": 39625
    },
    {
      "entropy": 6.037107467651367,
      "epoch": 3.499646767926528,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003807867791224809,
      "loss": 5.3102,
      "mean_token_accuracy": 0.21528023928403855,
      "num_tokens": 73877695.0,
      "step": 39630
    },
    {
      "entropy": 5.941096305847168,
      "epoch": 3.500088308018368,
      "grad_norm": 1.25,
      "learning_rate": 0.0003807589843332266,
      "loss": 5.2167,
      "mean_token_accuracy": 0.22238062918186188,
      "num_tokens": 73886792.0,
      "step": 39635
    },
    {
      "entropy": 6.020889616012573,
      "epoch": 3.5005298481102085,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003807311874722138,
      "loss": 5.2843,
      "mean_token_accuracy": 0.2132757693529129,
      "num_tokens": 73896567.0,
      "step": 39640
    },
    {
      "entropy": 6.070259666442871,
      "epoch": 3.500971388202049,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003807033885399873,
      "loss": 5.3358,
      "mean_token_accuracy": 0.20757968872785568,
      "num_tokens": 73905959.0,
      "step": 39645
    },
    {
      "entropy": 6.016506290435791,
      "epoch": 3.501412928293889,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038067558753709143,
      "loss": 5.2619,
      "mean_token_accuracy": 0.21611480861902238,
      "num_tokens": 73915389.0,
      "step": 39650
    },
    {
      "entropy": 6.002035617828369,
      "epoch": 3.5018544683857296,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000380647784464071,
      "loss": 5.36,
      "mean_token_accuracy": 0.20670391023159027,
      "num_tokens": 73924773.0,
      "step": 39655
    },
    {
      "entropy": 5.955693626403809,
      "epoch": 3.5022960084775696,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003806199793214705,
      "loss": 5.208,
      "mean_token_accuracy": 0.21976051479578018,
      "num_tokens": 73933809.0,
      "step": 39660
    },
    {
      "entropy": 5.979448032379151,
      "epoch": 3.50273754856941,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003805921721098347,
      "loss": 5.2425,
      "mean_token_accuracy": 0.22276560813188553,
      "num_tokens": 73942264.0,
      "step": 39665
    },
    {
      "entropy": 5.98626937866211,
      "epoch": 3.5031790886612506,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038056436282970824,
      "loss": 5.2197,
      "mean_token_accuracy": 0.217099691927433,
      "num_tokens": 73950912.0,
      "step": 39670
    },
    {
      "entropy": 5.911801147460937,
      "epoch": 3.5036206287530907,
      "grad_norm": 1.21875,
      "learning_rate": 0.000380536551481636,
      "loss": 5.1693,
      "mean_token_accuracy": 0.23524729609489442,
      "num_tokens": 73960082.0,
      "step": 39675
    },
    {
      "entropy": 6.015303325653076,
      "epoch": 3.504062168844931,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003805087380661628,
      "loss": 5.2621,
      "mean_token_accuracy": 0.21583391577005387,
      "num_tokens": 73969849.0,
      "step": 39680
    },
    {
      "entropy": 6.012163925170898,
      "epoch": 3.504503708936771,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003804809225838333,
      "loss": 5.2357,
      "mean_token_accuracy": 0.21732222735881807,
      "num_tokens": 73979129.0,
      "step": 39685
    },
    {
      "entropy": 6.000686264038086,
      "epoch": 3.5049452490286117,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038045310503519267,
      "loss": 5.2339,
      "mean_token_accuracy": 0.22281481474637985,
      "num_tokens": 73987692.0,
      "step": 39690
    },
    {
      "entropy": 5.967072010040283,
      "epoch": 3.505386789120452,
      "grad_norm": 1.125,
      "learning_rate": 0.0003804252854207855,
      "loss": 5.2646,
      "mean_token_accuracy": 0.2222661629319191,
      "num_tokens": 73997472.0,
      "step": 39695
    },
    {
      "entropy": 6.033125972747802,
      "epoch": 3.5058283292122923,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003803974637411571,
      "loss": 5.3538,
      "mean_token_accuracy": 0.21215066760778428,
      "num_tokens": 74006818.0,
      "step": 39700
    },
    {
      "entropy": 5.97237377166748,
      "epoch": 3.5062698693041328,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038036963999685226,
      "loss": 5.1969,
      "mean_token_accuracy": 0.22347831577062607,
      "num_tokens": 74016070.0,
      "step": 39705
    },
    {
      "entropy": 6.023147964477539,
      "epoch": 3.5067114093959733,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003803418141884162,
      "loss": 5.3213,
      "mean_token_accuracy": 0.2088426560163498,
      "num_tokens": 74025344.0,
      "step": 39710
    },
    {
      "entropy": 5.976116371154785,
      "epoch": 3.5071529494878133,
      "grad_norm": 1.328125,
      "learning_rate": 0.00038031398631639375,
      "loss": 5.1526,
      "mean_token_accuracy": 0.2252844750881195,
      "num_tokens": 74033788.0,
      "step": 39715
    },
    {
      "entropy": 6.0430042266845705,
      "epoch": 3.507594489579654,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003802861563813303,
      "loss": 5.2035,
      "mean_token_accuracy": 0.2225797101855278,
      "num_tokens": 74042641.0,
      "step": 39720
    },
    {
      "entropy": 5.958821153640747,
      "epoch": 3.5080360296714943,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003802583243837709,
      "loss": 5.2794,
      "mean_token_accuracy": 0.21726609617471696,
      "num_tokens": 74051260.0,
      "step": 39725
    },
    {
      "entropy": 6.014234638214111,
      "epoch": 3.5084775697633344,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003802304903242607,
      "loss": 5.3214,
      "mean_token_accuracy": 0.20417247265577315,
      "num_tokens": 74061416.0,
      "step": 39730
    },
    {
      "entropy": 6.067371988296509,
      "epoch": 3.508919109855175,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038020265420334515,
      "loss": 5.2848,
      "mean_token_accuracy": 0.21485998183488847,
      "num_tokens": 74070027.0,
      "step": 39735
    },
    {
      "entropy": 6.00372953414917,
      "epoch": 3.5093606499470154,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038017481602156943,
      "loss": 5.2054,
      "mean_token_accuracy": 0.22404781281948088,
      "num_tokens": 74079731.0,
      "step": 39740
    },
    {
      "entropy": 5.922117757797241,
      "epoch": 3.5098021900388554,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003801469757794789,
      "loss": 5.1372,
      "mean_token_accuracy": 0.22705452591180803,
      "num_tokens": 74088664.0,
      "step": 39745
    },
    {
      "entropy": 6.007049179077148,
      "epoch": 3.510243730130696,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003801191334776189,
      "loss": 5.1995,
      "mean_token_accuracy": 0.22336635887622833,
      "num_tokens": 74097846.0,
      "step": 39750
    },
    {
      "entropy": 6.059465503692627,
      "epoch": 3.5106852702225364,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038009128911653485,
      "loss": 5.351,
      "mean_token_accuracy": 0.2048035368323326,
      "num_tokens": 74107939.0,
      "step": 39755
    },
    {
      "entropy": 6.006894254684449,
      "epoch": 3.5111268103143765,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003800634426967722,
      "loss": 5.2297,
      "mean_token_accuracy": 0.22233900874853135,
      "num_tokens": 74118514.0,
      "step": 39760
    },
    {
      "entropy": 6.0526838302612305,
      "epoch": 3.511568350406217,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038003559421887645,
      "loss": 5.2898,
      "mean_token_accuracy": 0.21346719563007355,
      "num_tokens": 74127479.0,
      "step": 39765
    },
    {
      "entropy": 5.962346982955933,
      "epoch": 3.5120098904980575,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003800077436833932,
      "loss": 5.1874,
      "mean_token_accuracy": 0.22529797106981278,
      "num_tokens": 74136572.0,
      "step": 39770
    },
    {
      "entropy": 5.947238731384277,
      "epoch": 3.5124514305898975,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037997989109086794,
      "loss": 5.1912,
      "mean_token_accuracy": 0.22870098501443864,
      "num_tokens": 74145917.0,
      "step": 39775
    },
    {
      "entropy": 6.053023052215576,
      "epoch": 3.512892970681738,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003799520364418463,
      "loss": 5.291,
      "mean_token_accuracy": 0.2154805302619934,
      "num_tokens": 74154574.0,
      "step": 39780
    },
    {
      "entropy": 5.99984827041626,
      "epoch": 3.5133345107735785,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037992417973687406,
      "loss": 5.258,
      "mean_token_accuracy": 0.2206464886665344,
      "num_tokens": 74164263.0,
      "step": 39785
    },
    {
      "entropy": 6.041320705413819,
      "epoch": 3.5137760508654186,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037989632097649664,
      "loss": 5.3701,
      "mean_token_accuracy": 0.20803570449352266,
      "num_tokens": 74174823.0,
      "step": 39790
    },
    {
      "entropy": 5.975899171829224,
      "epoch": 3.514217590957259,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037986846016126,
      "loss": 5.1947,
      "mean_token_accuracy": 0.21731523871421815,
      "num_tokens": 74183571.0,
      "step": 39795
    },
    {
      "entropy": 5.998290252685547,
      "epoch": 3.514659131049099,
      "grad_norm": 1.296875,
      "learning_rate": 0.00037984059729171,
      "loss": 5.2145,
      "mean_token_accuracy": 0.22070469558238984,
      "num_tokens": 74192525.0,
      "step": 39800
    },
    {
      "entropy": 6.048928117752075,
      "epoch": 3.5151006711409396,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037981273236839216,
      "loss": 5.3162,
      "mean_token_accuracy": 0.20952077507972716,
      "num_tokens": 74202494.0,
      "step": 39805
    },
    {
      "entropy": 5.991612815856934,
      "epoch": 3.51554221123278,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037978486539185256,
      "loss": 5.2679,
      "mean_token_accuracy": 0.21404783874750138,
      "num_tokens": 74212509.0,
      "step": 39810
    },
    {
      "entropy": 6.03153624534607,
      "epoch": 3.51598375132462,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037975699636263706,
      "loss": 5.2565,
      "mean_token_accuracy": 0.21165734976530076,
      "num_tokens": 74222002.0,
      "step": 39815
    },
    {
      "entropy": 5.972796869277954,
      "epoch": 3.5164252914164607,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003797291252812916,
      "loss": 5.217,
      "mean_token_accuracy": 0.22478564083576202,
      "num_tokens": 74231411.0,
      "step": 39820
    },
    {
      "entropy": 5.917395114898682,
      "epoch": 3.5168668315083007,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037970125214836206,
      "loss": 5.269,
      "mean_token_accuracy": 0.2239825025200844,
      "num_tokens": 74240962.0,
      "step": 39825
    },
    {
      "entropy": 6.010361289978027,
      "epoch": 3.517308371600141,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003796733769643946,
      "loss": 5.3078,
      "mean_token_accuracy": 0.20743731558322906,
      "num_tokens": 74250867.0,
      "step": 39830
    },
    {
      "entropy": 6.057699155807495,
      "epoch": 3.5177499116919817,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037964549972993524,
      "loss": 5.3392,
      "mean_token_accuracy": 0.21388272047042847,
      "num_tokens": 74261105.0,
      "step": 39835
    },
    {
      "entropy": 5.958799552917481,
      "epoch": 3.5181914517838218,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037961762044553,
      "loss": 5.2202,
      "mean_token_accuracy": 0.22194215059280395,
      "num_tokens": 74271252.0,
      "step": 39840
    },
    {
      "entropy": 5.956078433990479,
      "epoch": 3.5186329918756623,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003795897391117251,
      "loss": 5.1988,
      "mean_token_accuracy": 0.2197341576218605,
      "num_tokens": 74280954.0,
      "step": 39845
    },
    {
      "entropy": 5.994592952728271,
      "epoch": 3.5190745319675028,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037956185572906665,
      "loss": 5.3058,
      "mean_token_accuracy": 0.2193632557988167,
      "num_tokens": 74290819.0,
      "step": 39850
    },
    {
      "entropy": 6.042195415496826,
      "epoch": 3.519516072059343,
      "grad_norm": 1.359375,
      "learning_rate": 0.000379533970298101,
      "loss": 5.4213,
      "mean_token_accuracy": 0.20448486804962157,
      "num_tokens": 74299363.0,
      "step": 39855
    },
    {
      "entropy": 6.015786790847779,
      "epoch": 3.5199576121511833,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003795060828193743,
      "loss": 5.2483,
      "mean_token_accuracy": 0.22052504420280455,
      "num_tokens": 74309360.0,
      "step": 39860
    },
    {
      "entropy": 5.976793670654297,
      "epoch": 3.520399152243024,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003794781932934328,
      "loss": 5.237,
      "mean_token_accuracy": 0.21986875236034392,
      "num_tokens": 74318694.0,
      "step": 39865
    },
    {
      "entropy": 6.007813262939453,
      "epoch": 3.520840692334864,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037945030172082295,
      "loss": 5.3357,
      "mean_token_accuracy": 0.2159380540251732,
      "num_tokens": 74328180.0,
      "step": 39870
    },
    {
      "entropy": 5.941378498077393,
      "epoch": 3.5212822324267044,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037942240810209116,
      "loss": 5.2848,
      "mean_token_accuracy": 0.21278072744607926,
      "num_tokens": 74338522.0,
      "step": 39875
    },
    {
      "entropy": 6.012328433990478,
      "epoch": 3.521723772518545,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003793945124377837,
      "loss": 5.2963,
      "mean_token_accuracy": 0.2156529262661934,
      "num_tokens": 74347968.0,
      "step": 39880
    },
    {
      "entropy": 6.001794481277466,
      "epoch": 3.522165312610385,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003793666147284472,
      "loss": 5.2457,
      "mean_token_accuracy": 0.22149786353111267,
      "num_tokens": 74356733.0,
      "step": 39885
    },
    {
      "entropy": 6.029110336303711,
      "epoch": 3.5226068527022254,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037933871497462796,
      "loss": 5.3664,
      "mean_token_accuracy": 0.21301701068878173,
      "num_tokens": 74365767.0,
      "step": 39890
    },
    {
      "entropy": 6.0304014682769775,
      "epoch": 3.523048392794066,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037931081317687273,
      "loss": 5.2331,
      "mean_token_accuracy": 0.21912465393543243,
      "num_tokens": 74375383.0,
      "step": 39895
    },
    {
      "entropy": 6.023105049133301,
      "epoch": 3.523489932885906,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037928290933572796,
      "loss": 5.2899,
      "mean_token_accuracy": 0.2148909881711006,
      "num_tokens": 74385513.0,
      "step": 39900
    },
    {
      "entropy": 6.03791356086731,
      "epoch": 3.5239314729777464,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003792550034517403,
      "loss": 5.3507,
      "mean_token_accuracy": 0.2141462028026581,
      "num_tokens": 74394805.0,
      "step": 39905
    },
    {
      "entropy": 5.973772716522217,
      "epoch": 3.524373013069587,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037922709552545635,
      "loss": 5.1943,
      "mean_token_accuracy": 0.22776255160570144,
      "num_tokens": 74403756.0,
      "step": 39910
    },
    {
      "entropy": 6.047693490982056,
      "epoch": 3.524814553161427,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037919918555742293,
      "loss": 5.3295,
      "mean_token_accuracy": 0.21177584379911424,
      "num_tokens": 74413198.0,
      "step": 39915
    },
    {
      "entropy": 6.061608934402466,
      "epoch": 3.5252560932532675,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003791712735481868,
      "loss": 5.2889,
      "mean_token_accuracy": 0.2173197641968727,
      "num_tokens": 74422732.0,
      "step": 39920
    },
    {
      "entropy": 5.939625549316406,
      "epoch": 3.525697633345108,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037914335949829455,
      "loss": 5.1062,
      "mean_token_accuracy": 0.23072348833084105,
      "num_tokens": 74431421.0,
      "step": 39925
    },
    {
      "entropy": 5.914484310150146,
      "epoch": 3.526139173436948,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037911544340829323,
      "loss": 5.1536,
      "mean_token_accuracy": 0.22594211846590043,
      "num_tokens": 74440432.0,
      "step": 39930
    },
    {
      "entropy": 5.963503217697143,
      "epoch": 3.5265807135287885,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003790875252787295,
      "loss": 5.295,
      "mean_token_accuracy": 0.2171752855181694,
      "num_tokens": 74449969.0,
      "step": 39935
    },
    {
      "entropy": 5.923338079452515,
      "epoch": 3.5270222536206286,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003790596051101504,
      "loss": 5.1942,
      "mean_token_accuracy": 0.21670351028442383,
      "num_tokens": 74458428.0,
      "step": 39940
    },
    {
      "entropy": 5.995209360122681,
      "epoch": 3.527463793712469,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037903168290310277,
      "loss": 5.2585,
      "mean_token_accuracy": 0.22157669365406035,
      "num_tokens": 74467625.0,
      "step": 39945
    },
    {
      "entropy": 5.980369949340821,
      "epoch": 3.5279053338043096,
      "grad_norm": 1.328125,
      "learning_rate": 0.00037900375865813365,
      "loss": 5.2488,
      "mean_token_accuracy": 0.21804854720830918,
      "num_tokens": 74477336.0,
      "step": 39950
    },
    {
      "entropy": 5.989242792129517,
      "epoch": 3.5283468738961496,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037897583237579,
      "loss": 5.1881,
      "mean_token_accuracy": 0.22082220762968063,
      "num_tokens": 74486615.0,
      "step": 39955
    },
    {
      "entropy": 5.943365621566772,
      "epoch": 3.52878841398799,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003789479040566191,
      "loss": 5.1875,
      "mean_token_accuracy": 0.23032441288232802,
      "num_tokens": 74495838.0,
      "step": 39960
    },
    {
      "entropy": 5.968229675292969,
      "epoch": 3.52922995407983,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003789199737011678,
      "loss": 5.1584,
      "mean_token_accuracy": 0.22178806364536285,
      "num_tokens": 74504231.0,
      "step": 39965
    },
    {
      "entropy": 6.007981634140014,
      "epoch": 3.5296714941716707,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037889204130998324,
      "loss": 5.2817,
      "mean_token_accuracy": 0.21553438305854797,
      "num_tokens": 74515021.0,
      "step": 39970
    },
    {
      "entropy": 6.04363260269165,
      "epoch": 3.530113034263511,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00037886410688361266,
      "loss": 5.33,
      "mean_token_accuracy": 0.20757727473974227,
      "num_tokens": 74524120.0,
      "step": 39975
    },
    {
      "entropy": 6.119251918792725,
      "epoch": 3.5305545743553512,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003788361704226034,
      "loss": 5.3739,
      "mean_token_accuracy": 0.21327370405197144,
      "num_tokens": 74534362.0,
      "step": 39980
    },
    {
      "entropy": 6.076960229873658,
      "epoch": 3.5309961144471917,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003788082319275026,
      "loss": 5.2887,
      "mean_token_accuracy": 0.2209983468055725,
      "num_tokens": 74543859.0,
      "step": 39985
    },
    {
      "entropy": 5.972440958023071,
      "epoch": 3.5314376545390322,
      "grad_norm": 1.25,
      "learning_rate": 0.0003787802913988575,
      "loss": 5.2276,
      "mean_token_accuracy": 0.22595645487308502,
      "num_tokens": 74553443.0,
      "step": 39990
    },
    {
      "entropy": 5.954843807220459,
      "epoch": 3.5318791946308723,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037875234883721556,
      "loss": 5.208,
      "mean_token_accuracy": 0.22286434322595597,
      "num_tokens": 74562221.0,
      "step": 39995
    },
    {
      "entropy": 6.021269416809082,
      "epoch": 3.532320734722713,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037872440424312415,
      "loss": 5.2239,
      "mean_token_accuracy": 0.22337252050638198,
      "num_tokens": 74571535.0,
      "step": 40000
    },
    {
      "entropy": 6.028404235839844,
      "epoch": 3.5327622748145533,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037869645761713064,
      "loss": 5.275,
      "mean_token_accuracy": 0.214714677631855,
      "num_tokens": 74580032.0,
      "step": 40005
    },
    {
      "entropy": 5.974641942977906,
      "epoch": 3.5332038149063933,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037866850895978244,
      "loss": 5.1784,
      "mean_token_accuracy": 0.2203855350613594,
      "num_tokens": 74588769.0,
      "step": 40010
    },
    {
      "entropy": 6.02975549697876,
      "epoch": 3.533645354998234,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037864055827162714,
      "loss": 5.3132,
      "mean_token_accuracy": 0.21142329424619674,
      "num_tokens": 74598859.0,
      "step": 40015
    },
    {
      "entropy": 5.961709642410279,
      "epoch": 3.5340868950900743,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037861260555321224,
      "loss": 5.2205,
      "mean_token_accuracy": 0.22083683162927628,
      "num_tokens": 74608268.0,
      "step": 40020
    },
    {
      "entropy": 6.0364892959594725,
      "epoch": 3.5345284351819144,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037858465080508534,
      "loss": 5.3713,
      "mean_token_accuracy": 0.20859326869249345,
      "num_tokens": 74617159.0,
      "step": 40025
    },
    {
      "entropy": 6.017865324020386,
      "epoch": 3.534969975273755,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000378556694027794,
      "loss": 5.2314,
      "mean_token_accuracy": 0.21236547231674194,
      "num_tokens": 74626271.0,
      "step": 40030
    },
    {
      "entropy": 6.0162591457366945,
      "epoch": 3.5354115153655954,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003785287352218861,
      "loss": 5.2459,
      "mean_token_accuracy": 0.21670022010803222,
      "num_tokens": 74636552.0,
      "step": 40035
    },
    {
      "entropy": 6.030885791778564,
      "epoch": 3.5358530554574354,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003785007743879089,
      "loss": 5.2655,
      "mean_token_accuracy": 0.2182398959994316,
      "num_tokens": 74646002.0,
      "step": 40040
    },
    {
      "entropy": 5.955898761749268,
      "epoch": 3.536294595549276,
      "grad_norm": 1.3125,
      "learning_rate": 0.00037847281152641056,
      "loss": 5.157,
      "mean_token_accuracy": 0.22119407653808593,
      "num_tokens": 74655634.0,
      "step": 40045
    },
    {
      "entropy": 6.040172004699707,
      "epoch": 3.5367361356411164,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037844484663793865,
      "loss": 5.2943,
      "mean_token_accuracy": 0.21360642611980438,
      "num_tokens": 74665554.0,
      "step": 40050
    },
    {
      "entropy": 6.069248867034912,
      "epoch": 3.5371776757329565,
      "grad_norm": 1.078125,
      "learning_rate": 0.000378416879723041,
      "loss": 5.3942,
      "mean_token_accuracy": 0.2082437589764595,
      "num_tokens": 74675510.0,
      "step": 40055
    },
    {
      "entropy": 6.0381299495697025,
      "epoch": 3.537619215824797,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003783889107822655,
      "loss": 5.2833,
      "mean_token_accuracy": 0.2245136708021164,
      "num_tokens": 74684032.0,
      "step": 40060
    },
    {
      "entropy": 5.989026737213135,
      "epoch": 3.5380607559166375,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003783609398161601,
      "loss": 5.229,
      "mean_token_accuracy": 0.21796538829803466,
      "num_tokens": 74694210.0,
      "step": 40065
    },
    {
      "entropy": 6.0116277694702145,
      "epoch": 3.5385022960084775,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003783329668252726,
      "loss": 5.2375,
      "mean_token_accuracy": 0.21821979135274888,
      "num_tokens": 74702940.0,
      "step": 40070
    },
    {
      "entropy": 6.003001165390015,
      "epoch": 3.538943836100318,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037830499181015103,
      "loss": 5.2475,
      "mean_token_accuracy": 0.21083034723997116,
      "num_tokens": 74711807.0,
      "step": 40075
    },
    {
      "entropy": 6.003211402893067,
      "epoch": 3.539385376192158,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037827701477134354,
      "loss": 5.3216,
      "mean_token_accuracy": 0.22210191041231156,
      "num_tokens": 74721634.0,
      "step": 40080
    },
    {
      "entropy": 6.027966594696045,
      "epoch": 3.5398269162839986,
      "grad_norm": 1.25,
      "learning_rate": 0.000378249035709398,
      "loss": 5.2371,
      "mean_token_accuracy": 0.22640684247016907,
      "num_tokens": 74730379.0,
      "step": 40085
    },
    {
      "entropy": 5.967738437652588,
      "epoch": 3.540268456375839,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037822105462486256,
      "loss": 5.1971,
      "mean_token_accuracy": 0.22347864061594008,
      "num_tokens": 74739255.0,
      "step": 40090
    },
    {
      "entropy": 5.980268383026123,
      "epoch": 3.540709996467679,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003781930715182853,
      "loss": 5.2599,
      "mean_token_accuracy": 0.21731191575527192,
      "num_tokens": 74747624.0,
      "step": 40095
    },
    {
      "entropy": 6.043707180023193,
      "epoch": 3.5411515365595196,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037816508639021456,
      "loss": 5.3127,
      "mean_token_accuracy": 0.20991480052471162,
      "num_tokens": 74756728.0,
      "step": 40100
    },
    {
      "entropy": 6.015752172470092,
      "epoch": 3.5415930766513597,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037813709924119844,
      "loss": 5.2951,
      "mean_token_accuracy": 0.20890674144029617,
      "num_tokens": 74765268.0,
      "step": 40105
    },
    {
      "entropy": 5.871838808059692,
      "epoch": 3.5420346167432,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003781091100717852,
      "loss": 5.1475,
      "mean_token_accuracy": 0.22681569010019303,
      "num_tokens": 74774599.0,
      "step": 40110
    },
    {
      "entropy": 5.987183904647827,
      "epoch": 3.5424761568350407,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003780811188825231,
      "loss": 5.3351,
      "mean_token_accuracy": 0.2065553233027458,
      "num_tokens": 74785114.0,
      "step": 40115
    },
    {
      "entropy": 6.039413261413574,
      "epoch": 3.5429176969268807,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003780531256739605,
      "loss": 5.2473,
      "mean_token_accuracy": 0.2187579721212387,
      "num_tokens": 74794646.0,
      "step": 40120
    },
    {
      "entropy": 5.9413291931152346,
      "epoch": 3.5433592370187212,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003780251304466458,
      "loss": 5.1326,
      "mean_token_accuracy": 0.23266733586788177,
      "num_tokens": 74803725.0,
      "step": 40125
    },
    {
      "entropy": 5.969723749160766,
      "epoch": 3.5438007771105617,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003779971332011274,
      "loss": 5.2263,
      "mean_token_accuracy": 0.21105937063694,
      "num_tokens": 74812564.0,
      "step": 40130
    },
    {
      "entropy": 6.020210886001587,
      "epoch": 3.544242317202402,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037796913393795374,
      "loss": 5.3113,
      "mean_token_accuracy": 0.2122286707162857,
      "num_tokens": 74821987.0,
      "step": 40135
    },
    {
      "entropy": 5.95532341003418,
      "epoch": 3.5446838572942423,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003779411326576733,
      "loss": 5.1634,
      "mean_token_accuracy": 0.22598518580198287,
      "num_tokens": 74831225.0,
      "step": 40140
    },
    {
      "entropy": 6.078421115875244,
      "epoch": 3.545125397386083,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003779131293608346,
      "loss": 5.3317,
      "mean_token_accuracy": 0.2161842554807663,
      "num_tokens": 74840394.0,
      "step": 40145
    },
    {
      "entropy": 5.993736934661865,
      "epoch": 3.545566937477923,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037788512404798624,
      "loss": 5.2256,
      "mean_token_accuracy": 0.21913499236106873,
      "num_tokens": 74849073.0,
      "step": 40150
    },
    {
      "entropy": 5.952115726470947,
      "epoch": 3.5460084775697633,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037785711671967684,
      "loss": 5.2023,
      "mean_token_accuracy": 0.21353587955236436,
      "num_tokens": 74858719.0,
      "step": 40155
    },
    {
      "entropy": 6.057963943481445,
      "epoch": 3.546450017661604,
      "grad_norm": 1.25,
      "learning_rate": 0.00037782910737645505,
      "loss": 5.24,
      "mean_token_accuracy": 0.21412381678819656,
      "num_tokens": 74868473.0,
      "step": 40160
    },
    {
      "entropy": 6.04938325881958,
      "epoch": 3.546891557753444,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037780109601886943,
      "loss": 5.2947,
      "mean_token_accuracy": 0.2155188724398613,
      "num_tokens": 74878116.0,
      "step": 40165
    },
    {
      "entropy": 6.054984855651855,
      "epoch": 3.5473330978452844,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003777730826474688,
      "loss": 5.2636,
      "mean_token_accuracy": 0.2189791738986969,
      "num_tokens": 74886931.0,
      "step": 40170
    },
    {
      "entropy": 5.987494087219238,
      "epoch": 3.547774637937125,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000377745067262802,
      "loss": 5.1825,
      "mean_token_accuracy": 0.21748789995908738,
      "num_tokens": 74895701.0,
      "step": 40175
    },
    {
      "entropy": 6.048718404769898,
      "epoch": 3.548216178028965,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003777170498654177,
      "loss": 5.2202,
      "mean_token_accuracy": 0.21920957118272782,
      "num_tokens": 74904658.0,
      "step": 40180
    },
    {
      "entropy": 5.9931262016296385,
      "epoch": 3.5486577181208054,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037768903045586484,
      "loss": 5.3395,
      "mean_token_accuracy": 0.21277833878993987,
      "num_tokens": 74914888.0,
      "step": 40185
    },
    {
      "entropy": 6.017413139343262,
      "epoch": 3.549099258212646,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003776610090346924,
      "loss": 5.3022,
      "mean_token_accuracy": 0.21165176928043367,
      "num_tokens": 74923622.0,
      "step": 40190
    },
    {
      "entropy": 6.020083999633789,
      "epoch": 3.549540798304486,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037763298560244897,
      "loss": 5.2437,
      "mean_token_accuracy": 0.21899598091840744,
      "num_tokens": 74933941.0,
      "step": 40195
    },
    {
      "entropy": 5.999558687210083,
      "epoch": 3.5499823383963265,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003776049601596839,
      "loss": 5.3239,
      "mean_token_accuracy": 0.21269194185733795,
      "num_tokens": 74943051.0,
      "step": 40200
    },
    {
      "entropy": 6.02444224357605,
      "epoch": 3.550423878488167,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003775769327069459,
      "loss": 5.2148,
      "mean_token_accuracy": 0.22431417256593705,
      "num_tokens": 74951510.0,
      "step": 40205
    },
    {
      "entropy": 5.992844343185425,
      "epoch": 3.550865418580007,
      "grad_norm": 1.25,
      "learning_rate": 0.0003775489032447842,
      "loss": 5.225,
      "mean_token_accuracy": 0.22577967643737792,
      "num_tokens": 74960452.0,
      "step": 40210
    },
    {
      "entropy": 5.909494209289551,
      "epoch": 3.5513069586718475,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003775208717737477,
      "loss": 5.1375,
      "mean_token_accuracy": 0.23299597799777985,
      "num_tokens": 74969241.0,
      "step": 40215
    },
    {
      "entropy": 5.980074310302735,
      "epoch": 3.5517484987636876,
      "grad_norm": 1.296875,
      "learning_rate": 0.00037749283829438573,
      "loss": 5.2244,
      "mean_token_accuracy": 0.2169053450226784,
      "num_tokens": 74978656.0,
      "step": 40220
    },
    {
      "entropy": 5.902518701553345,
      "epoch": 3.552190038855528,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003774648028072474,
      "loss": 5.0645,
      "mean_token_accuracy": 0.23682372272014618,
      "num_tokens": 74988261.0,
      "step": 40225
    },
    {
      "entropy": 6.022164344787598,
      "epoch": 3.5526315789473686,
      "grad_norm": 1.34375,
      "learning_rate": 0.00037743676531288173,
      "loss": 5.2442,
      "mean_token_accuracy": 0.22053935676813125,
      "num_tokens": 74996962.0,
      "step": 40230
    },
    {
      "entropy": 6.032794046401977,
      "epoch": 3.5530731190392086,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003774087258118381,
      "loss": 5.3019,
      "mean_token_accuracy": 0.2195504739880562,
      "num_tokens": 75006187.0,
      "step": 40235
    },
    {
      "entropy": 6.047561883926392,
      "epoch": 3.553514659131049,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037738068430466574,
      "loss": 5.2054,
      "mean_token_accuracy": 0.22519584149122238,
      "num_tokens": 75015908.0,
      "step": 40240
    },
    {
      "entropy": 5.955201768875122,
      "epoch": 3.553956199222889,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003773526407919142,
      "loss": 5.2308,
      "mean_token_accuracy": 0.21534061282873154,
      "num_tokens": 75025343.0,
      "step": 40245
    },
    {
      "entropy": 6.0419450283050535,
      "epoch": 3.5543977393147297,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037732459527413246,
      "loss": 5.2859,
      "mean_token_accuracy": 0.21652354300022125,
      "num_tokens": 75034688.0,
      "step": 40250
    },
    {
      "entropy": 5.992972087860108,
      "epoch": 3.55483927940657,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003772965477518702,
      "loss": 5.1834,
      "mean_token_accuracy": 0.22537823468446733,
      "num_tokens": 75043587.0,
      "step": 40255
    },
    {
      "entropy": 5.976587867736816,
      "epoch": 3.5552808194984102,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037726849822567664,
      "loss": 5.1852,
      "mean_token_accuracy": 0.22631260007619858,
      "num_tokens": 75052900.0,
      "step": 40260
    },
    {
      "entropy": 6.036372375488281,
      "epoch": 3.5557223595902507,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003772404466961015,
      "loss": 5.3458,
      "mean_token_accuracy": 0.20884989798069,
      "num_tokens": 75062902.0,
      "step": 40265
    },
    {
      "entropy": 6.0494836330413815,
      "epoch": 3.5561638996820912,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003772123931636939,
      "loss": 5.2892,
      "mean_token_accuracy": 0.21403950452804565,
      "num_tokens": 75071764.0,
      "step": 40270
    },
    {
      "entropy": 6.012067747116089,
      "epoch": 3.5566054397739313,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037718433762900394,
      "loss": 5.2614,
      "mean_token_accuracy": 0.21595583260059356,
      "num_tokens": 75080447.0,
      "step": 40275
    },
    {
      "entropy": 6.031564235687256,
      "epoch": 3.557046979865772,
      "grad_norm": 1.25,
      "learning_rate": 0.0003771562800925808,
      "loss": 5.2635,
      "mean_token_accuracy": 0.2239273175597191,
      "num_tokens": 75088957.0,
      "step": 40280
    },
    {
      "entropy": 5.957450294494629,
      "epoch": 3.5574885199576123,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003771282205549742,
      "loss": 5.2912,
      "mean_token_accuracy": 0.2158803015947342,
      "num_tokens": 75097999.0,
      "step": 40285
    },
    {
      "entropy": 6.0353141784667965,
      "epoch": 3.5579300600494523,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003771001590167338,
      "loss": 5.2768,
      "mean_token_accuracy": 0.21638451665639877,
      "num_tokens": 75105949.0,
      "step": 40290
    },
    {
      "entropy": 5.92271990776062,
      "epoch": 3.558371600141293,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003770720954784095,
      "loss": 5.2324,
      "mean_token_accuracy": 0.2269000992178917,
      "num_tokens": 75115212.0,
      "step": 40295
    },
    {
      "entropy": 6.054004430770874,
      "epoch": 3.5588131402331333,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003770440299405508,
      "loss": 5.3885,
      "mean_token_accuracy": 0.21041744947433472,
      "num_tokens": 75125133.0,
      "step": 40300
    },
    {
      "entropy": 6.0462483882904055,
      "epoch": 3.5592546803249734,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037701596240370764,
      "loss": 5.2588,
      "mean_token_accuracy": 0.22152296155691148,
      "num_tokens": 75134846.0,
      "step": 40305
    },
    {
      "entropy": 6.080868768692016,
      "epoch": 3.559696220416814,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037698789286842976,
      "loss": 5.3376,
      "mean_token_accuracy": 0.21327103823423385,
      "num_tokens": 75144284.0,
      "step": 40310
    },
    {
      "entropy": 5.992176723480225,
      "epoch": 3.5601377605086544,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037695982133526706,
      "loss": 5.3078,
      "mean_token_accuracy": 0.21489909887313843,
      "num_tokens": 75153485.0,
      "step": 40315
    },
    {
      "entropy": 6.001308107376099,
      "epoch": 3.5605793006004944,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003769317478047695,
      "loss": 5.2896,
      "mean_token_accuracy": 0.21984040588140488,
      "num_tokens": 75162677.0,
      "step": 40320
    },
    {
      "entropy": 6.075159025192261,
      "epoch": 3.561020840692335,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003769036722774869,
      "loss": 5.32,
      "mean_token_accuracy": 0.20686561912298201,
      "num_tokens": 75172335.0,
      "step": 40325
    },
    {
      "entropy": 6.091323757171631,
      "epoch": 3.5614623807841754,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003768755947539694,
      "loss": 5.2946,
      "mean_token_accuracy": 0.20762186050415038,
      "num_tokens": 75180963.0,
      "step": 40330
    },
    {
      "entropy": 6.083085441589356,
      "epoch": 3.5619039208760155,
      "grad_norm": 1.25,
      "learning_rate": 0.00037684751523476683,
      "loss": 5.3457,
      "mean_token_accuracy": 0.2090818077325821,
      "num_tokens": 75190174.0,
      "step": 40335
    },
    {
      "entropy": 5.92338376045227,
      "epoch": 3.562345460967856,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003768194337204295,
      "loss": 5.143,
      "mean_token_accuracy": 0.22949882596731186,
      "num_tokens": 75198829.0,
      "step": 40340
    },
    {
      "entropy": 5.975014162063599,
      "epoch": 3.5627870010596965,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003767913502115073,
      "loss": 5.2443,
      "mean_token_accuracy": 0.21865810304880143,
      "num_tokens": 75208106.0,
      "step": 40345
    },
    {
      "entropy": 6.052993583679199,
      "epoch": 3.5632285411515365,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037676326470855045,
      "loss": 5.2386,
      "mean_token_accuracy": 0.2164209485054016,
      "num_tokens": 75217232.0,
      "step": 40350
    },
    {
      "entropy": 6.019718360900879,
      "epoch": 3.563670081243377,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037673517721210903,
      "loss": 5.237,
      "mean_token_accuracy": 0.21644315421581267,
      "num_tokens": 75225792.0,
      "step": 40355
    },
    {
      "entropy": 5.951737260818481,
      "epoch": 3.564111621335217,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003767070877227335,
      "loss": 5.2083,
      "mean_token_accuracy": 0.22313550412654876,
      "num_tokens": 75234673.0,
      "step": 40360
    },
    {
      "entropy": 6.028862857818604,
      "epoch": 3.5645531614270576,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037667899624097386,
      "loss": 5.3435,
      "mean_token_accuracy": 0.2111271619796753,
      "num_tokens": 75244089.0,
      "step": 40365
    },
    {
      "entropy": 6.000273180007935,
      "epoch": 3.564994701518898,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003766509027673805,
      "loss": 5.207,
      "mean_token_accuracy": 0.2195389062166214,
      "num_tokens": 75253506.0,
      "step": 40370
    },
    {
      "entropy": 6.013524389266967,
      "epoch": 3.565436241610738,
      "grad_norm": 1.34375,
      "learning_rate": 0.00037662280730250383,
      "loss": 5.2667,
      "mean_token_accuracy": 0.21765804886817933,
      "num_tokens": 75263188.0,
      "step": 40375
    },
    {
      "entropy": 6.01783332824707,
      "epoch": 3.5658777817025786,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003765947098468941,
      "loss": 5.3573,
      "mean_token_accuracy": 0.21217057704925538,
      "num_tokens": 75273028.0,
      "step": 40380
    },
    {
      "entropy": 6.089385271072388,
      "epoch": 3.5663193217944187,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003765666104011019,
      "loss": 5.3857,
      "mean_token_accuracy": 0.21246006339788437,
      "num_tokens": 75282913.0,
      "step": 40385
    },
    {
      "entropy": 5.973013114929199,
      "epoch": 3.566760861886259,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00037653850896567743,
      "loss": 5.2036,
      "mean_token_accuracy": 0.2215322345495224,
      "num_tokens": 75291912.0,
      "step": 40390
    },
    {
      "entropy": 6.050421285629272,
      "epoch": 3.5672024019780997,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003765104055411714,
      "loss": 5.3034,
      "mean_token_accuracy": 0.20711027979850768,
      "num_tokens": 75301535.0,
      "step": 40395
    },
    {
      "entropy": 5.975517797470093,
      "epoch": 3.5676439420699397,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003764823001281343,
      "loss": 5.3063,
      "mean_token_accuracy": 0.21254506558179856,
      "num_tokens": 75310990.0,
      "step": 40400
    },
    {
      "entropy": 6.019581031799317,
      "epoch": 3.5680854821617802,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037645419272711657,
      "loss": 5.2505,
      "mean_token_accuracy": 0.21998805701732635,
      "num_tokens": 75320690.0,
      "step": 40405
    },
    {
      "entropy": 6.018678331375122,
      "epoch": 3.5685270222536207,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003764260833386689,
      "loss": 5.2046,
      "mean_token_accuracy": 0.22179261296987535,
      "num_tokens": 75330354.0,
      "step": 40410
    },
    {
      "entropy": 5.99930329322815,
      "epoch": 3.568968562345461,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003763979719633419,
      "loss": 5.2991,
      "mean_token_accuracy": 0.2203971892595291,
      "num_tokens": 75339599.0,
      "step": 40415
    },
    {
      "entropy": 6.047136116027832,
      "epoch": 3.5694101024373013,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037636985860168637,
      "loss": 5.3253,
      "mean_token_accuracy": 0.20301763862371444,
      "num_tokens": 75349667.0,
      "step": 40420
    },
    {
      "entropy": 6.0803382396698,
      "epoch": 3.5698516425291418,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037634174325425294,
      "loss": 5.3492,
      "mean_token_accuracy": 0.20801639407873154,
      "num_tokens": 75358862.0,
      "step": 40425
    },
    {
      "entropy": 5.9774048805236815,
      "epoch": 3.570293182620982,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037631362592159247,
      "loss": 5.1933,
      "mean_token_accuracy": 0.2252921536564827,
      "num_tokens": 75368451.0,
      "step": 40430
    },
    {
      "entropy": 5.989295578002929,
      "epoch": 3.5707347227128223,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003762855066042556,
      "loss": 5.2831,
      "mean_token_accuracy": 0.21193121522665023,
      "num_tokens": 75378734.0,
      "step": 40435
    },
    {
      "entropy": 6.0313643455505375,
      "epoch": 3.571176262804663,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003762573853027933,
      "loss": 5.2506,
      "mean_token_accuracy": 0.2149004876613617,
      "num_tokens": 75387959.0,
      "step": 40440
    },
    {
      "entropy": 6.066915559768677,
      "epoch": 3.571617802896503,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003762292620177564,
      "loss": 5.2741,
      "mean_token_accuracy": 0.21491328328847886,
      "num_tokens": 75396968.0,
      "step": 40445
    },
    {
      "entropy": 6.0523011684417725,
      "epoch": 3.5720593429883434,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003762011367496959,
      "loss": 5.2867,
      "mean_token_accuracy": 0.21570371389389037,
      "num_tokens": 75406378.0,
      "step": 40450
    },
    {
      "entropy": 6.019407749176025,
      "epoch": 3.572500883080184,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003761730094991625,
      "loss": 5.2412,
      "mean_token_accuracy": 0.2216253861784935,
      "num_tokens": 75416048.0,
      "step": 40455
    },
    {
      "entropy": 5.9177039623260494,
      "epoch": 3.572942423172024,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003761448802667076,
      "loss": 5.2109,
      "mean_token_accuracy": 0.2236751690506935,
      "num_tokens": 75425150.0,
      "step": 40460
    },
    {
      "entropy": 5.975475168228149,
      "epoch": 3.5733839632638644,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003761167490528819,
      "loss": 5.1776,
      "mean_token_accuracy": 0.22891451567411422,
      "num_tokens": 75434891.0,
      "step": 40465
    },
    {
      "entropy": 5.966288805007935,
      "epoch": 3.573825503355705,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003760886158582366,
      "loss": 5.1608,
      "mean_token_accuracy": 0.23286796659231185,
      "num_tokens": 75444221.0,
      "step": 40470
    },
    {
      "entropy": 6.027890872955322,
      "epoch": 3.574267043447545,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037606048068332284,
      "loss": 5.3306,
      "mean_token_accuracy": 0.2094999223947525,
      "num_tokens": 75454149.0,
      "step": 40475
    },
    {
      "entropy": 5.995287704467773,
      "epoch": 3.5747085835393855,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003760323435286917,
      "loss": 5.2436,
      "mean_token_accuracy": 0.22003562599420548,
      "num_tokens": 75462426.0,
      "step": 40480
    },
    {
      "entropy": 6.000324678421021,
      "epoch": 3.575150123631226,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003760042043948944,
      "loss": 5.3485,
      "mean_token_accuracy": 0.2148241728544235,
      "num_tokens": 75472513.0,
      "step": 40485
    },
    {
      "entropy": 6.01798939704895,
      "epoch": 3.575591663723066,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003759760632824822,
      "loss": 5.2377,
      "mean_token_accuracy": 0.21942866146564483,
      "num_tokens": 75481995.0,
      "step": 40490
    },
    {
      "entropy": 6.025844430923462,
      "epoch": 3.5760332038149065,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037594792019200637,
      "loss": 5.2644,
      "mean_token_accuracy": 0.21503855139017106,
      "num_tokens": 75491738.0,
      "step": 40495
    },
    {
      "entropy": 6.000866937637329,
      "epoch": 3.5764747439067466,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037591977512401814,
      "loss": 5.3043,
      "mean_token_accuracy": 0.21507494449615477,
      "num_tokens": 75502166.0,
      "step": 40500
    },
    {
      "entropy": 5.906515264511109,
      "epoch": 3.576916283998587,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000375891628079069,
      "loss": 5.0616,
      "mean_token_accuracy": 0.23867745995521544,
      "num_tokens": 75510500.0,
      "step": 40505
    },
    {
      "entropy": 5.999634170532227,
      "epoch": 3.5773578240904276,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003758634790577102,
      "loss": 5.2994,
      "mean_token_accuracy": 0.2213817521929741,
      "num_tokens": 75518801.0,
      "step": 40510
    },
    {
      "entropy": 5.953930044174195,
      "epoch": 3.5777993641822676,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037583532806049325,
      "loss": 5.2558,
      "mean_token_accuracy": 0.2164203032851219,
      "num_tokens": 75528524.0,
      "step": 40515
    },
    {
      "entropy": 6.029538679122925,
      "epoch": 3.578240904274108,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003758071750879696,
      "loss": 5.2582,
      "mean_token_accuracy": 0.2166221633553505,
      "num_tokens": 75537925.0,
      "step": 40520
    },
    {
      "entropy": 5.976492786407471,
      "epoch": 3.578682444365948,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003757790201406906,
      "loss": 5.2725,
      "mean_token_accuracy": 0.22201025038957595,
      "num_tokens": 75548075.0,
      "step": 40525
    },
    {
      "entropy": 6.062675666809082,
      "epoch": 3.5791239844577887,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037575086321920797,
      "loss": 5.269,
      "mean_token_accuracy": 0.21311791539192199,
      "num_tokens": 75556967.0,
      "step": 40530
    },
    {
      "entropy": 6.08052978515625,
      "epoch": 3.579565524549629,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003757227043240732,
      "loss": 5.3801,
      "mean_token_accuracy": 0.2062801644206047,
      "num_tokens": 75566584.0,
      "step": 40535
    },
    {
      "entropy": 6.0133754253387455,
      "epoch": 3.5800070646414692,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000375694543455838,
      "loss": 5.2923,
      "mean_token_accuracy": 0.20777398049831391,
      "num_tokens": 75576387.0,
      "step": 40540
    },
    {
      "entropy": 6.0413902759552,
      "epoch": 3.5804486047333097,
      "grad_norm": 1.53125,
      "learning_rate": 0.0003756663806150539,
      "loss": 5.3073,
      "mean_token_accuracy": 0.21391219645738602,
      "num_tokens": 75585390.0,
      "step": 40545
    },
    {
      "entropy": 5.929007911682129,
      "epoch": 3.5808901448251502,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037563821580227267,
      "loss": 5.1843,
      "mean_token_accuracy": 0.2310074269771576,
      "num_tokens": 75594178.0,
      "step": 40550
    },
    {
      "entropy": 6.014240980148315,
      "epoch": 3.5813316849169903,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000375610049018046,
      "loss": 5.2667,
      "mean_token_accuracy": 0.2189297154545784,
      "num_tokens": 75602712.0,
      "step": 40555
    },
    {
      "entropy": 6.098490619659424,
      "epoch": 3.5817732250088308,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003755818802629258,
      "loss": 5.3882,
      "mean_token_accuracy": 0.20451392531394957,
      "num_tokens": 75612727.0,
      "step": 40560
    },
    {
      "entropy": 6.05017466545105,
      "epoch": 3.5822147651006713,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037555370953746366,
      "loss": 5.3036,
      "mean_token_accuracy": 0.21641854345798492,
      "num_tokens": 75622443.0,
      "step": 40565
    },
    {
      "entropy": 6.056688547134399,
      "epoch": 3.5826563051925113,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037552553684221156,
      "loss": 5.3746,
      "mean_token_accuracy": 0.20595247000455857,
      "num_tokens": 75632966.0,
      "step": 40570
    },
    {
      "entropy": 6.092141962051391,
      "epoch": 3.583097845284352,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003754973621777213,
      "loss": 5.2778,
      "mean_token_accuracy": 0.215044042468071,
      "num_tokens": 75642987.0,
      "step": 40575
    },
    {
      "entropy": 6.008177804946899,
      "epoch": 3.5835393853761923,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003754691855445451,
      "loss": 5.1522,
      "mean_token_accuracy": 0.2280656799674034,
      "num_tokens": 75652407.0,
      "step": 40580
    },
    {
      "entropy": 6.03609676361084,
      "epoch": 3.5839809254680324,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003754410069432345,
      "loss": 5.3245,
      "mean_token_accuracy": 0.20957414358854293,
      "num_tokens": 75662258.0,
      "step": 40585
    },
    {
      "entropy": 5.895302724838257,
      "epoch": 3.584422465559873,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003754128263743418,
      "loss": 5.1645,
      "mean_token_accuracy": 0.222473905980587,
      "num_tokens": 75670836.0,
      "step": 40590
    },
    {
      "entropy": 5.963233089447021,
      "epoch": 3.5848640056517134,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037538464383841887,
      "loss": 5.1796,
      "mean_token_accuracy": 0.21839457303285598,
      "num_tokens": 75679733.0,
      "step": 40595
    },
    {
      "entropy": 6.019998550415039,
      "epoch": 3.5853055457435534,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003753564593360179,
      "loss": 5.2752,
      "mean_token_accuracy": 0.21135551780462264,
      "num_tokens": 75689549.0,
      "step": 40600
    },
    {
      "entropy": 6.053961753845215,
      "epoch": 3.585747085835394,
      "grad_norm": 1.3125,
      "learning_rate": 0.00037532827286769087,
      "loss": 5.2746,
      "mean_token_accuracy": 0.21343497633934022,
      "num_tokens": 75698321.0,
      "step": 40605
    },
    {
      "entropy": 6.0170714378356935,
      "epoch": 3.5861886259272344,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003753000844339901,
      "loss": 5.2964,
      "mean_token_accuracy": 0.21558504849672316,
      "num_tokens": 75707986.0,
      "step": 40610
    },
    {
      "entropy": 6.062403583526612,
      "epoch": 3.5866301660190745,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003752718940354677,
      "loss": 5.2702,
      "mean_token_accuracy": 0.2146832287311554,
      "num_tokens": 75717200.0,
      "step": 40615
    },
    {
      "entropy": 6.040963888168335,
      "epoch": 3.587071706110915,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037524370167267595,
      "loss": 5.351,
      "mean_token_accuracy": 0.20831598490476608,
      "num_tokens": 75727366.0,
      "step": 40620
    },
    {
      "entropy": 5.941378450393676,
      "epoch": 3.5875132462027555,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037521550734616717,
      "loss": 5.2035,
      "mean_token_accuracy": 0.22214534133672714,
      "num_tokens": 75736758.0,
      "step": 40625
    },
    {
      "entropy": 5.998485040664673,
      "epoch": 3.5879547862945955,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037518731105649347,
      "loss": 5.2597,
      "mean_token_accuracy": 0.2139914184808731,
      "num_tokens": 75745896.0,
      "step": 40630
    },
    {
      "entropy": 6.036436605453491,
      "epoch": 3.588396326386436,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003751591128042074,
      "loss": 5.2899,
      "mean_token_accuracy": 0.2199902668595314,
      "num_tokens": 75755460.0,
      "step": 40635
    },
    {
      "entropy": 6.082846212387085,
      "epoch": 3.588837866478276,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037513091258986126,
      "loss": 5.3142,
      "mean_token_accuracy": 0.21672751605510712,
      "num_tokens": 75764210.0,
      "step": 40640
    },
    {
      "entropy": 6.033925867080688,
      "epoch": 3.5892794065701166,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037510271041400745,
      "loss": 5.3365,
      "mean_token_accuracy": 0.21143637895584105,
      "num_tokens": 75773292.0,
      "step": 40645
    },
    {
      "entropy": 6.039337968826294,
      "epoch": 3.589720946661957,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037507450627719853,
      "loss": 5.2466,
      "mean_token_accuracy": 0.21495795696973802,
      "num_tokens": 75782607.0,
      "step": 40650
    },
    {
      "entropy": 6.003206539154053,
      "epoch": 3.590162486753797,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003750463001799869,
      "loss": 5.2329,
      "mean_token_accuracy": 0.21664899289608003,
      "num_tokens": 75791468.0,
      "step": 40655
    },
    {
      "entropy": 5.982588863372802,
      "epoch": 3.5906040268456376,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003750180921229252,
      "loss": 5.2243,
      "mean_token_accuracy": 0.2159337505698204,
      "num_tokens": 75800147.0,
      "step": 40660
    },
    {
      "entropy": 6.024096965789795,
      "epoch": 3.5910455669374777,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003749898821065659,
      "loss": 5.2397,
      "mean_token_accuracy": 0.21468513756990432,
      "num_tokens": 75809086.0,
      "step": 40665
    },
    {
      "entropy": 6.061948156356811,
      "epoch": 3.591487107029318,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037496167013146176,
      "loss": 5.2988,
      "mean_token_accuracy": 0.21000228226184844,
      "num_tokens": 75819146.0,
      "step": 40670
    },
    {
      "entropy": 5.9263832569122314,
      "epoch": 3.5919286471211587,
      "grad_norm": 1.3125,
      "learning_rate": 0.00037493345619816527,
      "loss": 5.1907,
      "mean_token_accuracy": 0.22666240781545638,
      "num_tokens": 75828590.0,
      "step": 40675
    },
    {
      "entropy": 5.973341178894043,
      "epoch": 3.5923701872129987,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003749052403072292,
      "loss": 5.187,
      "mean_token_accuracy": 0.2187003418803215,
      "num_tokens": 75837590.0,
      "step": 40680
    },
    {
      "entropy": 6.1032140254974365,
      "epoch": 3.592811727304839,
      "grad_norm": 1.375,
      "learning_rate": 0.00037487702245920634,
      "loss": 5.4042,
      "mean_token_accuracy": 0.20741328746080398,
      "num_tokens": 75846985.0,
      "step": 40685
    },
    {
      "entropy": 6.031117296218872,
      "epoch": 3.5932532673966797,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003748488026546494,
      "loss": 5.229,
      "mean_token_accuracy": 0.22071245014667512,
      "num_tokens": 75855411.0,
      "step": 40690
    },
    {
      "entropy": 6.0341535091400145,
      "epoch": 3.5936948074885198,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003748205808941112,
      "loss": 5.2559,
      "mean_token_accuracy": 0.2141103610396385,
      "num_tokens": 75865333.0,
      "step": 40695
    },
    {
      "entropy": 5.992195892333984,
      "epoch": 3.5941363475803603,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037479235717814463,
      "loss": 5.3202,
      "mean_token_accuracy": 0.21435325294733049,
      "num_tokens": 75875839.0,
      "step": 40700
    },
    {
      "entropy": 6.078212642669678,
      "epoch": 3.5945778876722008,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037476413150730253,
      "loss": 5.3632,
      "mean_token_accuracy": 0.21247205138206482,
      "num_tokens": 75885039.0,
      "step": 40705
    },
    {
      "entropy": 6.031851673126221,
      "epoch": 3.595019427764041,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037473590388213777,
      "loss": 5.2577,
      "mean_token_accuracy": 0.21714731603860854,
      "num_tokens": 75895326.0,
      "step": 40710
    },
    {
      "entropy": 5.99488468170166,
      "epoch": 3.5954609678558813,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00037470767430320345,
      "loss": 5.1896,
      "mean_token_accuracy": 0.2228793680667877,
      "num_tokens": 75903922.0,
      "step": 40715
    },
    {
      "entropy": 5.967334747314453,
      "epoch": 3.595902507947722,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037467944277105246,
      "loss": 5.2911,
      "mean_token_accuracy": 0.21776808500289918,
      "num_tokens": 75912949.0,
      "step": 40720
    },
    {
      "entropy": 5.998760986328125,
      "epoch": 3.596344048039562,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003746512092862379,
      "loss": 5.2772,
      "mean_token_accuracy": 0.21506306976079942,
      "num_tokens": 75922406.0,
      "step": 40725
    },
    {
      "entropy": 6.02526364326477,
      "epoch": 3.5967855881314024,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003746229738493128,
      "loss": 5.223,
      "mean_token_accuracy": 0.21496339440345763,
      "num_tokens": 75931608.0,
      "step": 40730
    },
    {
      "entropy": 5.98866605758667,
      "epoch": 3.597227128223243,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003745947364608303,
      "loss": 5.2718,
      "mean_token_accuracy": 0.21740919649600982,
      "num_tokens": 75940557.0,
      "step": 40735
    },
    {
      "entropy": 5.997849035263061,
      "epoch": 3.597668668315083,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037456649712134355,
      "loss": 5.3497,
      "mean_token_accuracy": 0.21214823573827743,
      "num_tokens": 75948847.0,
      "step": 40740
    },
    {
      "entropy": 6.002200746536255,
      "epoch": 3.5981102084069234,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003745382558314057,
      "loss": 5.2727,
      "mean_token_accuracy": 0.22495897263288497,
      "num_tokens": 75958014.0,
      "step": 40745
    },
    {
      "entropy": 5.99972972869873,
      "epoch": 3.598551748498764,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003745100125915701,
      "loss": 5.2367,
      "mean_token_accuracy": 0.2162540778517723,
      "num_tokens": 75967430.0,
      "step": 40750
    },
    {
      "entropy": 6.061254024505615,
      "epoch": 3.598993288590604,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003744817674023899,
      "loss": 5.2788,
      "mean_token_accuracy": 0.22028485536575318,
      "num_tokens": 75976731.0,
      "step": 40755
    },
    {
      "entropy": 5.9521832942962645,
      "epoch": 3.5994348286824445,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037445352026441853,
      "loss": 5.2907,
      "mean_token_accuracy": 0.2094278484582901,
      "num_tokens": 75986562.0,
      "step": 40760
    },
    {
      "entropy": 6.016607713699341,
      "epoch": 3.599876368774285,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037442527117820924,
      "loss": 5.3154,
      "mean_token_accuracy": 0.21187189519405364,
      "num_tokens": 75995415.0,
      "step": 40765
    },
    {
      "entropy": 6.04542818069458,
      "epoch": 3.600317908866125,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003743970201443154,
      "loss": 5.2933,
      "mean_token_accuracy": 0.21184273660182953,
      "num_tokens": 76004463.0,
      "step": 40770
    },
    {
      "entropy": 6.068217325210571,
      "epoch": 3.6007594489579655,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003743687671632905,
      "loss": 5.3023,
      "mean_token_accuracy": 0.2136110097169876,
      "num_tokens": 76013854.0,
      "step": 40775
    },
    {
      "entropy": 5.992674207687378,
      "epoch": 3.6012009890498056,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003743405122356879,
      "loss": 5.1852,
      "mean_token_accuracy": 0.22496361285448074,
      "num_tokens": 76023665.0,
      "step": 40780
    },
    {
      "entropy": 6.069551753997803,
      "epoch": 3.601642529141646,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00037431225536206127,
      "loss": 5.2086,
      "mean_token_accuracy": 0.2239723399281502,
      "num_tokens": 76031788.0,
      "step": 40785
    },
    {
      "entropy": 6.016032457351685,
      "epoch": 3.6020840692334866,
      "grad_norm": 1.171875,
      "learning_rate": 0.000374283996542964,
      "loss": 5.2014,
      "mean_token_accuracy": 0.22616005688905716,
      "num_tokens": 76040893.0,
      "step": 40790
    },
    {
      "entropy": 5.966346740722656,
      "epoch": 3.6025256093253266,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037425573577894964,
      "loss": 5.2215,
      "mean_token_accuracy": 0.21311252415180207,
      "num_tokens": 76050171.0,
      "step": 40795
    },
    {
      "entropy": 5.993173217773437,
      "epoch": 3.602967149417167,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037422747307057184,
      "loss": 5.2957,
      "mean_token_accuracy": 0.21058014482259751,
      "num_tokens": 76059071.0,
      "step": 40800
    },
    {
      "entropy": 5.988998174667358,
      "epoch": 3.603408689509007,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003741992084183844,
      "loss": 5.1889,
      "mean_token_accuracy": 0.2282044470310211,
      "num_tokens": 76067069.0,
      "step": 40805
    },
    {
      "entropy": 6.0758435249328615,
      "epoch": 3.6038502296008477,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003741709418229408,
      "loss": 5.2982,
      "mean_token_accuracy": 0.21315427422523497,
      "num_tokens": 76075699.0,
      "step": 40810
    },
    {
      "entropy": 6.041539049148559,
      "epoch": 3.604291769692688,
      "grad_norm": 1.4375,
      "learning_rate": 0.00037414267328479483,
      "loss": 5.2121,
      "mean_token_accuracy": 0.21845539510250092,
      "num_tokens": 76083837.0,
      "step": 40815
    },
    {
      "entropy": 5.958420133590698,
      "epoch": 3.604733309784528,
      "grad_norm": 1.25,
      "learning_rate": 0.00037411440280450036,
      "loss": 5.2705,
      "mean_token_accuracy": 0.21786748468875886,
      "num_tokens": 76093904.0,
      "step": 40820
    },
    {
      "entropy": 6.029642581939697,
      "epoch": 3.6051748498763687,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037408613038261116,
      "loss": 5.3813,
      "mean_token_accuracy": 0.21559922844171525,
      "num_tokens": 76104297.0,
      "step": 40825
    },
    {
      "entropy": 6.037792921066284,
      "epoch": 3.605616389968209,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003740578560196808,
      "loss": 5.2537,
      "mean_token_accuracy": 0.21754782795906066,
      "num_tokens": 76113767.0,
      "step": 40830
    },
    {
      "entropy": 6.013040208816529,
      "epoch": 3.6060579300600493,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037402957971626355,
      "loss": 5.3332,
      "mean_token_accuracy": 0.21154519021511078,
      "num_tokens": 76123133.0,
      "step": 40835
    },
    {
      "entropy": 5.9651405811309814,
      "epoch": 3.6064994701518898,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003740013014729131,
      "loss": 5.2424,
      "mean_token_accuracy": 0.22367806285619735,
      "num_tokens": 76132522.0,
      "step": 40840
    },
    {
      "entropy": 6.0399253368377686,
      "epoch": 3.6069410102437303,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037397302129018353,
      "loss": 5.2555,
      "mean_token_accuracy": 0.22173252254724501,
      "num_tokens": 76141421.0,
      "step": 40845
    },
    {
      "entropy": 6.014951944351196,
      "epoch": 3.6073825503355703,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003739447391686287,
      "loss": 5.2963,
      "mean_token_accuracy": 0.2138849675655365,
      "num_tokens": 76151134.0,
      "step": 40850
    },
    {
      "entropy": 5.9910603046417235,
      "epoch": 3.607824090427411,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037391645510880266,
      "loss": 5.2609,
      "mean_token_accuracy": 0.21881998032331468,
      "num_tokens": 76160443.0,
      "step": 40855
    },
    {
      "entropy": 6.006173992156983,
      "epoch": 3.6082656305192513,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003738881691112596,
      "loss": 5.225,
      "mean_token_accuracy": 0.21470891684293747,
      "num_tokens": 76169619.0,
      "step": 40860
    },
    {
      "entropy": 6.0078483581542965,
      "epoch": 3.6087071706110914,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037385988117655353,
      "loss": 5.2793,
      "mean_token_accuracy": 0.21445411443710327,
      "num_tokens": 76179227.0,
      "step": 40865
    },
    {
      "entropy": 5.878686714172363,
      "epoch": 3.609148710702932,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003738315913052386,
      "loss": 5.1671,
      "mean_token_accuracy": 0.22599320709705353,
      "num_tokens": 76187891.0,
      "step": 40870
    },
    {
      "entropy": 6.062799501419067,
      "epoch": 3.6095902507947724,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037380329949786897,
      "loss": 5.3719,
      "mean_token_accuracy": 0.2080405741930008,
      "num_tokens": 76197363.0,
      "step": 40875
    },
    {
      "entropy": 6.120318126678467,
      "epoch": 3.6100317908866124,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003737750057549989,
      "loss": 5.3181,
      "mean_token_accuracy": 0.21464070081710815,
      "num_tokens": 76206466.0,
      "step": 40880
    },
    {
      "entropy": 6.02061710357666,
      "epoch": 3.610473330978453,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003737467100771827,
      "loss": 5.2333,
      "mean_token_accuracy": 0.21542182117700576,
      "num_tokens": 76215857.0,
      "step": 40885
    },
    {
      "entropy": 5.997574377059936,
      "epoch": 3.6109148710702934,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003737184124649746,
      "loss": 5.2688,
      "mean_token_accuracy": 0.21611892879009248,
      "num_tokens": 76224929.0,
      "step": 40890
    },
    {
      "entropy": 5.996699380874634,
      "epoch": 3.6113564111621335,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003736901129189289,
      "loss": 5.2257,
      "mean_token_accuracy": 0.21897546499967574,
      "num_tokens": 76235109.0,
      "step": 40895
    },
    {
      "entropy": 5.951986646652221,
      "epoch": 3.611797951253974,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003736618114396001,
      "loss": 5.2307,
      "mean_token_accuracy": 0.21474871933460235,
      "num_tokens": 76244263.0,
      "step": 40900
    },
    {
      "entropy": 6.02395658493042,
      "epoch": 3.6122394913458145,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003736335080275424,
      "loss": 5.2951,
      "mean_token_accuracy": 0.21573172956705094,
      "num_tokens": 76253895.0,
      "step": 40905
    },
    {
      "entropy": 6.0476690292358395,
      "epoch": 3.6126810314376545,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003736052026833105,
      "loss": 5.2916,
      "mean_token_accuracy": 0.22373314648866655,
      "num_tokens": 76264479.0,
      "step": 40910
    },
    {
      "entropy": 5.883020496368408,
      "epoch": 3.613122571529495,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003735768954074587,
      "loss": 5.1292,
      "mean_token_accuracy": 0.23295954912900924,
      "num_tokens": 76274202.0,
      "step": 40915
    },
    {
      "entropy": 5.9578932762146,
      "epoch": 3.613564111621335,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003735485862005416,
      "loss": 5.1906,
      "mean_token_accuracy": 0.21897909939289092,
      "num_tokens": 76282518.0,
      "step": 40920
    },
    {
      "entropy": 5.992560768127442,
      "epoch": 3.6140056517131756,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003735202750631137,
      "loss": 5.199,
      "mean_token_accuracy": 0.2250184565782547,
      "num_tokens": 76291763.0,
      "step": 40925
    },
    {
      "entropy": 5.976190757751465,
      "epoch": 3.614447191805016,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037349196199572965,
      "loss": 5.2603,
      "mean_token_accuracy": 0.21830208599567413,
      "num_tokens": 76301991.0,
      "step": 40930
    },
    {
      "entropy": 5.995820713043213,
      "epoch": 3.614888731896856,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003734636469989441,
      "loss": 5.2978,
      "mean_token_accuracy": 0.21554333418607713,
      "num_tokens": 76311489.0,
      "step": 40935
    },
    {
      "entropy": 6.018421316146851,
      "epoch": 3.6153302719886966,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00037343533007331165,
      "loss": 5.2529,
      "mean_token_accuracy": 0.22215048372745513,
      "num_tokens": 76319409.0,
      "step": 40940
    },
    {
      "entropy": 5.9397745609283445,
      "epoch": 3.6157718120805367,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003734070112193872,
      "loss": 5.1793,
      "mean_token_accuracy": 0.22527479827404023,
      "num_tokens": 76329036.0,
      "step": 40945
    },
    {
      "entropy": 5.951801013946533,
      "epoch": 3.616213352172377,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003733786904377253,
      "loss": 5.189,
      "mean_token_accuracy": 0.22979983687400818,
      "num_tokens": 76338100.0,
      "step": 40950
    },
    {
      "entropy": 6.012299251556397,
      "epoch": 3.6166548922642177,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037335036772888076,
      "loss": 5.284,
      "mean_token_accuracy": 0.2129027232527733,
      "num_tokens": 76347657.0,
      "step": 40955
    },
    {
      "entropy": 6.017507886886596,
      "epoch": 3.6170964323560577,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003733220430934085,
      "loss": 5.27,
      "mean_token_accuracy": 0.21863668411970139,
      "num_tokens": 76357618.0,
      "step": 40960
    },
    {
      "entropy": 6.019603300094604,
      "epoch": 3.617537972447898,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037329371653186327,
      "loss": 5.2713,
      "mean_token_accuracy": 0.21097409278154372,
      "num_tokens": 76366702.0,
      "step": 40965
    },
    {
      "entropy": 5.974032258987426,
      "epoch": 3.6179795125397387,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003732653880448001,
      "loss": 5.1863,
      "mean_token_accuracy": 0.22855738699436187,
      "num_tokens": 76376742.0,
      "step": 40970
    },
    {
      "entropy": 6.041051912307739,
      "epoch": 3.6184210526315788,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003732370576327739,
      "loss": 5.3047,
      "mean_token_accuracy": 0.21270906180143356,
      "num_tokens": 76386379.0,
      "step": 40975
    },
    {
      "entropy": 6.037560701370239,
      "epoch": 3.6188625927234193,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003732087252963395,
      "loss": 5.3166,
      "mean_token_accuracy": 0.21356794089078904,
      "num_tokens": 76396113.0,
      "step": 40980
    },
    {
      "entropy": 5.995489311218262,
      "epoch": 3.6193041328152598,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037318039103605215,
      "loss": 5.2943,
      "mean_token_accuracy": 0.22467949539422988,
      "num_tokens": 76405408.0,
      "step": 40985
    },
    {
      "entropy": 6.011592197418213,
      "epoch": 3.6197456729071,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003731520548524667,
      "loss": 5.3119,
      "mean_token_accuracy": 0.20923762619495392,
      "num_tokens": 76415405.0,
      "step": 40990
    },
    {
      "entropy": 6.010025882720948,
      "epoch": 3.6201872129989403,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037312371674613824,
      "loss": 5.2636,
      "mean_token_accuracy": 0.21403160691261292,
      "num_tokens": 76425161.0,
      "step": 40995
    },
    {
      "entropy": 6.0961872100830075,
      "epoch": 3.620628753090781,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037309537671762215,
      "loss": 5.3672,
      "mean_token_accuracy": 0.20743662863969803,
      "num_tokens": 76435492.0,
      "step": 41000
    },
    {
      "entropy": 6.067695140838623,
      "epoch": 3.621070293182621,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037306703476747334,
      "loss": 5.2936,
      "mean_token_accuracy": 0.21799415946006775,
      "num_tokens": 76444236.0,
      "step": 41005
    },
    {
      "entropy": 5.990127468109131,
      "epoch": 3.6215118332744614,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003730386908962471,
      "loss": 5.2278,
      "mean_token_accuracy": 0.22212383449077605,
      "num_tokens": 76454144.0,
      "step": 41010
    },
    {
      "entropy": 5.91584038734436,
      "epoch": 3.621953373366302,
      "grad_norm": 1.25,
      "learning_rate": 0.0003730103451044986,
      "loss": 5.2142,
      "mean_token_accuracy": 0.2139120116829872,
      "num_tokens": 76463912.0,
      "step": 41015
    },
    {
      "entropy": 5.9681861877441404,
      "epoch": 3.622394913458142,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037298199739278326,
      "loss": 5.2051,
      "mean_token_accuracy": 0.22330947518348693,
      "num_tokens": 76473794.0,
      "step": 41020
    },
    {
      "entropy": 5.979453372955322,
      "epoch": 3.6228364535499824,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003729536477616563,
      "loss": 5.2868,
      "mean_token_accuracy": 0.21616660952568054,
      "num_tokens": 76483770.0,
      "step": 41025
    },
    {
      "entropy": 6.068654918670655,
      "epoch": 3.623277993641823,
      "grad_norm": 1.25,
      "learning_rate": 0.00037292529621167316,
      "loss": 5.365,
      "mean_token_accuracy": 0.20930698215961457,
      "num_tokens": 76493592.0,
      "step": 41030
    },
    {
      "entropy": 5.858396387100219,
      "epoch": 3.623719533733663,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037289694274338905,
      "loss": 5.0766,
      "mean_token_accuracy": 0.23911677151918412,
      "num_tokens": 76501574.0,
      "step": 41035
    },
    {
      "entropy": 5.971851444244384,
      "epoch": 3.6241610738255035,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003728685873573595,
      "loss": 5.2903,
      "mean_token_accuracy": 0.21378067433834075,
      "num_tokens": 76512174.0,
      "step": 41040
    },
    {
      "entropy": 6.037207365036011,
      "epoch": 3.624602613917344,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037284023005414,
      "loss": 5.2873,
      "mean_token_accuracy": 0.2137493759393692,
      "num_tokens": 76522000.0,
      "step": 41045
    },
    {
      "entropy": 5.972291803359985,
      "epoch": 3.625044154009184,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003728118708342861,
      "loss": 5.2708,
      "mean_token_accuracy": 0.2121549054980278,
      "num_tokens": 76531475.0,
      "step": 41050
    },
    {
      "entropy": 6.019604444503784,
      "epoch": 3.6254856941010245,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037278350969835325,
      "loss": 5.3137,
      "mean_token_accuracy": 0.20964637845754625,
      "num_tokens": 76541045.0,
      "step": 41055
    },
    {
      "entropy": 6.033460330963135,
      "epoch": 3.6259272341928646,
      "grad_norm": 1.3125,
      "learning_rate": 0.00037275514664689705,
      "loss": 5.2979,
      "mean_token_accuracy": 0.21739337146282195,
      "num_tokens": 76550357.0,
      "step": 41060
    },
    {
      "entropy": 6.069143056869507,
      "epoch": 3.626368774284705,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003727267816804731,
      "loss": 5.3174,
      "mean_token_accuracy": 0.2125117748975754,
      "num_tokens": 76559094.0,
      "step": 41065
    },
    {
      "entropy": 5.992920064926148,
      "epoch": 3.6268103143765456,
      "grad_norm": 1.125,
      "learning_rate": 0.0003726984147996371,
      "loss": 5.2835,
      "mean_token_accuracy": 0.2240440860390663,
      "num_tokens": 76567639.0,
      "step": 41070
    },
    {
      "entropy": 5.9914695739746096,
      "epoch": 3.6272518544683856,
      "grad_norm": 1.84375,
      "learning_rate": 0.0003726700460049448,
      "loss": 5.2614,
      "mean_token_accuracy": 0.22224708944559096,
      "num_tokens": 76577504.0,
      "step": 41075
    },
    {
      "entropy": 6.000213146209717,
      "epoch": 3.627693394560226,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037264167529695166,
      "loss": 5.2994,
      "mean_token_accuracy": 0.21444082260131836,
      "num_tokens": 76587116.0,
      "step": 41080
    },
    {
      "entropy": 5.986406230926514,
      "epoch": 3.628134934652066,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003726133026762138,
      "loss": 5.225,
      "mean_token_accuracy": 0.21861860454082488,
      "num_tokens": 76596332.0,
      "step": 41085
    },
    {
      "entropy": 5.989885091781616,
      "epoch": 3.6285764747439067,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037258492814328675,
      "loss": 5.2305,
      "mean_token_accuracy": 0.22599710077047347,
      "num_tokens": 76604414.0,
      "step": 41090
    },
    {
      "entropy": 6.000717639923096,
      "epoch": 3.629018014835747,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037255655169872654,
      "loss": 5.2031,
      "mean_token_accuracy": 0.2273504465818405,
      "num_tokens": 76613869.0,
      "step": 41095
    },
    {
      "entropy": 5.978997850418091,
      "epoch": 3.629459554927587,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037252817334308886,
      "loss": 5.1561,
      "mean_token_accuracy": 0.23183304518461229,
      "num_tokens": 76623016.0,
      "step": 41100
    },
    {
      "entropy": 5.9422486305236815,
      "epoch": 3.6299010950194277,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037249979307692985,
      "loss": 5.277,
      "mean_token_accuracy": 0.21565288603305816,
      "num_tokens": 76633108.0,
      "step": 41105
    },
    {
      "entropy": 6.061451053619384,
      "epoch": 3.630342635111268,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003724714109008053,
      "loss": 5.3797,
      "mean_token_accuracy": 0.2089252069592476,
      "num_tokens": 76642844.0,
      "step": 41110
    },
    {
      "entropy": 6.044785976409912,
      "epoch": 3.6307841752031083,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003724430268152712,
      "loss": 5.237,
      "mean_token_accuracy": 0.22241113930940629,
      "num_tokens": 76651836.0,
      "step": 41115
    },
    {
      "entropy": 6.000078296661377,
      "epoch": 3.6312257152949488,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037241464082088366,
      "loss": 5.1799,
      "mean_token_accuracy": 0.22404148280620576,
      "num_tokens": 76661045.0,
      "step": 41120
    },
    {
      "entropy": 6.053722667694092,
      "epoch": 3.6316672553867893,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003723862529181987,
      "loss": 5.3945,
      "mean_token_accuracy": 0.19960489571094514,
      "num_tokens": 76671161.0,
      "step": 41125
    },
    {
      "entropy": 6.050180196762085,
      "epoch": 3.6321087954786293,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037235786310777245,
      "loss": 5.3153,
      "mean_token_accuracy": 0.20597084760665893,
      "num_tokens": 76681668.0,
      "step": 41130
    },
    {
      "entropy": 6.000690650939942,
      "epoch": 3.63255033557047,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000372329471390161,
      "loss": 5.1852,
      "mean_token_accuracy": 0.22031749039888382,
      "num_tokens": 76691466.0,
      "step": 41135
    },
    {
      "entropy": 6.054096317291259,
      "epoch": 3.6329918756623103,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003723010777659206,
      "loss": 5.3163,
      "mean_token_accuracy": 0.212959186732769,
      "num_tokens": 76701921.0,
      "step": 41140
    },
    {
      "entropy": 5.905277109146118,
      "epoch": 3.6334334157541504,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003722726822356074,
      "loss": 5.1478,
      "mean_token_accuracy": 0.2296484962105751,
      "num_tokens": 76710384.0,
      "step": 41145
    },
    {
      "entropy": 6.016720151901245,
      "epoch": 3.633874955845991,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037224428479977766,
      "loss": 5.2714,
      "mean_token_accuracy": 0.21682512909173965,
      "num_tokens": 76719431.0,
      "step": 41150
    },
    {
      "entropy": 6.116383647918701,
      "epoch": 3.6343164959378313,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003722158854589877,
      "loss": 5.3874,
      "mean_token_accuracy": 0.20485586374998094,
      "num_tokens": 76728781.0,
      "step": 41155
    },
    {
      "entropy": 6.030366992950439,
      "epoch": 3.6347580360296714,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003721874842137939,
      "loss": 5.2941,
      "mean_token_accuracy": 0.21129718720912932,
      "num_tokens": 76737268.0,
      "step": 41160
    },
    {
      "entropy": 6.025602197647094,
      "epoch": 3.635199576121512,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003721590810647525,
      "loss": 5.2858,
      "mean_token_accuracy": 0.21471169143915175,
      "num_tokens": 76746439.0,
      "step": 41165
    },
    {
      "entropy": 5.966202449798584,
      "epoch": 3.6356411162133524,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037213067601241995,
      "loss": 5.1793,
      "mean_token_accuracy": 0.22706927210092545,
      "num_tokens": 76755839.0,
      "step": 41170
    },
    {
      "entropy": 5.907512760162353,
      "epoch": 3.6360826563051925,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037210226905735265,
      "loss": 5.135,
      "mean_token_accuracy": 0.22392502874135972,
      "num_tokens": 76765188.0,
      "step": 41175
    },
    {
      "entropy": 6.046259784698487,
      "epoch": 3.636524196397033,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037207386020010727,
      "loss": 5.3479,
      "mean_token_accuracy": 0.21789524257183074,
      "num_tokens": 76774808.0,
      "step": 41180
    },
    {
      "entropy": 6.125094604492188,
      "epoch": 3.6369657364888734,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037204544944124,
      "loss": 5.4287,
      "mean_token_accuracy": 0.20086912512779237,
      "num_tokens": 76783679.0,
      "step": 41185
    },
    {
      "entropy": 5.98892183303833,
      "epoch": 3.6374072765807135,
      "grad_norm": 1.375,
      "learning_rate": 0.0003720170367813077,
      "loss": 5.2619,
      "mean_token_accuracy": 0.22170437574386598,
      "num_tokens": 76793108.0,
      "step": 41190
    },
    {
      "entropy": 5.924862909317016,
      "epoch": 3.637848816672554,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003719886222208668,
      "loss": 5.2664,
      "mean_token_accuracy": 0.21619196832180024,
      "num_tokens": 76802410.0,
      "step": 41195
    },
    {
      "entropy": 5.989882946014404,
      "epoch": 3.638290356764394,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003719602057604739,
      "loss": 5.3063,
      "mean_token_accuracy": 0.22056856602430344,
      "num_tokens": 76811060.0,
      "step": 41200
    },
    {
      "entropy": 6.043942308425903,
      "epoch": 3.6387318968562345,
      "grad_norm": 1.359375,
      "learning_rate": 0.00037193178740068576,
      "loss": 5.3005,
      "mean_token_accuracy": 0.21807991713285446,
      "num_tokens": 76819884.0,
      "step": 41205
    },
    {
      "entropy": 6.031683254241943,
      "epoch": 3.639173436948075,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037190336714205903,
      "loss": 5.3135,
      "mean_token_accuracy": 0.21271411180496216,
      "num_tokens": 76828950.0,
      "step": 41210
    },
    {
      "entropy": 5.999802780151367,
      "epoch": 3.639614977039915,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003718749449851505,
      "loss": 5.3323,
      "mean_token_accuracy": 0.21222185492515563,
      "num_tokens": 76837770.0,
      "step": 41215
    },
    {
      "entropy": 6.07134256362915,
      "epoch": 3.6400565171317556,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037184652093051677,
      "loss": 5.344,
      "mean_token_accuracy": 0.21237178295850753,
      "num_tokens": 76847405.0,
      "step": 41220
    },
    {
      "entropy": 5.950365924835205,
      "epoch": 3.6404980572235957,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003718180949787149,
      "loss": 5.234,
      "mean_token_accuracy": 0.21901942938566207,
      "num_tokens": 76857271.0,
      "step": 41225
    },
    {
      "entropy": 6.096192264556885,
      "epoch": 3.640939597315436,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003717896671303015,
      "loss": 5.3872,
      "mean_token_accuracy": 0.2055180549621582,
      "num_tokens": 76867615.0,
      "step": 41230
    },
    {
      "entropy": 6.037059354782104,
      "epoch": 3.6413811374072766,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003717612373858337,
      "loss": 5.2387,
      "mean_token_accuracy": 0.22329726815223694,
      "num_tokens": 76875903.0,
      "step": 41235
    },
    {
      "entropy": 6.03580493927002,
      "epoch": 3.6418226774991167,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037173280574586814,
      "loss": 5.3999,
      "mean_token_accuracy": 0.20688697397708894,
      "num_tokens": 76887665.0,
      "step": 41240
    },
    {
      "entropy": 6.068441534042359,
      "epoch": 3.642264217590957,
      "grad_norm": 1.765625,
      "learning_rate": 0.000371704372210962,
      "loss": 5.3404,
      "mean_token_accuracy": 0.21263501793146133,
      "num_tokens": 76897347.0,
      "step": 41245
    },
    {
      "entropy": 6.0187554359436035,
      "epoch": 3.6427057576827977,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003716759367816722,
      "loss": 5.2973,
      "mean_token_accuracy": 0.21100156009197235,
      "num_tokens": 76907021.0,
      "step": 41250
    },
    {
      "entropy": 5.971277284622192,
      "epoch": 3.6431472977746377,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003716474994585558,
      "loss": 5.2364,
      "mean_token_accuracy": 0.22308458834886552,
      "num_tokens": 76916805.0,
      "step": 41255
    },
    {
      "entropy": 6.0206318378448485,
      "epoch": 3.6435888378664782,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037161906024216993,
      "loss": 5.2492,
      "mean_token_accuracy": 0.21903636753559114,
      "num_tokens": 76926092.0,
      "step": 41260
    },
    {
      "entropy": 5.9417235374450685,
      "epoch": 3.6440303779583187,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037159061913307146,
      "loss": 5.1556,
      "mean_token_accuracy": 0.2288821741938591,
      "num_tokens": 76936025.0,
      "step": 41265
    },
    {
      "entropy": 6.042243623733521,
      "epoch": 3.644471918050159,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003715621761318178,
      "loss": 5.2574,
      "mean_token_accuracy": 0.21696797460317613,
      "num_tokens": 76945432.0,
      "step": 41270
    },
    {
      "entropy": 5.975041294097901,
      "epoch": 3.6449134581419993,
      "grad_norm": 1.171875,
      "learning_rate": 0.000371533731238966,
      "loss": 5.228,
      "mean_token_accuracy": 0.22584901750087738,
      "num_tokens": 76954938.0,
      "step": 41275
    },
    {
      "entropy": 5.915478897094727,
      "epoch": 3.64535499823384,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037150528445507327,
      "loss": 5.1846,
      "mean_token_accuracy": 0.23097624778747558,
      "num_tokens": 76963881.0,
      "step": 41280
    },
    {
      "entropy": 6.0697434425354,
      "epoch": 3.64579653832568,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037147683578069696,
      "loss": 5.3057,
      "mean_token_accuracy": 0.20886266380548477,
      "num_tokens": 76972603.0,
      "step": 41285
    },
    {
      "entropy": 5.992646789550781,
      "epoch": 3.6462380784175203,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003714483852163943,
      "loss": 5.2086,
      "mean_token_accuracy": 0.22159686535596848,
      "num_tokens": 76981114.0,
      "step": 41290
    },
    {
      "entropy": 6.05653862953186,
      "epoch": 3.646679618509361,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003714199327627226,
      "loss": 5.3377,
      "mean_token_accuracy": 0.20936788916587828,
      "num_tokens": 76990145.0,
      "step": 41295
    },
    {
      "entropy": 6.085817337036133,
      "epoch": 3.647121158601201,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037139147842023924,
      "loss": 5.4201,
      "mean_token_accuracy": 0.20541784167289734,
      "num_tokens": 77000364.0,
      "step": 41300
    },
    {
      "entropy": 6.070298480987549,
      "epoch": 3.6475626986930414,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037136302218950163,
      "loss": 5.3162,
      "mean_token_accuracy": 0.21829595267772675,
      "num_tokens": 77009999.0,
      "step": 41305
    },
    {
      "entropy": 6.009690856933593,
      "epoch": 3.648004238784882,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003713345640710673,
      "loss": 5.2973,
      "mean_token_accuracy": 0.21633923649787903,
      "num_tokens": 77019993.0,
      "step": 41310
    },
    {
      "entropy": 6.045032691955567,
      "epoch": 3.648445778876722,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037130610406549354,
      "loss": 5.3405,
      "mean_token_accuracy": 0.20754066556692125,
      "num_tokens": 77029633.0,
      "step": 41315
    },
    {
      "entropy": 5.985825490951538,
      "epoch": 3.6488873189685624,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037127764217333804,
      "loss": 5.1715,
      "mean_token_accuracy": 0.22550591230392455,
      "num_tokens": 77038250.0,
      "step": 41320
    },
    {
      "entropy": 5.9818626880645756,
      "epoch": 3.649328859060403,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037124917839515833,
      "loss": 5.2878,
      "mean_token_accuracy": 0.21037957519292833,
      "num_tokens": 77046851.0,
      "step": 41325
    },
    {
      "entropy": 5.841588544845581,
      "epoch": 3.649770399152243,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037122071273151184,
      "loss": 5.0222,
      "mean_token_accuracy": 0.24588058590888978,
      "num_tokens": 77055877.0,
      "step": 41330
    },
    {
      "entropy": 5.997694301605224,
      "epoch": 3.6502119392440835,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003711922451829564,
      "loss": 5.1983,
      "mean_token_accuracy": 0.22447433620691298,
      "num_tokens": 77065122.0,
      "step": 41335
    },
    {
      "entropy": 6.045902585983276,
      "epoch": 3.6506534793359235,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037116377575004963,
      "loss": 5.2671,
      "mean_token_accuracy": 0.21709627658128738,
      "num_tokens": 77074949.0,
      "step": 41340
    },
    {
      "entropy": 6.063542985916138,
      "epoch": 3.651095019427764,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003711353044333491,
      "loss": 5.3298,
      "mean_token_accuracy": 0.20760814547538758,
      "num_tokens": 77083992.0,
      "step": 41345
    },
    {
      "entropy": 5.950414657592773,
      "epoch": 3.6515365595196045,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003711068312334126,
      "loss": 5.2255,
      "mean_token_accuracy": 0.22203479409217836,
      "num_tokens": 77093105.0,
      "step": 41350
    },
    {
      "entropy": 5.96161904335022,
      "epoch": 3.6519780996114446,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000371078356150798,
      "loss": 5.1635,
      "mean_token_accuracy": 0.23149460703134536,
      "num_tokens": 77101811.0,
      "step": 41355
    },
    {
      "entropy": 5.944381427764893,
      "epoch": 3.652419639703285,
      "grad_norm": 1.25,
      "learning_rate": 0.000371049879186063,
      "loss": 5.2798,
      "mean_token_accuracy": 0.22242685109376908,
      "num_tokens": 77111415.0,
      "step": 41360
    },
    {
      "entropy": 6.039681768417358,
      "epoch": 3.652861179795125,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037102140033976554,
      "loss": 5.2868,
      "mean_token_accuracy": 0.21710318475961685,
      "num_tokens": 77121910.0,
      "step": 41365
    },
    {
      "entropy": 6.035683298110962,
      "epoch": 3.6533027198869656,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037099291961246344,
      "loss": 5.2318,
      "mean_token_accuracy": 0.22486020177602767,
      "num_tokens": 77131625.0,
      "step": 41370
    },
    {
      "entropy": 6.055531311035156,
      "epoch": 3.653744259978806,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003709644370047146,
      "loss": 5.3216,
      "mean_token_accuracy": 0.20503088980913162,
      "num_tokens": 77141246.0,
      "step": 41375
    },
    {
      "entropy": 6.027341699600219,
      "epoch": 3.654185800070646,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037093595251707707,
      "loss": 5.3189,
      "mean_token_accuracy": 0.21476359963417052,
      "num_tokens": 77150911.0,
      "step": 41380
    },
    {
      "entropy": 6.0186717987060545,
      "epoch": 3.6546273401624867,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00037090746615010877,
      "loss": 5.227,
      "mean_token_accuracy": 0.2172963172197342,
      "num_tokens": 77159899.0,
      "step": 41385
    },
    {
      "entropy": 5.972275733947754,
      "epoch": 3.655068880254327,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003708789779043678,
      "loss": 5.2111,
      "mean_token_accuracy": 0.22901476919651031,
      "num_tokens": 77169247.0,
      "step": 41390
    },
    {
      "entropy": 5.987800359725952,
      "epoch": 3.6555104203461672,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003708504877804121,
      "loss": 5.168,
      "mean_token_accuracy": 0.2240384191274643,
      "num_tokens": 77177843.0,
      "step": 41395
    },
    {
      "entropy": 6.053216886520386,
      "epoch": 3.6559519604380077,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037082199577879993,
      "loss": 5.2971,
      "mean_token_accuracy": 0.21179691553115845,
      "num_tokens": 77187333.0,
      "step": 41400
    },
    {
      "entropy": 6.041362380981445,
      "epoch": 3.6563935005298482,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003707935019000893,
      "loss": 5.2713,
      "mean_token_accuracy": 0.21467529684305192,
      "num_tokens": 77196123.0,
      "step": 41405
    },
    {
      "entropy": 6.0064554691314695,
      "epoch": 3.6568350406216883,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003707650061448385,
      "loss": 5.2443,
      "mean_token_accuracy": 0.2218698039650917,
      "num_tokens": 77205368.0,
      "step": 41410
    },
    {
      "entropy": 6.094214534759521,
      "epoch": 3.657276580713529,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003707365085136056,
      "loss": 5.3893,
      "mean_token_accuracy": 0.20740968734025955,
      "num_tokens": 77215269.0,
      "step": 41415
    },
    {
      "entropy": 6.014744901657105,
      "epoch": 3.6577181208053693,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037070800900694913,
      "loss": 5.2537,
      "mean_token_accuracy": 0.22360860109329223,
      "num_tokens": 77224808.0,
      "step": 41420
    },
    {
      "entropy": 5.958381319046021,
      "epoch": 3.6581596608972093,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037067950762542705,
      "loss": 5.124,
      "mean_token_accuracy": 0.2334657445549965,
      "num_tokens": 77234073.0,
      "step": 41425
    },
    {
      "entropy": 5.994264459609985,
      "epoch": 3.65860120098905,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037065100436959784,
      "loss": 5.2358,
      "mean_token_accuracy": 0.2185708224773407,
      "num_tokens": 77243443.0,
      "step": 41430
    },
    {
      "entropy": 6.049420976638794,
      "epoch": 3.6590427410808903,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003706224992400199,
      "loss": 5.2341,
      "mean_token_accuracy": 0.21571906208992003,
      "num_tokens": 77252421.0,
      "step": 41435
    },
    {
      "entropy": 6.009492683410644,
      "epoch": 3.6594842811727304,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003705939922372515,
      "loss": 5.2702,
      "mean_token_accuracy": 0.2149125337600708,
      "num_tokens": 77262286.0,
      "step": 41440
    },
    {
      "entropy": 6.035197114944458,
      "epoch": 3.659925821264571,
      "grad_norm": 1.25,
      "learning_rate": 0.0003705654833618512,
      "loss": 5.3132,
      "mean_token_accuracy": 0.2105032205581665,
      "num_tokens": 77271805.0,
      "step": 41445
    },
    {
      "entropy": 6.109109020233154,
      "epoch": 3.6603673613564114,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037053697261437747,
      "loss": 5.2885,
      "mean_token_accuracy": 0.21421737521886824,
      "num_tokens": 77280925.0,
      "step": 41450
    },
    {
      "entropy": 5.988940906524658,
      "epoch": 3.6608089014482514,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003705084599953888,
      "loss": 5.2281,
      "mean_token_accuracy": 0.2252134457230568,
      "num_tokens": 77290861.0,
      "step": 41455
    },
    {
      "entropy": 6.007420015335083,
      "epoch": 3.661250441540092,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003704799455054437,
      "loss": 5.3058,
      "mean_token_accuracy": 0.21012211889028548,
      "num_tokens": 77300561.0,
      "step": 41460
    },
    {
      "entropy": 5.983600807189942,
      "epoch": 3.6616919816319324,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003704514291451007,
      "loss": 5.2824,
      "mean_token_accuracy": 0.2193104773759842,
      "num_tokens": 77309274.0,
      "step": 41465
    },
    {
      "entropy": 6.005647230148315,
      "epoch": 3.6621335217237725,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037042291091491855,
      "loss": 5.2703,
      "mean_token_accuracy": 0.21502462923526763,
      "num_tokens": 77318734.0,
      "step": 41470
    },
    {
      "entropy": 6.124268674850464,
      "epoch": 3.662575061815613,
      "grad_norm": 1.296875,
      "learning_rate": 0.00037039439081545586,
      "loss": 5.395,
      "mean_token_accuracy": 0.20193596482276915,
      "num_tokens": 77328920.0,
      "step": 41475
    },
    {
      "entropy": 6.031385135650635,
      "epoch": 3.663016601907453,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003703658688472713,
      "loss": 5.2737,
      "mean_token_accuracy": 0.21885279417037964,
      "num_tokens": 77338140.0,
      "step": 41480
    },
    {
      "entropy": 6.012501382827759,
      "epoch": 3.6634581419992935,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003703373450109235,
      "loss": 5.2464,
      "mean_token_accuracy": 0.21973757296800614,
      "num_tokens": 77346709.0,
      "step": 41485
    },
    {
      "entropy": 5.981048107147217,
      "epoch": 3.663899682091134,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003703088193069715,
      "loss": 5.3044,
      "mean_token_accuracy": 0.21729732751846315,
      "num_tokens": 77356102.0,
      "step": 41490
    },
    {
      "entropy": 6.065795087814331,
      "epoch": 3.664341222182974,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003702802917359738,
      "loss": 5.2542,
      "mean_token_accuracy": 0.21328988075256347,
      "num_tokens": 77364695.0,
      "step": 41495
    },
    {
      "entropy": 5.922579908370972,
      "epoch": 3.6647827622748146,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037025176229848946,
      "loss": 5.1858,
      "mean_token_accuracy": 0.22936673760414122,
      "num_tokens": 77374306.0,
      "step": 41500
    },
    {
      "entropy": 5.997726345062256,
      "epoch": 3.6652243023666546,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003702232309950772,
      "loss": 5.2335,
      "mean_token_accuracy": 0.22213684171438217,
      "num_tokens": 77383881.0,
      "step": 41505
    },
    {
      "entropy": 6.043284702301025,
      "epoch": 3.665665842458495,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003701946978262961,
      "loss": 5.317,
      "mean_token_accuracy": 0.20911959558725357,
      "num_tokens": 77393613.0,
      "step": 41510
    },
    {
      "entropy": 6.074390125274658,
      "epoch": 3.6661073825503356,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037016616279270494,
      "loss": 5.2996,
      "mean_token_accuracy": 0.21792113780975342,
      "num_tokens": 77403387.0,
      "step": 41515
    },
    {
      "entropy": 6.03606767654419,
      "epoch": 3.6665489226421757,
      "grad_norm": 1.421875,
      "learning_rate": 0.00037013762589486283,
      "loss": 5.2571,
      "mean_token_accuracy": 0.21912634670734404,
      "num_tokens": 77413087.0,
      "step": 41520
    },
    {
      "entropy": 5.912835788726807,
      "epoch": 3.666990462734016,
      "grad_norm": 1.328125,
      "learning_rate": 0.00037010908713332863,
      "loss": 5.115,
      "mean_token_accuracy": 0.2301020950078964,
      "num_tokens": 77422044.0,
      "step": 41525
    },
    {
      "entropy": 5.974138832092285,
      "epoch": 3.6674320028258567,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003700805465086616,
      "loss": 5.3068,
      "mean_token_accuracy": 0.21888009309768677,
      "num_tokens": 77431171.0,
      "step": 41530
    },
    {
      "entropy": 6.024148464202881,
      "epoch": 3.6678735429176967,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003700520040214206,
      "loss": 5.2441,
      "mean_token_accuracy": 0.21945209056138992,
      "num_tokens": 77440141.0,
      "step": 41535
    },
    {
      "entropy": 6.026817989349365,
      "epoch": 3.6683150830095372,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003700234596721651,
      "loss": 5.2711,
      "mean_token_accuracy": 0.2214582860469818,
      "num_tokens": 77449160.0,
      "step": 41540
    },
    {
      "entropy": 5.987123632431031,
      "epoch": 3.6687566231013777,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036999491346145406,
      "loss": 5.2698,
      "mean_token_accuracy": 0.2118513196706772,
      "num_tokens": 77458355.0,
      "step": 41545
    },
    {
      "entropy": 6.058993291854859,
      "epoch": 3.669198163193218,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036996636538984653,
      "loss": 5.3646,
      "mean_token_accuracy": 0.21042191684246064,
      "num_tokens": 77468456.0,
      "step": 41550
    },
    {
      "entropy": 6.0422900199890135,
      "epoch": 3.6696397032850583,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036993781545790203,
      "loss": 5.3115,
      "mean_token_accuracy": 0.2166571542620659,
      "num_tokens": 77477424.0,
      "step": 41555
    },
    {
      "entropy": 6.067286586761474,
      "epoch": 3.670081243376899,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003699092636661796,
      "loss": 5.2785,
      "mean_token_accuracy": 0.21163786798715592,
      "num_tokens": 77485886.0,
      "step": 41560
    },
    {
      "entropy": 5.928905582427978,
      "epoch": 3.670522783468739,
      "grad_norm": 1.25,
      "learning_rate": 0.0003698807100152387,
      "loss": 5.1943,
      "mean_token_accuracy": 0.21954996138811111,
      "num_tokens": 77494698.0,
      "step": 41565
    },
    {
      "entropy": 6.054824256896973,
      "epoch": 3.6709643235605793,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036985215450563875,
      "loss": 5.3978,
      "mean_token_accuracy": 0.21258438378572464,
      "num_tokens": 77503829.0,
      "step": 41570
    },
    {
      "entropy": 6.134568119049073,
      "epoch": 3.67140586365242,
      "grad_norm": 1.25,
      "learning_rate": 0.000369823597137939,
      "loss": 5.3756,
      "mean_token_accuracy": 0.2088463231921196,
      "num_tokens": 77514423.0,
      "step": 41575
    },
    {
      "entropy": 6.008603763580322,
      "epoch": 3.67184740374426,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003697950379126989,
      "loss": 5.1855,
      "mean_token_accuracy": 0.2304933935403824,
      "num_tokens": 77525345.0,
      "step": 41580
    },
    {
      "entropy": 5.980715322494507,
      "epoch": 3.6722889438361004,
      "grad_norm": 1.25,
      "learning_rate": 0.00036976647683047793,
      "loss": 5.255,
      "mean_token_accuracy": 0.22254034727811814,
      "num_tokens": 77534689.0,
      "step": 41585
    },
    {
      "entropy": 6.023245811462402,
      "epoch": 3.672730483927941,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036973791389183555,
      "loss": 5.2537,
      "mean_token_accuracy": 0.2198873832821846,
      "num_tokens": 77543679.0,
      "step": 41590
    },
    {
      "entropy": 5.953178119659424,
      "epoch": 3.673172024019781,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036970934909733136,
      "loss": 5.176,
      "mean_token_accuracy": 0.22438991367816924,
      "num_tokens": 77552324.0,
      "step": 41595
    },
    {
      "entropy": 5.998449087142944,
      "epoch": 3.6736135641116214,
      "grad_norm": 1.25,
      "learning_rate": 0.00036968078244752485,
      "loss": 5.2531,
      "mean_token_accuracy": 0.2221282199025154,
      "num_tokens": 77562013.0,
      "step": 41600
    },
    {
      "entropy": 5.98947377204895,
      "epoch": 3.674055104203462,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036965221394297563,
      "loss": 5.2004,
      "mean_token_accuracy": 0.22316838651895524,
      "num_tokens": 77571915.0,
      "step": 41605
    },
    {
      "entropy": 6.018424463272095,
      "epoch": 3.674496644295302,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036962364358424343,
      "loss": 5.2909,
      "mean_token_accuracy": 0.22311420738697052,
      "num_tokens": 77581786.0,
      "step": 41610
    },
    {
      "entropy": 6.039031314849853,
      "epoch": 3.6749381843871425,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036959507137188785,
      "loss": 5.3205,
      "mean_token_accuracy": 0.2108864575624466,
      "num_tokens": 77590681.0,
      "step": 41615
    },
    {
      "entropy": 6.048904514312744,
      "epoch": 3.6753797244789825,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003695664973064686,
      "loss": 5.2209,
      "mean_token_accuracy": 0.21624691486358644,
      "num_tokens": 77599450.0,
      "step": 41620
    },
    {
      "entropy": 5.995367336273193,
      "epoch": 3.675821264570823,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036953792138854545,
      "loss": 5.2433,
      "mean_token_accuracy": 0.21458617597818375,
      "num_tokens": 77609325.0,
      "step": 41625
    },
    {
      "entropy": 5.986234283447265,
      "epoch": 3.6762628046626635,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036950934361867823,
      "loss": 5.2556,
      "mean_token_accuracy": 0.2143192782998085,
      "num_tokens": 77619479.0,
      "step": 41630
    },
    {
      "entropy": 5.983013772964478,
      "epoch": 3.6767043447545036,
      "grad_norm": 1.25,
      "learning_rate": 0.0003694807639974267,
      "loss": 5.3277,
      "mean_token_accuracy": 0.21080234199762343,
      "num_tokens": 77629752.0,
      "step": 41635
    },
    {
      "entropy": 6.033335971832275,
      "epoch": 3.677145884846344,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003694521825253507,
      "loss": 5.2864,
      "mean_token_accuracy": 0.21687623113393784,
      "num_tokens": 77638630.0,
      "step": 41640
    },
    {
      "entropy": 5.9518509864807125,
      "epoch": 3.677587424938184,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003694235992030102,
      "loss": 5.2111,
      "mean_token_accuracy": 0.22598367631435395,
      "num_tokens": 77647623.0,
      "step": 41645
    },
    {
      "entropy": 6.04039421081543,
      "epoch": 3.6780289650300246,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003693950140309651,
      "loss": 5.3159,
      "mean_token_accuracy": 0.21505086421966552,
      "num_tokens": 77657437.0,
      "step": 41650
    },
    {
      "entropy": 6.056277465820313,
      "epoch": 3.678470505121865,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036936642700977534,
      "loss": 5.3143,
      "mean_token_accuracy": 0.2096608340740204,
      "num_tokens": 77667481.0,
      "step": 41655
    },
    {
      "entropy": 6.032918262481689,
      "epoch": 3.678912045213705,
      "grad_norm": 1.21875,
      "learning_rate": 0.000369337838140001,
      "loss": 5.3359,
      "mean_token_accuracy": 0.2145427167415619,
      "num_tokens": 77676431.0,
      "step": 41660
    },
    {
      "entropy": 6.001050233840942,
      "epoch": 3.6793535853055457,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000369309247422202,
      "loss": 5.2611,
      "mean_token_accuracy": 0.22794784605503082,
      "num_tokens": 77685320.0,
      "step": 41665
    },
    {
      "entropy": 6.028917026519776,
      "epoch": 3.679795125397386,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036928065485693847,
      "loss": 5.2936,
      "mean_token_accuracy": 0.2115004241466522,
      "num_tokens": 77695335.0,
      "step": 41670
    },
    {
      "entropy": 5.951930809020996,
      "epoch": 3.6802366654892262,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003692520604447706,
      "loss": 5.1511,
      "mean_token_accuracy": 0.2324732333421707,
      "num_tokens": 77704775.0,
      "step": 41675
    },
    {
      "entropy": 6.013600015640259,
      "epoch": 3.6806782055810667,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003692234641862584,
      "loss": 5.1864,
      "mean_token_accuracy": 0.22185406982898712,
      "num_tokens": 77713747.0,
      "step": 41680
    },
    {
      "entropy": 5.981971073150635,
      "epoch": 3.6811197456729072,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036919486608196216,
      "loss": 5.205,
      "mean_token_accuracy": 0.22351097017526628,
      "num_tokens": 77723763.0,
      "step": 41685
    },
    {
      "entropy": 6.042652797698975,
      "epoch": 3.6815612857647473,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000369166266132442,
      "loss": 5.4092,
      "mean_token_accuracy": 0.2043102726340294,
      "num_tokens": 77734117.0,
      "step": 41690
    },
    {
      "entropy": 5.955663204193115,
      "epoch": 3.682002825856588,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003691376643382583,
      "loss": 5.2234,
      "mean_token_accuracy": 0.2218985766172409,
      "num_tokens": 77744146.0,
      "step": 41695
    },
    {
      "entropy": 5.9346545219421385,
      "epoch": 3.6824443659484283,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003691090606999713,
      "loss": 5.1849,
      "mean_token_accuracy": 0.2196182742714882,
      "num_tokens": 77753646.0,
      "step": 41700
    },
    {
      "entropy": 5.950384426116943,
      "epoch": 3.6828859060402683,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003690804552181413,
      "loss": 5.1996,
      "mean_token_accuracy": 0.22772760540246964,
      "num_tokens": 77763439.0,
      "step": 41705
    },
    {
      "entropy": 6.112442684173584,
      "epoch": 3.683327446132109,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003690518478933287,
      "loss": 5.3562,
      "mean_token_accuracy": 0.2090508833527565,
      "num_tokens": 77772398.0,
      "step": 41710
    },
    {
      "entropy": 6.030537462234497,
      "epoch": 3.6837689862239493,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003690232387260939,
      "loss": 5.309,
      "mean_token_accuracy": 0.21255876868963242,
      "num_tokens": 77781265.0,
      "step": 41715
    },
    {
      "entropy": 5.945437908172607,
      "epoch": 3.6842105263157894,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003689946277169974,
      "loss": 5.2116,
      "mean_token_accuracy": 0.22228961884975434,
      "num_tokens": 77789990.0,
      "step": 41720
    },
    {
      "entropy": 5.966455078125,
      "epoch": 3.68465206640763,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003689660148665994,
      "loss": 5.3138,
      "mean_token_accuracy": 0.2163165658712387,
      "num_tokens": 77799504.0,
      "step": 41725
    },
    {
      "entropy": 6.018873071670532,
      "epoch": 3.6850936064994704,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003689374001754608,
      "loss": 5.2686,
      "mean_token_accuracy": 0.21650283336639403,
      "num_tokens": 77809332.0,
      "step": 41730
    },
    {
      "entropy": 6.024634790420532,
      "epoch": 3.6855351465913104,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036890878364414186,
      "loss": 5.2011,
      "mean_token_accuracy": 0.21883153021335602,
      "num_tokens": 77817704.0,
      "step": 41735
    },
    {
      "entropy": 5.822226524353027,
      "epoch": 3.685976686683151,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036888016527320336,
      "loss": 4.9902,
      "mean_token_accuracy": 0.24743401110172272,
      "num_tokens": 77827037.0,
      "step": 41740
    },
    {
      "entropy": 6.00948657989502,
      "epoch": 3.6864182267749914,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036885154506320573,
      "loss": 5.311,
      "mean_token_accuracy": 0.20819534957408906,
      "num_tokens": 77836416.0,
      "step": 41745
    },
    {
      "entropy": 5.999962425231933,
      "epoch": 3.6868597668668315,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003688229230147097,
      "loss": 5.2954,
      "mean_token_accuracy": 0.21605041474103928,
      "num_tokens": 77845297.0,
      "step": 41750
    },
    {
      "entropy": 5.987960863113403,
      "epoch": 3.687301306958672,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036879429912827595,
      "loss": 5.2341,
      "mean_token_accuracy": 0.21845365613698958,
      "num_tokens": 77854720.0,
      "step": 41755
    },
    {
      "entropy": 6.045339155197143,
      "epoch": 3.687742847050512,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003687656734044653,
      "loss": 5.2744,
      "mean_token_accuracy": 0.21331124752759933,
      "num_tokens": 77863980.0,
      "step": 41760
    },
    {
      "entropy": 6.00418872833252,
      "epoch": 3.6881843871423525,
      "grad_norm": 1.40625,
      "learning_rate": 0.00036873704584383833,
      "loss": 5.2777,
      "mean_token_accuracy": 0.21674338430166246,
      "num_tokens": 77873188.0,
      "step": 41765
    },
    {
      "entropy": 5.987074565887451,
      "epoch": 3.688625927234193,
      "grad_norm": 1.1875,
      "learning_rate": 0.000368708416446956,
      "loss": 5.2877,
      "mean_token_accuracy": 0.21833640038967134,
      "num_tokens": 77883359.0,
      "step": 41770
    },
    {
      "entropy": 6.081725263595581,
      "epoch": 3.689067467326033,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003686797852143791,
      "loss": 5.3227,
      "mean_token_accuracy": 0.20723979026079178,
      "num_tokens": 77892965.0,
      "step": 41775
    },
    {
      "entropy": 6.063504600524903,
      "epoch": 3.6895090074178736,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003686511521466685,
      "loss": 5.3165,
      "mean_token_accuracy": 0.21088393181562423,
      "num_tokens": 77902009.0,
      "step": 41780
    },
    {
      "entropy": 6.04690728187561,
      "epoch": 3.6899505475097136,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003686225172443851,
      "loss": 5.2865,
      "mean_token_accuracy": 0.21592076867818832,
      "num_tokens": 77910033.0,
      "step": 41785
    },
    {
      "entropy": 5.959406900405884,
      "epoch": 3.690392087601554,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003685938805080898,
      "loss": 5.2248,
      "mean_token_accuracy": 0.2248488411307335,
      "num_tokens": 77919175.0,
      "step": 41790
    },
    {
      "entropy": 6.008219289779663,
      "epoch": 3.6908336276933946,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003685652419383436,
      "loss": 5.3538,
      "mean_token_accuracy": 0.20490991473197936,
      "num_tokens": 77928821.0,
      "step": 41795
    },
    {
      "entropy": 6.068520879745483,
      "epoch": 3.6912751677852347,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036853660153570757,
      "loss": 5.3259,
      "mean_token_accuracy": 0.21413565874099733,
      "num_tokens": 77938450.0,
      "step": 41800
    },
    {
      "entropy": 6.048783779144287,
      "epoch": 3.691716707877075,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036850795930074264,
      "loss": 5.262,
      "mean_token_accuracy": 0.2165166661143303,
      "num_tokens": 77948356.0,
      "step": 41805
    },
    {
      "entropy": 6.010520648956299,
      "epoch": 3.6921582479689157,
      "grad_norm": 1.3125,
      "learning_rate": 0.00036847931523401,
      "loss": 5.2362,
      "mean_token_accuracy": 0.22082072347402573,
      "num_tokens": 77956140.0,
      "step": 41810
    },
    {
      "entropy": 6.1003693580627445,
      "epoch": 3.6925997880607557,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003684506693360708,
      "loss": 5.3086,
      "mean_token_accuracy": 0.21396929323673247,
      "num_tokens": 77965822.0,
      "step": 41815
    },
    {
      "entropy": 6.016507339477539,
      "epoch": 3.6930413281525962,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00036842202160748607,
      "loss": 5.3222,
      "mean_token_accuracy": 0.2110392689704895,
      "num_tokens": 77975563.0,
      "step": 41820
    },
    {
      "entropy": 6.018904161453247,
      "epoch": 3.6934828682444367,
      "grad_norm": 1.3125,
      "learning_rate": 0.00036839337204881706,
      "loss": 5.3423,
      "mean_token_accuracy": 0.21653753221035005,
      "num_tokens": 77984613.0,
      "step": 41825
    },
    {
      "entropy": 6.060272455215454,
      "epoch": 3.6939244083362768,
      "grad_norm": 1.375,
      "learning_rate": 0.000368364720660625,
      "loss": 5.2393,
      "mean_token_accuracy": 0.22138720005750656,
      "num_tokens": 77993408.0,
      "step": 41830
    },
    {
      "entropy": 6.055892038345337,
      "epoch": 3.6943659484281173,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003683360674434712,
      "loss": 5.2542,
      "mean_token_accuracy": 0.2184482231736183,
      "num_tokens": 78001967.0,
      "step": 41835
    },
    {
      "entropy": 5.9645373821258545,
      "epoch": 3.6948074885199578,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000368307412397917,
      "loss": 5.3393,
      "mean_token_accuracy": 0.20974552631378174,
      "num_tokens": 78011687.0,
      "step": 41840
    },
    {
      "entropy": 6.028473997116089,
      "epoch": 3.695249028611798,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003682787555245235,
      "loss": 5.304,
      "mean_token_accuracy": 0.21726565659046174,
      "num_tokens": 78021195.0,
      "step": 41845
    },
    {
      "entropy": 6.041517543792724,
      "epoch": 3.6956905687036383,
      "grad_norm": 1.3125,
      "learning_rate": 0.00036825009682385235,
      "loss": 5.2571,
      "mean_token_accuracy": 0.21627792865037918,
      "num_tokens": 78030388.0,
      "step": 41850
    },
    {
      "entropy": 5.955438184738159,
      "epoch": 3.696132108795479,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003682214362964648,
      "loss": 5.1921,
      "mean_token_accuracy": 0.22136341482400895,
      "num_tokens": 78038901.0,
      "step": 41855
    },
    {
      "entropy": 5.975899362564087,
      "epoch": 3.696573648887319,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003681927739429223,
      "loss": 5.277,
      "mean_token_accuracy": 0.21799213141202928,
      "num_tokens": 78047158.0,
      "step": 41860
    },
    {
      "entropy": 5.9491541385650635,
      "epoch": 3.6970151889791594,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003681641097637864,
      "loss": 5.2323,
      "mean_token_accuracy": 0.22285337448120118,
      "num_tokens": 78057116.0,
      "step": 41865
    },
    {
      "entropy": 6.126093959808349,
      "epoch": 3.697456729071,
      "grad_norm": 1.25,
      "learning_rate": 0.0003681354437596185,
      "loss": 5.3487,
      "mean_token_accuracy": 0.2101105347275734,
      "num_tokens": 78066437.0,
      "step": 41870
    },
    {
      "entropy": 6.088962650299072,
      "epoch": 3.69789826916284,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003681067759309803,
      "loss": 5.286,
      "mean_token_accuracy": 0.21259111613035203,
      "num_tokens": 78075045.0,
      "step": 41875
    },
    {
      "entropy": 5.956013774871826,
      "epoch": 3.6983398092546804,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003680781062784333,
      "loss": 5.2081,
      "mean_token_accuracy": 0.22201838940382004,
      "num_tokens": 78084340.0,
      "step": 41880
    },
    {
      "entropy": 6.067495918273925,
      "epoch": 3.698781349346521,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003680494348025391,
      "loss": 5.3235,
      "mean_token_accuracy": 0.2125403732061386,
      "num_tokens": 78094119.0,
      "step": 41885
    },
    {
      "entropy": 6.013126182556152,
      "epoch": 3.699222889438361,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003680207615038594,
      "loss": 5.2923,
      "mean_token_accuracy": 0.21712397933006286,
      "num_tokens": 78103598.0,
      "step": 41890
    },
    {
      "entropy": 6.049494457244873,
      "epoch": 3.6996644295302015,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036799208638295595,
      "loss": 5.3487,
      "mean_token_accuracy": 0.21038172394037247,
      "num_tokens": 78112686.0,
      "step": 41895
    },
    {
      "entropy": 6.128420305252075,
      "epoch": 3.7001059696220415,
      "grad_norm": 1.375,
      "learning_rate": 0.0003679634094403904,
      "loss": 5.3715,
      "mean_token_accuracy": 0.20682986825704575,
      "num_tokens": 78122009.0,
      "step": 41900
    },
    {
      "entropy": 6.0941239356994625,
      "epoch": 3.700547509713882,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003679347306767245,
      "loss": 5.2777,
      "mean_token_accuracy": 0.21085772514343262,
      "num_tokens": 78131267.0,
      "step": 41905
    },
    {
      "entropy": 5.983356380462647,
      "epoch": 3.7009890498057225,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036790605009252007,
      "loss": 5.322,
      "mean_token_accuracy": 0.2078641176223755,
      "num_tokens": 78141265.0,
      "step": 41910
    },
    {
      "entropy": 5.957098817825317,
      "epoch": 3.7014305898975626,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036787736768833894,
      "loss": 5.2922,
      "mean_token_accuracy": 0.22685550302267074,
      "num_tokens": 78151217.0,
      "step": 41915
    },
    {
      "entropy": 6.04235348701477,
      "epoch": 3.701872129989403,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000367848683464743,
      "loss": 5.3194,
      "mean_token_accuracy": 0.2132650464773178,
      "num_tokens": 78161781.0,
      "step": 41920
    },
    {
      "entropy": 5.982724571228028,
      "epoch": 3.702313670081243,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036781999742229425,
      "loss": 5.2293,
      "mean_token_accuracy": 0.21542841643095018,
      "num_tokens": 78170619.0,
      "step": 41925
    },
    {
      "entropy": 6.038200044631958,
      "epoch": 3.7027552101730836,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036779130956155447,
      "loss": 5.3304,
      "mean_token_accuracy": 0.21068172603845597,
      "num_tokens": 78180385.0,
      "step": 41930
    },
    {
      "entropy": 6.0958061695098875,
      "epoch": 3.703196750264924,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003677626198830857,
      "loss": 5.3176,
      "mean_token_accuracy": 0.2152378499507904,
      "num_tokens": 78190732.0,
      "step": 41935
    },
    {
      "entropy": 6.0252283096313475,
      "epoch": 3.703638290356764,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036773392838745,
      "loss": 5.2806,
      "mean_token_accuracy": 0.20994695872068406,
      "num_tokens": 78200029.0,
      "step": 41940
    },
    {
      "entropy": 6.026421308517456,
      "epoch": 3.7040798304486047,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003677052350752093,
      "loss": 5.2917,
      "mean_token_accuracy": 0.21319807022809983,
      "num_tokens": 78209364.0,
      "step": 41945
    },
    {
      "entropy": 5.861460733413696,
      "epoch": 3.704521370540445,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036767653994692587,
      "loss": 5.1419,
      "mean_token_accuracy": 0.2422653079032898,
      "num_tokens": 78219184.0,
      "step": 41950
    },
    {
      "entropy": 6.034869194030762,
      "epoch": 3.7049629106322852,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003676478430031617,
      "loss": 5.2671,
      "mean_token_accuracy": 0.2142718866467476,
      "num_tokens": 78228101.0,
      "step": 41955
    },
    {
      "entropy": 5.99943585395813,
      "epoch": 3.7054044507241257,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036761914424447896,
      "loss": 5.2791,
      "mean_token_accuracy": 0.22500259131193162,
      "num_tokens": 78237161.0,
      "step": 41960
    },
    {
      "entropy": 6.043912935256958,
      "epoch": 3.705845990815966,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036759044367143986,
      "loss": 5.203,
      "mean_token_accuracy": 0.22660729736089708,
      "num_tokens": 78245587.0,
      "step": 41965
    },
    {
      "entropy": 5.9669825553894045,
      "epoch": 3.7062875309078063,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003675617412846067,
      "loss": 5.2677,
      "mean_token_accuracy": 0.21196317225694655,
      "num_tokens": 78255270.0,
      "step": 41970
    },
    {
      "entropy": 5.932211065292359,
      "epoch": 3.7067290709996468,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036753303708454157,
      "loss": 5.1529,
      "mean_token_accuracy": 0.23166748583316804,
      "num_tokens": 78264740.0,
      "step": 41975
    },
    {
      "entropy": 5.933231592178345,
      "epoch": 3.7071706110914873,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003675043310718069,
      "loss": 5.1229,
      "mean_token_accuracy": 0.22621422111988068,
      "num_tokens": 78273541.0,
      "step": 41980
    },
    {
      "entropy": 6.045287227630615,
      "epoch": 3.7076121511833273,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036747562324696497,
      "loss": 5.2782,
      "mean_token_accuracy": 0.21136512607336044,
      "num_tokens": 78283193.0,
      "step": 41985
    },
    {
      "entropy": 6.020685005187988,
      "epoch": 3.708053691275168,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036744691361057825,
      "loss": 5.2299,
      "mean_token_accuracy": 0.21798670440912246,
      "num_tokens": 78292896.0,
      "step": 41990
    },
    {
      "entropy": 5.992304897308349,
      "epoch": 3.7084952313670083,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000367418202163209,
      "loss": 5.3041,
      "mean_token_accuracy": 0.2154141217470169,
      "num_tokens": 78302332.0,
      "step": 41995
    },
    {
      "entropy": 6.011648464202881,
      "epoch": 3.7089367714588484,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003673894889054198,
      "loss": 5.2905,
      "mean_token_accuracy": 0.21562178879976274,
      "num_tokens": 78312219.0,
      "step": 42000
    },
    {
      "epoch": 3.7089367714588484,
      "eval_entropy": 5.706960968037019,
      "eval_loss": 5.325263500213623,
      "eval_mean_token_accuracy": 0.22286920644688848,
      "eval_num_tokens": 78312219.0,
      "eval_runtime": 20.2364,
      "eval_samples_per_second": 1745.124,
      "eval_steps_per_second": 218.171,
      "step": 42000
    },
    {
      "entropy": 5.953034830093384,
      "epoch": 3.709378311550689,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036736077383777295,
      "loss": 5.2091,
      "mean_token_accuracy": 0.21941746920347213,
      "num_tokens": 78321775.0,
      "step": 42005
    },
    {
      "entropy": 5.9776472568511965,
      "epoch": 3.7098198516425294,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003673320569608311,
      "loss": 5.2297,
      "mean_token_accuracy": 0.22915003597736358,
      "num_tokens": 78330704.0,
      "step": 42010
    },
    {
      "entropy": 6.119410800933838,
      "epoch": 3.7102613917343694,
      "grad_norm": 1.25,
      "learning_rate": 0.00036730333827515683,
      "loss": 5.3866,
      "mean_token_accuracy": 0.21188260316848756,
      "num_tokens": 78339768.0,
      "step": 42015
    },
    {
      "entropy": 6.008576202392578,
      "epoch": 3.71070293182621,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036727461778131266,
      "loss": 5.2613,
      "mean_token_accuracy": 0.22078511118888855,
      "num_tokens": 78348879.0,
      "step": 42020
    },
    {
      "entropy": 6.053712320327759,
      "epoch": 3.7111444719180504,
      "grad_norm": 1.40625,
      "learning_rate": 0.00036724589547986115,
      "loss": 5.2692,
      "mean_token_accuracy": 0.2199752375483513,
      "num_tokens": 78357620.0,
      "step": 42025
    },
    {
      "entropy": 5.951960659027099,
      "epoch": 3.7115860120098905,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000367217171371365,
      "loss": 5.2467,
      "mean_token_accuracy": 0.2166401743888855,
      "num_tokens": 78366476.0,
      "step": 42030
    },
    {
      "entropy": 5.9569169044494625,
      "epoch": 3.712027552101731,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000367188445456387,
      "loss": 5.1611,
      "mean_token_accuracy": 0.2234411507844925,
      "num_tokens": 78375594.0,
      "step": 42035
    },
    {
      "entropy": 5.988476037979126,
      "epoch": 3.712469092193571,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003671597177354897,
      "loss": 5.2198,
      "mean_token_accuracy": 0.22224095165729524,
      "num_tokens": 78384740.0,
      "step": 42040
    },
    {
      "entropy": 6.0482635498046875,
      "epoch": 3.7129106322854115,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036713098820923603,
      "loss": 5.3146,
      "mean_token_accuracy": 0.21450986713171005,
      "num_tokens": 78394018.0,
      "step": 42045
    },
    {
      "entropy": 5.963355302810669,
      "epoch": 3.713352172377252,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003671022568781886,
      "loss": 5.2212,
      "mean_token_accuracy": 0.22306836992502213,
      "num_tokens": 78402670.0,
      "step": 42050
    },
    {
      "entropy": 6.063767433166504,
      "epoch": 3.713793712469092,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003670735237429104,
      "loss": 5.3022,
      "mean_token_accuracy": 0.2135046139359474,
      "num_tokens": 78412085.0,
      "step": 42055
    },
    {
      "entropy": 6.015009260177612,
      "epoch": 3.7142352525609326,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036704478880396417,
      "loss": 5.3034,
      "mean_token_accuracy": 0.21305010467767715,
      "num_tokens": 78421214.0,
      "step": 42060
    },
    {
      "entropy": 6.044259977340698,
      "epoch": 3.7146767926527726,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036701605206191295,
      "loss": 5.3035,
      "mean_token_accuracy": 0.2150068536400795,
      "num_tokens": 78430371.0,
      "step": 42065
    },
    {
      "entropy": 5.968030023574829,
      "epoch": 3.715118332744613,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036698731351731957,
      "loss": 5.2143,
      "mean_token_accuracy": 0.2217412158846855,
      "num_tokens": 78439502.0,
      "step": 42070
    },
    {
      "entropy": 5.957730960845947,
      "epoch": 3.7155598728364536,
      "grad_norm": 1.265625,
      "learning_rate": 0.000366958573170747,
      "loss": 5.3003,
      "mean_token_accuracy": 0.22327180206775665,
      "num_tokens": 78448619.0,
      "step": 42075
    },
    {
      "entropy": 5.980364608764648,
      "epoch": 3.7160014129282937,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036692983102275836,
      "loss": 5.2231,
      "mean_token_accuracy": 0.22974613606929778,
      "num_tokens": 78458215.0,
      "step": 42080
    },
    {
      "entropy": 5.976739168167114,
      "epoch": 3.716442953020134,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003669010870739165,
      "loss": 5.2198,
      "mean_token_accuracy": 0.22618685215711593,
      "num_tokens": 78468404.0,
      "step": 42085
    },
    {
      "entropy": 6.031637954711914,
      "epoch": 3.7168844931119747,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036687234132478475,
      "loss": 5.3307,
      "mean_token_accuracy": 0.2137994721531868,
      "num_tokens": 78478521.0,
      "step": 42090
    },
    {
      "entropy": 6.018298149108887,
      "epoch": 3.7173260332038147,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000366843593775926,
      "loss": 5.3323,
      "mean_token_accuracy": 0.2150583803653717,
      "num_tokens": 78488557.0,
      "step": 42095
    },
    {
      "entropy": 5.981817102432251,
      "epoch": 3.717767573295655,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036681484442790355,
      "loss": 5.2358,
      "mean_token_accuracy": 0.2223811998963356,
      "num_tokens": 78497066.0,
      "step": 42100
    },
    {
      "entropy": 6.016824626922608,
      "epoch": 3.7182091133874957,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003667860932812804,
      "loss": 5.2032,
      "mean_token_accuracy": 0.22347557544708252,
      "num_tokens": 78506122.0,
      "step": 42105
    },
    {
      "entropy": 5.974647426605225,
      "epoch": 3.7186506534793358,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036675734033662,
      "loss": 5.2134,
      "mean_token_accuracy": 0.21972811222076416,
      "num_tokens": 78515833.0,
      "step": 42110
    },
    {
      "entropy": 6.024445962905884,
      "epoch": 3.7190921935711763,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003667285855944854,
      "loss": 5.2378,
      "mean_token_accuracy": 0.22498371601104736,
      "num_tokens": 78524300.0,
      "step": 42115
    },
    {
      "entropy": 6.070438194274902,
      "epoch": 3.7195337336630168,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036669982905544003,
      "loss": 5.3692,
      "mean_token_accuracy": 0.2103586331009865,
      "num_tokens": 78533751.0,
      "step": 42120
    },
    {
      "entropy": 6.0179564476013185,
      "epoch": 3.719975273754857,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00036667107072004717,
      "loss": 5.2408,
      "mean_token_accuracy": 0.21358773559331895,
      "num_tokens": 78542556.0,
      "step": 42125
    },
    {
      "entropy": 6.062253046035766,
      "epoch": 3.7204168138466973,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003666423105888701,
      "loss": 5.3338,
      "mean_token_accuracy": 0.21144355684518815,
      "num_tokens": 78552343.0,
      "step": 42130
    },
    {
      "entropy": 5.910145235061646,
      "epoch": 3.720858353938538,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00036661354866247234,
      "loss": 5.2105,
      "mean_token_accuracy": 0.22221680581569672,
      "num_tokens": 78561289.0,
      "step": 42135
    },
    {
      "entropy": 6.017061614990235,
      "epoch": 3.721299894030378,
      "grad_norm": 1.375,
      "learning_rate": 0.00036658478494141724,
      "loss": 5.2344,
      "mean_token_accuracy": 0.22171377390623093,
      "num_tokens": 78570803.0,
      "step": 42140
    },
    {
      "entropy": 5.963635206222534,
      "epoch": 3.7217414341222184,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003665560194262683,
      "loss": 5.1957,
      "mean_token_accuracy": 0.2213050588965416,
      "num_tokens": 78579465.0,
      "step": 42145
    },
    {
      "entropy": 6.073292446136475,
      "epoch": 3.722182974214059,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000366527252117589,
      "loss": 5.3092,
      "mean_token_accuracy": 0.21089359670877456,
      "num_tokens": 78587732.0,
      "step": 42150
    },
    {
      "entropy": 6.037198257446289,
      "epoch": 3.722624514305899,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000366498483015943,
      "loss": 5.3002,
      "mean_token_accuracy": 0.20779816508293153,
      "num_tokens": 78596573.0,
      "step": 42155
    },
    {
      "entropy": 6.05093150138855,
      "epoch": 3.7230660543977394,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036646971212189354,
      "loss": 5.3177,
      "mean_token_accuracy": 0.2196388140320778,
      "num_tokens": 78605682.0,
      "step": 42160
    },
    {
      "entropy": 6.047093343734741,
      "epoch": 3.72350759448958,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036644093943600457,
      "loss": 5.3261,
      "mean_token_accuracy": 0.21578431278467178,
      "num_tokens": 78615399.0,
      "step": 42165
    },
    {
      "entropy": 6.044371557235718,
      "epoch": 3.72394913458142,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003664121649588396,
      "loss": 5.2775,
      "mean_token_accuracy": 0.20848629921674727,
      "num_tokens": 78625030.0,
      "step": 42170
    },
    {
      "entropy": 6.039622259140015,
      "epoch": 3.7243906746732605,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036638338869096236,
      "loss": 5.3177,
      "mean_token_accuracy": 0.21266670823097228,
      "num_tokens": 78634492.0,
      "step": 42175
    },
    {
      "entropy": 5.996959686279297,
      "epoch": 3.7248322147651005,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003663546106329364,
      "loss": 5.2632,
      "mean_token_accuracy": 0.2207649603486061,
      "num_tokens": 78643597.0,
      "step": 42180
    },
    {
      "entropy": 5.978757905960083,
      "epoch": 3.725273754856941,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036632583078532565,
      "loss": 5.25,
      "mean_token_accuracy": 0.21145643889904023,
      "num_tokens": 78654083.0,
      "step": 42185
    },
    {
      "entropy": 6.019081020355225,
      "epoch": 3.7257152949487815,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003662970491486938,
      "loss": 5.2766,
      "mean_token_accuracy": 0.2132173255085945,
      "num_tokens": 78663465.0,
      "step": 42190
    },
    {
      "entropy": 6.021770334243774,
      "epoch": 3.7261568350406216,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003662682657236047,
      "loss": 5.2633,
      "mean_token_accuracy": 0.2178608074784279,
      "num_tokens": 78673048.0,
      "step": 42195
    },
    {
      "entropy": 6.055482053756714,
      "epoch": 3.726598375132462,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003662394805106221,
      "loss": 5.2671,
      "mean_token_accuracy": 0.22020864933729173,
      "num_tokens": 78681989.0,
      "step": 42200
    },
    {
      "entropy": 6.053921461105347,
      "epoch": 3.727039915224302,
      "grad_norm": 1.3125,
      "learning_rate": 0.00036621069351031,
      "loss": 5.3111,
      "mean_token_accuracy": 0.20579284131526948,
      "num_tokens": 78691346.0,
      "step": 42205
    },
    {
      "entropy": 6.039604234695434,
      "epoch": 3.7274814553161426,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003661819047232323,
      "loss": 5.3293,
      "mean_token_accuracy": 0.21260005682706834,
      "num_tokens": 78701357.0,
      "step": 42210
    },
    {
      "entropy": 6.024290132522583,
      "epoch": 3.727922995407983,
      "grad_norm": 1.3125,
      "learning_rate": 0.000366153114149953,
      "loss": 5.2811,
      "mean_token_accuracy": 0.21190930753946305,
      "num_tokens": 78710364.0,
      "step": 42215
    },
    {
      "entropy": 6.026815414428711,
      "epoch": 3.728364535499823,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036612432179103594,
      "loss": 5.3011,
      "mean_token_accuracy": 0.21598343104124068,
      "num_tokens": 78719982.0,
      "step": 42220
    },
    {
      "entropy": 6.0415936470031735,
      "epoch": 3.7288060755916637,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003660955276470453,
      "loss": 5.2686,
      "mean_token_accuracy": 0.21843071281909943,
      "num_tokens": 78727925.0,
      "step": 42225
    },
    {
      "entropy": 6.042771196365356,
      "epoch": 3.729247615683504,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036606673171854504,
      "loss": 5.2841,
      "mean_token_accuracy": 0.2185239389538765,
      "num_tokens": 78737528.0,
      "step": 42230
    },
    {
      "entropy": 5.968287515640259,
      "epoch": 3.729689155775344,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003660379340060993,
      "loss": 5.2148,
      "mean_token_accuracy": 0.22543177157640457,
      "num_tokens": 78746453.0,
      "step": 42235
    },
    {
      "entropy": 6.060806798934936,
      "epoch": 3.7301306958671847,
      "grad_norm": 1.375,
      "learning_rate": 0.0003660091345102723,
      "loss": 5.3208,
      "mean_token_accuracy": 0.21217268258333205,
      "num_tokens": 78755768.0,
      "step": 42240
    },
    {
      "entropy": 5.949831104278564,
      "epoch": 3.730572235959025,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000365980333231628,
      "loss": 5.2177,
      "mean_token_accuracy": 0.2209671586751938,
      "num_tokens": 78764688.0,
      "step": 42245
    },
    {
      "entropy": 5.987271881103515,
      "epoch": 3.7310137760508653,
      "grad_norm": 1.34375,
      "learning_rate": 0.00036595153017073077,
      "loss": 5.2406,
      "mean_token_accuracy": 0.21964129209518432,
      "num_tokens": 78773690.0,
      "step": 42250
    },
    {
      "entropy": 6.040000247955322,
      "epoch": 3.7314553161427058,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003659227253281448,
      "loss": 5.2721,
      "mean_token_accuracy": 0.21928405463695527,
      "num_tokens": 78783136.0,
      "step": 42255
    },
    {
      "entropy": 6.049388122558594,
      "epoch": 3.7318968562345463,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00036589391870443436,
      "loss": 5.3734,
      "mean_token_accuracy": 0.20236010998487472,
      "num_tokens": 78792618.0,
      "step": 42260
    },
    {
      "entropy": 6.071663522720337,
      "epoch": 3.7323383963263863,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003658651103001637,
      "loss": 5.3394,
      "mean_token_accuracy": 0.21233391463756562,
      "num_tokens": 78801629.0,
      "step": 42265
    },
    {
      "entropy": 6.053658676147461,
      "epoch": 3.732779936418227,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036583630011589717,
      "loss": 5.2441,
      "mean_token_accuracy": 0.2120149627327919,
      "num_tokens": 78810525.0,
      "step": 42270
    },
    {
      "entropy": 6.015283823013306,
      "epoch": 3.7332214765100673,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003658074881521993,
      "loss": 5.2354,
      "mean_token_accuracy": 0.21788023561239242,
      "num_tokens": 78818999.0,
      "step": 42275
    },
    {
      "entropy": 5.979992818832398,
      "epoch": 3.7336630166019074,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036577867440963424,
      "loss": 5.2209,
      "mean_token_accuracy": 0.22544351667165757,
      "num_tokens": 78828550.0,
      "step": 42280
    },
    {
      "entropy": 6.0741105556488035,
      "epoch": 3.734104556693748,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003657498588887667,
      "loss": 5.2962,
      "mean_token_accuracy": 0.21168177872896193,
      "num_tokens": 78837430.0,
      "step": 42285
    },
    {
      "entropy": 5.948882341384888,
      "epoch": 3.7345460967855884,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00036572104159016095,
      "loss": 5.2144,
      "mean_token_accuracy": 0.2175568386912346,
      "num_tokens": 78845969.0,
      "step": 42290
    },
    {
      "entropy": 6.031601047515869,
      "epoch": 3.7349876368774284,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003656922225143817,
      "loss": 5.2412,
      "mean_token_accuracy": 0.22092944681644439,
      "num_tokens": 78854492.0,
      "step": 42295
    },
    {
      "entropy": 6.0291985988616945,
      "epoch": 3.735429176969269,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003656634016619933,
      "loss": 5.2455,
      "mean_token_accuracy": 0.22695847004652023,
      "num_tokens": 78863554.0,
      "step": 42300
    },
    {
      "entropy": 6.053267192840576,
      "epoch": 3.7358707170611094,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003656345790335605,
      "loss": 5.3051,
      "mean_token_accuracy": 0.21121251285076142,
      "num_tokens": 78871833.0,
      "step": 42305
    },
    {
      "entropy": 5.997975301742554,
      "epoch": 3.7363122571529495,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003656057546296478,
      "loss": 5.1855,
      "mean_token_accuracy": 0.22190462797880173,
      "num_tokens": 78882432.0,
      "step": 42310
    },
    {
      "entropy": 6.0608323097229,
      "epoch": 3.73675379724479,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003655769284508199,
      "loss": 5.2542,
      "mean_token_accuracy": 0.21941542327404023,
      "num_tokens": 78891317.0,
      "step": 42315
    },
    {
      "entropy": 5.968648958206177,
      "epoch": 3.73719533733663,
      "grad_norm": 1.46875,
      "learning_rate": 0.00036554810049764145,
      "loss": 5.2521,
      "mean_token_accuracy": 0.21753930002450944,
      "num_tokens": 78901168.0,
      "step": 42320
    },
    {
      "entropy": 6.01164026260376,
      "epoch": 3.7376368774284705,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003655192707706772,
      "loss": 5.2827,
      "mean_token_accuracy": 0.21752189993858337,
      "num_tokens": 78911816.0,
      "step": 42325
    },
    {
      "entropy": 6.098807764053345,
      "epoch": 3.738078417520311,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000365490439270492,
      "loss": 5.3298,
      "mean_token_accuracy": 0.21556907445192336,
      "num_tokens": 78921794.0,
      "step": 42330
    },
    {
      "entropy": 6.09018087387085,
      "epoch": 3.738519957612151,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003654616059976505,
      "loss": 5.284,
      "mean_token_accuracy": 0.21288902163505555,
      "num_tokens": 78931776.0,
      "step": 42335
    },
    {
      "entropy": 6.0239299774169925,
      "epoch": 3.7389614977039916,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003654327709527176,
      "loss": 5.2234,
      "mean_token_accuracy": 0.21966502964496612,
      "num_tokens": 78940944.0,
      "step": 42340
    },
    {
      "entropy": 6.028392744064331,
      "epoch": 3.7394030377958316,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036540393413625813,
      "loss": 5.2798,
      "mean_token_accuracy": 0.2190561607480049,
      "num_tokens": 78950666.0,
      "step": 42345
    },
    {
      "entropy": 5.973885917663575,
      "epoch": 3.739844577887672,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036537509554883705,
      "loss": 5.2686,
      "mean_token_accuracy": 0.21499705612659453,
      "num_tokens": 78960888.0,
      "step": 42350
    },
    {
      "entropy": 5.997585201263428,
      "epoch": 3.7402861179795126,
      "grad_norm": 1.25,
      "learning_rate": 0.00036534625519101915,
      "loss": 5.2291,
      "mean_token_accuracy": 0.22346763163805008,
      "num_tokens": 78970023.0,
      "step": 42355
    },
    {
      "entropy": 6.006108951568604,
      "epoch": 3.7407276580713527,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003653174130633696,
      "loss": 5.2466,
      "mean_token_accuracy": 0.2205234184861183,
      "num_tokens": 78979132.0,
      "step": 42360
    },
    {
      "entropy": 5.9957257270812985,
      "epoch": 3.741169198163193,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00036528856916645324,
      "loss": 5.1879,
      "mean_token_accuracy": 0.22474960386753082,
      "num_tokens": 78988183.0,
      "step": 42365
    },
    {
      "entropy": 6.008842134475708,
      "epoch": 3.7416107382550337,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036525972350083525,
      "loss": 5.1922,
      "mean_token_accuracy": 0.22721321433782576,
      "num_tokens": 78996867.0,
      "step": 42370
    },
    {
      "entropy": 5.885774850845337,
      "epoch": 3.7420522783468737,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003652308760670805,
      "loss": 5.1753,
      "mean_token_accuracy": 0.22629380226135254,
      "num_tokens": 79006326.0,
      "step": 42375
    },
    {
      "entropy": 6.013526439666748,
      "epoch": 3.742493818438714,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036520202686575427,
      "loss": 5.2507,
      "mean_token_accuracy": 0.22315470576286317,
      "num_tokens": 79015300.0,
      "step": 42380
    },
    {
      "entropy": 6.072935152053833,
      "epoch": 3.7429353585305547,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003651731758974216,
      "loss": 5.3324,
      "mean_token_accuracy": 0.21027711033821106,
      "num_tokens": 79024881.0,
      "step": 42385
    },
    {
      "entropy": 6.046615552902222,
      "epoch": 3.7433768986223948,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036514432316264766,
      "loss": 5.273,
      "mean_token_accuracy": 0.21336543560028076,
      "num_tokens": 79034840.0,
      "step": 42390
    },
    {
      "entropy": 6.032408952713013,
      "epoch": 3.7438184387142353,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003651154686619977,
      "loss": 5.3246,
      "mean_token_accuracy": 0.21609075516462325,
      "num_tokens": 79043537.0,
      "step": 42395
    },
    {
      "entropy": 5.916296672821045,
      "epoch": 3.7442599788060758,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000365086612396037,
      "loss": 5.2397,
      "mean_token_accuracy": 0.220825432240963,
      "num_tokens": 79052868.0,
      "step": 42400
    },
    {
      "entropy": 6.077965831756591,
      "epoch": 3.744701518897916,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036505775436533085,
      "loss": 5.3838,
      "mean_token_accuracy": 0.20482812225818633,
      "num_tokens": 79062491.0,
      "step": 42405
    },
    {
      "entropy": 6.050261926651001,
      "epoch": 3.7451430589897563,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036502889457044446,
      "loss": 5.2763,
      "mean_token_accuracy": 0.22007493674755096,
      "num_tokens": 79071511.0,
      "step": 42410
    },
    {
      "entropy": 6.055062532424927,
      "epoch": 3.745584599081597,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003650000330119433,
      "loss": 5.3615,
      "mean_token_accuracy": 0.20482504665851592,
      "num_tokens": 79080956.0,
      "step": 42415
    },
    {
      "entropy": 6.042124223709107,
      "epoch": 3.746026139173437,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003649711696903925,
      "loss": 5.245,
      "mean_token_accuracy": 0.22252153605222702,
      "num_tokens": 79089751.0,
      "step": 42420
    },
    {
      "entropy": 6.00844988822937,
      "epoch": 3.7464676792652774,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036494230460635787,
      "loss": 5.2712,
      "mean_token_accuracy": 0.2145080581307411,
      "num_tokens": 79098586.0,
      "step": 42425
    },
    {
      "entropy": 6.058753442764282,
      "epoch": 3.746909219357118,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003649134377604045,
      "loss": 5.3725,
      "mean_token_accuracy": 0.21298624128103255,
      "num_tokens": 79108171.0,
      "step": 42430
    },
    {
      "entropy": 6.053863525390625,
      "epoch": 3.747350759448958,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003648845691530981,
      "loss": 5.3134,
      "mean_token_accuracy": 0.21144293993711472,
      "num_tokens": 79117659.0,
      "step": 42435
    },
    {
      "entropy": 6.066207361221314,
      "epoch": 3.7477922995407984,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036485569878500407,
      "loss": 5.2962,
      "mean_token_accuracy": 0.21808723658323287,
      "num_tokens": 79127159.0,
      "step": 42440
    },
    {
      "entropy": 5.973908138275147,
      "epoch": 3.748233839632639,
      "grad_norm": 1.171875,
      "learning_rate": 0.000364826826656688,
      "loss": 5.1909,
      "mean_token_accuracy": 0.22485121637582778,
      "num_tokens": 79136641.0,
      "step": 42445
    },
    {
      "entropy": 6.05207405090332,
      "epoch": 3.748675379724479,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036479795276871547,
      "loss": 5.3195,
      "mean_token_accuracy": 0.22154427766799928,
      "num_tokens": 79145746.0,
      "step": 42450
    },
    {
      "entropy": 5.91261100769043,
      "epoch": 3.7491169198163194,
      "grad_norm": 1.46875,
      "learning_rate": 0.00036476907712165214,
      "loss": 5.1642,
      "mean_token_accuracy": 0.21921120285987855,
      "num_tokens": 79154880.0,
      "step": 42455
    },
    {
      "entropy": 5.982310438156128,
      "epoch": 3.7495584599081595,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003647401997160636,
      "loss": 5.243,
      "mean_token_accuracy": 0.22259870618581773,
      "num_tokens": 79164428.0,
      "step": 42460
    },
    {
      "entropy": 5.94856595993042,
      "epoch": 3.75,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003647113205525156,
      "loss": 5.1216,
      "mean_token_accuracy": 0.22769768089056014,
      "num_tokens": 79173367.0,
      "step": 42465
    },
    {
      "entropy": 5.9379270553588865,
      "epoch": 3.7504415400918405,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036468243963157395,
      "loss": 5.2133,
      "mean_token_accuracy": 0.21733782291412354,
      "num_tokens": 79182383.0,
      "step": 42470
    },
    {
      "entropy": 5.921837902069091,
      "epoch": 3.7508830801836806,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003646535569538042,
      "loss": 5.1914,
      "mean_token_accuracy": 0.22284788638353348,
      "num_tokens": 79191403.0,
      "step": 42475
    },
    {
      "entropy": 6.025750970840454,
      "epoch": 3.751324620275521,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036462467251977226,
      "loss": 5.3043,
      "mean_token_accuracy": 0.2061627298593521,
      "num_tokens": 79201055.0,
      "step": 42480
    },
    {
      "entropy": 6.063861608505249,
      "epoch": 3.751766160367361,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036459578633004405,
      "loss": 5.3121,
      "mean_token_accuracy": 0.21651774495840073,
      "num_tokens": 79209596.0,
      "step": 42485
    },
    {
      "entropy": 6.059188890457153,
      "epoch": 3.7522077004592016,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003645668983851853,
      "loss": 5.2311,
      "mean_token_accuracy": 0.2210493192076683,
      "num_tokens": 79218982.0,
      "step": 42490
    },
    {
      "entropy": 6.061312818527222,
      "epoch": 3.752649240551042,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036453800868576193,
      "loss": 5.3189,
      "mean_token_accuracy": 0.202720744907856,
      "num_tokens": 79228872.0,
      "step": 42495
    },
    {
      "entropy": 5.9811207294464115,
      "epoch": 3.753090780642882,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036450911723233994,
      "loss": 5.2106,
      "mean_token_accuracy": 0.22361805737018586,
      "num_tokens": 79238469.0,
      "step": 42500
    },
    {
      "entropy": 5.940560293197632,
      "epoch": 3.7535323207347226,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003644802240254852,
      "loss": 5.2072,
      "mean_token_accuracy": 0.22580990940332413,
      "num_tokens": 79247194.0,
      "step": 42505
    },
    {
      "entropy": 6.0248627185821535,
      "epoch": 3.753973860826563,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036445132906576374,
      "loss": 5.3366,
      "mean_token_accuracy": 0.2142035946249962,
      "num_tokens": 79256925.0,
      "step": 42510
    },
    {
      "entropy": 6.0263772964477536,
      "epoch": 3.754415400918403,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003644224323537416,
      "loss": 5.2867,
      "mean_token_accuracy": 0.21112608313560485,
      "num_tokens": 79266530.0,
      "step": 42515
    },
    {
      "entropy": 5.984225463867188,
      "epoch": 3.7548569410102437,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003643935338899849,
      "loss": 5.251,
      "mean_token_accuracy": 0.2219371095299721,
      "num_tokens": 79275328.0,
      "step": 42520
    },
    {
      "entropy": 6.007853651046753,
      "epoch": 3.755298481102084,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003643646336750598,
      "loss": 5.2127,
      "mean_token_accuracy": 0.2227490484714508,
      "num_tokens": 79285084.0,
      "step": 42525
    },
    {
      "entropy": 5.999251747131348,
      "epoch": 3.7557400211939242,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036433573170953224,
      "loss": 5.2001,
      "mean_token_accuracy": 0.22026511728763581,
      "num_tokens": 79294595.0,
      "step": 42530
    },
    {
      "entropy": 6.002381992340088,
      "epoch": 3.7561815612857647,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003643068279939686,
      "loss": 5.311,
      "mean_token_accuracy": 0.21393502801656722,
      "num_tokens": 79304198.0,
      "step": 42535
    },
    {
      "entropy": 6.032972097396851,
      "epoch": 3.7566231013776052,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003642779225289349,
      "loss": 5.2739,
      "mean_token_accuracy": 0.2146414890885353,
      "num_tokens": 79313528.0,
      "step": 42540
    },
    {
      "entropy": 6.076814317703247,
      "epoch": 3.7570646414694453,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036424901531499756,
      "loss": 5.3189,
      "mean_token_accuracy": 0.21336955726146697,
      "num_tokens": 79323039.0,
      "step": 42545
    },
    {
      "entropy": 6.099421072006225,
      "epoch": 3.757506181561286,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036422010635272264,
      "loss": 5.3953,
      "mean_token_accuracy": 0.21163086146116256,
      "num_tokens": 79332564.0,
      "step": 42550
    },
    {
      "entropy": 5.9722192764282225,
      "epoch": 3.7579477216531263,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003641911956426767,
      "loss": 5.2631,
      "mean_token_accuracy": 0.2199915960431099,
      "num_tokens": 79342206.0,
      "step": 42555
    },
    {
      "entropy": 5.939135503768921,
      "epoch": 3.7583892617449663,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036416228318542586,
      "loss": 5.1386,
      "mean_token_accuracy": 0.23180091828107835,
      "num_tokens": 79350992.0,
      "step": 42560
    },
    {
      "entropy": 6.056774854660034,
      "epoch": 3.758830801836807,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036413336898153667,
      "loss": 5.3142,
      "mean_token_accuracy": 0.21173805445432664,
      "num_tokens": 79360284.0,
      "step": 42565
    },
    {
      "entropy": 5.98705677986145,
      "epoch": 3.7592723419286473,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036410445303157555,
      "loss": 5.2329,
      "mean_token_accuracy": 0.22108222246170045,
      "num_tokens": 79370404.0,
      "step": 42570
    },
    {
      "entropy": 5.9929403305053714,
      "epoch": 3.7597138820204874,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003640755353361087,
      "loss": 5.3497,
      "mean_token_accuracy": 0.21552008539438247,
      "num_tokens": 79379031.0,
      "step": 42575
    },
    {
      "entropy": 6.078564882278442,
      "epoch": 3.760155422112328,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036404661589570285,
      "loss": 5.3339,
      "mean_token_accuracy": 0.21047959625720977,
      "num_tokens": 79388405.0,
      "step": 42580
    },
    {
      "entropy": 6.050285482406617,
      "epoch": 3.7605969622041684,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036401769471092443,
      "loss": 5.3278,
      "mean_token_accuracy": 0.21293817162513734,
      "num_tokens": 79398497.0,
      "step": 42585
    },
    {
      "entropy": 6.002913904190064,
      "epoch": 3.7610385022960084,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036398877178234,
      "loss": 5.2416,
      "mean_token_accuracy": 0.22033692747354508,
      "num_tokens": 79408579.0,
      "step": 42590
    },
    {
      "entropy": 6.005420398712158,
      "epoch": 3.761480042387849,
      "grad_norm": 1.25,
      "learning_rate": 0.00036395984711051624,
      "loss": 5.2921,
      "mean_token_accuracy": 0.20948113054037093,
      "num_tokens": 79418038.0,
      "step": 42595
    },
    {
      "entropy": 6.0122535705566404,
      "epoch": 3.761921582479689,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003639309206960195,
      "loss": 5.2796,
      "mean_token_accuracy": 0.21450708210468292,
      "num_tokens": 79427428.0,
      "step": 42600
    },
    {
      "entropy": 6.072609996795654,
      "epoch": 3.7623631225715295,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003639019925394167,
      "loss": 5.3271,
      "mean_token_accuracy": 0.20510325580835342,
      "num_tokens": 79437006.0,
      "step": 42605
    },
    {
      "entropy": 5.982869863510132,
      "epoch": 3.76280466266337,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036387306264127446,
      "loss": 5.2134,
      "mean_token_accuracy": 0.22605197131633759,
      "num_tokens": 79447284.0,
      "step": 42610
    },
    {
      "entropy": 6.0553563117980955,
      "epoch": 3.76324620275521,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036384413100215936,
      "loss": 5.2639,
      "mean_token_accuracy": 0.219516721367836,
      "num_tokens": 79457024.0,
      "step": 42615
    },
    {
      "entropy": 5.991971969604492,
      "epoch": 3.7636877428470505,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003638151976226384,
      "loss": 5.2326,
      "mean_token_accuracy": 0.21878402531147004,
      "num_tokens": 79465664.0,
      "step": 42620
    },
    {
      "entropy": 5.962288475036621,
      "epoch": 3.7641292829388906,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003637862625032781,
      "loss": 5.195,
      "mean_token_accuracy": 0.22510105222463608,
      "num_tokens": 79475110.0,
      "step": 42625
    },
    {
      "entropy": 6.069219779968262,
      "epoch": 3.764570823030731,
      "grad_norm": 1.125,
      "learning_rate": 0.00036375732564464545,
      "loss": 5.3241,
      "mean_token_accuracy": 0.20596041083335875,
      "num_tokens": 79485151.0,
      "step": 42630
    },
    {
      "entropy": 6.061473655700683,
      "epoch": 3.7650123631225716,
      "grad_norm": 1.390625,
      "learning_rate": 0.00036372838704730736,
      "loss": 5.332,
      "mean_token_accuracy": 0.21082351356744766,
      "num_tokens": 79494751.0,
      "step": 42635
    },
    {
      "entropy": 6.047118234634399,
      "epoch": 3.7654539032144116,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003636994467118305,
      "loss": 5.2448,
      "mean_token_accuracy": 0.22332147508859634,
      "num_tokens": 79503703.0,
      "step": 42640
    },
    {
      "entropy": 5.909613418579101,
      "epoch": 3.765895443306252,
      "grad_norm": 1.265625,
      "learning_rate": 0.000363670504638782,
      "loss": 5.1736,
      "mean_token_accuracy": 0.22443289458751678,
      "num_tokens": 79513665.0,
      "step": 42645
    },
    {
      "entropy": 5.999354028701783,
      "epoch": 3.7663369833980926,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00036364156082872867,
      "loss": 5.1261,
      "mean_token_accuracy": 0.2309316262602806,
      "num_tokens": 79522432.0,
      "step": 42650
    },
    {
      "entropy": 6.090775060653686,
      "epoch": 3.7667785234899327,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003636126152822377,
      "loss": 5.343,
      "mean_token_accuracy": 0.21793788224458693,
      "num_tokens": 79531561.0,
      "step": 42655
    },
    {
      "entropy": 6.063354921340943,
      "epoch": 3.767220063581773,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00036358366799987585,
      "loss": 5.4103,
      "mean_token_accuracy": 0.20587964206933976,
      "num_tokens": 79540299.0,
      "step": 42660
    },
    {
      "entropy": 6.00131893157959,
      "epoch": 3.7676616036736137,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036355471898221045,
      "loss": 5.304,
      "mean_token_accuracy": 0.20842873454093933,
      "num_tokens": 79550171.0,
      "step": 42665
    },
    {
      "entropy": 6.008993434906006,
      "epoch": 3.7681031437654537,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003635257682298084,
      "loss": 5.2047,
      "mean_token_accuracy": 0.21647921800613404,
      "num_tokens": 79559806.0,
      "step": 42670
    },
    {
      "entropy": 6.007473468780518,
      "epoch": 3.7685446838572942,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003634968157432369,
      "loss": 5.2851,
      "mean_token_accuracy": 0.21302532404661179,
      "num_tokens": 79568986.0,
      "step": 42675
    },
    {
      "entropy": 5.974026536941528,
      "epoch": 3.7689862239491347,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036346786152306314,
      "loss": 5.2298,
      "mean_token_accuracy": 0.22233688533306123,
      "num_tokens": 79577962.0,
      "step": 42680
    },
    {
      "entropy": 5.967868375778198,
      "epoch": 3.769427764040975,
      "grad_norm": 1.40625,
      "learning_rate": 0.00036343890556985424,
      "loss": 5.2555,
      "mean_token_accuracy": 0.22147727012634277,
      "num_tokens": 79587043.0,
      "step": 42685
    },
    {
      "entropy": 6.040402603149414,
      "epoch": 3.7698693041328153,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036340994788417754,
      "loss": 5.3463,
      "mean_token_accuracy": 0.20593555122613907,
      "num_tokens": 79596812.0,
      "step": 42690
    },
    {
      "entropy": 6.045345878601074,
      "epoch": 3.770310844224656,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036338098846660025,
      "loss": 5.3217,
      "mean_token_accuracy": 0.21011202782392502,
      "num_tokens": 79606058.0,
      "step": 42695
    },
    {
      "entropy": 6.007207727432251,
      "epoch": 3.770752384316496,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036335202731768956,
      "loss": 5.2223,
      "mean_token_accuracy": 0.21821150928735733,
      "num_tokens": 79616118.0,
      "step": 42700
    },
    {
      "entropy": 6.0629534244537355,
      "epoch": 3.7711939244083363,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000363323064438013,
      "loss": 5.2592,
      "mean_token_accuracy": 0.21659672558307647,
      "num_tokens": 79626223.0,
      "step": 42705
    },
    {
      "entropy": 6.059642028808594,
      "epoch": 3.771635464500177,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003632940998281378,
      "loss": 5.3455,
      "mean_token_accuracy": 0.21521352231502533,
      "num_tokens": 79634692.0,
      "step": 42710
    },
    {
      "entropy": 6.026362466812134,
      "epoch": 3.772077004592017,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003632651334886314,
      "loss": 5.263,
      "mean_token_accuracy": 0.21268602460622787,
      "num_tokens": 79644323.0,
      "step": 42715
    },
    {
      "entropy": 6.028447914123535,
      "epoch": 3.7725185446838574,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036323616542006124,
      "loss": 5.3218,
      "mean_token_accuracy": 0.2152400031685829,
      "num_tokens": 79653802.0,
      "step": 42720
    },
    {
      "entropy": 5.99633936882019,
      "epoch": 3.772960084775698,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036320719562299474,
      "loss": 5.2442,
      "mean_token_accuracy": 0.2146581992506981,
      "num_tokens": 79663206.0,
      "step": 42725
    },
    {
      "entropy": 6.044142484664917,
      "epoch": 3.773401624867538,
      "grad_norm": 1.25,
      "learning_rate": 0.0003631782240979995,
      "loss": 5.311,
      "mean_token_accuracy": 0.21213002651929855,
      "num_tokens": 79673026.0,
      "step": 42730
    },
    {
      "entropy": 6.048744535446167,
      "epoch": 3.7738431649593784,
      "grad_norm": 1.21875,
      "learning_rate": 0.000363149250845643,
      "loss": 5.2984,
      "mean_token_accuracy": 0.2114138960838318,
      "num_tokens": 79682504.0,
      "step": 42735
    },
    {
      "entropy": 6.030384588241577,
      "epoch": 3.7742847050512185,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003631202758664928,
      "loss": 5.2813,
      "mean_token_accuracy": 0.22007232308387756,
      "num_tokens": 79692443.0,
      "step": 42740
    },
    {
      "entropy": 6.015874338150025,
      "epoch": 3.774726245143059,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036309129916111657,
      "loss": 5.2861,
      "mean_token_accuracy": 0.21749907433986665,
      "num_tokens": 79701820.0,
      "step": 42745
    },
    {
      "entropy": 6.002307319641114,
      "epoch": 3.7751677852348995,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00036306232073008175,
      "loss": 5.2884,
      "mean_token_accuracy": 0.21766636669635772,
      "num_tokens": 79710176.0,
      "step": 42750
    },
    {
      "entropy": 6.0123796463012695,
      "epoch": 3.7756093253267395,
      "grad_norm": 1.3125,
      "learning_rate": 0.00036303334057395626,
      "loss": 5.2598,
      "mean_token_accuracy": 0.22078149169683456,
      "num_tokens": 79719886.0,
      "step": 42755
    },
    {
      "entropy": 6.0587201595306395,
      "epoch": 3.77605086541858,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003630043586933077,
      "loss": 5.2989,
      "mean_token_accuracy": 0.21591682136058807,
      "num_tokens": 79729432.0,
      "step": 42760
    },
    {
      "entropy": 6.015498208999634,
      "epoch": 3.77649240551042,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003629753750887037,
      "loss": 5.2684,
      "mean_token_accuracy": 0.21499119699001312,
      "num_tokens": 79738832.0,
      "step": 42765
    },
    {
      "entropy": 5.970015668869019,
      "epoch": 3.7769339456022606,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003629463897607122,
      "loss": 5.2347,
      "mean_token_accuracy": 0.23146018534898757,
      "num_tokens": 79748554.0,
      "step": 42770
    },
    {
      "entropy": 6.085123777389526,
      "epoch": 3.777375485694101,
      "grad_norm": 1.296875,
      "learning_rate": 0.000362917402709901,
      "loss": 5.3631,
      "mean_token_accuracy": 0.20595103055238723,
      "num_tokens": 79757263.0,
      "step": 42775
    },
    {
      "entropy": 6.0661060333251955,
      "epoch": 3.777817025785941,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036288841393683773,
      "loss": 5.3668,
      "mean_token_accuracy": 0.21098582446575165,
      "num_tokens": 79765909.0,
      "step": 42780
    },
    {
      "entropy": 5.994005727767944,
      "epoch": 3.7782585658777816,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036285942344209057,
      "loss": 5.2293,
      "mean_token_accuracy": 0.21113922894001008,
      "num_tokens": 79775128.0,
      "step": 42785
    },
    {
      "entropy": 6.05211992263794,
      "epoch": 3.778700105969622,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036283043122622717,
      "loss": 5.3253,
      "mean_token_accuracy": 0.2128269299864769,
      "num_tokens": 79784921.0,
      "step": 42790
    },
    {
      "entropy": 6.073528194427491,
      "epoch": 3.779141646061462,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003628014372898157,
      "loss": 5.3209,
      "mean_token_accuracy": 0.2135973334312439,
      "num_tokens": 79793644.0,
      "step": 42795
    },
    {
      "entropy": 6.026942729949951,
      "epoch": 3.7795831861533027,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003627724416334239,
      "loss": 5.2732,
      "mean_token_accuracy": 0.22303889840841293,
      "num_tokens": 79803131.0,
      "step": 42800
    },
    {
      "entropy": 6.088052988052368,
      "epoch": 3.780024726245143,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003627434442576199,
      "loss": 5.343,
      "mean_token_accuracy": 0.20822370648384095,
      "num_tokens": 79812322.0,
      "step": 42805
    },
    {
      "entropy": 6.017120122909546,
      "epoch": 3.7804662663369832,
      "grad_norm": 1.25,
      "learning_rate": 0.0003627144451629718,
      "loss": 5.2612,
      "mean_token_accuracy": 0.21562101393938066,
      "num_tokens": 79821634.0,
      "step": 42810
    },
    {
      "entropy": 6.078363275527954,
      "epoch": 3.7809078064288237,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00036268544435004755,
      "loss": 5.2977,
      "mean_token_accuracy": 0.22090030014514922,
      "num_tokens": 79830229.0,
      "step": 42815
    },
    {
      "entropy": 6.028746032714844,
      "epoch": 3.7813493465206642,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003626564418194153,
      "loss": 5.2193,
      "mean_token_accuracy": 0.21773789823055267,
      "num_tokens": 79839678.0,
      "step": 42820
    },
    {
      "entropy": 5.9603461742401125,
      "epoch": 3.7817908866125043,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036262743757164317,
      "loss": 5.2144,
      "mean_token_accuracy": 0.2206832319498062,
      "num_tokens": 79848743.0,
      "step": 42825
    },
    {
      "entropy": 6.080780982971191,
      "epoch": 3.782232426704345,
      "grad_norm": 1.375,
      "learning_rate": 0.00036259843160729943,
      "loss": 5.4299,
      "mean_token_accuracy": 0.20534070134162902,
      "num_tokens": 79858131.0,
      "step": 42830
    },
    {
      "entropy": 6.048813915252685,
      "epoch": 3.7826739667961853,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003625694239269522,
      "loss": 5.2947,
      "mean_token_accuracy": 0.21107049733400346,
      "num_tokens": 79867027.0,
      "step": 42835
    },
    {
      "entropy": 6.009788084030151,
      "epoch": 3.7831155068880253,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036254041453116987,
      "loss": 5.195,
      "mean_token_accuracy": 0.22012798637151718,
      "num_tokens": 79875545.0,
      "step": 42840
    },
    {
      "entropy": 6.045275354385376,
      "epoch": 3.783557046979866,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036251140342052045,
      "loss": 5.2951,
      "mean_token_accuracy": 0.2151389941573143,
      "num_tokens": 79884907.0,
      "step": 42845
    },
    {
      "entropy": 5.974017143249512,
      "epoch": 3.7839985870717063,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003624823905955726,
      "loss": 5.1722,
      "mean_token_accuracy": 0.22549605816602708,
      "num_tokens": 79893626.0,
      "step": 42850
    },
    {
      "entropy": 6.041332101821899,
      "epoch": 3.7844401271635464,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003624533760568943,
      "loss": 5.2374,
      "mean_token_accuracy": 0.22274807393550872,
      "num_tokens": 79902205.0,
      "step": 42855
    },
    {
      "entropy": 6.0076171398162845,
      "epoch": 3.784881667255387,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036242435980505416,
      "loss": 5.2297,
      "mean_token_accuracy": 0.21624011397361756,
      "num_tokens": 79912346.0,
      "step": 42860
    },
    {
      "entropy": 5.966396379470825,
      "epoch": 3.7853232073472274,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003623953418406205,
      "loss": 5.2693,
      "mean_token_accuracy": 0.2197701156139374,
      "num_tokens": 79921661.0,
      "step": 42865
    },
    {
      "entropy": 5.9423267364501955,
      "epoch": 3.7857647474390674,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036236632216416186,
      "loss": 5.2693,
      "mean_token_accuracy": 0.2268640324473381,
      "num_tokens": 79931004.0,
      "step": 42870
    },
    {
      "entropy": 5.997977304458618,
      "epoch": 3.786206287530908,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00036233730077624664,
      "loss": 5.1917,
      "mean_token_accuracy": 0.22303149253129959,
      "num_tokens": 79939852.0,
      "step": 42875
    },
    {
      "entropy": 6.011513519287109,
      "epoch": 3.786647827622748,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036230827767744336,
      "loss": 5.3327,
      "mean_token_accuracy": 0.21597013473510743,
      "num_tokens": 79949365.0,
      "step": 42880
    },
    {
      "entropy": 5.922567892074585,
      "epoch": 3.7870893677145885,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003622792528683206,
      "loss": 5.1823,
      "mean_token_accuracy": 0.2255784884095192,
      "num_tokens": 79958583.0,
      "step": 42885
    },
    {
      "entropy": 6.024000120162964,
      "epoch": 3.787530907806429,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036225022634944683,
      "loss": 5.2378,
      "mean_token_accuracy": 0.2234050840139389,
      "num_tokens": 79966907.0,
      "step": 42890
    },
    {
      "entropy": 6.046203851699829,
      "epoch": 3.787972447898269,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036222119812139074,
      "loss": 5.4024,
      "mean_token_accuracy": 0.20578263700008392,
      "num_tokens": 79976267.0,
      "step": 42895
    },
    {
      "entropy": 6.017287349700927,
      "epoch": 3.7884139879901095,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00036219216818472104,
      "loss": 5.2557,
      "mean_token_accuracy": 0.2192061349749565,
      "num_tokens": 79986618.0,
      "step": 42900
    },
    {
      "entropy": 5.9912947654724125,
      "epoch": 3.7888555280819496,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003621631365400064,
      "loss": 5.2087,
      "mean_token_accuracy": 0.22247947454452516,
      "num_tokens": 79996292.0,
      "step": 42905
    },
    {
      "entropy": 5.958834648132324,
      "epoch": 3.78929706817379,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003621341031878154,
      "loss": 5.2688,
      "mean_token_accuracy": 0.2235727310180664,
      "num_tokens": 80006128.0,
      "step": 42910
    },
    {
      "entropy": 5.977275276184082,
      "epoch": 3.7897386082656306,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036210506812871696,
      "loss": 5.3022,
      "mean_token_accuracy": 0.2152072712779045,
      "num_tokens": 80015163.0,
      "step": 42915
    },
    {
      "entropy": 6.120819044113159,
      "epoch": 3.7901801483574706,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003620760313632798,
      "loss": 5.3619,
      "mean_token_accuracy": 0.20944087952375412,
      "num_tokens": 80023932.0,
      "step": 42920
    },
    {
      "entropy": 6.055482196807861,
      "epoch": 3.790621688449311,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003620469928920726,
      "loss": 5.31,
      "mean_token_accuracy": 0.20958310067653657,
      "num_tokens": 80033676.0,
      "step": 42925
    },
    {
      "entropy": 6.075841569900513,
      "epoch": 3.7910632285411516,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003620179527156644,
      "loss": 5.3934,
      "mean_token_accuracy": 0.20960612446069718,
      "num_tokens": 80045215.0,
      "step": 42930
    },
    {
      "entropy": 6.013024425506591,
      "epoch": 3.7915047686329917,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003619889108346239,
      "loss": 5.2767,
      "mean_token_accuracy": 0.217362442612648,
      "num_tokens": 80055065.0,
      "step": 42935
    },
    {
      "entropy": 5.967290258407592,
      "epoch": 3.791946308724832,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00036195986724952016,
      "loss": 5.2095,
      "mean_token_accuracy": 0.22437662184238433,
      "num_tokens": 80064230.0,
      "step": 42940
    },
    {
      "entropy": 6.032168579101563,
      "epoch": 3.7923878488166727,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036193082196092207,
      "loss": 5.2818,
      "mean_token_accuracy": 0.21188434809446335,
      "num_tokens": 80074656.0,
      "step": 42945
    },
    {
      "entropy": 5.978481101989746,
      "epoch": 3.7928293889085127,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003619017749693987,
      "loss": 5.2203,
      "mean_token_accuracy": 0.22059565782546997,
      "num_tokens": 80084822.0,
      "step": 42950
    },
    {
      "entropy": 6.002022075653076,
      "epoch": 3.7932709290003532,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003618727262755189,
      "loss": 5.2137,
      "mean_token_accuracy": 0.21875778585672379,
      "num_tokens": 80093646.0,
      "step": 42955
    },
    {
      "entropy": 5.984192991256714,
      "epoch": 3.7937124690921937,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003618436758798518,
      "loss": 5.2469,
      "mean_token_accuracy": 0.21721965372562407,
      "num_tokens": 80103170.0,
      "step": 42960
    },
    {
      "entropy": 5.982781934738159,
      "epoch": 3.794154009184034,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036181462378296643,
      "loss": 5.1776,
      "mean_token_accuracy": 0.21566992700099946,
      "num_tokens": 80113539.0,
      "step": 42965
    },
    {
      "entropy": 5.992352104187011,
      "epoch": 3.7945955492758743,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036178556998543206,
      "loss": 5.2608,
      "mean_token_accuracy": 0.22190568447113038,
      "num_tokens": 80122494.0,
      "step": 42970
    },
    {
      "entropy": 6.014445209503174,
      "epoch": 3.7950370893677148,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00036175651448781766,
      "loss": 5.2323,
      "mean_token_accuracy": 0.22312407791614533,
      "num_tokens": 80130946.0,
      "step": 42975
    },
    {
      "entropy": 6.086071538925171,
      "epoch": 3.795478629459555,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036172745729069256,
      "loss": 5.2785,
      "mean_token_accuracy": 0.21585483402013778,
      "num_tokens": 80140974.0,
      "step": 42980
    },
    {
      "entropy": 6.003747081756591,
      "epoch": 3.7959201695513953,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003616983983946258,
      "loss": 5.2569,
      "mean_token_accuracy": 0.21474365293979644,
      "num_tokens": 80151080.0,
      "step": 42985
    },
    {
      "entropy": 5.967628860473633,
      "epoch": 3.796361709643236,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003616693378001868,
      "loss": 5.2518,
      "mean_token_accuracy": 0.212323297560215,
      "num_tokens": 80159452.0,
      "step": 42990
    },
    {
      "entropy": 6.037431859970093,
      "epoch": 3.796803249735076,
      "grad_norm": 1.375,
      "learning_rate": 0.0003616402755079448,
      "loss": 5.206,
      "mean_token_accuracy": 0.2198104590177536,
      "num_tokens": 80168264.0,
      "step": 42995
    },
    {
      "entropy": 6.071934080123901,
      "epoch": 3.7972447898269164,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036161121151846894,
      "loss": 5.2721,
      "mean_token_accuracy": 0.216286101937294,
      "num_tokens": 80177495.0,
      "step": 43000
    },
    {
      "entropy": 6.023909664154052,
      "epoch": 3.797686329918757,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003615821458323288,
      "loss": 5.3326,
      "mean_token_accuracy": 0.21751774847507477,
      "num_tokens": 80187285.0,
      "step": 43005
    },
    {
      "entropy": 6.0116535186767575,
      "epoch": 3.798127870010597,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003615530784500936,
      "loss": 5.2759,
      "mean_token_accuracy": 0.21989172995090484,
      "num_tokens": 80196612.0,
      "step": 43010
    },
    {
      "entropy": 5.983617687225342,
      "epoch": 3.7985694101024374,
      "grad_norm": 1.359375,
      "learning_rate": 0.00036152400937233294,
      "loss": 5.2398,
      "mean_token_accuracy": 0.22071364820003508,
      "num_tokens": 80205073.0,
      "step": 43015
    },
    {
      "entropy": 5.932385730743408,
      "epoch": 3.7990109501942775,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000361494938599616,
      "loss": 5.2114,
      "mean_token_accuracy": 0.21799306869506835,
      "num_tokens": 80215562.0,
      "step": 43020
    },
    {
      "entropy": 5.9725830078125,
      "epoch": 3.799452490286118,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003614658661325125,
      "loss": 5.172,
      "mean_token_accuracy": 0.2193455755710602,
      "num_tokens": 80225337.0,
      "step": 43025
    },
    {
      "entropy": 6.0085169792175295,
      "epoch": 3.7998940303779585,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003614367919715918,
      "loss": 5.2401,
      "mean_token_accuracy": 0.22063093930482863,
      "num_tokens": 80234865.0,
      "step": 43030
    },
    {
      "entropy": 5.941318941116333,
      "epoch": 3.8003355704697985,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003614077161174235,
      "loss": 5.264,
      "mean_token_accuracy": 0.2170095533132553,
      "num_tokens": 80244026.0,
      "step": 43035
    },
    {
      "entropy": 5.9841207504272464,
      "epoch": 3.800777110561639,
      "grad_norm": 1.578125,
      "learning_rate": 0.00036137863857057726,
      "loss": 5.2859,
      "mean_token_accuracy": 0.21982468217611312,
      "num_tokens": 80252854.0,
      "step": 43040
    },
    {
      "entropy": 6.074405670166016,
      "epoch": 3.801218650653479,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036134955933162253,
      "loss": 5.3078,
      "mean_token_accuracy": 0.217137148976326,
      "num_tokens": 80263257.0,
      "step": 43045
    },
    {
      "entropy": 5.93114333152771,
      "epoch": 3.8016601907453196,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003613204784011292,
      "loss": 5.1577,
      "mean_token_accuracy": 0.23252501487731933,
      "num_tokens": 80271894.0,
      "step": 43050
    },
    {
      "entropy": 6.00022258758545,
      "epoch": 3.80210173083716,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036129139577966666,
      "loss": 5.2176,
      "mean_token_accuracy": 0.22252071499824524,
      "num_tokens": 80281695.0,
      "step": 43055
    },
    {
      "entropy": 6.0376482009887695,
      "epoch": 3.802543270929,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003612623114678047,
      "loss": 5.3085,
      "mean_token_accuracy": 0.21180357038974762,
      "num_tokens": 80291105.0,
      "step": 43060
    },
    {
      "entropy": 6.012930488586425,
      "epoch": 3.8029848110208406,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003612332254661132,
      "loss": 5.2268,
      "mean_token_accuracy": 0.21793885082006453,
      "num_tokens": 80299320.0,
      "step": 43065
    },
    {
      "entropy": 5.97972617149353,
      "epoch": 3.803426351112681,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003612041377751619,
      "loss": 5.262,
      "mean_token_accuracy": 0.21988650411367416,
      "num_tokens": 80307678.0,
      "step": 43070
    },
    {
      "entropy": 5.931871509552002,
      "epoch": 3.803867891204521,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036117504839552043,
      "loss": 5.1767,
      "mean_token_accuracy": 0.22750315815210342,
      "num_tokens": 80317317.0,
      "step": 43075
    },
    {
      "entropy": 6.0700804710388185,
      "epoch": 3.8043094312963617,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003611459573277589,
      "loss": 5.2304,
      "mean_token_accuracy": 0.22140474915504454,
      "num_tokens": 80326562.0,
      "step": 43080
    },
    {
      "entropy": 6.008482885360718,
      "epoch": 3.804750971388202,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036111686457244703,
      "loss": 5.2877,
      "mean_token_accuracy": 0.21238065659999847,
      "num_tokens": 80335546.0,
      "step": 43085
    },
    {
      "entropy": 6.013809394836426,
      "epoch": 3.8051925114800422,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003610877701301547,
      "loss": 5.313,
      "mean_token_accuracy": 0.21693359017372132,
      "num_tokens": 80344773.0,
      "step": 43090
    },
    {
      "entropy": 5.946885347366333,
      "epoch": 3.8056340515718827,
      "grad_norm": 1.25,
      "learning_rate": 0.0003610586740014519,
      "loss": 5.2036,
      "mean_token_accuracy": 0.22217418998479843,
      "num_tokens": 80355036.0,
      "step": 43095
    },
    {
      "entropy": 6.063947057723999,
      "epoch": 3.8060755916637232,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036102957618690875,
      "loss": 5.2989,
      "mean_token_accuracy": 0.2115948900580406,
      "num_tokens": 80364485.0,
      "step": 43100
    },
    {
      "entropy": 6.005980634689331,
      "epoch": 3.8065171317555633,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036100047668709506,
      "loss": 5.2589,
      "mean_token_accuracy": 0.21985725462436675,
      "num_tokens": 80373429.0,
      "step": 43105
    },
    {
      "entropy": 5.953507328033448,
      "epoch": 3.8069586718474038,
      "grad_norm": 1.25,
      "learning_rate": 0.000360971375502581,
      "loss": 5.196,
      "mean_token_accuracy": 0.22026986330747605,
      "num_tokens": 80382951.0,
      "step": 43110
    },
    {
      "entropy": 5.977100801467896,
      "epoch": 3.8074002119392443,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036094227263393653,
      "loss": 5.1846,
      "mean_token_accuracy": 0.22608063519001007,
      "num_tokens": 80391445.0,
      "step": 43115
    },
    {
      "entropy": 6.065096616744995,
      "epoch": 3.8078417520310843,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003609131680817317,
      "loss": 5.329,
      "mean_token_accuracy": 0.21303049325942994,
      "num_tokens": 80401119.0,
      "step": 43120
    },
    {
      "entropy": 6.033899307250977,
      "epoch": 3.808283292122925,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000360884061846537,
      "loss": 5.2708,
      "mean_token_accuracy": 0.2160505935549736,
      "num_tokens": 80410014.0,
      "step": 43125
    },
    {
      "entropy": 6.070316886901855,
      "epoch": 3.8087248322147653,
      "grad_norm": 1.25,
      "learning_rate": 0.0003608549539289222,
      "loss": 5.2684,
      "mean_token_accuracy": 0.2173441842198372,
      "num_tokens": 80419432.0,
      "step": 43130
    },
    {
      "entropy": 6.025496625900269,
      "epoch": 3.8091663723066054,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036082584432945775,
      "loss": 5.2229,
      "mean_token_accuracy": 0.22292453944683074,
      "num_tokens": 80428334.0,
      "step": 43135
    },
    {
      "entropy": 5.96901273727417,
      "epoch": 3.809607912398446,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036079673304871383,
      "loss": 5.2987,
      "mean_token_accuracy": 0.2141331300139427,
      "num_tokens": 80437985.0,
      "step": 43140
    },
    {
      "entropy": 6.103000545501709,
      "epoch": 3.8100494524902864,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036076762008726074,
      "loss": 5.3321,
      "mean_token_accuracy": 0.20559173822402954,
      "num_tokens": 80446952.0,
      "step": 43145
    },
    {
      "entropy": 6.0906177997589115,
      "epoch": 3.8104909925821264,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036073850544566865,
      "loss": 5.317,
      "mean_token_accuracy": 0.22308984249830247,
      "num_tokens": 80456683.0,
      "step": 43150
    },
    {
      "entropy": 5.9699396133422855,
      "epoch": 3.810932532673967,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000360709389124508,
      "loss": 5.2203,
      "mean_token_accuracy": 0.22047397792339324,
      "num_tokens": 80466554.0,
      "step": 43155
    },
    {
      "entropy": 5.942391490936279,
      "epoch": 3.811374072765807,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036068027112434926,
      "loss": 5.2169,
      "mean_token_accuracy": 0.2264366939663887,
      "num_tokens": 80475980.0,
      "step": 43160
    },
    {
      "entropy": 5.992697525024414,
      "epoch": 3.8118156128576475,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003606511514457627,
      "loss": 5.2656,
      "mean_token_accuracy": 0.21782431304454802,
      "num_tokens": 80485523.0,
      "step": 43165
    },
    {
      "entropy": 6.120897340774536,
      "epoch": 3.812257152949488,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003606220300893188,
      "loss": 5.4034,
      "mean_token_accuracy": 0.20877128094434738,
      "num_tokens": 80495044.0,
      "step": 43170
    },
    {
      "entropy": 6.049223327636719,
      "epoch": 3.812698693041328,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036059290705558794,
      "loss": 5.2408,
      "mean_token_accuracy": 0.21807715892791749,
      "num_tokens": 80504037.0,
      "step": 43175
    },
    {
      "entropy": 6.013270378112793,
      "epoch": 3.8131402331331685,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036056378234514087,
      "loss": 5.1938,
      "mean_token_accuracy": 0.2222325995564461,
      "num_tokens": 80512711.0,
      "step": 43180
    },
    {
      "entropy": 5.958714866638184,
      "epoch": 3.8135817732250086,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003605346559585478,
      "loss": 5.2065,
      "mean_token_accuracy": 0.2212669000029564,
      "num_tokens": 80521098.0,
      "step": 43185
    },
    {
      "entropy": 5.968178129196167,
      "epoch": 3.814023313316849,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036050552789637966,
      "loss": 5.2475,
      "mean_token_accuracy": 0.22310220897197724,
      "num_tokens": 80529781.0,
      "step": 43190
    },
    {
      "entropy": 5.989362812042236,
      "epoch": 3.8144648534086896,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036047639815920665,
      "loss": 5.3348,
      "mean_token_accuracy": 0.21631790101528167,
      "num_tokens": 80539655.0,
      "step": 43195
    },
    {
      "entropy": 5.942022848129272,
      "epoch": 3.8149063935005296,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036044726674759977,
      "loss": 5.2212,
      "mean_token_accuracy": 0.22623943090438842,
      "num_tokens": 80548261.0,
      "step": 43200
    },
    {
      "entropy": 6.067667627334595,
      "epoch": 3.81534793359237,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036041813366212937,
      "loss": 5.2849,
      "mean_token_accuracy": 0.2163387805223465,
      "num_tokens": 80557119.0,
      "step": 43205
    },
    {
      "entropy": 6.039463806152344,
      "epoch": 3.8157894736842106,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036038899890336644,
      "loss": 5.2061,
      "mean_token_accuracy": 0.21824061572551728,
      "num_tokens": 80566562.0,
      "step": 43210
    },
    {
      "entropy": 6.133769702911377,
      "epoch": 3.8162310137760507,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036035986247188155,
      "loss": 5.3259,
      "mean_token_accuracy": 0.21063434034585954,
      "num_tokens": 80575434.0,
      "step": 43215
    },
    {
      "entropy": 5.987796449661255,
      "epoch": 3.816672553867891,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036033072436824547,
      "loss": 5.1599,
      "mean_token_accuracy": 0.22821572721004485,
      "num_tokens": 80584575.0,
      "step": 43220
    },
    {
      "entropy": 5.921192693710327,
      "epoch": 3.8171140939597317,
      "grad_norm": 1.125,
      "learning_rate": 0.0003603015845930291,
      "loss": 5.2657,
      "mean_token_accuracy": 0.22097226083278657,
      "num_tokens": 80594555.0,
      "step": 43225
    },
    {
      "entropy": 5.957421779632568,
      "epoch": 3.8175556340515717,
      "grad_norm": 1.3125,
      "learning_rate": 0.00036027244314680313,
      "loss": 5.1907,
      "mean_token_accuracy": 0.22255463749170304,
      "num_tokens": 80603545.0,
      "step": 43230
    },
    {
      "entropy": 5.949598693847657,
      "epoch": 3.817997174143412,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036024330003013854,
      "loss": 5.1548,
      "mean_token_accuracy": 0.2298674017190933,
      "num_tokens": 80613921.0,
      "step": 43235
    },
    {
      "entropy": 6.006989240646362,
      "epoch": 3.8184387142352527,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003602141552436062,
      "loss": 5.2602,
      "mean_token_accuracy": 0.21678022295236588,
      "num_tokens": 80623076.0,
      "step": 43240
    },
    {
      "entropy": 5.923131799697876,
      "epoch": 3.8188802543270928,
      "grad_norm": 1.484375,
      "learning_rate": 0.00036018500878777705,
      "loss": 5.1823,
      "mean_token_accuracy": 0.21951804757118226,
      "num_tokens": 80631122.0,
      "step": 43245
    },
    {
      "entropy": 6.029859256744385,
      "epoch": 3.8193217944189333,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003601558606632219,
      "loss": 5.352,
      "mean_token_accuracy": 0.20956872552633285,
      "num_tokens": 80640893.0,
      "step": 43250
    },
    {
      "entropy": 6.0120738506317135,
      "epoch": 3.8197633345107738,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000360126710870512,
      "loss": 5.3285,
      "mean_token_accuracy": 0.21320124864578247,
      "num_tokens": 80649801.0,
      "step": 43255
    },
    {
      "entropy": 6.065294933319092,
      "epoch": 3.820204874602614,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036009755941021816,
      "loss": 5.3144,
      "mean_token_accuracy": 0.22055518627166748,
      "num_tokens": 80659440.0,
      "step": 43260
    },
    {
      "entropy": 6.005476474761963,
      "epoch": 3.8206464146944543,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036006840628291157,
      "loss": 5.2468,
      "mean_token_accuracy": 0.22210200428962706,
      "num_tokens": 80668573.0,
      "step": 43265
    },
    {
      "entropy": 6.020976305007935,
      "epoch": 3.821087954786295,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003600392514891633,
      "loss": 5.2581,
      "mean_token_accuracy": 0.21442724019289017,
      "num_tokens": 80676945.0,
      "step": 43270
    },
    {
      "entropy": 6.031437253952026,
      "epoch": 3.821529494878135,
      "grad_norm": 1.453125,
      "learning_rate": 0.00036001009502954436,
      "loss": 5.2374,
      "mean_token_accuracy": 0.22121127992868422,
      "num_tokens": 80685995.0,
      "step": 43275
    },
    {
      "entropy": 5.908753395080566,
      "epoch": 3.8219710349699754,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035998093690462607,
      "loss": 5.1803,
      "mean_token_accuracy": 0.22275364100933076,
      "num_tokens": 80696075.0,
      "step": 43280
    },
    {
      "entropy": 6.06276273727417,
      "epoch": 3.822412575061816,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003599517771149796,
      "loss": 5.2742,
      "mean_token_accuracy": 0.21532830744981765,
      "num_tokens": 80705678.0,
      "step": 43285
    },
    {
      "entropy": 6.074077558517456,
      "epoch": 3.822854115153656,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035992261566117605,
      "loss": 5.3216,
      "mean_token_accuracy": 0.20875124335289003,
      "num_tokens": 80715225.0,
      "step": 43290
    },
    {
      "entropy": 6.087213945388794,
      "epoch": 3.8232956552454964,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035989345254378686,
      "loss": 5.3415,
      "mean_token_accuracy": 0.2050445482134819,
      "num_tokens": 80723663.0,
      "step": 43295
    },
    {
      "entropy": 5.986309766769409,
      "epoch": 3.8237371953373365,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035986428776338316,
      "loss": 5.2112,
      "mean_token_accuracy": 0.22926914244890212,
      "num_tokens": 80734341.0,
      "step": 43300
    },
    {
      "entropy": 6.019325208663941,
      "epoch": 3.824178735429177,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003598351213205363,
      "loss": 5.232,
      "mean_token_accuracy": 0.21956840753555298,
      "num_tokens": 80743190.0,
      "step": 43305
    },
    {
      "entropy": 6.044756078720093,
      "epoch": 3.8246202755210175,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035980595321581774,
      "loss": 5.3033,
      "mean_token_accuracy": 0.2108157217502594,
      "num_tokens": 80752062.0,
      "step": 43310
    },
    {
      "entropy": 6.00514407157898,
      "epoch": 3.8250618156128575,
      "grad_norm": 1.125,
      "learning_rate": 0.0003597767834497987,
      "loss": 5.2412,
      "mean_token_accuracy": 0.22475461810827255,
      "num_tokens": 80761921.0,
      "step": 43315
    },
    {
      "entropy": 6.059457015991211,
      "epoch": 3.825503355704698,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035974761202305067,
      "loss": 5.3363,
      "mean_token_accuracy": 0.2135033756494522,
      "num_tokens": 80771715.0,
      "step": 43320
    },
    {
      "entropy": 6.059922122955323,
      "epoch": 3.825944895796538,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003597184389361452,
      "loss": 5.2801,
      "mean_token_accuracy": 0.2138896942138672,
      "num_tokens": 80780748.0,
      "step": 43325
    },
    {
      "entropy": 6.021622371673584,
      "epoch": 3.8263864358883786,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035968926418965367,
      "loss": 5.2685,
      "mean_token_accuracy": 0.214019276201725,
      "num_tokens": 80790087.0,
      "step": 43330
    },
    {
      "entropy": 6.0585980892181395,
      "epoch": 3.826827975980219,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003596600877841476,
      "loss": 5.2932,
      "mean_token_accuracy": 0.21504271179437637,
      "num_tokens": 80798856.0,
      "step": 43335
    },
    {
      "entropy": 6.113998794555664,
      "epoch": 3.827269516072059,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003596309097201986,
      "loss": 5.3263,
      "mean_token_accuracy": 0.20576089173555373,
      "num_tokens": 80808988.0,
      "step": 43340
    },
    {
      "entropy": 6.03040132522583,
      "epoch": 3.8277110561638996,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003596017299983781,
      "loss": 5.26,
      "mean_token_accuracy": 0.21416073739528657,
      "num_tokens": 80817781.0,
      "step": 43345
    },
    {
      "entropy": 5.957181835174561,
      "epoch": 3.82815259625574,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035957254861925794,
      "loss": 5.3043,
      "mean_token_accuracy": 0.21751437038183213,
      "num_tokens": 80827610.0,
      "step": 43350
    },
    {
      "entropy": 5.985765600204468,
      "epoch": 3.82859413634758,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00035954336558340966,
      "loss": 5.2684,
      "mean_token_accuracy": 0.21784651279449463,
      "num_tokens": 80836202.0,
      "step": 43355
    },
    {
      "entropy": 6.038657236099243,
      "epoch": 3.8290356764394207,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035951418089140484,
      "loss": 5.2465,
      "mean_token_accuracy": 0.22364134192466736,
      "num_tokens": 80844953.0,
      "step": 43360
    },
    {
      "entropy": 5.907186841964721,
      "epoch": 3.829477216531261,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003594849945438154,
      "loss": 5.1501,
      "mean_token_accuracy": 0.22612613290548325,
      "num_tokens": 80854189.0,
      "step": 43365
    },
    {
      "entropy": 6.038308811187744,
      "epoch": 3.829918756623101,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035945580654121284,
      "loss": 5.3768,
      "mean_token_accuracy": 0.20340159237384797,
      "num_tokens": 80863520.0,
      "step": 43370
    },
    {
      "entropy": 5.985066699981689,
      "epoch": 3.8303602967149417,
      "grad_norm": 1.25,
      "learning_rate": 0.0003594266168841691,
      "loss": 5.2282,
      "mean_token_accuracy": 0.22047821134328843,
      "num_tokens": 80873108.0,
      "step": 43375
    },
    {
      "entropy": 5.958801317214966,
      "epoch": 3.830801836806782,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000359397425573256,
      "loss": 5.1919,
      "mean_token_accuracy": 0.2221863940358162,
      "num_tokens": 80881960.0,
      "step": 43380
    },
    {
      "entropy": 6.034222555160523,
      "epoch": 3.8312433768986223,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035936823260904535,
      "loss": 5.2521,
      "mean_token_accuracy": 0.21977051496505737,
      "num_tokens": 80891577.0,
      "step": 43385
    },
    {
      "entropy": 6.029996633529663,
      "epoch": 3.8316849169904628,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035933903799210894,
      "loss": 5.2556,
      "mean_token_accuracy": 0.22253739833831787,
      "num_tokens": 80900933.0,
      "step": 43390
    },
    {
      "entropy": 6.023517894744873,
      "epoch": 3.8321264570823033,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003593098417230188,
      "loss": 5.3425,
      "mean_token_accuracy": 0.20846591591835023,
      "num_tokens": 80910732.0,
      "step": 43395
    },
    {
      "entropy": 6.0230076789855955,
      "epoch": 3.8325679971741433,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003592806438023467,
      "loss": 5.3367,
      "mean_token_accuracy": 0.21506793648004532,
      "num_tokens": 80920184.0,
      "step": 43400
    },
    {
      "entropy": 6.097343492507934,
      "epoch": 3.833009537265984,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003592514442306648,
      "loss": 5.4241,
      "mean_token_accuracy": 0.2102203071117401,
      "num_tokens": 80930437.0,
      "step": 43405
    },
    {
      "entropy": 6.06840558052063,
      "epoch": 3.8334510773578243,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035922224300854506,
      "loss": 5.2593,
      "mean_token_accuracy": 0.2164929673075676,
      "num_tokens": 80940167.0,
      "step": 43410
    },
    {
      "entropy": 6.071955871582031,
      "epoch": 3.8338926174496644,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003591930401365594,
      "loss": 5.3211,
      "mean_token_accuracy": 0.2135607197880745,
      "num_tokens": 80949880.0,
      "step": 43415
    },
    {
      "entropy": 6.064805793762207,
      "epoch": 3.834334157541505,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035916383561528006,
      "loss": 5.2696,
      "mean_token_accuracy": 0.2199770599603653,
      "num_tokens": 80958965.0,
      "step": 43420
    },
    {
      "entropy": 5.954156064987183,
      "epoch": 3.8347756976333454,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00035913462944527893,
      "loss": 5.1796,
      "mean_token_accuracy": 0.22770227640867233,
      "num_tokens": 80968367.0,
      "step": 43425
    },
    {
      "entropy": 6.074294328689575,
      "epoch": 3.8352172377251854,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003591054216271283,
      "loss": 5.38,
      "mean_token_accuracy": 0.20972346663475036,
      "num_tokens": 80977628.0,
      "step": 43430
    },
    {
      "entropy": 6.1015605449676515,
      "epoch": 3.835658777817026,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003590762121614004,
      "loss": 5.3313,
      "mean_token_accuracy": 0.21990984827280044,
      "num_tokens": 80987563.0,
      "step": 43435
    },
    {
      "entropy": 6.055803442001343,
      "epoch": 3.836100317908866,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003590470010486672,
      "loss": 5.2864,
      "mean_token_accuracy": 0.21547551304101945,
      "num_tokens": 80996311.0,
      "step": 43440
    },
    {
      "entropy": 6.035711336135864,
      "epoch": 3.8365418580007065,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035901778828950105,
      "loss": 5.2837,
      "mean_token_accuracy": 0.21645474731922149,
      "num_tokens": 81006761.0,
      "step": 43445
    },
    {
      "entropy": 6.0007366180419925,
      "epoch": 3.836983398092547,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035898857388447426,
      "loss": 5.2546,
      "mean_token_accuracy": 0.21730376780033112,
      "num_tokens": 81017273.0,
      "step": 43450
    },
    {
      "entropy": 6.036715936660767,
      "epoch": 3.837424938184387,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000358959357834159,
      "loss": 5.3106,
      "mean_token_accuracy": 0.21676707565784453,
      "num_tokens": 81026816.0,
      "step": 43455
    },
    {
      "entropy": 6.05206618309021,
      "epoch": 3.8378664782762275,
      "grad_norm": 1.34375,
      "learning_rate": 0.00035893014013912777,
      "loss": 5.2981,
      "mean_token_accuracy": 0.2213979110121727,
      "num_tokens": 81036749.0,
      "step": 43460
    },
    {
      "entropy": 6.054798126220703,
      "epoch": 3.8383080183680676,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003589009207999528,
      "loss": 5.3146,
      "mean_token_accuracy": 0.21770083606243135,
      "num_tokens": 81046416.0,
      "step": 43465
    },
    {
      "entropy": 5.982965993881225,
      "epoch": 3.838749558459908,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003588716998172065,
      "loss": 5.2581,
      "mean_token_accuracy": 0.22175773680210115,
      "num_tokens": 81055390.0,
      "step": 43470
    },
    {
      "entropy": 5.947188520431519,
      "epoch": 3.8391910985517486,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003588424771914613,
      "loss": 5.2098,
      "mean_token_accuracy": 0.22465428113937377,
      "num_tokens": 81065408.0,
      "step": 43475
    },
    {
      "entropy": 5.979728937149048,
      "epoch": 3.8396326386435886,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035881325292328966,
      "loss": 5.2025,
      "mean_token_accuracy": 0.21899257451295853,
      "num_tokens": 81074874.0,
      "step": 43480
    },
    {
      "entropy": 6.103550910949707,
      "epoch": 3.840074178735429,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003587840270132641,
      "loss": 5.3313,
      "mean_token_accuracy": 0.20698728263378144,
      "num_tokens": 81083813.0,
      "step": 43485
    },
    {
      "entropy": 5.979351377487182,
      "epoch": 3.8405157188272696,
      "grad_norm": 1.328125,
      "learning_rate": 0.000358754799461957,
      "loss": 5.1865,
      "mean_token_accuracy": 0.22293912917375563,
      "num_tokens": 81092925.0,
      "step": 43490
    },
    {
      "entropy": 5.947723913192749,
      "epoch": 3.8409572589191097,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003587255702699411,
      "loss": 5.2152,
      "mean_token_accuracy": 0.22412340939044953,
      "num_tokens": 81102360.0,
      "step": 43495
    },
    {
      "entropy": 6.032581186294555,
      "epoch": 3.84139879901095,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035869633943778884,
      "loss": 5.3114,
      "mean_token_accuracy": 0.21992335915565492,
      "num_tokens": 81112127.0,
      "step": 43500
    },
    {
      "entropy": 6.000271511077881,
      "epoch": 3.8418403391027907,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003586671069660729,
      "loss": 5.2335,
      "mean_token_accuracy": 0.21232149600982667,
      "num_tokens": 81121963.0,
      "step": 43505
    },
    {
      "entropy": 6.071288156509399,
      "epoch": 3.8422818791946307,
      "grad_norm": 1.34375,
      "learning_rate": 0.000358637872855366,
      "loss": 5.2806,
      "mean_token_accuracy": 0.21223467886447905,
      "num_tokens": 81130601.0,
      "step": 43510
    },
    {
      "entropy": 5.982119703292847,
      "epoch": 3.842723419286471,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035860863710624065,
      "loss": 5.2752,
      "mean_token_accuracy": 0.22105062156915664,
      "num_tokens": 81141467.0,
      "step": 43515
    },
    {
      "entropy": 6.090457391738892,
      "epoch": 3.8431649593783117,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035857939971926966,
      "loss": 5.2912,
      "mean_token_accuracy": 0.21782675832509996,
      "num_tokens": 81151020.0,
      "step": 43520
    },
    {
      "entropy": 6.040722846984863,
      "epoch": 3.8436064994701518,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035855016069502577,
      "loss": 5.331,
      "mean_token_accuracy": 0.2093273863196373,
      "num_tokens": 81159967.0,
      "step": 43525
    },
    {
      "entropy": 5.98053503036499,
      "epoch": 3.8440480395619923,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003585209200340817,
      "loss": 5.2497,
      "mean_token_accuracy": 0.21620708256959914,
      "num_tokens": 81169326.0,
      "step": 43530
    },
    {
      "entropy": 6.022316360473633,
      "epoch": 3.8444895796538328,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035849167773701035,
      "loss": 5.2516,
      "mean_token_accuracy": 0.2253073051571846,
      "num_tokens": 81178001.0,
      "step": 43535
    },
    {
      "entropy": 6.034397172927856,
      "epoch": 3.844931119745673,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003584624338043846,
      "loss": 5.2502,
      "mean_token_accuracy": 0.22236330360174178,
      "num_tokens": 81187010.0,
      "step": 43540
    },
    {
      "entropy": 5.916353940963745,
      "epoch": 3.8453726598375133,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003584331882367771,
      "loss": 5.2157,
      "mean_token_accuracy": 0.2188978374004364,
      "num_tokens": 81196215.0,
      "step": 43545
    },
    {
      "entropy": 5.999292945861816,
      "epoch": 3.845814199929354,
      "grad_norm": 1.25,
      "learning_rate": 0.000358403941034761,
      "loss": 5.2247,
      "mean_token_accuracy": 0.22529589533805847,
      "num_tokens": 81205757.0,
      "step": 43550
    },
    {
      "entropy": 5.891422176361084,
      "epoch": 3.846255740021194,
      "grad_norm": 1.25,
      "learning_rate": 0.00035837469219890904,
      "loss": 5.1667,
      "mean_token_accuracy": 0.23416064083576202,
      "num_tokens": 81214648.0,
      "step": 43555
    },
    {
      "entropy": 6.0747878074646,
      "epoch": 3.8466972801130344,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003583454417297944,
      "loss": 5.2959,
      "mean_token_accuracy": 0.21364967226982118,
      "num_tokens": 81222967.0,
      "step": 43560
    },
    {
      "entropy": 6.014092683792114,
      "epoch": 3.847138820204875,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003583161896279899,
      "loss": 5.2891,
      "mean_token_accuracy": 0.21528176963329315,
      "num_tokens": 81231840.0,
      "step": 43565
    },
    {
      "entropy": 6.065160751342773,
      "epoch": 3.847580360296715,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035828693589406866,
      "loss": 5.2836,
      "mean_token_accuracy": 0.21313550919294358,
      "num_tokens": 81241577.0,
      "step": 43570
    },
    {
      "entropy": 6.100767135620117,
      "epoch": 3.8480219003885554,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003582576805286036,
      "loss": 5.2916,
      "mean_token_accuracy": 0.21663534194231032,
      "num_tokens": 81250041.0,
      "step": 43575
    },
    {
      "entropy": 5.9850939273834225,
      "epoch": 3.8484634404803955,
      "grad_norm": 1.296875,
      "learning_rate": 0.000358228423532168,
      "loss": 5.2448,
      "mean_token_accuracy": 0.2258100762963295,
      "num_tokens": 81259036.0,
      "step": 43580
    },
    {
      "entropy": 5.945687913894654,
      "epoch": 3.848904980572236,
      "grad_norm": 1.453125,
      "learning_rate": 0.000358199164905335,
      "loss": 5.1873,
      "mean_token_accuracy": 0.22884017825126649,
      "num_tokens": 81267533.0,
      "step": 43585
    },
    {
      "entropy": 6.0045552253723145,
      "epoch": 3.8493465206640765,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003581699046486775,
      "loss": 5.2709,
      "mean_token_accuracy": 0.22147178798913955,
      "num_tokens": 81275635.0,
      "step": 43590
    },
    {
      "entropy": 6.067506742477417,
      "epoch": 3.8497880607559165,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035814064276276896,
      "loss": 5.3179,
      "mean_token_accuracy": 0.2112848863005638,
      "num_tokens": 81285939.0,
      "step": 43595
    },
    {
      "entropy": 5.937997484207154,
      "epoch": 3.850229600847757,
      "grad_norm": 1.25,
      "learning_rate": 0.0003581113792481825,
      "loss": 5.1226,
      "mean_token_accuracy": 0.2374778524041176,
      "num_tokens": 81295186.0,
      "step": 43600
    },
    {
      "entropy": 6.040243482589721,
      "epoch": 3.850671140939597,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003580821141054914,
      "loss": 5.3162,
      "mean_token_accuracy": 0.21193426698446274,
      "num_tokens": 81305910.0,
      "step": 43605
    },
    {
      "entropy": 5.96397066116333,
      "epoch": 3.8511126810314376,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000358052847335269,
      "loss": 5.1993,
      "mean_token_accuracy": 0.22629195898771287,
      "num_tokens": 81314461.0,
      "step": 43610
    },
    {
      "entropy": 5.944454860687256,
      "epoch": 3.851554221123278,
      "grad_norm": 1.25,
      "learning_rate": 0.00035802357893808845,
      "loss": 5.2898,
      "mean_token_accuracy": 0.21762025952339173,
      "num_tokens": 81324296.0,
      "step": 43615
    },
    {
      "entropy": 6.0544579982757565,
      "epoch": 3.851995761215118,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003579943089145233,
      "loss": 5.2599,
      "mean_token_accuracy": 0.2189017951488495,
      "num_tokens": 81333367.0,
      "step": 43620
    },
    {
      "entropy": 5.977033710479736,
      "epoch": 3.8524373013069586,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035796503726514677,
      "loss": 5.1733,
      "mean_token_accuracy": 0.22513466328382492,
      "num_tokens": 81343161.0,
      "step": 43625
    },
    {
      "entropy": 5.904634809494018,
      "epoch": 3.852878841398799,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035793576399053233,
      "loss": 5.2076,
      "mean_token_accuracy": 0.22493790537118913,
      "num_tokens": 81352877.0,
      "step": 43630
    },
    {
      "entropy": 5.992755699157715,
      "epoch": 3.853320381490639,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003579064890912535,
      "loss": 5.3638,
      "mean_token_accuracy": 0.21188012957572938,
      "num_tokens": 81362018.0,
      "step": 43635
    },
    {
      "entropy": 6.060629272460938,
      "epoch": 3.8537619215824797,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035787721256788375,
      "loss": 5.3169,
      "mean_token_accuracy": 0.20948658585548402,
      "num_tokens": 81371933.0,
      "step": 43640
    },
    {
      "entropy": 6.047772169113159,
      "epoch": 3.85420346167432,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035784793442099637,
      "loss": 5.2626,
      "mean_token_accuracy": 0.2210078164935112,
      "num_tokens": 81381917.0,
      "step": 43645
    },
    {
      "entropy": 6.015163850784302,
      "epoch": 3.85464500176616,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003578186546511652,
      "loss": 5.2648,
      "mean_token_accuracy": 0.21629477739334108,
      "num_tokens": 81392096.0,
      "step": 43650
    },
    {
      "entropy": 6.080514192581177,
      "epoch": 3.8550865418580007,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00035778937325896373,
      "loss": 5.2448,
      "mean_token_accuracy": 0.21733590960502625,
      "num_tokens": 81401135.0,
      "step": 43655
    },
    {
      "entropy": 6.012237405776977,
      "epoch": 3.855528081949841,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003577600902449654,
      "loss": 5.2928,
      "mean_token_accuracy": 0.21621742248535156,
      "num_tokens": 81410449.0,
      "step": 43660
    },
    {
      "entropy": 6.005366277694702,
      "epoch": 3.8559696220416813,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00035773080560974396,
      "loss": 5.3088,
      "mean_token_accuracy": 0.2143358215689659,
      "num_tokens": 81419475.0,
      "step": 43665
    },
    {
      "entropy": 5.9939275741577145,
      "epoch": 3.8564111621335218,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035770151935387326,
      "loss": 5.2328,
      "mean_token_accuracy": 0.21788452714681625,
      "num_tokens": 81428187.0,
      "step": 43670
    },
    {
      "entropy": 5.920358753204345,
      "epoch": 3.8568527022253623,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003576722314779267,
      "loss": 5.1366,
      "mean_token_accuracy": 0.2285088211297989,
      "num_tokens": 81437327.0,
      "step": 43675
    },
    {
      "entropy": 6.022119235992432,
      "epoch": 3.8572942423172023,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00035764294198247816,
      "loss": 5.2563,
      "mean_token_accuracy": 0.21203745156526566,
      "num_tokens": 81446915.0,
      "step": 43680
    },
    {
      "entropy": 5.996302270889283,
      "epoch": 3.857735782409043,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003576136508681014,
      "loss": 5.2247,
      "mean_token_accuracy": 0.21518728733062745,
      "num_tokens": 81456003.0,
      "step": 43685
    },
    {
      "entropy": 6.0537982940673825,
      "epoch": 3.8581773225008833,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035758435813537014,
      "loss": 5.3026,
      "mean_token_accuracy": 0.21217511594295502,
      "num_tokens": 81465694.0,
      "step": 43690
    },
    {
      "entropy": 5.911627292633057,
      "epoch": 3.8586188625927234,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003575550637848584,
      "loss": 5.1972,
      "mean_token_accuracy": 0.21914947777986526,
      "num_tokens": 81474691.0,
      "step": 43695
    },
    {
      "entropy": 6.091728019714355,
      "epoch": 3.859060402684564,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035752576781713987,
      "loss": 5.2924,
      "mean_token_accuracy": 0.21296086609363557,
      "num_tokens": 81483779.0,
      "step": 43700
    },
    {
      "entropy": 5.973643350601196,
      "epoch": 3.8595019427764043,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003574964702327884,
      "loss": 5.2311,
      "mean_token_accuracy": 0.22238195240497588,
      "num_tokens": 81492404.0,
      "step": 43705
    },
    {
      "entropy": 6.041108274459839,
      "epoch": 3.8599434828682444,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003574671710323781,
      "loss": 5.2746,
      "mean_token_accuracy": 0.2156492665410042,
      "num_tokens": 81501204.0,
      "step": 43710
    },
    {
      "entropy": 5.999344205856323,
      "epoch": 3.860385022960085,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003574378702164827,
      "loss": 5.3251,
      "mean_token_accuracy": 0.21341822445392608,
      "num_tokens": 81509278.0,
      "step": 43715
    },
    {
      "entropy": 6.053410291671753,
      "epoch": 3.860826563051925,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003574085677856764,
      "loss": 5.305,
      "mean_token_accuracy": 0.219347782433033,
      "num_tokens": 81518550.0,
      "step": 43720
    },
    {
      "entropy": 6.013616704940796,
      "epoch": 3.8612681031437655,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035737926374053316,
      "loss": 5.2445,
      "mean_token_accuracy": 0.21226266175508499,
      "num_tokens": 81527406.0,
      "step": 43725
    },
    {
      "entropy": 6.029502820968628,
      "epoch": 3.861709643235606,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035734995808162686,
      "loss": 5.2489,
      "mean_token_accuracy": 0.21608269661664964,
      "num_tokens": 81536741.0,
      "step": 43730
    },
    {
      "entropy": 6.062789821624756,
      "epoch": 3.862151183327446,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035732065080953186,
      "loss": 5.2709,
      "mean_token_accuracy": 0.21500158309936523,
      "num_tokens": 81545159.0,
      "step": 43735
    },
    {
      "entropy": 6.052179384231567,
      "epoch": 3.8625927234192865,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035729134192482204,
      "loss": 5.2936,
      "mean_token_accuracy": 0.21641255617141725,
      "num_tokens": 81554291.0,
      "step": 43740
    },
    {
      "entropy": 5.858110141754151,
      "epoch": 3.8630342635111266,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003572620314280717,
      "loss": 5.0706,
      "mean_token_accuracy": 0.2340275004506111,
      "num_tokens": 81562976.0,
      "step": 43745
    },
    {
      "entropy": 5.978762006759643,
      "epoch": 3.863475803602967,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035723271931985484,
      "loss": 5.2411,
      "mean_token_accuracy": 0.21740575283765792,
      "num_tokens": 81572756.0,
      "step": 43750
    },
    {
      "entropy": 6.0621133804321286,
      "epoch": 3.8639173436948075,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035720340560074586,
      "loss": 5.3349,
      "mean_token_accuracy": 0.21295953541994095,
      "num_tokens": 81581775.0,
      "step": 43755
    },
    {
      "entropy": 6.017093086242676,
      "epoch": 3.8643588837866476,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035717409027131894,
      "loss": 5.2543,
      "mean_token_accuracy": 0.2193053185939789,
      "num_tokens": 81591289.0,
      "step": 43760
    },
    {
      "entropy": 5.945333671569824,
      "epoch": 3.864800423878488,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003571447733321483,
      "loss": 5.22,
      "mean_token_accuracy": 0.22865577638149262,
      "num_tokens": 81600486.0,
      "step": 43765
    },
    {
      "entropy": 6.037507343292236,
      "epoch": 3.8652419639703286,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003571154547838083,
      "loss": 5.3034,
      "mean_token_accuracy": 0.21304320991039277,
      "num_tokens": 81609917.0,
      "step": 43770
    },
    {
      "entropy": 6.072610139846802,
      "epoch": 3.8656835040621687,
      "grad_norm": 1.25,
      "learning_rate": 0.0003570861346268732,
      "loss": 5.2642,
      "mean_token_accuracy": 0.21385544538497925,
      "num_tokens": 81618758.0,
      "step": 43775
    },
    {
      "entropy": 6.086416912078858,
      "epoch": 3.866125044154009,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035705681286191733,
      "loss": 5.273,
      "mean_token_accuracy": 0.21409757882356645,
      "num_tokens": 81628529.0,
      "step": 43780
    },
    {
      "entropy": 6.066524934768677,
      "epoch": 3.8665665842458496,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035702748948951527,
      "loss": 5.3012,
      "mean_token_accuracy": 0.2174595534801483,
      "num_tokens": 81639128.0,
      "step": 43785
    },
    {
      "entropy": 6.075344467163086,
      "epoch": 3.8670081243376897,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003569981645102413,
      "loss": 5.3268,
      "mean_token_accuracy": 0.21094316244125366,
      "num_tokens": 81648865.0,
      "step": 43790
    },
    {
      "entropy": 5.92785234451294,
      "epoch": 3.86744966442953,
      "grad_norm": 1.34375,
      "learning_rate": 0.00035696883792466996,
      "loss": 5.2138,
      "mean_token_accuracy": 0.21839619278907776,
      "num_tokens": 81658068.0,
      "step": 43795
    },
    {
      "entropy": 5.9663323879241945,
      "epoch": 3.8678912045213707,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00035693950973337565,
      "loss": 5.2339,
      "mean_token_accuracy": 0.22316295355558396,
      "num_tokens": 81667266.0,
      "step": 43800
    },
    {
      "entropy": 5.987966442108155,
      "epoch": 3.8683327446132107,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003569101799369329,
      "loss": 5.2481,
      "mean_token_accuracy": 0.22442806959152223,
      "num_tokens": 81677576.0,
      "step": 43805
    },
    {
      "entropy": 5.959464454650879,
      "epoch": 3.8687742847050512,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035688084853591646,
      "loss": 5.2478,
      "mean_token_accuracy": 0.2191699579358101,
      "num_tokens": 81687350.0,
      "step": 43810
    },
    {
      "entropy": 6.060076570510864,
      "epoch": 3.8692158247968917,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035685151553090067,
      "loss": 5.3396,
      "mean_token_accuracy": 0.21087126582860946,
      "num_tokens": 81696881.0,
      "step": 43815
    },
    {
      "entropy": 6.060290431976318,
      "epoch": 3.869657364888732,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035682218092246025,
      "loss": 5.2357,
      "mean_token_accuracy": 0.21780274361371993,
      "num_tokens": 81706357.0,
      "step": 43820
    },
    {
      "entropy": 6.045257568359375,
      "epoch": 3.8700989049805723,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003567928447111699,
      "loss": 5.2809,
      "mean_token_accuracy": 0.21442119479179383,
      "num_tokens": 81715672.0,
      "step": 43825
    },
    {
      "entropy": 6.070092105865479,
      "epoch": 3.870540445072413,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003567635068976042,
      "loss": 5.3849,
      "mean_token_accuracy": 0.21045074313879014,
      "num_tokens": 81724495.0,
      "step": 43830
    },
    {
      "entropy": 6.0102966785430905,
      "epoch": 3.870981985164253,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003567341674823379,
      "loss": 5.2408,
      "mean_token_accuracy": 0.2227312833070755,
      "num_tokens": 81733383.0,
      "step": 43835
    },
    {
      "entropy": 6.015229177474976,
      "epoch": 3.8714235252560933,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003567048264659457,
      "loss": 5.2864,
      "mean_token_accuracy": 0.21993706822395326,
      "num_tokens": 81742927.0,
      "step": 43840
    },
    {
      "entropy": 6.038743543624878,
      "epoch": 3.871865065347934,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035667548384900246,
      "loss": 5.3434,
      "mean_token_accuracy": 0.21085081696510316,
      "num_tokens": 81752766.0,
      "step": 43845
    },
    {
      "entropy": 6.158266210556031,
      "epoch": 3.872306605439774,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035664613963208293,
      "loss": 5.3814,
      "mean_token_accuracy": 0.21264486014842987,
      "num_tokens": 81762984.0,
      "step": 43850
    },
    {
      "entropy": 6.036128234863281,
      "epoch": 3.8727481455316144,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000356616793815762,
      "loss": 5.3223,
      "mean_token_accuracy": 0.21353858560323716,
      "num_tokens": 81772087.0,
      "step": 43855
    },
    {
      "entropy": 5.990540838241577,
      "epoch": 3.8731896856234544,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035658744640061445,
      "loss": 5.2486,
      "mean_token_accuracy": 0.21793775558471679,
      "num_tokens": 81781428.0,
      "step": 43860
    },
    {
      "entropy": 5.92001781463623,
      "epoch": 3.873631225715295,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035655809738721525,
      "loss": 5.185,
      "mean_token_accuracy": 0.21717156171798707,
      "num_tokens": 81790786.0,
      "step": 43865
    },
    {
      "entropy": 6.089915704727173,
      "epoch": 3.8740727658071354,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003565287467761393,
      "loss": 5.3313,
      "mean_token_accuracy": 0.20976323783397674,
      "num_tokens": 81799801.0,
      "step": 43870
    },
    {
      "entropy": 6.181219911575317,
      "epoch": 3.8745143058989755,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035649939456796156,
      "loss": 5.3966,
      "mean_token_accuracy": 0.20647420436143876,
      "num_tokens": 81808546.0,
      "step": 43875
    },
    {
      "entropy": 6.042022371292115,
      "epoch": 3.874955845990816,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003564700407632571,
      "loss": 5.3024,
      "mean_token_accuracy": 0.21479354202747344,
      "num_tokens": 81817921.0,
      "step": 43880
    },
    {
      "entropy": 6.023239850997925,
      "epoch": 3.875397386082656,
      "grad_norm": 1.25,
      "learning_rate": 0.0003564406853626008,
      "loss": 5.2446,
      "mean_token_accuracy": 0.21870691180229188,
      "num_tokens": 81826803.0,
      "step": 43885
    },
    {
      "entropy": 6.0923424243927,
      "epoch": 3.8758389261744965,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003564113283665678,
      "loss": 5.2947,
      "mean_token_accuracy": 0.22074586898088455,
      "num_tokens": 81835946.0,
      "step": 43890
    },
    {
      "entropy": 6.072552251815796,
      "epoch": 3.876280466266337,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003563819697757331,
      "loss": 5.3505,
      "mean_token_accuracy": 0.2173222541809082,
      "num_tokens": 81845426.0,
      "step": 43895
    },
    {
      "entropy": 6.021532678604126,
      "epoch": 3.876722006358177,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035635260959067204,
      "loss": 5.2619,
      "mean_token_accuracy": 0.21822309643030166,
      "num_tokens": 81854980.0,
      "step": 43900
    },
    {
      "entropy": 6.005882024765015,
      "epoch": 3.8771635464500176,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035632324781195956,
      "loss": 5.2163,
      "mean_token_accuracy": 0.2209432139992714,
      "num_tokens": 81864807.0,
      "step": 43905
    },
    {
      "entropy": 6.001540184020996,
      "epoch": 3.877605086541858,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00035629388444017093,
      "loss": 5.3484,
      "mean_token_accuracy": 0.21309403479099273,
      "num_tokens": 81874359.0,
      "step": 43910
    },
    {
      "entropy": 6.044634771347046,
      "epoch": 3.878046626633698,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035626451947588127,
      "loss": 5.3494,
      "mean_token_accuracy": 0.2117778480052948,
      "num_tokens": 81884095.0,
      "step": 43915
    },
    {
      "entropy": 6.025137186050415,
      "epoch": 3.8784881667255386,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035623515291966596,
      "loss": 5.2138,
      "mean_token_accuracy": 0.21962784826755524,
      "num_tokens": 81893163.0,
      "step": 43920
    },
    {
      "entropy": 6.000915050506592,
      "epoch": 3.878929706817379,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003562057847721002,
      "loss": 5.2153,
      "mean_token_accuracy": 0.22436934411525727,
      "num_tokens": 81902342.0,
      "step": 43925
    },
    {
      "entropy": 5.976465177536011,
      "epoch": 3.879371246909219,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003561764150337593,
      "loss": 5.3072,
      "mean_token_accuracy": 0.21759456545114517,
      "num_tokens": 81911553.0,
      "step": 43930
    },
    {
      "entropy": 6.0686040878295895,
      "epoch": 3.8798127870010597,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035614704370521855,
      "loss": 5.2307,
      "mean_token_accuracy": 0.21654965579509736,
      "num_tokens": 81919719.0,
      "step": 43935
    },
    {
      "entropy": 6.027402067184449,
      "epoch": 3.8802543270929,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035611767078705343,
      "loss": 5.3383,
      "mean_token_accuracy": 0.2130897969007492,
      "num_tokens": 81928337.0,
      "step": 43940
    },
    {
      "entropy": 6.0535361766815186,
      "epoch": 3.8806958671847402,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003560882962798392,
      "loss": 5.3672,
      "mean_token_accuracy": 0.20613840073347092,
      "num_tokens": 81938340.0,
      "step": 43945
    },
    {
      "entropy": 5.991926288604736,
      "epoch": 3.8811374072765807,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035605892018415145,
      "loss": 5.2327,
      "mean_token_accuracy": 0.22589986473321916,
      "num_tokens": 81948175.0,
      "step": 43950
    },
    {
      "entropy": 5.990320682525635,
      "epoch": 3.8815789473684212,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003560295425005655,
      "loss": 5.2151,
      "mean_token_accuracy": 0.22207016199827195,
      "num_tokens": 81958018.0,
      "step": 43955
    },
    {
      "entropy": 5.993079757690429,
      "epoch": 3.8820204874602613,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003560001632296569,
      "loss": 5.2203,
      "mean_token_accuracy": 0.2238598272204399,
      "num_tokens": 81967560.0,
      "step": 43960
    },
    {
      "entropy": 6.0456404209136965,
      "epoch": 3.882462027552102,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003559707823720012,
      "loss": 5.2805,
      "mean_token_accuracy": 0.21654489785432815,
      "num_tokens": 81977713.0,
      "step": 43965
    },
    {
      "entropy": 5.956272220611572,
      "epoch": 3.8829035676439423,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035594139992817394,
      "loss": 5.1697,
      "mean_token_accuracy": 0.22398622334003448,
      "num_tokens": 81987150.0,
      "step": 43970
    },
    {
      "entropy": 6.026166391372681,
      "epoch": 3.8833451077357823,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003559120158987506,
      "loss": 5.2241,
      "mean_token_accuracy": 0.2175970569252968,
      "num_tokens": 81995469.0,
      "step": 43975
    },
    {
      "entropy": 6.062734794616699,
      "epoch": 3.883786647827623,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003558826302843069,
      "loss": 5.2479,
      "mean_token_accuracy": 0.21605999171733856,
      "num_tokens": 82003596.0,
      "step": 43980
    },
    {
      "entropy": 5.987908267974854,
      "epoch": 3.8842281879194633,
      "grad_norm": 1.375,
      "learning_rate": 0.0003558532430854185,
      "loss": 5.2267,
      "mean_token_accuracy": 0.2203853577375412,
      "num_tokens": 82012012.0,
      "step": 43985
    },
    {
      "entropy": 5.985701704025269,
      "epoch": 3.8846697280113034,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035582385430266107,
      "loss": 5.2965,
      "mean_token_accuracy": 0.21338431984186174,
      "num_tokens": 82020725.0,
      "step": 43990
    },
    {
      "entropy": 5.977908563613892,
      "epoch": 3.885111268103144,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003557944639366103,
      "loss": 5.3265,
      "mean_token_accuracy": 0.21741789132356643,
      "num_tokens": 82030204.0,
      "step": 43995
    },
    {
      "entropy": 6.008548974990845,
      "epoch": 3.885552808194984,
      "grad_norm": 1.328125,
      "learning_rate": 0.000355765071987842,
      "loss": 5.2275,
      "mean_token_accuracy": 0.21875914186239243,
      "num_tokens": 82038413.0,
      "step": 44000
    },
    {
      "entropy": 6.0047845363616945,
      "epoch": 3.8859943482868244,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035573567845693173,
      "loss": 5.2617,
      "mean_token_accuracy": 0.2107418179512024,
      "num_tokens": 82047258.0,
      "step": 44005
    },
    {
      "entropy": 5.999988842010498,
      "epoch": 3.886435888378665,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035570628334445555,
      "loss": 5.3019,
      "mean_token_accuracy": 0.21290665864944458,
      "num_tokens": 82057528.0,
      "step": 44010
    },
    {
      "entropy": 6.095284175872803,
      "epoch": 3.886877428470505,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003556768866509891,
      "loss": 5.2279,
      "mean_token_accuracy": 0.21368225812911987,
      "num_tokens": 82066351.0,
      "step": 44015
    },
    {
      "entropy": 6.00396466255188,
      "epoch": 3.8873189685623455,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035564748837710834,
      "loss": 5.23,
      "mean_token_accuracy": 0.22082876712083815,
      "num_tokens": 82076480.0,
      "step": 44020
    },
    {
      "entropy": 5.970465898513794,
      "epoch": 3.8877605086541855,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003556180885233892,
      "loss": 5.2713,
      "mean_token_accuracy": 0.22223780751228334,
      "num_tokens": 82086316.0,
      "step": 44025
    },
    {
      "entropy": 5.9895031452178955,
      "epoch": 3.888202048746026,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003555886870904075,
      "loss": 5.2053,
      "mean_token_accuracy": 0.2271646663546562,
      "num_tokens": 82096357.0,
      "step": 44030
    },
    {
      "entropy": 5.9616063117980955,
      "epoch": 3.8886435888378665,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003555592840787393,
      "loss": 5.2006,
      "mean_token_accuracy": 0.2250925526022911,
      "num_tokens": 82106449.0,
      "step": 44035
    },
    {
      "entropy": 5.941616582870483,
      "epoch": 3.8890851289297066,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035552987948896044,
      "loss": 5.164,
      "mean_token_accuracy": 0.22798434793949127,
      "num_tokens": 82116078.0,
      "step": 44040
    },
    {
      "entropy": 5.958127164840699,
      "epoch": 3.889526669021547,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035550047332164714,
      "loss": 5.2209,
      "mean_token_accuracy": 0.22314728647470475,
      "num_tokens": 82126649.0,
      "step": 44045
    },
    {
      "entropy": 6.09056510925293,
      "epoch": 3.8899682091133876,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035547106557737534,
      "loss": 5.3585,
      "mean_token_accuracy": 0.20629846900701523,
      "num_tokens": 82136538.0,
      "step": 44050
    },
    {
      "entropy": 6.066563940048217,
      "epoch": 3.8904097492052276,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035544165625672107,
      "loss": 5.2497,
      "mean_token_accuracy": 0.2174815222620964,
      "num_tokens": 82145309.0,
      "step": 44055
    },
    {
      "entropy": 6.029371929168701,
      "epoch": 3.890851289297068,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035541224536026064,
      "loss": 5.2032,
      "mean_token_accuracy": 0.22210506051778794,
      "num_tokens": 82155368.0,
      "step": 44060
    },
    {
      "entropy": 6.029401540756226,
      "epoch": 3.8912928293889086,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003553828328885699,
      "loss": 5.2539,
      "mean_token_accuracy": 0.21308321207761766,
      "num_tokens": 82164915.0,
      "step": 44065
    },
    {
      "entropy": 5.88796329498291,
      "epoch": 3.8917343694807487,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00035535341884222526,
      "loss": 5.078,
      "mean_token_accuracy": 0.2307225227355957,
      "num_tokens": 82174125.0,
      "step": 44070
    },
    {
      "entropy": 6.096642541885376,
      "epoch": 3.892175909572589,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003553240032218029,
      "loss": 5.3881,
      "mean_token_accuracy": 0.20844802409410476,
      "num_tokens": 82184002.0,
      "step": 44075
    },
    {
      "entropy": 5.966463565826416,
      "epoch": 3.8926174496644297,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035529458602787887,
      "loss": 5.1294,
      "mean_token_accuracy": 0.2335033968091011,
      "num_tokens": 82193871.0,
      "step": 44080
    },
    {
      "entropy": 6.086993837356568,
      "epoch": 3.8930589897562697,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003552651672610297,
      "loss": 5.3229,
      "mean_token_accuracy": 0.21357047855854033,
      "num_tokens": 82203053.0,
      "step": 44085
    },
    {
      "entropy": 5.9958137512207035,
      "epoch": 3.8935005298481102,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003552357469218315,
      "loss": 5.2943,
      "mean_token_accuracy": 0.2175476923584938,
      "num_tokens": 82213003.0,
      "step": 44090
    },
    {
      "entropy": 5.938380765914917,
      "epoch": 3.8939420699399507,
      "grad_norm": 1.25,
      "learning_rate": 0.00035520632501086056,
      "loss": 5.2017,
      "mean_token_accuracy": 0.22022455632686616,
      "num_tokens": 82222391.0,
      "step": 44095
    },
    {
      "entropy": 6.0235871315002445,
      "epoch": 3.894383610031791,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003551769015286934,
      "loss": 5.291,
      "mean_token_accuracy": 0.21581198126077653,
      "num_tokens": 82231832.0,
      "step": 44100
    },
    {
      "entropy": 6.031273460388183,
      "epoch": 3.8948251501236313,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035514747647590636,
      "loss": 5.273,
      "mean_token_accuracy": 0.2186792492866516,
      "num_tokens": 82241962.0,
      "step": 44105
    },
    {
      "entropy": 6.078340339660644,
      "epoch": 3.895266690215472,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003551180498530758,
      "loss": 5.2709,
      "mean_token_accuracy": 0.21517967730760573,
      "num_tokens": 82250391.0,
      "step": 44110
    },
    {
      "entropy": 5.995383405685425,
      "epoch": 3.895708230307312,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00035508862166077823,
      "loss": 5.2099,
      "mean_token_accuracy": 0.22053136676549911,
      "num_tokens": 82259606.0,
      "step": 44115
    },
    {
      "entropy": 6.066356039047241,
      "epoch": 3.8961497703991523,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035505919189959015,
      "loss": 5.3335,
      "mean_token_accuracy": 0.2080269142985344,
      "num_tokens": 82269088.0,
      "step": 44120
    },
    {
      "entropy": 6.013703203201294,
      "epoch": 3.896591310490993,
      "grad_norm": 1.375,
      "learning_rate": 0.0003550297605700879,
      "loss": 5.2667,
      "mean_token_accuracy": 0.2220728412270546,
      "num_tokens": 82277376.0,
      "step": 44125
    },
    {
      "entropy": 5.955901002883911,
      "epoch": 3.897032850582833,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003550003276728483,
      "loss": 5.14,
      "mean_token_accuracy": 0.2322755604982376,
      "num_tokens": 82285857.0,
      "step": 44130
    },
    {
      "entropy": 6.017395782470703,
      "epoch": 3.8974743906746734,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035497089320844766,
      "loss": 5.2643,
      "mean_token_accuracy": 0.21562798023223878,
      "num_tokens": 82295628.0,
      "step": 44135
    },
    {
      "entropy": 6.016577911376953,
      "epoch": 3.8979159307665134,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035494145717746283,
      "loss": 5.2488,
      "mean_token_accuracy": 0.2206182971596718,
      "num_tokens": 82304245.0,
      "step": 44140
    },
    {
      "entropy": 6.004542064666748,
      "epoch": 3.898357470858354,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035491201958047025,
      "loss": 5.1921,
      "mean_token_accuracy": 0.22912857830524444,
      "num_tokens": 82313215.0,
      "step": 44145
    },
    {
      "entropy": 5.945651865005493,
      "epoch": 3.8987990109501944,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003548825804180467,
      "loss": 5.1945,
      "mean_token_accuracy": 0.22673737555742263,
      "num_tokens": 82322516.0,
      "step": 44150
    },
    {
      "entropy": 6.029306173324585,
      "epoch": 3.8992405510420345,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003548531396907687,
      "loss": 5.3712,
      "mean_token_accuracy": 0.20877432376146315,
      "num_tokens": 82332642.0,
      "step": 44155
    },
    {
      "entropy": 6.023918104171753,
      "epoch": 3.899682091133875,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035482369739921324,
      "loss": 5.2611,
      "mean_token_accuracy": 0.2224413588643074,
      "num_tokens": 82341771.0,
      "step": 44160
    },
    {
      "entropy": 6.033157634735107,
      "epoch": 3.900123631225715,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003547942535439569,
      "loss": 5.274,
      "mean_token_accuracy": 0.2268818661570549,
      "num_tokens": 82350460.0,
      "step": 44165
    },
    {
      "entropy": 6.048027801513672,
      "epoch": 3.9005651713175555,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035476480812557653,
      "loss": 5.28,
      "mean_token_accuracy": 0.21409918367862701,
      "num_tokens": 82359118.0,
      "step": 44170
    },
    {
      "entropy": 6.0380211353302,
      "epoch": 3.901006711409396,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035473536114464887,
      "loss": 5.2908,
      "mean_token_accuracy": 0.21104481667280198,
      "num_tokens": 82367877.0,
      "step": 44175
    },
    {
      "entropy": 6.062379360198975,
      "epoch": 3.901448251501236,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003547059126017509,
      "loss": 5.2959,
      "mean_token_accuracy": 0.21762061566114427,
      "num_tokens": 82377227.0,
      "step": 44180
    },
    {
      "entropy": 5.9528059482574465,
      "epoch": 3.9018897915930766,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003546764624974595,
      "loss": 5.1942,
      "mean_token_accuracy": 0.2245165154337883,
      "num_tokens": 82386375.0,
      "step": 44185
    },
    {
      "entropy": 6.016768121719361,
      "epoch": 3.902331331684917,
      "grad_norm": 1.34375,
      "learning_rate": 0.00035464701083235133,
      "loss": 5.287,
      "mean_token_accuracy": 0.21729191839694978,
      "num_tokens": 82395302.0,
      "step": 44190
    },
    {
      "entropy": 6.010980844497681,
      "epoch": 3.902772871776757,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003546175576070037,
      "loss": 5.2131,
      "mean_token_accuracy": 0.220414000749588,
      "num_tokens": 82404675.0,
      "step": 44195
    },
    {
      "entropy": 6.005852603912354,
      "epoch": 3.9032144118685976,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003545881028219933,
      "loss": 5.2461,
      "mean_token_accuracy": 0.2151898056268692,
      "num_tokens": 82413568.0,
      "step": 44200
    },
    {
      "entropy": 5.95063328742981,
      "epoch": 3.903655951960438,
      "grad_norm": 1.25,
      "learning_rate": 0.00035455864647789725,
      "loss": 5.2564,
      "mean_token_accuracy": 0.22674217075109482,
      "num_tokens": 82423000.0,
      "step": 44205
    },
    {
      "entropy": 6.07244291305542,
      "epoch": 3.904097492052278,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035452918857529255,
      "loss": 5.4153,
      "mean_token_accuracy": 0.20367512106895447,
      "num_tokens": 82432014.0,
      "step": 44210
    },
    {
      "entropy": 6.055272722244263,
      "epoch": 3.9045390321441187,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003544997291147563,
      "loss": 5.2989,
      "mean_token_accuracy": 0.21367886215448378,
      "num_tokens": 82441428.0,
      "step": 44215
    },
    {
      "entropy": 5.9945745944976805,
      "epoch": 3.904980572235959,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003544702680968656,
      "loss": 5.2538,
      "mean_token_accuracy": 0.22046552449464799,
      "num_tokens": 82451069.0,
      "step": 44220
    },
    {
      "entropy": 6.015207529067993,
      "epoch": 3.9054221123277992,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003544408055221974,
      "loss": 5.2595,
      "mean_token_accuracy": 0.21384406238794326,
      "num_tokens": 82459886.0,
      "step": 44225
    },
    {
      "entropy": 6.011767864227295,
      "epoch": 3.9058636524196397,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035441134139132917,
      "loss": 5.2194,
      "mean_token_accuracy": 0.21255114525556565,
      "num_tokens": 82468252.0,
      "step": 44230
    },
    {
      "entropy": 6.038792371749878,
      "epoch": 3.9063051925114802,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003543818757048379,
      "loss": 5.2382,
      "mean_token_accuracy": 0.214597250521183,
      "num_tokens": 82476669.0,
      "step": 44235
    },
    {
      "entropy": 5.964660263061523,
      "epoch": 3.9067467326033203,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003543524084633007,
      "loss": 5.2807,
      "mean_token_accuracy": 0.22261089980602264,
      "num_tokens": 82486025.0,
      "step": 44240
    },
    {
      "entropy": 5.985046243667602,
      "epoch": 3.907188272695161,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035432293966729505,
      "loss": 5.235,
      "mean_token_accuracy": 0.22429085969924928,
      "num_tokens": 82495292.0,
      "step": 44245
    },
    {
      "entropy": 5.993054628372192,
      "epoch": 3.9076298127870013,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003542934693173982,
      "loss": 5.1929,
      "mean_token_accuracy": 0.22426402121782302,
      "num_tokens": 82504042.0,
      "step": 44250
    },
    {
      "entropy": 6.041909885406494,
      "epoch": 3.9080713528788413,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003542639974141872,
      "loss": 5.215,
      "mean_token_accuracy": 0.22036901116371155,
      "num_tokens": 82513200.0,
      "step": 44255
    },
    {
      "entropy": 6.094451379776001,
      "epoch": 3.908512892970682,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035423452395823973,
      "loss": 5.3491,
      "mean_token_accuracy": 0.20398102104663848,
      "num_tokens": 82523641.0,
      "step": 44260
    },
    {
      "entropy": 5.959281063079834,
      "epoch": 3.9089544330625223,
      "grad_norm": 1.53125,
      "learning_rate": 0.0003542050489501329,
      "loss": 5.2064,
      "mean_token_accuracy": 0.22584726810455322,
      "num_tokens": 82532779.0,
      "step": 44265
    },
    {
      "entropy": 6.0857738018035885,
      "epoch": 3.9093959731543624,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003541755723904442,
      "loss": 5.3055,
      "mean_token_accuracy": 0.2098919555544853,
      "num_tokens": 82542308.0,
      "step": 44270
    },
    {
      "entropy": 6.061996555328369,
      "epoch": 3.909837513246203,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003541460942797512,
      "loss": 5.261,
      "mean_token_accuracy": 0.2186134785413742,
      "num_tokens": 82551648.0,
      "step": 44275
    },
    {
      "entropy": 6.027879333496093,
      "epoch": 3.910279053338043,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003541166146186311,
      "loss": 5.3101,
      "mean_token_accuracy": 0.22122278064489365,
      "num_tokens": 82560908.0,
      "step": 44280
    },
    {
      "entropy": 5.9932568073272705,
      "epoch": 3.9107205934298834,
      "grad_norm": 1.25,
      "learning_rate": 0.0003540871334076616,
      "loss": 5.2291,
      "mean_token_accuracy": 0.22245445251464843,
      "num_tokens": 82570972.0,
      "step": 44285
    },
    {
      "entropy": 6.139543867111206,
      "epoch": 3.911162133521724,
      "grad_norm": 1.578125,
      "learning_rate": 0.00035405765064742,
      "loss": 5.3413,
      "mean_token_accuracy": 0.21934475153684616,
      "num_tokens": 82579675.0,
      "step": 44290
    },
    {
      "entropy": 5.937576341629028,
      "epoch": 3.911603673613564,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003540281663384841,
      "loss": 5.0985,
      "mean_token_accuracy": 0.2363465189933777,
      "num_tokens": 82588261.0,
      "step": 44295
    },
    {
      "entropy": 5.991480159759521,
      "epoch": 3.9120452137054045,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035399868048143136,
      "loss": 5.2958,
      "mean_token_accuracy": 0.22101247757673265,
      "num_tokens": 82597531.0,
      "step": 44300
    },
    {
      "entropy": 6.022162389755249,
      "epoch": 3.9124867537972445,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003539691930768394,
      "loss": 5.3658,
      "mean_token_accuracy": 0.21055143773555757,
      "num_tokens": 82607104.0,
      "step": 44305
    },
    {
      "entropy": 6.05572361946106,
      "epoch": 3.912928293889085,
      "grad_norm": 1.421875,
      "learning_rate": 0.00035393970412528586,
      "loss": 5.2562,
      "mean_token_accuracy": 0.22459523379802704,
      "num_tokens": 82615476.0,
      "step": 44310
    },
    {
      "entropy": 6.015165901184082,
      "epoch": 3.9133698339809255,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003539102136273484,
      "loss": 5.3003,
      "mean_token_accuracy": 0.22032299041748046,
      "num_tokens": 82624245.0,
      "step": 44315
    },
    {
      "entropy": 6.102042388916016,
      "epoch": 3.9138113740727656,
      "grad_norm": 1.125,
      "learning_rate": 0.0003538807215836046,
      "loss": 5.3086,
      "mean_token_accuracy": 0.22214835733175278,
      "num_tokens": 82633923.0,
      "step": 44320
    },
    {
      "entropy": 6.078921270370484,
      "epoch": 3.914252914164606,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035385122799463253,
      "loss": 5.27,
      "mean_token_accuracy": 0.21912813931703568,
      "num_tokens": 82642934.0,
      "step": 44325
    },
    {
      "entropy": 6.079595851898193,
      "epoch": 3.9146944542564466,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035382173286100966,
      "loss": 5.2893,
      "mean_token_accuracy": 0.2142021745443344,
      "num_tokens": 82652625.0,
      "step": 44330
    },
    {
      "entropy": 6.043612098693847,
      "epoch": 3.9151359943482866,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003537922361833139,
      "loss": 5.2843,
      "mean_token_accuracy": 0.21405851393938063,
      "num_tokens": 82661468.0,
      "step": 44335
    },
    {
      "entropy": 6.026862001419067,
      "epoch": 3.915577534440127,
      "grad_norm": 1.34375,
      "learning_rate": 0.000353762737962123,
      "loss": 5.2732,
      "mean_token_accuracy": 0.22029314786195756,
      "num_tokens": 82670722.0,
      "step": 44340
    },
    {
      "entropy": 5.980880546569824,
      "epoch": 3.9160190745319676,
      "grad_norm": 1.25,
      "learning_rate": 0.0003537332381980149,
      "loss": 5.2191,
      "mean_token_accuracy": 0.2224508449435234,
      "num_tokens": 82680294.0,
      "step": 44345
    },
    {
      "entropy": 6.011997127532959,
      "epoch": 3.9164606146238077,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035370373689156743,
      "loss": 5.2483,
      "mean_token_accuracy": 0.22334983199834824,
      "num_tokens": 82689237.0,
      "step": 44350
    },
    {
      "entropy": 6.046076488494873,
      "epoch": 3.916902154715648,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035367423404335853,
      "loss": 5.392,
      "mean_token_accuracy": 0.21163323819637297,
      "num_tokens": 82698493.0,
      "step": 44355
    },
    {
      "entropy": 6.054538774490356,
      "epoch": 3.9173436948074887,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003536447296539661,
      "loss": 5.2889,
      "mean_token_accuracy": 0.21274371296167374,
      "num_tokens": 82707364.0,
      "step": 44360
    },
    {
      "entropy": 6.069072246551514,
      "epoch": 3.9177852348993287,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035361522372396817,
      "loss": 5.2626,
      "mean_token_accuracy": 0.2165999948978424,
      "num_tokens": 82717399.0,
      "step": 44365
    },
    {
      "entropy": 6.009288215637207,
      "epoch": 3.9182267749911692,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003535857162539427,
      "loss": 5.2251,
      "mean_token_accuracy": 0.21845061779022218,
      "num_tokens": 82727357.0,
      "step": 44370
    },
    {
      "entropy": 5.9532544136047365,
      "epoch": 3.9186683150830097,
      "grad_norm": 1.25,
      "learning_rate": 0.00035355620724446767,
      "loss": 5.192,
      "mean_token_accuracy": 0.21890112906694412,
      "num_tokens": 82736816.0,
      "step": 44375
    },
    {
      "entropy": 6.041455554962158,
      "epoch": 3.9191098551748498,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035352669669612137,
      "loss": 5.2909,
      "mean_token_accuracy": 0.2111929714679718,
      "num_tokens": 82746474.0,
      "step": 44380
    },
    {
      "entropy": 5.908472681045533,
      "epoch": 3.9195513952666903,
      "grad_norm": 1.25,
      "learning_rate": 0.0003534971846094816,
      "loss": 5.1441,
      "mean_token_accuracy": 0.22190767526626587,
      "num_tokens": 82756380.0,
      "step": 44385
    },
    {
      "entropy": 6.013777017593384,
      "epoch": 3.9199929353585308,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003534676709851267,
      "loss": 5.2773,
      "mean_token_accuracy": 0.2137884944677353,
      "num_tokens": 82765522.0,
      "step": 44390
    },
    {
      "entropy": 6.061391162872314,
      "epoch": 3.920434475450371,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035343815582363473,
      "loss": 5.3438,
      "mean_token_accuracy": 0.21480797827243805,
      "num_tokens": 82775015.0,
      "step": 44395
    },
    {
      "entropy": 6.013245630264282,
      "epoch": 3.9208760155422113,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000353408639125584,
      "loss": 5.2895,
      "mean_token_accuracy": 0.22064314484596254,
      "num_tokens": 82784104.0,
      "step": 44400
    },
    {
      "entropy": 6.04103717803955,
      "epoch": 3.921317555634052,
      "grad_norm": 1.40625,
      "learning_rate": 0.00035337912089155243,
      "loss": 5.2445,
      "mean_token_accuracy": 0.21462804824113846,
      "num_tokens": 82793233.0,
      "step": 44405
    },
    {
      "entropy": 6.07046217918396,
      "epoch": 3.921759095725892,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035334960112211864,
      "loss": 5.3586,
      "mean_token_accuracy": 0.21396900713443756,
      "num_tokens": 82803106.0,
      "step": 44410
    },
    {
      "entropy": 5.943084621429444,
      "epoch": 3.9222006358177324,
      "grad_norm": 1.25,
      "learning_rate": 0.0003533200798178606,
      "loss": 5.1791,
      "mean_token_accuracy": 0.22376867681741713,
      "num_tokens": 82811776.0,
      "step": 44415
    },
    {
      "entropy": 5.949887800216675,
      "epoch": 3.9226421759095724,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003532905569793568,
      "loss": 5.1403,
      "mean_token_accuracy": 0.2309111088514328,
      "num_tokens": 82820943.0,
      "step": 44420
    },
    {
      "entropy": 6.034988498687744,
      "epoch": 3.923083716001413,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003532610326071855,
      "loss": 5.1954,
      "mean_token_accuracy": 0.230599108338356,
      "num_tokens": 82830605.0,
      "step": 44425
    },
    {
      "entropy": 6.005761575698853,
      "epoch": 3.9235252560932534,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003532315067019251,
      "loss": 5.2041,
      "mean_token_accuracy": 0.22486667931079865,
      "num_tokens": 82839930.0,
      "step": 44430
    },
    {
      "entropy": 6.021865510940552,
      "epoch": 3.9239667961850935,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035320197926415396,
      "loss": 5.194,
      "mean_token_accuracy": 0.22333887368440627,
      "num_tokens": 82849455.0,
      "step": 44435
    },
    {
      "entropy": 6.057667303085327,
      "epoch": 3.924408336276934,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003531724502944505,
      "loss": 5.3323,
      "mean_token_accuracy": 0.20970822274684905,
      "num_tokens": 82858366.0,
      "step": 44440
    },
    {
      "entropy": 5.965437364578247,
      "epoch": 3.924849876368774,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003531429197933932,
      "loss": 5.308,
      "mean_token_accuracy": 0.21449284851551056,
      "num_tokens": 82867171.0,
      "step": 44445
    },
    {
      "entropy": 6.089388513565064,
      "epoch": 3.9252914164606145,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003531133877615606,
      "loss": 5.2682,
      "mean_token_accuracy": 0.22084086537361144,
      "num_tokens": 82875927.0,
      "step": 44450
    },
    {
      "entropy": 6.10954065322876,
      "epoch": 3.925732956552455,
      "grad_norm": 1.5,
      "learning_rate": 0.00035308385419953117,
      "loss": 5.2965,
      "mean_token_accuracy": 0.20904716849327087,
      "num_tokens": 82885807.0,
      "step": 44455
    },
    {
      "entropy": 5.92997727394104,
      "epoch": 3.926174496644295,
      "grad_norm": 1.390625,
      "learning_rate": 0.00035305431910788334,
      "loss": 5.2424,
      "mean_token_accuracy": 0.2210040882229805,
      "num_tokens": 82894910.0,
      "step": 44460
    },
    {
      "entropy": 5.9176685333251955,
      "epoch": 3.9266160367361356,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003530247824871959,
      "loss": 5.1171,
      "mean_token_accuracy": 0.2389889106154442,
      "num_tokens": 82904272.0,
      "step": 44465
    },
    {
      "entropy": 5.976398658752442,
      "epoch": 3.927057576827976,
      "grad_norm": 1.359375,
      "learning_rate": 0.00035299524433804724,
      "loss": 5.2082,
      "mean_token_accuracy": 0.22177051156759262,
      "num_tokens": 82912676.0,
      "step": 44470
    },
    {
      "entropy": 5.993638038635254,
      "epoch": 3.927499116919816,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003529657046610162,
      "loss": 5.2562,
      "mean_token_accuracy": 0.2190802663564682,
      "num_tokens": 82922052.0,
      "step": 44475
    },
    {
      "entropy": 6.079806184768676,
      "epoch": 3.9279406570116566,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003529361634566813,
      "loss": 5.3612,
      "mean_token_accuracy": 0.21009182631969453,
      "num_tokens": 82931132.0,
      "step": 44480
    },
    {
      "entropy": 5.889812898635864,
      "epoch": 3.928382197103497,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035290662072562124,
      "loss": 5.1048,
      "mean_token_accuracy": 0.23762754797935487,
      "num_tokens": 82940754.0,
      "step": 44485
    },
    {
      "entropy": 6.048758935928345,
      "epoch": 3.928823737195337,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003528770764684148,
      "loss": 5.2804,
      "mean_token_accuracy": 0.21819046884775162,
      "num_tokens": 82950298.0,
      "step": 44490
    },
    {
      "entropy": 6.047828340530396,
      "epoch": 3.9292652772871777,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035284753068564077,
      "loss": 5.2767,
      "mean_token_accuracy": 0.21840884387493134,
      "num_tokens": 82959163.0,
      "step": 44495
    },
    {
      "entropy": 5.985724306106567,
      "epoch": 3.929706817379018,
      "grad_norm": 1.34375,
      "learning_rate": 0.00035281798337787794,
      "loss": 5.1303,
      "mean_token_accuracy": 0.23308484852313996,
      "num_tokens": 82968538.0,
      "step": 44500
    },
    {
      "entropy": 6.011068868637085,
      "epoch": 3.9301483574708582,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003527884345457049,
      "loss": 5.3594,
      "mean_token_accuracy": 0.21459730416536332,
      "num_tokens": 82976968.0,
      "step": 44505
    },
    {
      "entropy": 6.0504810333251955,
      "epoch": 3.9305898975626987,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003527588841897009,
      "loss": 5.2109,
      "mean_token_accuracy": 0.21705062687397003,
      "num_tokens": 82985810.0,
      "step": 44510
    },
    {
      "entropy": 6.040159034729004,
      "epoch": 3.931031437654539,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003527293323104444,
      "loss": 5.2594,
      "mean_token_accuracy": 0.22080627381801604,
      "num_tokens": 82996104.0,
      "step": 44515
    },
    {
      "entropy": 6.032768583297729,
      "epoch": 3.9314729777463793,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003526997789085145,
      "loss": 5.3147,
      "mean_token_accuracy": 0.21043235808610916,
      "num_tokens": 83006770.0,
      "step": 44520
    },
    {
      "entropy": 5.908075332641602,
      "epoch": 3.9319145178382198,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003526702239844902,
      "loss": 5.1144,
      "mean_token_accuracy": 0.2353791907429695,
      "num_tokens": 83015954.0,
      "step": 44525
    },
    {
      "entropy": 6.016772317886352,
      "epoch": 3.9323560579300603,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035264066753895037,
      "loss": 5.1995,
      "mean_token_accuracy": 0.21948954910039903,
      "num_tokens": 83025448.0,
      "step": 44530
    },
    {
      "entropy": 5.8917820930480955,
      "epoch": 3.9327975980219003,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035261110957247394,
      "loss": 5.1983,
      "mean_token_accuracy": 0.23087005019187928,
      "num_tokens": 83034912.0,
      "step": 44535
    },
    {
      "entropy": 6.065963172912598,
      "epoch": 3.933239138113741,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035258155008564016,
      "loss": 5.2887,
      "mean_token_accuracy": 0.21845447570085524,
      "num_tokens": 83043574.0,
      "step": 44540
    },
    {
      "entropy": 6.027032709121704,
      "epoch": 3.9336806782055813,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003525519890790278,
      "loss": 5.2659,
      "mean_token_accuracy": 0.21860438585281372,
      "num_tokens": 83052473.0,
      "step": 44545
    },
    {
      "entropy": 5.987795114517212,
      "epoch": 3.9341222182974214,
      "grad_norm": 1.359375,
      "learning_rate": 0.00035252242655321617,
      "loss": 5.2056,
      "mean_token_accuracy": 0.2216299593448639,
      "num_tokens": 83060766.0,
      "step": 44550
    },
    {
      "entropy": 5.993598747253418,
      "epoch": 3.934563758389262,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003524928625087842,
      "loss": 5.2231,
      "mean_token_accuracy": 0.22068991363048554,
      "num_tokens": 83070011.0,
      "step": 44555
    },
    {
      "entropy": 6.08459882736206,
      "epoch": 3.935005298481102,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003524632969463112,
      "loss": 5.3743,
      "mean_token_accuracy": 0.2059927463531494,
      "num_tokens": 83079118.0,
      "step": 44560
    },
    {
      "entropy": 5.943555402755737,
      "epoch": 3.9354468385729424,
      "grad_norm": 1.40625,
      "learning_rate": 0.00035243372986637624,
      "loss": 5.1317,
      "mean_token_accuracy": 0.24136220067739486,
      "num_tokens": 83087474.0,
      "step": 44565
    },
    {
      "entropy": 5.957268190383911,
      "epoch": 3.935888378664783,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035240416126955853,
      "loss": 5.151,
      "mean_token_accuracy": 0.2309529885649681,
      "num_tokens": 83096497.0,
      "step": 44570
    },
    {
      "entropy": 5.973049068450928,
      "epoch": 3.936329918756623,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003523745911564374,
      "loss": 5.2081,
      "mean_token_accuracy": 0.22389961630105973,
      "num_tokens": 83105083.0,
      "step": 44575
    },
    {
      "entropy": 6.099007844924927,
      "epoch": 3.9367714588484635,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035234501952759195,
      "loss": 5.2883,
      "mean_token_accuracy": 0.212787064909935,
      "num_tokens": 83114260.0,
      "step": 44580
    },
    {
      "entropy": 5.981989240646362,
      "epoch": 3.9372129989403035,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035231544638360165,
      "loss": 5.2579,
      "mean_token_accuracy": 0.21572421491146088,
      "num_tokens": 83122037.0,
      "step": 44585
    },
    {
      "entropy": 6.099932861328125,
      "epoch": 3.937654539032144,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003522858717250456,
      "loss": 5.4179,
      "mean_token_accuracy": 0.20925847738981246,
      "num_tokens": 83131756.0,
      "step": 44590
    },
    {
      "entropy": 6.014395904541016,
      "epoch": 3.9380960791239845,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003522562955525034,
      "loss": 5.2352,
      "mean_token_accuracy": 0.2176181435585022,
      "num_tokens": 83140645.0,
      "step": 44595
    },
    {
      "entropy": 6.05191559791565,
      "epoch": 3.9385376192158246,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035222671786655426,
      "loss": 5.26,
      "mean_token_accuracy": 0.21710438132286072,
      "num_tokens": 83150054.0,
      "step": 44600
    },
    {
      "entropy": 5.930578231811523,
      "epoch": 3.938979159307665,
      "grad_norm": 1.375,
      "learning_rate": 0.0003521971386677776,
      "loss": 5.1453,
      "mean_token_accuracy": 0.2244912564754486,
      "num_tokens": 83159100.0,
      "step": 44605
    },
    {
      "entropy": 6.066023254394532,
      "epoch": 3.9394206993995056,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003521675579567529,
      "loss": 5.322,
      "mean_token_accuracy": 0.21517992317676543,
      "num_tokens": 83168046.0,
      "step": 44610
    },
    {
      "entropy": 6.099901151657105,
      "epoch": 3.9398622394913456,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003521379757340597,
      "loss": 5.2748,
      "mean_token_accuracy": 0.22708053588867189,
      "num_tokens": 83178413.0,
      "step": 44615
    },
    {
      "entropy": 6.083432674407959,
      "epoch": 3.940303779583186,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003521083920002774,
      "loss": 5.3235,
      "mean_token_accuracy": 0.21442606896162034,
      "num_tokens": 83188091.0,
      "step": 44620
    },
    {
      "entropy": 5.930837106704712,
      "epoch": 3.9407453196750266,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035207880675598545,
      "loss": 5.2401,
      "mean_token_accuracy": 0.22309821844100952,
      "num_tokens": 83199142.0,
      "step": 44625
    },
    {
      "entropy": 6.060051918029785,
      "epoch": 3.9411868597668667,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003520492200017637,
      "loss": 5.321,
      "mean_token_accuracy": 0.21405799984931945,
      "num_tokens": 83208840.0,
      "step": 44630
    },
    {
      "entropy": 6.026650238037109,
      "epoch": 3.941628399858707,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035201963173819145,
      "loss": 5.2841,
      "mean_token_accuracy": 0.21385698467493058,
      "num_tokens": 83218176.0,
      "step": 44635
    },
    {
      "entropy": 5.943852138519287,
      "epoch": 3.9420699399505477,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003519900419658484,
      "loss": 5.24,
      "mean_token_accuracy": 0.22436939030885697,
      "num_tokens": 83227848.0,
      "step": 44640
    },
    {
      "entropy": 6.07407774925232,
      "epoch": 3.9425114800423877,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035196045068531417,
      "loss": 5.333,
      "mean_token_accuracy": 0.21020565927028656,
      "num_tokens": 83237804.0,
      "step": 44645
    },
    {
      "entropy": 5.952978610992432,
      "epoch": 3.942953020134228,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003519308578971686,
      "loss": 5.1942,
      "mean_token_accuracy": 0.2247151702642441,
      "num_tokens": 83246257.0,
      "step": 44650
    },
    {
      "entropy": 6.088982057571411,
      "epoch": 3.9433945602260687,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035190126360199126,
      "loss": 5.3294,
      "mean_token_accuracy": 0.207074174284935,
      "num_tokens": 83256247.0,
      "step": 44655
    },
    {
      "entropy": 6.018608427047729,
      "epoch": 3.9438361003179088,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035187166780036185,
      "loss": 5.3012,
      "mean_token_accuracy": 0.21209842562675477,
      "num_tokens": 83265481.0,
      "step": 44660
    },
    {
      "entropy": 6.002248811721802,
      "epoch": 3.9442776404097493,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00035184207049286014,
      "loss": 5.1719,
      "mean_token_accuracy": 0.22124868333339692,
      "num_tokens": 83274455.0,
      "step": 44665
    },
    {
      "entropy": 6.11311092376709,
      "epoch": 3.9447191805015898,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003518124716800661,
      "loss": 5.4339,
      "mean_token_accuracy": 0.20586782693862915,
      "num_tokens": 83284017.0,
      "step": 44670
    },
    {
      "entropy": 6.120937824249268,
      "epoch": 3.94516072059343,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035178287136255934,
      "loss": 5.3433,
      "mean_token_accuracy": 0.21519638150930404,
      "num_tokens": 83292577.0,
      "step": 44675
    },
    {
      "entropy": 6.021289968490601,
      "epoch": 3.9456022606852703,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003517532695409198,
      "loss": 5.2033,
      "mean_token_accuracy": 0.22385138273239136,
      "num_tokens": 83302384.0,
      "step": 44680
    },
    {
      "entropy": 5.948854160308838,
      "epoch": 3.946043800777111,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003517236662157274,
      "loss": 5.1659,
      "mean_token_accuracy": 0.23019636869430543,
      "num_tokens": 83311184.0,
      "step": 44685
    },
    {
      "entropy": 6.021206569671631,
      "epoch": 3.946485340868951,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035169406138756197,
      "loss": 5.3404,
      "mean_token_accuracy": 0.21144280731678008,
      "num_tokens": 83321400.0,
      "step": 44690
    },
    {
      "entropy": 6.0370580673217775,
      "epoch": 3.9469268809607914,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003516644550570035,
      "loss": 5.3325,
      "mean_token_accuracy": 0.2019073784351349,
      "num_tokens": 83330475.0,
      "step": 44695
    },
    {
      "entropy": 6.036035060882568,
      "epoch": 3.9473684210526314,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000351634847224632,
      "loss": 5.316,
      "mean_token_accuracy": 0.21722423583269118,
      "num_tokens": 83340104.0,
      "step": 44700
    },
    {
      "entropy": 6.0337748527526855,
      "epoch": 3.947809961144472,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035160523789102747,
      "loss": 5.1944,
      "mean_token_accuracy": 0.22120240926742554,
      "num_tokens": 83349066.0,
      "step": 44705
    },
    {
      "entropy": 6.045544624328613,
      "epoch": 3.9482515012363124,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035157562705676976,
      "loss": 5.3215,
      "mean_token_accuracy": 0.21237749457359315,
      "num_tokens": 83358144.0,
      "step": 44710
    },
    {
      "entropy": 6.026306438446045,
      "epoch": 3.9486930413281525,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035154601472243914,
      "loss": 5.2436,
      "mean_token_accuracy": 0.21816513389348985,
      "num_tokens": 83366876.0,
      "step": 44715
    },
    {
      "entropy": 6.110620784759521,
      "epoch": 3.949134581419993,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003515164008886157,
      "loss": 5.3407,
      "mean_token_accuracy": 0.21266090869903564,
      "num_tokens": 83375317.0,
      "step": 44720
    },
    {
      "entropy": 5.952551746368409,
      "epoch": 3.949576121511833,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003514867855558793,
      "loss": 5.1936,
      "mean_token_accuracy": 0.22598046213388442,
      "num_tokens": 83384433.0,
      "step": 44725
    },
    {
      "entropy": 6.031879520416259,
      "epoch": 3.9500176616036735,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035145716872481044,
      "loss": 5.2691,
      "mean_token_accuracy": 0.21957691311836242,
      "num_tokens": 83392876.0,
      "step": 44730
    },
    {
      "entropy": 5.990634298324585,
      "epoch": 3.950459201695514,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000351427550395989,
      "loss": 5.2025,
      "mean_token_accuracy": 0.2231363132596016,
      "num_tokens": 83401913.0,
      "step": 44735
    },
    {
      "entropy": 5.997908592224121,
      "epoch": 3.950900741787354,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003513979305699954,
      "loss": 5.2422,
      "mean_token_accuracy": 0.2139204576611519,
      "num_tokens": 83411399.0,
      "step": 44740
    },
    {
      "entropy": 6.034974193572998,
      "epoch": 3.9513422818791946,
      "grad_norm": 1.375,
      "learning_rate": 0.0003513683092474098,
      "loss": 5.3118,
      "mean_token_accuracy": 0.21970219016075135,
      "num_tokens": 83420269.0,
      "step": 44745
    },
    {
      "entropy": 5.964357137680054,
      "epoch": 3.951783821971035,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035133868642881237,
      "loss": 5.2608,
      "mean_token_accuracy": 0.21721086651086807,
      "num_tokens": 83429975.0,
      "step": 44750
    },
    {
      "entropy": 5.996291160583496,
      "epoch": 3.952225362062875,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003513090621147836,
      "loss": 5.2381,
      "mean_token_accuracy": 0.21875801980495452,
      "num_tokens": 83438845.0,
      "step": 44755
    },
    {
      "entropy": 6.012143087387085,
      "epoch": 3.9526669021547156,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035127943630590365,
      "loss": 5.3501,
      "mean_token_accuracy": 0.2130981296300888,
      "num_tokens": 83447595.0,
      "step": 44760
    },
    {
      "entropy": 6.014147663116455,
      "epoch": 3.953108442246556,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00035124980900275296,
      "loss": 5.2413,
      "mean_token_accuracy": 0.21881105601787568,
      "num_tokens": 83457072.0,
      "step": 44765
    },
    {
      "entropy": 5.9869708061218265,
      "epoch": 3.953549982338396,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003512201802059118,
      "loss": 5.3152,
      "mean_token_accuracy": 0.21336784809827805,
      "num_tokens": 83467435.0,
      "step": 44770
    },
    {
      "entropy": 6.048624324798584,
      "epoch": 3.9539915224302367,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035119054991596067,
      "loss": 5.234,
      "mean_token_accuracy": 0.2205306589603424,
      "num_tokens": 83475968.0,
      "step": 44775
    },
    {
      "entropy": 6.07154598236084,
      "epoch": 3.954433062522077,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035116091813348004,
      "loss": 5.2404,
      "mean_token_accuracy": 0.21948420852422715,
      "num_tokens": 83485199.0,
      "step": 44780
    },
    {
      "entropy": 6.062142276763916,
      "epoch": 3.954874602613917,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003511312848590503,
      "loss": 5.3242,
      "mean_token_accuracy": 0.21755024045705795,
      "num_tokens": 83495016.0,
      "step": 44785
    },
    {
      "entropy": 6.0773584842681885,
      "epoch": 3.9553161427057577,
      "grad_norm": 1.171875,
      "learning_rate": 0.000351101650093252,
      "loss": 5.3666,
      "mean_token_accuracy": 0.21180338263511658,
      "num_tokens": 83504927.0,
      "step": 44790
    },
    {
      "entropy": 6.0833676815032955,
      "epoch": 3.955757682797598,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003510720138366657,
      "loss": 5.2395,
      "mean_token_accuracy": 0.22078608721494675,
      "num_tokens": 83513666.0,
      "step": 44795
    },
    {
      "entropy": 6.001941156387329,
      "epoch": 3.9561992228894383,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003510423760898719,
      "loss": 5.1946,
      "mean_token_accuracy": 0.22983457446098327,
      "num_tokens": 83523242.0,
      "step": 44800
    },
    {
      "entropy": 6.020172166824341,
      "epoch": 3.9566407629812788,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003510127368534512,
      "loss": 5.253,
      "mean_token_accuracy": 0.21851015836000443,
      "num_tokens": 83532243.0,
      "step": 44805
    },
    {
      "entropy": 6.099425792694092,
      "epoch": 3.9570823030731193,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003509830961279842,
      "loss": 5.3718,
      "mean_token_accuracy": 0.20498846024274825,
      "num_tokens": 83541326.0,
      "step": 44810
    },
    {
      "entropy": 6.08664493560791,
      "epoch": 3.9575238431649593,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003509534539140515,
      "loss": 5.3559,
      "mean_token_accuracy": 0.20942816287279128,
      "num_tokens": 83549560.0,
      "step": 44815
    },
    {
      "entropy": 6.009989452362061,
      "epoch": 3.9579653832568,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003509238102122339,
      "loss": 5.2446,
      "mean_token_accuracy": 0.21834950596094133,
      "num_tokens": 83559115.0,
      "step": 44820
    },
    {
      "entropy": 6.050127220153809,
      "epoch": 3.9584069233486403,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035089416502311207,
      "loss": 5.3452,
      "mean_token_accuracy": 0.2087724506855011,
      "num_tokens": 83569053.0,
      "step": 44825
    },
    {
      "entropy": 6.086886024475097,
      "epoch": 3.9588484634404804,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035086451834726663,
      "loss": 5.3399,
      "mean_token_accuracy": 0.20802310556173326,
      "num_tokens": 83578429.0,
      "step": 44830
    },
    {
      "entropy": 5.959472942352295,
      "epoch": 3.959290003532321,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003508348701852785,
      "loss": 5.2308,
      "mean_token_accuracy": 0.22347248643636702,
      "num_tokens": 83588844.0,
      "step": 44835
    },
    {
      "entropy": 5.940687465667724,
      "epoch": 3.959731543624161,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003508052205377284,
      "loss": 5.2209,
      "mean_token_accuracy": 0.21629827171564103,
      "num_tokens": 83598184.0,
      "step": 44840
    },
    {
      "entropy": 5.999181270599365,
      "epoch": 3.9601730837160014,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035077556940519704,
      "loss": 5.2387,
      "mean_token_accuracy": 0.21486677527427672,
      "num_tokens": 83607970.0,
      "step": 44845
    },
    {
      "entropy": 6.020187568664551,
      "epoch": 3.960614623807842,
      "grad_norm": 1.25,
      "learning_rate": 0.00035074591678826544,
      "loss": 5.317,
      "mean_token_accuracy": 0.2135197788476944,
      "num_tokens": 83617069.0,
      "step": 44850
    },
    {
      "entropy": 6.143280267715454,
      "epoch": 3.961056163899682,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003507162626875144,
      "loss": 5.4156,
      "mean_token_accuracy": 0.20730241686105727,
      "num_tokens": 83627102.0,
      "step": 44855
    },
    {
      "entropy": 6.106833171844483,
      "epoch": 3.9614977039915225,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035068660710352485,
      "loss": 5.3237,
      "mean_token_accuracy": 0.2148787945508957,
      "num_tokens": 83636308.0,
      "step": 44860
    },
    {
      "entropy": 6.077988815307617,
      "epoch": 3.9619392440833625,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035065695003687764,
      "loss": 5.2885,
      "mean_token_accuracy": 0.21327048689126968,
      "num_tokens": 83646085.0,
      "step": 44865
    },
    {
      "entropy": 6.068980026245117,
      "epoch": 3.962380784175203,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035062729148815393,
      "loss": 5.224,
      "mean_token_accuracy": 0.21774369329214097,
      "num_tokens": 83655543.0,
      "step": 44870
    },
    {
      "entropy": 6.080832004547119,
      "epoch": 3.9628223242670435,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003505976314579345,
      "loss": 5.3753,
      "mean_token_accuracy": 0.2079795330762863,
      "num_tokens": 83665157.0,
      "step": 44875
    },
    {
      "entropy": 6.070641899108887,
      "epoch": 3.9632638643588836,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003505679699468005,
      "loss": 5.3255,
      "mean_token_accuracy": 0.21322414875030518,
      "num_tokens": 83674429.0,
      "step": 44880
    },
    {
      "entropy": 6.069449520111084,
      "epoch": 3.963705404450724,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003505383069553329,
      "loss": 5.3669,
      "mean_token_accuracy": 0.21035897135734558,
      "num_tokens": 83682970.0,
      "step": 44885
    },
    {
      "entropy": 6.014232444763183,
      "epoch": 3.9641469445425646,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003505086424841129,
      "loss": 5.3011,
      "mean_token_accuracy": 0.21794302463531495,
      "num_tokens": 83692705.0,
      "step": 44890
    },
    {
      "entropy": 6.024931859970093,
      "epoch": 3.9645884846344046,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035047897653372144,
      "loss": 5.2558,
      "mean_token_accuracy": 0.22697740644216538,
      "num_tokens": 83701747.0,
      "step": 44895
    },
    {
      "entropy": 6.028133726119995,
      "epoch": 3.965030024726245,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035044930910473987,
      "loss": 5.3286,
      "mean_token_accuracy": 0.2156145840883255,
      "num_tokens": 83711937.0,
      "step": 44900
    },
    {
      "entropy": 6.104268646240234,
      "epoch": 3.9654715648180856,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035041964019774923,
      "loss": 5.3618,
      "mean_token_accuracy": 0.21044577658176422,
      "num_tokens": 83721516.0,
      "step": 44905
    },
    {
      "entropy": 6.0260901927948,
      "epoch": 3.9659131049099257,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035038996981333066,
      "loss": 5.2059,
      "mean_token_accuracy": 0.22535073906183242,
      "num_tokens": 83730666.0,
      "step": 44910
    },
    {
      "entropy": 5.98084568977356,
      "epoch": 3.966354645001766,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035036029795206555,
      "loss": 5.241,
      "mean_token_accuracy": 0.2194148987531662,
      "num_tokens": 83739253.0,
      "step": 44915
    },
    {
      "entropy": 6.061598443984986,
      "epoch": 3.9667961850936067,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000350330624614535,
      "loss": 5.2219,
      "mean_token_accuracy": 0.2202388420701027,
      "num_tokens": 83747556.0,
      "step": 44920
    },
    {
      "entropy": 5.911270523071289,
      "epoch": 3.9672377251854467,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035030094980132033,
      "loss": 5.222,
      "mean_token_accuracy": 0.22914663404226304,
      "num_tokens": 83757040.0,
      "step": 44925
    },
    {
      "entropy": 5.965656566619873,
      "epoch": 3.967679265277287,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035027127351300286,
      "loss": 5.2409,
      "mean_token_accuracy": 0.2216604918241501,
      "num_tokens": 83766372.0,
      "step": 44930
    },
    {
      "entropy": 6.044836568832397,
      "epoch": 3.9681208053691277,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035024159575016404,
      "loss": 5.2695,
      "mean_token_accuracy": 0.2152898982167244,
      "num_tokens": 83775498.0,
      "step": 44935
    },
    {
      "entropy": 6.015338516235351,
      "epoch": 3.9685623454609678,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003502119165133851,
      "loss": 5.2074,
      "mean_token_accuracy": 0.22254157662391663,
      "num_tokens": 83784634.0,
      "step": 44940
    },
    {
      "entropy": 5.955641412734986,
      "epoch": 3.9690038855528083,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003501822358032475,
      "loss": 5.1938,
      "mean_token_accuracy": 0.2255952090024948,
      "num_tokens": 83793554.0,
      "step": 44945
    },
    {
      "entropy": 6.035187387466431,
      "epoch": 3.9694454256446488,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035015255362033274,
      "loss": 5.3651,
      "mean_token_accuracy": 0.20666819512844087,
      "num_tokens": 83804411.0,
      "step": 44950
    },
    {
      "entropy": 6.036164474487305,
      "epoch": 3.969886965736489,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035012286996522206,
      "loss": 5.2825,
      "mean_token_accuracy": 0.2131996527314186,
      "num_tokens": 83814386.0,
      "step": 44955
    },
    {
      "entropy": 6.142437314987182,
      "epoch": 3.9703285058283293,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003500931848384972,
      "loss": 5.3877,
      "mean_token_accuracy": 0.2035341590642929,
      "num_tokens": 83824127.0,
      "step": 44960
    },
    {
      "entropy": 5.929211521148682,
      "epoch": 3.97077004592017,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003500634982407395,
      "loss": 5.083,
      "mean_token_accuracy": 0.2289426252245903,
      "num_tokens": 83832641.0,
      "step": 44965
    },
    {
      "entropy": 6.040754938125611,
      "epoch": 3.97121158601201,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035003381017253055,
      "loss": 5.3387,
      "mean_token_accuracy": 0.20700230598449706,
      "num_tokens": 83842408.0,
      "step": 44970
    },
    {
      "entropy": 6.001801872253418,
      "epoch": 3.9716531261038504,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003500041206344519,
      "loss": 5.2479,
      "mean_token_accuracy": 0.21914913654327392,
      "num_tokens": 83851684.0,
      "step": 44975
    },
    {
      "entropy": 6.033646583557129,
      "epoch": 3.9720946661956904,
      "grad_norm": 1.3125,
      "learning_rate": 0.00034997442962708523,
      "loss": 5.2016,
      "mean_token_accuracy": 0.22827744781970977,
      "num_tokens": 83859894.0,
      "step": 44980
    },
    {
      "entropy": 6.07735800743103,
      "epoch": 3.972536206287531,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034994473715101214,
      "loss": 5.3804,
      "mean_token_accuracy": 0.1992912396788597,
      "num_tokens": 83869642.0,
      "step": 44985
    },
    {
      "entropy": 6.100507020950317,
      "epoch": 3.9729777463793714,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003499150432068143,
      "loss": 5.356,
      "mean_token_accuracy": 0.20908645391464234,
      "num_tokens": 83880672.0,
      "step": 44990
    },
    {
      "entropy": 6.020199728012085,
      "epoch": 3.9734192864712115,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034988534779507326,
      "loss": 5.2579,
      "mean_token_accuracy": 0.23408968299627303,
      "num_tokens": 83890375.0,
      "step": 44995
    },
    {
      "entropy": 6.052239990234375,
      "epoch": 3.973860826563052,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003498556509163709,
      "loss": 5.2828,
      "mean_token_accuracy": 0.22154953330755234,
      "num_tokens": 83899859.0,
      "step": 45000
    },
    {
      "epoch": 3.973860826563052,
      "eval_entropy": 5.737819992294398,
      "eval_loss": 5.321580410003662,
      "eval_mean_token_accuracy": 0.22393635238182233,
      "eval_num_tokens": 83899859.0,
      "eval_runtime": 20.2168,
      "eval_samples_per_second": 1746.818,
      "eval_steps_per_second": 218.383,
      "step": 45000
    },
    {
      "entropy": 6.025277805328369,
      "epoch": 3.974302366654892,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034982595257128894,
      "loss": 5.3053,
      "mean_token_accuracy": 0.21770953387022018,
      "num_tokens": 83908413.0,
      "step": 45005
    },
    {
      "entropy": 6.121549654006958,
      "epoch": 3.9747439067467325,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003497962527604091,
      "loss": 5.3905,
      "mean_token_accuracy": 0.213508939743042,
      "num_tokens": 83918450.0,
      "step": 45010
    },
    {
      "entropy": 6.044871854782104,
      "epoch": 3.975185446838573,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034976655148431324,
      "loss": 5.2761,
      "mean_token_accuracy": 0.21203491985797882,
      "num_tokens": 83927361.0,
      "step": 45015
    },
    {
      "entropy": 6.047916507720947,
      "epoch": 3.975626986930413,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00034973684874358313,
      "loss": 5.3785,
      "mean_token_accuracy": 0.2134945034980774,
      "num_tokens": 83937616.0,
      "step": 45020
    },
    {
      "entropy": 6.031595516204834,
      "epoch": 3.9760685270222536,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003497071445388007,
      "loss": 5.2748,
      "mean_token_accuracy": 0.2146395891904831,
      "num_tokens": 83947729.0,
      "step": 45025
    },
    {
      "entropy": 5.99030122756958,
      "epoch": 3.976510067114094,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003496774388705477,
      "loss": 5.2229,
      "mean_token_accuracy": 0.22388364225625992,
      "num_tokens": 83956942.0,
      "step": 45030
    },
    {
      "entropy": 6.0642862796783445,
      "epoch": 3.976951607205934,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034964773173940623,
      "loss": 5.3243,
      "mean_token_accuracy": 0.20853824466466903,
      "num_tokens": 83966224.0,
      "step": 45035
    },
    {
      "entropy": 6.0108174800872805,
      "epoch": 3.9773931472977746,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00034961802314595813,
      "loss": 5.2256,
      "mean_token_accuracy": 0.22425764352083205,
      "num_tokens": 83976234.0,
      "step": 45040
    },
    {
      "entropy": 6.066013669967651,
      "epoch": 3.977834687389615,
      "grad_norm": 1.28125,
      "learning_rate": 0.00034958831309078535,
      "loss": 5.2766,
      "mean_token_accuracy": 0.21524969786405562,
      "num_tokens": 83985340.0,
      "step": 45045
    },
    {
      "entropy": 5.97447509765625,
      "epoch": 3.978276227481455,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034955860157447,
      "loss": 5.2557,
      "mean_token_accuracy": 0.22435519099235535,
      "num_tokens": 83994090.0,
      "step": 45050
    },
    {
      "entropy": 5.985933399200439,
      "epoch": 3.9787177675732956,
      "grad_norm": 1.453125,
      "learning_rate": 0.00034952888859759404,
      "loss": 5.2757,
      "mean_token_accuracy": 0.2139851778745651,
      "num_tokens": 84003146.0,
      "step": 45055
    },
    {
      "entropy": 6.067541456222534,
      "epoch": 3.979159307665136,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003494991741607395,
      "loss": 5.285,
      "mean_token_accuracy": 0.21749711334705352,
      "num_tokens": 84012620.0,
      "step": 45060
    },
    {
      "entropy": 6.002041244506836,
      "epoch": 3.979600847756976,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003494694582644885,
      "loss": 5.3141,
      "mean_token_accuracy": 0.218638314306736,
      "num_tokens": 84022592.0,
      "step": 45065
    },
    {
      "entropy": 6.076465797424317,
      "epoch": 3.9800423878488167,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034943974090942317,
      "loss": 5.2718,
      "mean_token_accuracy": 0.21307071447372436,
      "num_tokens": 84032289.0,
      "step": 45070
    },
    {
      "entropy": 6.010618829727173,
      "epoch": 3.980483927940657,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003494100220961257,
      "loss": 5.2484,
      "mean_token_accuracy": 0.21668341010808945,
      "num_tokens": 84040450.0,
      "step": 45075
    },
    {
      "entropy": 6.046470689773559,
      "epoch": 3.9809254680324972,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034938030182517823,
      "loss": 5.3262,
      "mean_token_accuracy": 0.2079145535826683,
      "num_tokens": 84049998.0,
      "step": 45080
    },
    {
      "entropy": 6.065830326080322,
      "epoch": 3.9813670081243377,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034935058009716286,
      "loss": 5.2475,
      "mean_token_accuracy": 0.21557382196187974,
      "num_tokens": 84059881.0,
      "step": 45085
    },
    {
      "entropy": 6.034912061691284,
      "epoch": 3.9818085482161782,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000349320856912662,
      "loss": 5.2412,
      "mean_token_accuracy": 0.21545311212539672,
      "num_tokens": 84069138.0,
      "step": 45090
    },
    {
      "entropy": 6.085068035125732,
      "epoch": 3.9822500883080183,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034929113227225776,
      "loss": 5.3147,
      "mean_token_accuracy": 0.21417201608419417,
      "num_tokens": 84077596.0,
      "step": 45095
    },
    {
      "entropy": 6.014547538757324,
      "epoch": 3.982691628399859,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003492614061765326,
      "loss": 5.2623,
      "mean_token_accuracy": 0.2171581655740738,
      "num_tokens": 84087172.0,
      "step": 45100
    },
    {
      "entropy": 6.078246831893921,
      "epoch": 3.9831331684916993,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034923167862606857,
      "loss": 5.3369,
      "mean_token_accuracy": 0.21160465180873872,
      "num_tokens": 84097025.0,
      "step": 45105
    },
    {
      "entropy": 6.011866569519043,
      "epoch": 3.9835747085835393,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003492019496214483,
      "loss": 5.2422,
      "mean_token_accuracy": 0.21790120303630828,
      "num_tokens": 84105478.0,
      "step": 45110
    },
    {
      "entropy": 5.9654128074646,
      "epoch": 3.98401624867538,
      "grad_norm": 1.34375,
      "learning_rate": 0.000349172219163254,
      "loss": 5.1626,
      "mean_token_accuracy": 0.23029535561800002,
      "num_tokens": 84114441.0,
      "step": 45115
    },
    {
      "entropy": 5.984992742538452,
      "epoch": 3.98445778876722,
      "grad_norm": 1.359375,
      "learning_rate": 0.000349142487252068,
      "loss": 5.2684,
      "mean_token_accuracy": 0.2146920755505562,
      "num_tokens": 84123155.0,
      "step": 45120
    },
    {
      "entropy": 5.991797637939453,
      "epoch": 3.9848993288590604,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00034911275388847297,
      "loss": 5.2199,
      "mean_token_accuracy": 0.2200811356306076,
      "num_tokens": 84131447.0,
      "step": 45125
    },
    {
      "entropy": 6.0094624042510985,
      "epoch": 3.985340868950901,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003490830190730512,
      "loss": 5.2377,
      "mean_token_accuracy": 0.21828147768974304,
      "num_tokens": 84140641.0,
      "step": 45130
    },
    {
      "entropy": 5.936993503570557,
      "epoch": 3.985782409042741,
      "grad_norm": 1.328125,
      "learning_rate": 0.00034905328280638524,
      "loss": 5.172,
      "mean_token_accuracy": 0.22933163791894912,
      "num_tokens": 84149548.0,
      "step": 45135
    },
    {
      "entropy": 5.970210170745849,
      "epoch": 3.9862239491345814,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003490235450890576,
      "loss": 5.1693,
      "mean_token_accuracy": 0.22185318022966385,
      "num_tokens": 84158197.0,
      "step": 45140
    },
    {
      "entropy": 6.108661890029907,
      "epoch": 3.9866654892264215,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034899380592165073,
      "loss": 5.3637,
      "mean_token_accuracy": 0.20079794973134996,
      "num_tokens": 84168075.0,
      "step": 45145
    },
    {
      "entropy": 6.019127655029297,
      "epoch": 3.987107029318262,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034896406530474734,
      "loss": 5.2105,
      "mean_token_accuracy": 0.22543721348047258,
      "num_tokens": 84177784.0,
      "step": 45150
    },
    {
      "entropy": 5.992271757125854,
      "epoch": 3.9875485694101025,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00034893432323893005,
      "loss": 5.1556,
      "mean_token_accuracy": 0.22994588911533356,
      "num_tokens": 84186684.0,
      "step": 45155
    },
    {
      "entropy": 6.038823556900025,
      "epoch": 3.9879901095019425,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00034890457972478134,
      "loss": 5.3535,
      "mean_token_accuracy": 0.21070061326026918,
      "num_tokens": 84195519.0,
      "step": 45160
    },
    {
      "entropy": 6.015270662307739,
      "epoch": 3.988431649593783,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003488748347628839,
      "loss": 5.3849,
      "mean_token_accuracy": 0.20609830915927888,
      "num_tokens": 84205343.0,
      "step": 45165
    },
    {
      "entropy": 6.059975242614746,
      "epoch": 3.9888731896856235,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003488450883538205,
      "loss": 5.2456,
      "mean_token_accuracy": 0.22786748856306077,
      "num_tokens": 84214881.0,
      "step": 45170
    },
    {
      "entropy": 6.03755292892456,
      "epoch": 3.9893147297774636,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003488153404981738,
      "loss": 5.2633,
      "mean_token_accuracy": 0.22356326580047609,
      "num_tokens": 84224892.0,
      "step": 45175
    },
    {
      "entropy": 6.097801589965821,
      "epoch": 3.989756269869304,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034878559119652663,
      "loss": 5.2158,
      "mean_token_accuracy": 0.22056454718112944,
      "num_tokens": 84234209.0,
      "step": 45180
    },
    {
      "entropy": 6.1007520198822025,
      "epoch": 3.9901978099611446,
      "grad_norm": 2.203125,
      "learning_rate": 0.0003487558404494616,
      "loss": 5.3344,
      "mean_token_accuracy": 0.21001892834901809,
      "num_tokens": 84245091.0,
      "step": 45185
    },
    {
      "entropy": 6.007193613052368,
      "epoch": 3.9906393500529846,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003487260882575616,
      "loss": 5.1912,
      "mean_token_accuracy": 0.22218522280454636,
      "num_tokens": 84254141.0,
      "step": 45190
    },
    {
      "entropy": 5.9796302795410154,
      "epoch": 3.991080890144825,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003486963346214096,
      "loss": 5.2303,
      "mean_token_accuracy": 0.21600744873285294,
      "num_tokens": 84263219.0,
      "step": 45195
    },
    {
      "entropy": 6.002531051635742,
      "epoch": 3.9915224302366656,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00034866657954158825,
      "loss": 5.2462,
      "mean_token_accuracy": 0.21916204541921616,
      "num_tokens": 84271707.0,
      "step": 45200
    },
    {
      "entropy": 5.993917417526245,
      "epoch": 3.9919639703285057,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003486368230186805,
      "loss": 5.1877,
      "mean_token_accuracy": 0.22008079737424852,
      "num_tokens": 84281040.0,
      "step": 45205
    },
    {
      "entropy": 6.106966638565064,
      "epoch": 3.992405510420346,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034860706505326927,
      "loss": 5.345,
      "mean_token_accuracy": 0.21615461260080338,
      "num_tokens": 84290593.0,
      "step": 45210
    },
    {
      "entropy": 6.0404784202575685,
      "epoch": 3.9928470505121867,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034857730564593747,
      "loss": 5.2831,
      "mean_token_accuracy": 0.20784064680337905,
      "num_tokens": 84298820.0,
      "step": 45215
    },
    {
      "entropy": 6.0067566394805905,
      "epoch": 3.9932885906040267,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003485475447972681,
      "loss": 5.3015,
      "mean_token_accuracy": 0.211746446788311,
      "num_tokens": 84307861.0,
      "step": 45220
    },
    {
      "entropy": 6.1105663776397705,
      "epoch": 3.9937301306958672,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003485177825078442,
      "loss": 5.2775,
      "mean_token_accuracy": 0.21714983135461807,
      "num_tokens": 84316539.0,
      "step": 45225
    },
    {
      "entropy": 6.059420919418335,
      "epoch": 3.9941716707877077,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003484880187782487,
      "loss": 5.3682,
      "mean_token_accuracy": 0.21449725180864335,
      "num_tokens": 84326288.0,
      "step": 45230
    },
    {
      "entropy": 6.038682842254639,
      "epoch": 3.994613210879548,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003484582536090647,
      "loss": 5.3078,
      "mean_token_accuracy": 0.20961325913667678,
      "num_tokens": 84336573.0,
      "step": 45235
    },
    {
      "entropy": 6.025557804107666,
      "epoch": 3.9950547509713883,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034842848700087525,
      "loss": 5.2868,
      "mean_token_accuracy": 0.22166690230369568,
      "num_tokens": 84346136.0,
      "step": 45240
    },
    {
      "entropy": 6.0119000434875485,
      "epoch": 3.995496291063229,
      "grad_norm": 1.25,
      "learning_rate": 0.0003483987189542636,
      "loss": 5.2862,
      "mean_token_accuracy": 0.21793947070837022,
      "num_tokens": 84355445.0,
      "step": 45245
    },
    {
      "entropy": 6.048709297180176,
      "epoch": 3.995937831155069,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003483689494698127,
      "loss": 5.2908,
      "mean_token_accuracy": 0.21745062470436097,
      "num_tokens": 84365383.0,
      "step": 45250
    },
    {
      "entropy": 5.956735849380493,
      "epoch": 3.9963793712469093,
      "grad_norm": 1.25,
      "learning_rate": 0.00034833917854810585,
      "loss": 5.2016,
      "mean_token_accuracy": 0.22716621607542037,
      "num_tokens": 84375014.0,
      "step": 45255
    },
    {
      "entropy": 6.090403747558594,
      "epoch": 3.9968209113387494,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034830940618972627,
      "loss": 5.3578,
      "mean_token_accuracy": 0.21063620448112488,
      "num_tokens": 84384322.0,
      "step": 45260
    },
    {
      "entropy": 6.073917245864868,
      "epoch": 3.99726245143059,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003482796323952571,
      "loss": 5.3852,
      "mean_token_accuracy": 0.21193755716085433,
      "num_tokens": 84395391.0,
      "step": 45265
    },
    {
      "entropy": 6.068159103393555,
      "epoch": 3.9977039915224304,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003482498571652816,
      "loss": 5.2906,
      "mean_token_accuracy": 0.2189597263932228,
      "num_tokens": 84405321.0,
      "step": 45270
    },
    {
      "entropy": 6.015040922164917,
      "epoch": 3.9981455316142704,
      "grad_norm": 1.359375,
      "learning_rate": 0.00034822008050038294,
      "loss": 5.2419,
      "mean_token_accuracy": 0.21862588673830033,
      "num_tokens": 84415022.0,
      "step": 45275
    },
    {
      "entropy": 6.072760152816772,
      "epoch": 3.998587071706111,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003481903024011446,
      "loss": 5.2766,
      "mean_token_accuracy": 0.21619438529014587,
      "num_tokens": 84423552.0,
      "step": 45280
    },
    {
      "entropy": 6.00975489616394,
      "epoch": 3.999028611797951,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034816052286814995,
      "loss": 5.1428,
      "mean_token_accuracy": 0.22803988456726074,
      "num_tokens": 84433038.0,
      "step": 45285
    },
    {
      "entropy": 6.058194923400879,
      "epoch": 3.9994701518897915,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003481307419019822,
      "loss": 5.2919,
      "mean_token_accuracy": 0.21641863584518434,
      "num_tokens": 84441875.0,
      "step": 45290
    },
    {
      "entropy": 6.071742105484009,
      "epoch": 3.999911691981632,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034810095950322487,
      "loss": 5.3002,
      "mean_token_accuracy": 0.21755418479442595,
      "num_tokens": 84451255.0,
      "step": 45295
    },
    {
      "entropy": 5.980913686752319,
      "epoch": 4.000353232073472,
      "grad_norm": 1.375,
      "learning_rate": 0.0003480711756724612,
      "loss": 5.166,
      "mean_token_accuracy": 0.22997811585664749,
      "num_tokens": 84459507.0,
      "step": 45300
    },
    {
      "entropy": 6.056581735610962,
      "epoch": 4.000794772165313,
      "grad_norm": 1.3125,
      "learning_rate": 0.00034804139041027485,
      "loss": 5.2804,
      "mean_token_accuracy": 0.21484961211681367,
      "num_tokens": 84468620.0,
      "step": 45305
    },
    {
      "entropy": 5.9949616432189945,
      "epoch": 4.001236312257153,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00034801160371724917,
      "loss": 5.2824,
      "mean_token_accuracy": 0.21566374897956847,
      "num_tokens": 84477761.0,
      "step": 45310
    },
    {
      "entropy": 6.03219780921936,
      "epoch": 4.001677852348993,
      "grad_norm": 1.359375,
      "learning_rate": 0.00034798181559396766,
      "loss": 5.2398,
      "mean_token_accuracy": 0.22175712138414383,
      "num_tokens": 84487102.0,
      "step": 45315
    },
    {
      "entropy": 6.066239356994629,
      "epoch": 4.002119392440834,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034795202604101397,
      "loss": 5.1875,
      "mean_token_accuracy": 0.22394466996192933,
      "num_tokens": 84496394.0,
      "step": 45320
    },
    {
      "entropy": 6.032477951049804,
      "epoch": 4.002560932532674,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00034792223505897155,
      "loss": 5.2701,
      "mean_token_accuracy": 0.20981077253818511,
      "num_tokens": 84505952.0,
      "step": 45325
    },
    {
      "entropy": 6.013494491577148,
      "epoch": 4.003002472624514,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034789244264842403,
      "loss": 5.2265,
      "mean_token_accuracy": 0.22526264041662217,
      "num_tokens": 84515106.0,
      "step": 45330
    },
    {
      "entropy": 6.021367168426513,
      "epoch": 4.003444012716355,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003478626488099549,
      "loss": 5.1864,
      "mean_token_accuracy": 0.22498246133327485,
      "num_tokens": 84525567.0,
      "step": 45335
    },
    {
      "entropy": 5.967926263809204,
      "epoch": 4.003885552808195,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003478328535441481,
      "loss": 5.2023,
      "mean_token_accuracy": 0.2276553213596344,
      "num_tokens": 84534547.0,
      "step": 45340
    },
    {
      "entropy": 5.89364800453186,
      "epoch": 4.004327092900035,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00034780305685158696,
      "loss": 5.1145,
      "mean_token_accuracy": 0.23477641344070435,
      "num_tokens": 84544672.0,
      "step": 45345
    },
    {
      "entropy": 6.0797497749328615,
      "epoch": 4.004768632991875,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034777325873285546,
      "loss": 5.1793,
      "mean_token_accuracy": 0.22567088752985,
      "num_tokens": 84553623.0,
      "step": 45350
    },
    {
      "entropy": 6.002631616592407,
      "epoch": 4.005210173083716,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003477434591885371,
      "loss": 5.2323,
      "mean_token_accuracy": 0.22757741212844848,
      "num_tokens": 84563708.0,
      "step": 45355
    },
    {
      "entropy": 6.020001077651978,
      "epoch": 4.005651713175556,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003477136582192159,
      "loss": 5.2559,
      "mean_token_accuracy": 0.21594179719686507,
      "num_tokens": 84572800.0,
      "step": 45360
    },
    {
      "entropy": 5.975028562545776,
      "epoch": 4.006093253267396,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034768385582547543,
      "loss": 5.1402,
      "mean_token_accuracy": 0.23341232687234878,
      "num_tokens": 84581450.0,
      "step": 45365
    },
    {
      "entropy": 6.062916851043701,
      "epoch": 4.006534793359237,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003476540520078995,
      "loss": 5.2879,
      "mean_token_accuracy": 0.22015472650527954,
      "num_tokens": 84590662.0,
      "step": 45370
    },
    {
      "entropy": 6.021073246002198,
      "epoch": 4.006976333451077,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034762424676707194,
      "loss": 5.2286,
      "mean_token_accuracy": 0.2121891051530838,
      "num_tokens": 84599269.0,
      "step": 45375
    },
    {
      "entropy": 5.951429748535157,
      "epoch": 4.007417873542917,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00034759444010357685,
      "loss": 5.1359,
      "mean_token_accuracy": 0.22768983542919158,
      "num_tokens": 84608497.0,
      "step": 45380
    },
    {
      "entropy": 6.041575002670288,
      "epoch": 4.007859413634758,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003475646320179979,
      "loss": 5.2219,
      "mean_token_accuracy": 0.2237247958779335,
      "num_tokens": 84617371.0,
      "step": 45385
    },
    {
      "entropy": 6.043727684020996,
      "epoch": 4.008300953726598,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003475348225109191,
      "loss": 5.1833,
      "mean_token_accuracy": 0.22340904623270036,
      "num_tokens": 84626366.0,
      "step": 45390
    },
    {
      "entropy": 6.075990486145019,
      "epoch": 4.008742493818438,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003475050115829244,
      "loss": 5.2492,
      "mean_token_accuracy": 0.2240932524204254,
      "num_tokens": 84635430.0,
      "step": 45395
    },
    {
      "entropy": 6.059129285812378,
      "epoch": 4.009184033910279,
      "grad_norm": 1.375,
      "learning_rate": 0.0003474751992345977,
      "loss": 5.2704,
      "mean_token_accuracy": 0.21829981058835984,
      "num_tokens": 84645253.0,
      "step": 45400
    },
    {
      "entropy": 6.018649244308472,
      "epoch": 4.009625574002119,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003474453854665231,
      "loss": 5.2156,
      "mean_token_accuracy": 0.2224625512957573,
      "num_tokens": 84653582.0,
      "step": 45405
    },
    {
      "entropy": 6.046276569366455,
      "epoch": 4.010067114093959,
      "grad_norm": 1.359375,
      "learning_rate": 0.00034741557027928463,
      "loss": 5.2418,
      "mean_token_accuracy": 0.22092559039592743,
      "num_tokens": 84662119.0,
      "step": 45410
    },
    {
      "entropy": 6.037004852294922,
      "epoch": 4.0105086541858,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034738575367346627,
      "loss": 5.2014,
      "mean_token_accuracy": 0.23074933290481567,
      "num_tokens": 84671387.0,
      "step": 45415
    },
    {
      "entropy": 5.945198106765747,
      "epoch": 4.01095019427764,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003473559356496522,
      "loss": 5.2114,
      "mean_token_accuracy": 0.22667208611965178,
      "num_tokens": 84680903.0,
      "step": 45420
    },
    {
      "entropy": 5.972325181961059,
      "epoch": 4.0113917343694805,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003473261162084264,
      "loss": 5.1509,
      "mean_token_accuracy": 0.2331261917948723,
      "num_tokens": 84689724.0,
      "step": 45425
    },
    {
      "entropy": 6.074733257293701,
      "epoch": 4.011833274461321,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003472962953503732,
      "loss": 5.2762,
      "mean_token_accuracy": 0.2098717361688614,
      "num_tokens": 84698791.0,
      "step": 45430
    },
    {
      "entropy": 6.03940896987915,
      "epoch": 4.0122748145531615,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003472664730760767,
      "loss": 5.1431,
      "mean_token_accuracy": 0.23067331612110137,
      "num_tokens": 84707489.0,
      "step": 45435
    },
    {
      "entropy": 6.092252779006958,
      "epoch": 4.0127163546450015,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003472366493861211,
      "loss": 5.2834,
      "mean_token_accuracy": 0.22041306793689727,
      "num_tokens": 84717682.0,
      "step": 45440
    },
    {
      "entropy": 5.944389057159424,
      "epoch": 4.0131578947368425,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003472068242810906,
      "loss": 5.1391,
      "mean_token_accuracy": 0.22988066971302032,
      "num_tokens": 84726270.0,
      "step": 45445
    },
    {
      "entropy": 6.009178018569946,
      "epoch": 4.0135994348286825,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003471769977615695,
      "loss": 5.2071,
      "mean_token_accuracy": 0.22513783276081084,
      "num_tokens": 84735094.0,
      "step": 45450
    },
    {
      "entropy": 6.019597721099854,
      "epoch": 4.014040974920523,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00034714716982814196,
      "loss": 5.2761,
      "mean_token_accuracy": 0.21718994975090028,
      "num_tokens": 84744934.0,
      "step": 45455
    },
    {
      "entropy": 6.071618747711182,
      "epoch": 4.0144825150123635,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003471173404813925,
      "loss": 5.23,
      "mean_token_accuracy": 0.2144773617386818,
      "num_tokens": 84754183.0,
      "step": 45460
    },
    {
      "entropy": 6.026746606826782,
      "epoch": 4.014924055104204,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003470875097219054,
      "loss": 5.2503,
      "mean_token_accuracy": 0.22219666689634324,
      "num_tokens": 84762784.0,
      "step": 45465
    },
    {
      "entropy": 6.038256978988647,
      "epoch": 4.015365595196044,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003470576775502649,
      "loss": 5.1831,
      "mean_token_accuracy": 0.22318761199712753,
      "num_tokens": 84771822.0,
      "step": 45470
    },
    {
      "entropy": 6.025847768783569,
      "epoch": 4.015807135287885,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003470278439670555,
      "loss": 5.2447,
      "mean_token_accuracy": 0.21939079463481903,
      "num_tokens": 84781378.0,
      "step": 45475
    },
    {
      "entropy": 6.079157972335816,
      "epoch": 4.016248675379725,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003469980089728616,
      "loss": 5.3523,
      "mean_token_accuracy": 0.206982783973217,
      "num_tokens": 84791142.0,
      "step": 45480
    },
    {
      "entropy": 6.03229284286499,
      "epoch": 4.016690215471565,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034696817256826763,
      "loss": 5.2213,
      "mean_token_accuracy": 0.22395772188901902,
      "num_tokens": 84801304.0,
      "step": 45485
    },
    {
      "entropy": 6.053192377090454,
      "epoch": 4.017131755563405,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003469383347538582,
      "loss": 5.216,
      "mean_token_accuracy": 0.23089343160390854,
      "num_tokens": 84811044.0,
      "step": 45490
    },
    {
      "entropy": 6.041147613525391,
      "epoch": 4.017573295655246,
      "grad_norm": 1.3125,
      "learning_rate": 0.00034690849553021755,
      "loss": 5.2703,
      "mean_token_accuracy": 0.21809124648571016,
      "num_tokens": 84820598.0,
      "step": 45495
    },
    {
      "entropy": 5.9927146434783936,
      "epoch": 4.018014835747086,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003468786548979304,
      "loss": 5.1522,
      "mean_token_accuracy": 0.22188531309366227,
      "num_tokens": 84830088.0,
      "step": 45500
    },
    {
      "entropy": 6.0212489604949955,
      "epoch": 4.018456375838926,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003468488128575813,
      "loss": 5.2514,
      "mean_token_accuracy": 0.22388871312141417,
      "num_tokens": 84839814.0,
      "step": 45505
    },
    {
      "entropy": 6.120859622955322,
      "epoch": 4.018897915930767,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003468189694097548,
      "loss": 5.3144,
      "mean_token_accuracy": 0.2169924259185791,
      "num_tokens": 84850070.0,
      "step": 45510
    },
    {
      "entropy": 6.007102203369141,
      "epoch": 4.019339456022607,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003467891245550356,
      "loss": 5.1721,
      "mean_token_accuracy": 0.2313310518860817,
      "num_tokens": 84859769.0,
      "step": 45515
    },
    {
      "entropy": 6.000762128829956,
      "epoch": 4.019780996114447,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003467592782940082,
      "loss": 5.2067,
      "mean_token_accuracy": 0.2256634846329689,
      "num_tokens": 84868592.0,
      "step": 45520
    },
    {
      "entropy": 6.10133638381958,
      "epoch": 4.020222536206288,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003467294306272574,
      "loss": 5.2422,
      "mean_token_accuracy": 0.22380170971155167,
      "num_tokens": 84877944.0,
      "step": 45525
    },
    {
      "entropy": 5.991440677642823,
      "epoch": 4.020664076298128,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003466995815553679,
      "loss": 5.2339,
      "mean_token_accuracy": 0.222163525223732,
      "num_tokens": 84887076.0,
      "step": 45530
    },
    {
      "entropy": 5.9878990650177,
      "epoch": 4.021105616389968,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034666973107892437,
      "loss": 5.2033,
      "mean_token_accuracy": 0.21896566152572633,
      "num_tokens": 84896876.0,
      "step": 45535
    },
    {
      "entropy": 6.007861709594726,
      "epoch": 4.021547156481809,
      "grad_norm": 1.375,
      "learning_rate": 0.00034663987919851145,
      "loss": 5.1513,
      "mean_token_accuracy": 0.22174738645553588,
      "num_tokens": 84905413.0,
      "step": 45540
    },
    {
      "entropy": 6.130804491043091,
      "epoch": 4.021988696573649,
      "grad_norm": 1.359375,
      "learning_rate": 0.00034661002591471413,
      "loss": 5.3373,
      "mean_token_accuracy": 0.2045507848262787,
      "num_tokens": 84914581.0,
      "step": 45545
    },
    {
      "entropy": 5.969032192230225,
      "epoch": 4.022430236665489,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00034658017122811706,
      "loss": 5.1346,
      "mean_token_accuracy": 0.2338632896542549,
      "num_tokens": 84923932.0,
      "step": 45550
    },
    {
      "entropy": 5.973731946945191,
      "epoch": 4.02287177675733,
      "grad_norm": 1.25,
      "learning_rate": 0.00034655031513930527,
      "loss": 5.1537,
      "mean_token_accuracy": 0.22582750171422958,
      "num_tokens": 84932447.0,
      "step": 45555
    },
    {
      "entropy": 5.868187379837036,
      "epoch": 4.02331331684917,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003465204576488634,
      "loss": 5.1414,
      "mean_token_accuracy": 0.22902274280786514,
      "num_tokens": 84942461.0,
      "step": 45560
    },
    {
      "entropy": 6.096148300170898,
      "epoch": 4.02375485694101,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034649059875737644,
      "loss": 5.3216,
      "mean_token_accuracy": 0.21616707593202591,
      "num_tokens": 84952634.0,
      "step": 45565
    },
    {
      "entropy": 6.050629043579102,
      "epoch": 4.024196397032851,
      "grad_norm": 1.28125,
      "learning_rate": 0.00034646073846542936,
      "loss": 5.2674,
      "mean_token_accuracy": 0.2234114706516266,
      "num_tokens": 84961898.0,
      "step": 45570
    },
    {
      "entropy": 5.992895460128784,
      "epoch": 4.024637937124691,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000346430876773607,
      "loss": 5.2027,
      "mean_token_accuracy": 0.2227422222495079,
      "num_tokens": 84971929.0,
      "step": 45575
    },
    {
      "entropy": 5.9995934009552006,
      "epoch": 4.025079477216531,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034640101368249446,
      "loss": 5.287,
      "mean_token_accuracy": 0.21778619587421416,
      "num_tokens": 84981146.0,
      "step": 45580
    },
    {
      "entropy": 6.038600444793701,
      "epoch": 4.025521017308372,
      "grad_norm": 1.375,
      "learning_rate": 0.0003463711491926766,
      "loss": 5.2846,
      "mean_token_accuracy": 0.21634224504232408,
      "num_tokens": 84990762.0,
      "step": 45585
    },
    {
      "entropy": 6.031782531738282,
      "epoch": 4.025962557400212,
      "grad_norm": 1.34375,
      "learning_rate": 0.00034634128330473857,
      "loss": 5.2691,
      "mean_token_accuracy": 0.21662976443767548,
      "num_tokens": 84999301.0,
      "step": 45590
    },
    {
      "entropy": 6.02599835395813,
      "epoch": 4.026404097492052,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003463114160192653,
      "loss": 5.1981,
      "mean_token_accuracy": 0.22195932269096375,
      "num_tokens": 85009271.0,
      "step": 45595
    },
    {
      "entropy": 6.074722099304199,
      "epoch": 4.026845637583893,
      "grad_norm": 1.234375,
      "learning_rate": 0.000346281547336842,
      "loss": 5.253,
      "mean_token_accuracy": 0.21382103860378265,
      "num_tokens": 85018607.0,
      "step": 45600
    },
    {
      "entropy": 5.970374011993409,
      "epoch": 4.027287177675733,
      "grad_norm": 1.328125,
      "learning_rate": 0.00034625167725805373,
      "loss": 5.1318,
      "mean_token_accuracy": 0.23321763724088668,
      "num_tokens": 85028114.0,
      "step": 45605
    },
    {
      "entropy": 5.984044933319092,
      "epoch": 4.027728717767573,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003462218057834857,
      "loss": 5.1727,
      "mean_token_accuracy": 0.22439542412757874,
      "num_tokens": 85037667.0,
      "step": 45610
    },
    {
      "entropy": 6.017594480514527,
      "epoch": 4.028170257859414,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00034619193291372287,
      "loss": 5.3016,
      "mean_token_accuracy": 0.2204879567027092,
      "num_tokens": 85046905.0,
      "step": 45615
    },
    {
      "entropy": 6.011033964157105,
      "epoch": 4.028611797951254,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00034616205864935063,
      "loss": 5.2084,
      "mean_token_accuracy": 0.21994798332452775,
      "num_tokens": 85056478.0,
      "step": 45620
    },
    {
      "entropy": 5.9799628257751465,
      "epoch": 4.029053338043094,
      "grad_norm": 1.328125,
      "learning_rate": 0.00034613218299095407,
      "loss": 5.1544,
      "mean_token_accuracy": 0.23596559911966325,
      "num_tokens": 85066311.0,
      "step": 45625
    },
    {
      "entropy": 6.007596492767334,
      "epoch": 4.029494878134934,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034610230593911854,
      "loss": 5.1995,
      "mean_token_accuracy": 0.22549901753664017,
      "num_tokens": 85076955.0,
      "step": 45630
    },
    {
      "entropy": 6.033570671081543,
      "epoch": 4.029936418226775,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00034607242749442934,
      "loss": 5.3173,
      "mean_token_accuracy": 0.20904891192913055,
      "num_tokens": 85086271.0,
      "step": 45635
    },
    {
      "entropy": 6.0675740242004395,
      "epoch": 4.030377958318615,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034604254765747164,
      "loss": 5.2587,
      "mean_token_accuracy": 0.21850921511650084,
      "num_tokens": 85095445.0,
      "step": 45640
    },
    {
      "entropy": 5.983114194869995,
      "epoch": 4.030819498410455,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00034601266642883083,
      "loss": 5.1627,
      "mean_token_accuracy": 0.22654741555452346,
      "num_tokens": 85105038.0,
      "step": 45645
    },
    {
      "entropy": 6.045617485046387,
      "epoch": 4.031261038502296,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003459827838090923,
      "loss": 5.2318,
      "mean_token_accuracy": 0.2187518894672394,
      "num_tokens": 85113859.0,
      "step": 45650
    },
    {
      "entropy": 6.073772335052491,
      "epoch": 4.031702578594136,
      "grad_norm": 1.28125,
      "learning_rate": 0.00034595289979884135,
      "loss": 5.3078,
      "mean_token_accuracy": 0.21677394807338715,
      "num_tokens": 85123932.0,
      "step": 45655
    },
    {
      "entropy": 6.019299602508545,
      "epoch": 4.032144118685976,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034592301439866346,
      "loss": 5.2479,
      "mean_token_accuracy": 0.2192234843969345,
      "num_tokens": 85132681.0,
      "step": 45660
    },
    {
      "entropy": 5.982638549804688,
      "epoch": 4.032585658777817,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003458931276091442,
      "loss": 5.1237,
      "mean_token_accuracy": 0.2330656960606575,
      "num_tokens": 85140869.0,
      "step": 45665
    },
    {
      "entropy": 5.979846096038818,
      "epoch": 4.033027198869657,
      "grad_norm": 1.3125,
      "learning_rate": 0.00034586323943086865,
      "loss": 5.19,
      "mean_token_accuracy": 0.22318578660488128,
      "num_tokens": 85149864.0,
      "step": 45670
    },
    {
      "entropy": 6.021295547485352,
      "epoch": 4.033468738961497,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00034583334986442273,
      "loss": 5.1935,
      "mean_token_accuracy": 0.2229855850338936,
      "num_tokens": 85158903.0,
      "step": 45675
    },
    {
      "entropy": 6.016885757446289,
      "epoch": 4.033910279053338,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034580345891039167,
      "loss": 5.1483,
      "mean_token_accuracy": 0.22928842306137084,
      "num_tokens": 85167507.0,
      "step": 45680
    },
    {
      "entropy": 5.991143846511841,
      "epoch": 4.034351819145178,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034577356656936115,
      "loss": 5.1256,
      "mean_token_accuracy": 0.2298918828368187,
      "num_tokens": 85176775.0,
      "step": 45685
    },
    {
      "entropy": 5.965619230270386,
      "epoch": 4.034793359237018,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003457436728419167,
      "loss": 5.125,
      "mean_token_accuracy": 0.2312069207429886,
      "num_tokens": 85186292.0,
      "step": 45690
    },
    {
      "entropy": 5.976528930664062,
      "epoch": 4.035234899328859,
      "grad_norm": 1.390625,
      "learning_rate": 0.00034571377772864395,
      "loss": 5.1889,
      "mean_token_accuracy": 0.22738512605428696,
      "num_tokens": 85196272.0,
      "step": 45695
    },
    {
      "entropy": 6.0001201152801515,
      "epoch": 4.035676439420699,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034568388123012853,
      "loss": 5.0983,
      "mean_token_accuracy": 0.22630978375673294,
      "num_tokens": 85205019.0,
      "step": 45700
    },
    {
      "entropy": 6.059381628036499,
      "epoch": 4.0361179795125395,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000345653983346956,
      "loss": 5.2586,
      "mean_token_accuracy": 0.21625475734472274,
      "num_tokens": 85214553.0,
      "step": 45705
    },
    {
      "entropy": 6.038079309463501,
      "epoch": 4.03655951960438,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034562408407971226,
      "loss": 5.2482,
      "mean_token_accuracy": 0.21557446271181108,
      "num_tokens": 85224864.0,
      "step": 45710
    },
    {
      "entropy": 6.011667060852051,
      "epoch": 4.0370010596962205,
      "grad_norm": 1.3125,
      "learning_rate": 0.00034559418342898274,
      "loss": 5.2055,
      "mean_token_accuracy": 0.2274302676320076,
      "num_tokens": 85233468.0,
      "step": 45715
    },
    {
      "entropy": 6.0517768383026125,
      "epoch": 4.0374425997880605,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003455642813953534,
      "loss": 5.2473,
      "mean_token_accuracy": 0.2199545457959175,
      "num_tokens": 85242823.0,
      "step": 45720
    },
    {
      "entropy": 6.1046919345855715,
      "epoch": 4.0378841398799015,
      "grad_norm": 1.34375,
      "learning_rate": 0.00034553437797940986,
      "loss": 5.3227,
      "mean_token_accuracy": 0.21580790579319,
      "num_tokens": 85252358.0,
      "step": 45725
    },
    {
      "entropy": 6.0303129196167,
      "epoch": 4.0383256799717415,
      "grad_norm": 1.1875,
      "learning_rate": 0.000345504473181738,
      "loss": 5.198,
      "mean_token_accuracy": 0.22231885492801667,
      "num_tokens": 85261875.0,
      "step": 45730
    },
    {
      "entropy": 6.048499727249146,
      "epoch": 4.038767220063582,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003454745670029236,
      "loss": 5.1515,
      "mean_token_accuracy": 0.2266262099146843,
      "num_tokens": 85270650.0,
      "step": 45735
    },
    {
      "entropy": 6.100598335266113,
      "epoch": 4.0392087601554225,
      "grad_norm": 1.453125,
      "learning_rate": 0.00034544465944355253,
      "loss": 5.319,
      "mean_token_accuracy": 0.21460433304309845,
      "num_tokens": 85279897.0,
      "step": 45740
    },
    {
      "entropy": 5.990235090255737,
      "epoch": 4.039650300247263,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003454147505042107,
      "loss": 5.1999,
      "mean_token_accuracy": 0.2318572998046875,
      "num_tokens": 85289139.0,
      "step": 45745
    },
    {
      "entropy": 5.969772005081177,
      "epoch": 4.040091840339103,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003453848401854838,
      "loss": 5.1534,
      "mean_token_accuracy": 0.22717931866645813,
      "num_tokens": 85298223.0,
      "step": 45750
    },
    {
      "entropy": 6.004457235336304,
      "epoch": 4.040533380430944,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00034535492848795804,
      "loss": 5.176,
      "mean_token_accuracy": 0.23245576173067092,
      "num_tokens": 85306771.0,
      "step": 45755
    },
    {
      "entropy": 5.9422074317932125,
      "epoch": 4.040974920522784,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003453250154122192,
      "loss": 5.1114,
      "mean_token_accuracy": 0.23993971049785615,
      "num_tokens": 85316278.0,
      "step": 45760
    },
    {
      "entropy": 6.03349814414978,
      "epoch": 4.041416460614624,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003452951009588532,
      "loss": 5.2629,
      "mean_token_accuracy": 0.2162030428647995,
      "num_tokens": 85326107.0,
      "step": 45765
    },
    {
      "entropy": 6.077238893508911,
      "epoch": 4.041858000706464,
      "grad_norm": 1.375,
      "learning_rate": 0.0003452651851284463,
      "loss": 5.2841,
      "mean_token_accuracy": 0.2157113164663315,
      "num_tokens": 85335393.0,
      "step": 45770
    },
    {
      "entropy": 6.052962589263916,
      "epoch": 4.042299540798305,
      "grad_norm": 1.3125,
      "learning_rate": 0.00034523526792158444,
      "loss": 5.1889,
      "mean_token_accuracy": 0.22597548514604568,
      "num_tokens": 85344972.0,
      "step": 45775
    },
    {
      "entropy": 6.012907028198242,
      "epoch": 4.042741080890145,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034520534933885343,
      "loss": 5.204,
      "mean_token_accuracy": 0.22498733401298524,
      "num_tokens": 85354407.0,
      "step": 45780
    },
    {
      "entropy": 5.950084543228149,
      "epoch": 4.043182620981985,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003451754293808397,
      "loss": 5.1428,
      "mean_token_accuracy": 0.22825745791196822,
      "num_tokens": 85364120.0,
      "step": 45785
    },
    {
      "entropy": 6.0233818054199215,
      "epoch": 4.043624161073826,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00034514550804812916,
      "loss": 5.2106,
      "mean_token_accuracy": 0.22439976334571837,
      "num_tokens": 85372331.0,
      "step": 45790
    },
    {
      "entropy": 5.962687635421753,
      "epoch": 4.044065701165666,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003451155853413081,
      "loss": 5.1282,
      "mean_token_accuracy": 0.2287181109189987,
      "num_tokens": 85380178.0,
      "step": 45795
    },
    {
      "entropy": 5.974563789367676,
      "epoch": 4.044507241257506,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003450856612609625,
      "loss": 5.2666,
      "mean_token_accuracy": 0.21527080982923508,
      "num_tokens": 85389621.0,
      "step": 45800
    },
    {
      "entropy": 6.105629777908325,
      "epoch": 4.044948781349347,
      "grad_norm": 1.3125,
      "learning_rate": 0.00034505573580767866,
      "loss": 5.2794,
      "mean_token_accuracy": 0.21085913628339767,
      "num_tokens": 85398374.0,
      "step": 45805
    },
    {
      "entropy": 6.061788749694824,
      "epoch": 4.045390321441187,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003450258089820428,
      "loss": 5.2093,
      "mean_token_accuracy": 0.21585270762443542,
      "num_tokens": 85407504.0,
      "step": 45810
    },
    {
      "entropy": 6.117653894424438,
      "epoch": 4.045831861533027,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034499588078464114,
      "loss": 5.3295,
      "mean_token_accuracy": 0.21208755522966385,
      "num_tokens": 85417095.0,
      "step": 45815
    },
    {
      "entropy": 6.071929168701172,
      "epoch": 4.046273401624868,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003449659512160601,
      "loss": 5.2157,
      "mean_token_accuracy": 0.21923642009496688,
      "num_tokens": 85427358.0,
      "step": 45820
    },
    {
      "entropy": 5.984328508377075,
      "epoch": 4.046714941716708,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003449360202768858,
      "loss": 5.2746,
      "mean_token_accuracy": 0.2223903328180313,
      "num_tokens": 85436572.0,
      "step": 45825
    },
    {
      "entropy": 6.041511583328247,
      "epoch": 4.047156481808548,
      "grad_norm": 1.25,
      "learning_rate": 0.00034490608796770465,
      "loss": 5.1901,
      "mean_token_accuracy": 0.23061669170856475,
      "num_tokens": 85446784.0,
      "step": 45830
    },
    {
      "entropy": 5.941690587997437,
      "epoch": 4.047598021900389,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034487615428910296,
      "loss": 5.1359,
      "mean_token_accuracy": 0.22317383140325547,
      "num_tokens": 85456411.0,
      "step": 45835
    },
    {
      "entropy": 6.1056797981262205,
      "epoch": 4.048039561992229,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00034484621924166714,
      "loss": 5.2549,
      "mean_token_accuracy": 0.22579850107431412,
      "num_tokens": 85465227.0,
      "step": 45840
    },
    {
      "entropy": 5.9327394485473635,
      "epoch": 4.048481102084069,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00034481628282598373,
      "loss": 5.1608,
      "mean_token_accuracy": 0.22384582459926605,
      "num_tokens": 85474986.0,
      "step": 45845
    },
    {
      "entropy": 5.993351316452026,
      "epoch": 4.04892264217591,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003447863450426389,
      "loss": 5.2292,
      "mean_token_accuracy": 0.21897172033786774,
      "num_tokens": 85484329.0,
      "step": 45850
    },
    {
      "entropy": 6.098811912536621,
      "epoch": 4.04936418226775,
      "grad_norm": 1.375,
      "learning_rate": 0.0003447564058922193,
      "loss": 5.3105,
      "mean_token_accuracy": 0.21696172803640365,
      "num_tokens": 85493850.0,
      "step": 45855
    },
    {
      "entropy": 6.057078409194946,
      "epoch": 4.04980572235959,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003447264653753114,
      "loss": 5.2352,
      "mean_token_accuracy": 0.21936322003602982,
      "num_tokens": 85503847.0,
      "step": 45860
    },
    {
      "entropy": 6.053471374511719,
      "epoch": 4.050247262451431,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034469652349250176,
      "loss": 5.2493,
      "mean_token_accuracy": 0.2174942523241043,
      "num_tokens": 85513543.0,
      "step": 45865
    },
    {
      "entropy": 6.007662296295166,
      "epoch": 4.050688802543271,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003446665802443767,
      "loss": 5.1776,
      "mean_token_accuracy": 0.22646371871232987,
      "num_tokens": 85522680.0,
      "step": 45870
    },
    {
      "entropy": 5.97169303894043,
      "epoch": 4.051130342635111,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003446366356315231,
      "loss": 5.1461,
      "mean_token_accuracy": 0.22126318961381913,
      "num_tokens": 85531331.0,
      "step": 45875
    },
    {
      "entropy": 5.990429067611695,
      "epoch": 4.051571882726952,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034460668965452735,
      "loss": 5.2718,
      "mean_token_accuracy": 0.21272738128900529,
      "num_tokens": 85540439.0,
      "step": 45880
    },
    {
      "entropy": 6.021009540557861,
      "epoch": 4.052013422818792,
      "grad_norm": 1.359375,
      "learning_rate": 0.00034457674231397616,
      "loss": 5.2116,
      "mean_token_accuracy": 0.21591506898403168,
      "num_tokens": 85549685.0,
      "step": 45885
    },
    {
      "entropy": 5.971744346618652,
      "epoch": 4.052454962910632,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034454679361045615,
      "loss": 5.123,
      "mean_token_accuracy": 0.22667415589094161,
      "num_tokens": 85557840.0,
      "step": 45890
    },
    {
      "entropy": 6.03105058670044,
      "epoch": 4.052896503002473,
      "grad_norm": 1.125,
      "learning_rate": 0.00034451684354455404,
      "loss": 5.2803,
      "mean_token_accuracy": 0.21983147114515306,
      "num_tokens": 85567388.0,
      "step": 45895
    },
    {
      "entropy": 6.03747067451477,
      "epoch": 4.053338043094313,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034448689211685643,
      "loss": 5.2258,
      "mean_token_accuracy": 0.22080033421516418,
      "num_tokens": 85576815.0,
      "step": 45900
    },
    {
      "entropy": 5.995801734924316,
      "epoch": 4.053779583186153,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003444569393279502,
      "loss": 5.2602,
      "mean_token_accuracy": 0.22311142086982727,
      "num_tokens": 85587439.0,
      "step": 45905
    },
    {
      "entropy": 5.9936244010925295,
      "epoch": 4.054221123277993,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003444269851784219,
      "loss": 5.1881,
      "mean_token_accuracy": 0.22179583013057708,
      "num_tokens": 85596985.0,
      "step": 45910
    },
    {
      "entropy": 6.0054826736450195,
      "epoch": 4.054662663369834,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003443970296688586,
      "loss": 5.2118,
      "mean_token_accuracy": 0.2219054028391838,
      "num_tokens": 85605812.0,
      "step": 45915
    },
    {
      "entropy": 6.022563743591308,
      "epoch": 4.055104203461674,
      "grad_norm": 1.421875,
      "learning_rate": 0.000344367072799847,
      "loss": 5.2101,
      "mean_token_accuracy": 0.22439063638448714,
      "num_tokens": 85615067.0,
      "step": 45920
    },
    {
      "entropy": 6.00335111618042,
      "epoch": 4.055545743553514,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003443371145719737,
      "loss": 5.2288,
      "mean_token_accuracy": 0.2176429182291031,
      "num_tokens": 85624140.0,
      "step": 45925
    },
    {
      "entropy": 6.002627992630005,
      "epoch": 4.055987283645355,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003443071549858259,
      "loss": 5.2018,
      "mean_token_accuracy": 0.2214554324746132,
      "num_tokens": 85634154.0,
      "step": 45930
    },
    {
      "entropy": 5.9745337009429935,
      "epoch": 4.056428823737195,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003442771940419903,
      "loss": 5.2343,
      "mean_token_accuracy": 0.22472371459007262,
      "num_tokens": 85643305.0,
      "step": 45935
    },
    {
      "entropy": 6.018222951889038,
      "epoch": 4.056870363829035,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003442472317410538,
      "loss": 5.3099,
      "mean_token_accuracy": 0.2104371592402458,
      "num_tokens": 85652731.0,
      "step": 45940
    },
    {
      "entropy": 6.094095897674561,
      "epoch": 4.057311903920876,
      "grad_norm": 1.25,
      "learning_rate": 0.0003442172680836035,
      "loss": 5.2641,
      "mean_token_accuracy": 0.215056349337101,
      "num_tokens": 85661626.0,
      "step": 45945
    },
    {
      "entropy": 6.042269229888916,
      "epoch": 4.057753444012716,
      "grad_norm": 1.375,
      "learning_rate": 0.0003441873030702262,
      "loss": 5.2677,
      "mean_token_accuracy": 0.21433379948139192,
      "num_tokens": 85671294.0,
      "step": 45950
    },
    {
      "entropy": 6.141558408737183,
      "epoch": 4.058194984104556,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003441573367015091,
      "loss": 5.2814,
      "mean_token_accuracy": 0.2165001466870308,
      "num_tokens": 85680630.0,
      "step": 45955
    },
    {
      "entropy": 6.055333042144776,
      "epoch": 4.058636524196397,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034412736897803905,
      "loss": 5.2793,
      "mean_token_accuracy": 0.2154654696583748,
      "num_tokens": 85689840.0,
      "step": 45960
    },
    {
      "entropy": 6.020235204696656,
      "epoch": 4.059078064288237,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003440973999004031,
      "loss": 5.2315,
      "mean_token_accuracy": 0.22157532423734666,
      "num_tokens": 85699323.0,
      "step": 45965
    },
    {
      "entropy": 6.035408592224121,
      "epoch": 4.059519604380077,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034406742946918855,
      "loss": 5.259,
      "mean_token_accuracy": 0.21656618714332582,
      "num_tokens": 85708469.0,
      "step": 45970
    },
    {
      "entropy": 6.039740800857544,
      "epoch": 4.059961144471918,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003440374576849822,
      "loss": 5.2318,
      "mean_token_accuracy": 0.22284298539161682,
      "num_tokens": 85716963.0,
      "step": 45975
    },
    {
      "entropy": 5.957383012771606,
      "epoch": 4.060402684563758,
      "grad_norm": 1.25,
      "learning_rate": 0.0003440074845483714,
      "loss": 5.1653,
      "mean_token_accuracy": 0.22238486558198928,
      "num_tokens": 85726424.0,
      "step": 45980
    },
    {
      "entropy": 6.004732751846314,
      "epoch": 4.0608442246555985,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00034397751005994315,
      "loss": 5.2262,
      "mean_token_accuracy": 0.22423979192972182,
      "num_tokens": 85734738.0,
      "step": 45985
    },
    {
      "entropy": 6.053892421722412,
      "epoch": 4.061285764747439,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003439475342202847,
      "loss": 5.223,
      "mean_token_accuracy": 0.21718688905239106,
      "num_tokens": 85743710.0,
      "step": 45990
    },
    {
      "entropy": 6.061928033828735,
      "epoch": 4.0617273048392795,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00034391755702998334,
      "loss": 5.229,
      "mean_token_accuracy": 0.21743571609258652,
      "num_tokens": 85752927.0,
      "step": 45995
    },
    {
      "entropy": 6.084523630142212,
      "epoch": 4.0621688449311195,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003438875784896261,
      "loss": 5.3366,
      "mean_token_accuracy": 0.21269995868206024,
      "num_tokens": 85762602.0,
      "step": 46000
    },
    {
      "entropy": 6.0847522735595705,
      "epoch": 4.0626103850229605,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003438575985998005,
      "loss": 5.3189,
      "mean_token_accuracy": 0.21783996373414993,
      "num_tokens": 85771853.0,
      "step": 46005
    },
    {
      "entropy": 6.12772569656372,
      "epoch": 4.0630519251148005,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003438276173610936,
      "loss": 5.3236,
      "mean_token_accuracy": 0.2155100867152214,
      "num_tokens": 85781487.0,
      "step": 46010
    },
    {
      "entropy": 6.033100271224976,
      "epoch": 4.063493465206641,
      "grad_norm": 1.109375,
      "learning_rate": 0.000343797634774093,
      "loss": 5.29,
      "mean_token_accuracy": 0.22060343474149705,
      "num_tokens": 85791716.0,
      "step": 46015
    },
    {
      "entropy": 6.079574918746948,
      "epoch": 4.0639350052984815,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003437676508393858,
      "loss": 5.2892,
      "mean_token_accuracy": 0.21856808811426162,
      "num_tokens": 85801759.0,
      "step": 46020
    },
    {
      "entropy": 6.118456077575684,
      "epoch": 4.064376545390322,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034373766555755936,
      "loss": 5.3315,
      "mean_token_accuracy": 0.21040456742048264,
      "num_tokens": 85811508.0,
      "step": 46025
    },
    {
      "entropy": 6.037736129760742,
      "epoch": 4.064818085482162,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003437076789292011,
      "loss": 5.1923,
      "mean_token_accuracy": 0.22675348222255706,
      "num_tokens": 85820458.0,
      "step": 46030
    },
    {
      "entropy": 6.002013349533081,
      "epoch": 4.0652596255740026,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003436776909548986,
      "loss": 5.2892,
      "mean_token_accuracy": 0.21728213280439376,
      "num_tokens": 85829501.0,
      "step": 46035
    },
    {
      "entropy": 6.034308910369873,
      "epoch": 4.065701165665843,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034364770163523915,
      "loss": 5.295,
      "mean_token_accuracy": 0.21589999198913573,
      "num_tokens": 85838974.0,
      "step": 46040
    },
    {
      "entropy": 6.064281463623047,
      "epoch": 4.066142705757683,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003436177109708102,
      "loss": 5.2099,
      "mean_token_accuracy": 0.21521705240011216,
      "num_tokens": 85847700.0,
      "step": 46045
    },
    {
      "entropy": 6.050464534759522,
      "epoch": 4.066584245849523,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00034358771896219947,
      "loss": 5.2721,
      "mean_token_accuracy": 0.2169858232140541,
      "num_tokens": 85855951.0,
      "step": 46050
    },
    {
      "entropy": 6.08761773109436,
      "epoch": 4.067025785941364,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003435577256099942,
      "loss": 5.2553,
      "mean_token_accuracy": 0.21547945737838745,
      "num_tokens": 85864648.0,
      "step": 46055
    },
    {
      "entropy": 5.993184423446655,
      "epoch": 4.067467326033204,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034352773091478214,
      "loss": 5.1547,
      "mean_token_accuracy": 0.22772542536258697,
      "num_tokens": 85874333.0,
      "step": 46060
    },
    {
      "entropy": 5.993645668029785,
      "epoch": 4.067908866125044,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034349773487715084,
      "loss": 5.2486,
      "mean_token_accuracy": 0.2209472358226776,
      "num_tokens": 85885233.0,
      "step": 46065
    },
    {
      "entropy": 6.036723947525024,
      "epoch": 4.068350406216885,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003434677374976878,
      "loss": 5.2284,
      "mean_token_accuracy": 0.22581839859485625,
      "num_tokens": 85893872.0,
      "step": 46070
    },
    {
      "entropy": 6.025001811981201,
      "epoch": 4.068791946308725,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003434377387769807,
      "loss": 5.1198,
      "mean_token_accuracy": 0.22995758354663848,
      "num_tokens": 85903331.0,
      "step": 46075
    },
    {
      "entropy": 6.028918170928955,
      "epoch": 4.069233486400565,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003434077387156173,
      "loss": 5.2413,
      "mean_token_accuracy": 0.21844387501478196,
      "num_tokens": 85912764.0,
      "step": 46080
    },
    {
      "entropy": 6.058193302154541,
      "epoch": 4.069675026492406,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00034337773731418515,
      "loss": 5.2251,
      "mean_token_accuracy": 0.22376702278852462,
      "num_tokens": 85921970.0,
      "step": 46085
    },
    {
      "entropy": 6.046959066390992,
      "epoch": 4.070116566584246,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034334773457327203,
      "loss": 5.2285,
      "mean_token_accuracy": 0.22604520320892335,
      "num_tokens": 85931341.0,
      "step": 46090
    },
    {
      "entropy": 5.941412782669067,
      "epoch": 4.070558106676086,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003433177304934657,
      "loss": 5.1428,
      "mean_token_accuracy": 0.23304441422224045,
      "num_tokens": 85939643.0,
      "step": 46095
    },
    {
      "entropy": 6.023074436187744,
      "epoch": 4.070999646767927,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003432877250753538,
      "loss": 5.2244,
      "mean_token_accuracy": 0.22042263001203538,
      "num_tokens": 85948264.0,
      "step": 46100
    },
    {
      "entropy": 6.075756454467774,
      "epoch": 4.071441186859767,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003432577183195243,
      "loss": 5.2302,
      "mean_token_accuracy": 0.213609716296196,
      "num_tokens": 85957283.0,
      "step": 46105
    },
    {
      "entropy": 6.042297649383545,
      "epoch": 4.071882726951607,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034322771022656483,
      "loss": 5.289,
      "mean_token_accuracy": 0.2160578265786171,
      "num_tokens": 85966818.0,
      "step": 46110
    },
    {
      "entropy": 6.0532537460327145,
      "epoch": 4.072324267043448,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003431977007970634,
      "loss": 5.2452,
      "mean_token_accuracy": 0.22021753787994386,
      "num_tokens": 85977243.0,
      "step": 46115
    },
    {
      "entropy": 5.993740701675415,
      "epoch": 4.072765807135288,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003431676900316077,
      "loss": 5.1925,
      "mean_token_accuracy": 0.23251734375953675,
      "num_tokens": 85986392.0,
      "step": 46120
    },
    {
      "entropy": 6.085222434997559,
      "epoch": 4.073207347227128,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034313767793078576,
      "loss": 5.3551,
      "mean_token_accuracy": 0.21293370574712753,
      "num_tokens": 85995796.0,
      "step": 46125
    },
    {
      "entropy": 6.124733352661133,
      "epoch": 4.073648887318969,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00034310766449518554,
      "loss": 5.2323,
      "mean_token_accuracy": 0.21827073395252228,
      "num_tokens": 86004816.0,
      "step": 46130
    },
    {
      "entropy": 6.035002708435059,
      "epoch": 4.074090427410809,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003430776497253948,
      "loss": 5.2572,
      "mean_token_accuracy": 0.22537948787212372,
      "num_tokens": 86014395.0,
      "step": 46135
    },
    {
      "entropy": 5.978442239761352,
      "epoch": 4.074531967502649,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003430476336220016,
      "loss": 5.1668,
      "mean_token_accuracy": 0.22804515212774276,
      "num_tokens": 86023568.0,
      "step": 46140
    },
    {
      "entropy": 6.072990131378174,
      "epoch": 4.07497350759449,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003430176161855941,
      "loss": 5.2441,
      "mean_token_accuracy": 0.20997285842895508,
      "num_tokens": 86032872.0,
      "step": 46145
    },
    {
      "entropy": 6.026016473770142,
      "epoch": 4.07541504768633,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034298759741676005,
      "loss": 5.2781,
      "mean_token_accuracy": 0.21929776966571807,
      "num_tokens": 86043366.0,
      "step": 46150
    },
    {
      "entropy": 6.031500434875488,
      "epoch": 4.07585658777817,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00034295757731608775,
      "loss": 5.2076,
      "mean_token_accuracy": 0.22554149031639098,
      "num_tokens": 86052701.0,
      "step": 46155
    },
    {
      "entropy": 6.083640098571777,
      "epoch": 4.076298127870011,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003429275558841651,
      "loss": 5.2517,
      "mean_token_accuracy": 0.22307583093643188,
      "num_tokens": 86061495.0,
      "step": 46160
    },
    {
      "entropy": 5.974522256851197,
      "epoch": 4.076739667961851,
      "grad_norm": 1.390625,
      "learning_rate": 0.00034289753312158024,
      "loss": 5.1886,
      "mean_token_accuracy": 0.2258812814950943,
      "num_tokens": 86070712.0,
      "step": 46165
    },
    {
      "entropy": 6.05516037940979,
      "epoch": 4.077181208053691,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003428675090289214,
      "loss": 5.2335,
      "mean_token_accuracy": 0.2237399086356163,
      "num_tokens": 86080129.0,
      "step": 46170
    },
    {
      "entropy": 6.112754964828492,
      "epoch": 4.077622748145532,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003428374836067765,
      "loss": 5.3025,
      "mean_token_accuracy": 0.21694792956113815,
      "num_tokens": 86089415.0,
      "step": 46175
    },
    {
      "entropy": 6.171350955963135,
      "epoch": 4.078064288237372,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000342807456855734,
      "loss": 5.3289,
      "mean_token_accuracy": 0.2127427339553833,
      "num_tokens": 86098450.0,
      "step": 46180
    },
    {
      "entropy": 6.015309238433838,
      "epoch": 4.078505828329212,
      "grad_norm": 1.40625,
      "learning_rate": 0.00034277742877638197,
      "loss": 5.1417,
      "mean_token_accuracy": 0.2311128169298172,
      "num_tokens": 86107066.0,
      "step": 46185
    },
    {
      "entropy": 5.95137209892273,
      "epoch": 4.078947368421052,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003427473993693086,
      "loss": 5.1916,
      "mean_token_accuracy": 0.22700754404067994,
      "num_tokens": 86115686.0,
      "step": 46190
    },
    {
      "entropy": 5.989948749542236,
      "epoch": 4.079388908512893,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034271736863510235,
      "loss": 5.1918,
      "mean_token_accuracy": 0.2238135278224945,
      "num_tokens": 86125016.0,
      "step": 46195
    },
    {
      "entropy": 6.030992460250855,
      "epoch": 4.079830448604733,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003426873365743512,
      "loss": 5.2114,
      "mean_token_accuracy": 0.2235543116927147,
      "num_tokens": 86135115.0,
      "step": 46200
    },
    {
      "entropy": 5.981351613998413,
      "epoch": 4.080271988696573,
      "grad_norm": 1.4375,
      "learning_rate": 0.00034265730318764373,
      "loss": 5.1771,
      "mean_token_accuracy": 0.22798052728176116,
      "num_tokens": 86143723.0,
      "step": 46205
    },
    {
      "entropy": 6.0181769847869875,
      "epoch": 4.080713528788414,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003426272684755681,
      "loss": 5.2937,
      "mean_token_accuracy": 0.2156779482960701,
      "num_tokens": 86152402.0,
      "step": 46210
    },
    {
      "entropy": 6.0052581310272215,
      "epoch": 4.081155068880254,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00034259723243871274,
      "loss": 5.1886,
      "mean_token_accuracy": 0.22245227098464965,
      "num_tokens": 86161497.0,
      "step": 46215
    },
    {
      "entropy": 6.005712175369263,
      "epoch": 4.081596608972094,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034256719507766615,
      "loss": 5.2724,
      "mean_token_accuracy": 0.21786702126264573,
      "num_tokens": 86171198.0,
      "step": 46220
    },
    {
      "entropy": 6.115357875823975,
      "epoch": 4.082038149063935,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003425371563930165,
      "loss": 5.2799,
      "mean_token_accuracy": 0.21717361956834794,
      "num_tokens": 86180405.0,
      "step": 46225
    },
    {
      "entropy": 6.041113710403442,
      "epoch": 4.082479689155775,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003425071163853525,
      "loss": 5.161,
      "mean_token_accuracy": 0.22019142061471939,
      "num_tokens": 86189405.0,
      "step": 46230
    },
    {
      "entropy": 6.047818803787232,
      "epoch": 4.082921229247615,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003424770750552624,
      "loss": 5.2713,
      "mean_token_accuracy": 0.21679375022649766,
      "num_tokens": 86198658.0,
      "step": 46235
    },
    {
      "entropy": 6.081414413452149,
      "epoch": 4.083362769339456,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034244703240333475,
      "loss": 5.284,
      "mean_token_accuracy": 0.2178828239440918,
      "num_tokens": 86208220.0,
      "step": 46240
    },
    {
      "entropy": 6.0684624195098875,
      "epoch": 4.083804309431296,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003424169884301581,
      "loss": 5.2721,
      "mean_token_accuracy": 0.2198503792285919,
      "num_tokens": 86218479.0,
      "step": 46245
    },
    {
      "entropy": 6.050108814239502,
      "epoch": 4.084245849523136,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000342386943136321,
      "loss": 5.2306,
      "mean_token_accuracy": 0.22327733784914017,
      "num_tokens": 86227629.0,
      "step": 46250
    },
    {
      "entropy": 6.023873424530029,
      "epoch": 4.084687389614977,
      "grad_norm": 1.359375,
      "learning_rate": 0.000342356896522412,
      "loss": 5.2051,
      "mean_token_accuracy": 0.22716898322105408,
      "num_tokens": 86236373.0,
      "step": 46255
    },
    {
      "entropy": 5.98418869972229,
      "epoch": 4.085128929706817,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00034232684858901975,
      "loss": 5.144,
      "mean_token_accuracy": 0.23177446871995927,
      "num_tokens": 86245456.0,
      "step": 46260
    },
    {
      "entropy": 6.042062044143677,
      "epoch": 4.0855704697986575,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034229679933673283,
      "loss": 5.2106,
      "mean_token_accuracy": 0.22155874520540236,
      "num_tokens": 86255070.0,
      "step": 46265
    },
    {
      "entropy": 6.080567026138306,
      "epoch": 4.086012009890498,
      "grad_norm": 1.46875,
      "learning_rate": 0.00034226674876613984,
      "loss": 5.2535,
      "mean_token_accuracy": 0.2192096456885338,
      "num_tokens": 86264167.0,
      "step": 46270
    },
    {
      "entropy": 6.095953416824341,
      "epoch": 4.0864535499823385,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003422366968778295,
      "loss": 5.2811,
      "mean_token_accuracy": 0.21364083290100097,
      "num_tokens": 86273371.0,
      "step": 46275
    },
    {
      "entropy": 6.02670431137085,
      "epoch": 4.0868950900741785,
      "grad_norm": 1.34375,
      "learning_rate": 0.00034220664367239057,
      "loss": 5.1769,
      "mean_token_accuracy": 0.21934508681297302,
      "num_tokens": 86281941.0,
      "step": 46280
    },
    {
      "entropy": 5.952284431457519,
      "epoch": 4.0873366301660194,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00034217658915041164,
      "loss": 5.0586,
      "mean_token_accuracy": 0.23120543658733367,
      "num_tokens": 86290720.0,
      "step": 46285
    },
    {
      "entropy": 5.986656236648559,
      "epoch": 4.0877781702578595,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003421465333124816,
      "loss": 5.1934,
      "mean_token_accuracy": 0.2208116054534912,
      "num_tokens": 86299724.0,
      "step": 46290
    },
    {
      "entropy": 6.0255786895751955,
      "epoch": 4.0882197103496996,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003421164761591891,
      "loss": 5.288,
      "mean_token_accuracy": 0.222448468208313,
      "num_tokens": 86308644.0,
      "step": 46295
    },
    {
      "entropy": 6.080521202087402,
      "epoch": 4.0886612504415405,
      "grad_norm": 1.453125,
      "learning_rate": 0.000342086417691123,
      "loss": 5.2826,
      "mean_token_accuracy": 0.22389502823352814,
      "num_tokens": 86318279.0,
      "step": 46300
    },
    {
      "entropy": 6.042315053939819,
      "epoch": 4.0891027905333805,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00034205635790887217,
      "loss": 5.2299,
      "mean_token_accuracy": 0.22492376416921617,
      "num_tokens": 86326556.0,
      "step": 46305
    },
    {
      "entropy": 5.85365743637085,
      "epoch": 4.089544330625221,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034202629681302543,
      "loss": 5.0248,
      "mean_token_accuracy": 0.25096840113401414,
      "num_tokens": 86335664.0,
      "step": 46310
    },
    {
      "entropy": 6.066942882537842,
      "epoch": 4.0899858707170615,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034199623440417165,
      "loss": 5.295,
      "mean_token_accuracy": 0.2190383180975914,
      "num_tokens": 86344636.0,
      "step": 46315
    },
    {
      "entropy": 6.042390823364258,
      "epoch": 4.090427410808902,
      "grad_norm": 1.25,
      "learning_rate": 0.0003419661706828997,
      "loss": 5.2796,
      "mean_token_accuracy": 0.21270343512296677,
      "num_tokens": 86353860.0,
      "step": 46320
    },
    {
      "entropy": 6.060139608383179,
      "epoch": 4.090868950900742,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003419361056497986,
      "loss": 5.2531,
      "mean_token_accuracy": 0.2185185059905052,
      "num_tokens": 86363081.0,
      "step": 46325
    },
    {
      "entropy": 6.03204231262207,
      "epoch": 4.091310490992582,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003419060393054573,
      "loss": 5.2504,
      "mean_token_accuracy": 0.2167677715420723,
      "num_tokens": 86372275.0,
      "step": 46330
    },
    {
      "entropy": 6.037454986572266,
      "epoch": 4.091752031084423,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00034187597165046457,
      "loss": 5.221,
      "mean_token_accuracy": 0.2208596020936966,
      "num_tokens": 86381037.0,
      "step": 46335
    },
    {
      "entropy": 6.076590204238892,
      "epoch": 4.092193571176263,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003418459026854098,
      "loss": 5.2454,
      "mean_token_accuracy": 0.22074569910764694,
      "num_tokens": 86389947.0,
      "step": 46340
    },
    {
      "entropy": 6.112495851516724,
      "epoch": 4.092635111268103,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003418158324108817,
      "loss": 5.3561,
      "mean_token_accuracy": 0.21500703692436218,
      "num_tokens": 86399499.0,
      "step": 46345
    },
    {
      "entropy": 5.866162538528442,
      "epoch": 4.093076651359944,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034178576082746943,
      "loss": 5.0665,
      "mean_token_accuracy": 0.22726663202047348,
      "num_tokens": 86408941.0,
      "step": 46350
    },
    {
      "entropy": 5.95369782447815,
      "epoch": 4.093518191451784,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034175568793576213,
      "loss": 5.2615,
      "mean_token_accuracy": 0.21984466314315795,
      "num_tokens": 86418992.0,
      "step": 46355
    },
    {
      "entropy": 6.003301954269409,
      "epoch": 4.093959731543624,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003417256137363489,
      "loss": 5.2214,
      "mean_token_accuracy": 0.2175860196352005,
      "num_tokens": 86428613.0,
      "step": 46360
    },
    {
      "entropy": 6.075820016860962,
      "epoch": 4.094401271635465,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00034169553822981883,
      "loss": 5.2504,
      "mean_token_accuracy": 0.21899478435516356,
      "num_tokens": 86438060.0,
      "step": 46365
    },
    {
      "entropy": 5.96404390335083,
      "epoch": 4.094842811727305,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00034166546141676103,
      "loss": 5.0935,
      "mean_token_accuracy": 0.23142267614603043,
      "num_tokens": 86447022.0,
      "step": 46370
    },
    {
      "entropy": 6.037503290176391,
      "epoch": 4.095284351819145,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034163538329776486,
      "loss": 5.1791,
      "mean_token_accuracy": 0.21695266366004945,
      "num_tokens": 86455122.0,
      "step": 46375
    },
    {
      "entropy": 5.990430784225464,
      "epoch": 4.095725891910986,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003416053038734193,
      "loss": 5.2445,
      "mean_token_accuracy": 0.22572651356458664,
      "num_tokens": 86464851.0,
      "step": 46380
    },
    {
      "entropy": 6.007442808151245,
      "epoch": 4.096167432002826,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003415752231443138,
      "loss": 5.2152,
      "mean_token_accuracy": 0.21553845852613449,
      "num_tokens": 86473895.0,
      "step": 46385
    },
    {
      "entropy": 6.092041254043579,
      "epoch": 4.096608972094666,
      "grad_norm": 1.25,
      "learning_rate": 0.00034154514111103747,
      "loss": 5.2868,
      "mean_token_accuracy": 0.21532027870416642,
      "num_tokens": 86483237.0,
      "step": 46390
    },
    {
      "entropy": 6.039560842514038,
      "epoch": 4.097050512186507,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003415150577741797,
      "loss": 5.3311,
      "mean_token_accuracy": 0.2234566792845726,
      "num_tokens": 86492796.0,
      "step": 46395
    },
    {
      "entropy": 6.137242317199707,
      "epoch": 4.097492052278347,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034148497313432987,
      "loss": 5.3763,
      "mean_token_accuracy": 0.20878811776638032,
      "num_tokens": 86502861.0,
      "step": 46400
    },
    {
      "entropy": 6.109652853012085,
      "epoch": 4.097933592370187,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034145488719207717,
      "loss": 5.248,
      "mean_token_accuracy": 0.2180774077773094,
      "num_tokens": 86512568.0,
      "step": 46405
    },
    {
      "entropy": 6.0615908145904545,
      "epoch": 4.098375132462028,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034142479994801105,
      "loss": 5.2309,
      "mean_token_accuracy": 0.22762700021266938,
      "num_tokens": 86521331.0,
      "step": 46410
    },
    {
      "entropy": 5.931298112869262,
      "epoch": 4.098816672553868,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00034139471140272077,
      "loss": 5.1499,
      "mean_token_accuracy": 0.2259072795510292,
      "num_tokens": 86530556.0,
      "step": 46415
    },
    {
      "entropy": 5.9744391441345215,
      "epoch": 4.099258212645708,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003413646215567959,
      "loss": 5.1931,
      "mean_token_accuracy": 0.22672607600688935,
      "num_tokens": 86539692.0,
      "step": 46420
    },
    {
      "entropy": 6.100841522216797,
      "epoch": 4.099699752737549,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003413345304108259,
      "loss": 5.2973,
      "mean_token_accuracy": 0.21763567626476288,
      "num_tokens": 86549398.0,
      "step": 46425
    },
    {
      "entropy": 6.171937084197998,
      "epoch": 4.100141292829389,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003413044379654002,
      "loss": 5.399,
      "mean_token_accuracy": 0.1996545046567917,
      "num_tokens": 86560404.0,
      "step": 46430
    },
    {
      "entropy": 6.067877006530762,
      "epoch": 4.100582832921229,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003412743442211082,
      "loss": 5.3388,
      "mean_token_accuracy": 0.21507746279239653,
      "num_tokens": 86569498.0,
      "step": 46435
    },
    {
      "entropy": 6.0595964908599855,
      "epoch": 4.10102437301307,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003412442491785395,
      "loss": 5.2145,
      "mean_token_accuracy": 0.22090430706739425,
      "num_tokens": 86579337.0,
      "step": 46440
    },
    {
      "entropy": 6.058010005950928,
      "epoch": 4.10146591310491,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003412141528382836,
      "loss": 5.2779,
      "mean_token_accuracy": 0.2196697399020195,
      "num_tokens": 86588523.0,
      "step": 46445
    },
    {
      "entropy": 5.966662836074829,
      "epoch": 4.10190745319675,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00034118405520093016,
      "loss": 5.1643,
      "mean_token_accuracy": 0.23021695911884307,
      "num_tokens": 86597502.0,
      "step": 46450
    },
    {
      "entropy": 6.010409498214722,
      "epoch": 4.102348993288591,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003411539562670687,
      "loss": 5.1919,
      "mean_token_accuracy": 0.22920526713132858,
      "num_tokens": 86607520.0,
      "step": 46455
    },
    {
      "entropy": 6.044626379013062,
      "epoch": 4.102790533380431,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00034112385603728883,
      "loss": 5.2688,
      "mean_token_accuracy": 0.2140714317560196,
      "num_tokens": 86617816.0,
      "step": 46460
    },
    {
      "entropy": 6.031726598739624,
      "epoch": 4.103232073472271,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034109375451218027,
      "loss": 5.2428,
      "mean_token_accuracy": 0.21055955290794373,
      "num_tokens": 86626765.0,
      "step": 46465
    },
    {
      "entropy": 6.07670316696167,
      "epoch": 4.103673613564111,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00034106365169233265,
      "loss": 5.3026,
      "mean_token_accuracy": 0.21498771607875825,
      "num_tokens": 86635687.0,
      "step": 46470
    },
    {
      "entropy": 6.0891059875488285,
      "epoch": 4.104115153655952,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003410335475783357,
      "loss": 5.2686,
      "mean_token_accuracy": 0.21730803251266478,
      "num_tokens": 86646133.0,
      "step": 46475
    },
    {
      "entropy": 6.0673987865448,
      "epoch": 4.104556693747792,
      "grad_norm": 1.390625,
      "learning_rate": 0.000341003442170779,
      "loss": 5.3139,
      "mean_token_accuracy": 0.21111902594566345,
      "num_tokens": 86655695.0,
      "step": 46480
    },
    {
      "entropy": 6.044810247421265,
      "epoch": 4.104998233839632,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003409733354702525,
      "loss": 5.2755,
      "mean_token_accuracy": 0.219763445854187,
      "num_tokens": 86665392.0,
      "step": 46485
    },
    {
      "entropy": 5.914094352722168,
      "epoch": 4.105439773931473,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003409432274773458,
      "loss": 5.0563,
      "mean_token_accuracy": 0.23788079768419265,
      "num_tokens": 86674319.0,
      "step": 46490
    },
    {
      "entropy": 6.013232803344726,
      "epoch": 4.105881314023313,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003409131181926488,
      "loss": 5.2645,
      "mean_token_accuracy": 0.21535738706588745,
      "num_tokens": 86683441.0,
      "step": 46495
    },
    {
      "entropy": 6.078309154510498,
      "epoch": 4.106322854115153,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034088300761675135,
      "loss": 5.2952,
      "mean_token_accuracy": 0.2148442879319191,
      "num_tokens": 86692533.0,
      "step": 46500
    },
    {
      "entropy": 6.059471225738525,
      "epoch": 4.106764394206994,
      "grad_norm": 1.28125,
      "learning_rate": 0.00034085289575024323,
      "loss": 5.2724,
      "mean_token_accuracy": 0.21418047845363616,
      "num_tokens": 86702877.0,
      "step": 46505
    },
    {
      "entropy": 6.103561878204346,
      "epoch": 4.107205934298834,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003408227825937143,
      "loss": 5.2954,
      "mean_token_accuracy": 0.21347306817770004,
      "num_tokens": 86711479.0,
      "step": 46510
    },
    {
      "entropy": 6.00793948173523,
      "epoch": 4.107647474390674,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034079266814775454,
      "loss": 5.2087,
      "mean_token_accuracy": 0.22080521136522294,
      "num_tokens": 86721599.0,
      "step": 46515
    },
    {
      "entropy": 6.048004055023194,
      "epoch": 4.108089014482515,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034076255241295386,
      "loss": 5.2734,
      "mean_token_accuracy": 0.21503284573554993,
      "num_tokens": 86730256.0,
      "step": 46520
    },
    {
      "entropy": 6.074232196807861,
      "epoch": 4.108530554574355,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003407324353899021,
      "loss": 5.2215,
      "mean_token_accuracy": 0.2192171961069107,
      "num_tokens": 86739198.0,
      "step": 46525
    },
    {
      "entropy": 6.00707483291626,
      "epoch": 4.108972094666195,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003407023170791893,
      "loss": 5.2309,
      "mean_token_accuracy": 0.2201836809515953,
      "num_tokens": 86748302.0,
      "step": 46530
    },
    {
      "entropy": 6.03673152923584,
      "epoch": 4.109413634758036,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003406721974814055,
      "loss": 5.2927,
      "mean_token_accuracy": 0.21272365003824234,
      "num_tokens": 86759142.0,
      "step": 46535
    },
    {
      "entropy": 6.055937814712524,
      "epoch": 4.109855174849876,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003406420765971408,
      "loss": 5.2585,
      "mean_token_accuracy": 0.21994090229272842,
      "num_tokens": 86768076.0,
      "step": 46540
    },
    {
      "entropy": 6.0311956882476805,
      "epoch": 4.1102967149417164,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034061195442698503,
      "loss": 5.1492,
      "mean_token_accuracy": 0.22777457386255265,
      "num_tokens": 86777635.0,
      "step": 46545
    },
    {
      "entropy": 5.942142152786255,
      "epoch": 4.110738255033557,
      "grad_norm": 1.28125,
      "learning_rate": 0.00034058183097152847,
      "loss": 5.1118,
      "mean_token_accuracy": 0.23699157387018205,
      "num_tokens": 86786477.0,
      "step": 46550
    },
    {
      "entropy": 6.0330931663513185,
      "epoch": 4.111179795125397,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034055170623136107,
      "loss": 5.2088,
      "mean_token_accuracy": 0.21653988659381868,
      "num_tokens": 86796191.0,
      "step": 46555
    },
    {
      "entropy": 5.939274549484253,
      "epoch": 4.1116213352172375,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003405215802070731,
      "loss": 5.1567,
      "mean_token_accuracy": 0.2226681962609291,
      "num_tokens": 86804691.0,
      "step": 46560
    },
    {
      "entropy": 6.049469947814941,
      "epoch": 4.112062875309078,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003404914528992546,
      "loss": 5.312,
      "mean_token_accuracy": 0.2144138529896736,
      "num_tokens": 86814083.0,
      "step": 46565
    },
    {
      "entropy": 6.040106773376465,
      "epoch": 4.1125044154009185,
      "grad_norm": 1.25,
      "learning_rate": 0.00034046132430849577,
      "loss": 5.3279,
      "mean_token_accuracy": 0.21956436336040497,
      "num_tokens": 86823465.0,
      "step": 46570
    },
    {
      "entropy": 6.034784841537475,
      "epoch": 4.1129459554927585,
      "grad_norm": 1.390625,
      "learning_rate": 0.00034043119443538677,
      "loss": 5.273,
      "mean_token_accuracy": 0.21523347795009612,
      "num_tokens": 86831629.0,
      "step": 46575
    },
    {
      "entropy": 5.975419044494629,
      "epoch": 4.1133874955845995,
      "grad_norm": 1.28125,
      "learning_rate": 0.000340401063280518,
      "loss": 5.1917,
      "mean_token_accuracy": 0.22162755578756332,
      "num_tokens": 86840111.0,
      "step": 46580
    },
    {
      "entropy": 6.00600152015686,
      "epoch": 4.1138290356764395,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034037093084447956,
      "loss": 5.1526,
      "mean_token_accuracy": 0.23045524954795837,
      "num_tokens": 86849348.0,
      "step": 46585
    },
    {
      "entropy": 6.120313358306885,
      "epoch": 4.11427057576828,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003403407971278617,
      "loss": 5.3529,
      "mean_token_accuracy": 0.21042579859495164,
      "num_tokens": 86859161.0,
      "step": 46590
    },
    {
      "entropy": 6.027136564254761,
      "epoch": 4.1147121158601205,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003403106621312549,
      "loss": 5.2071,
      "mean_token_accuracy": 0.22200062423944472,
      "num_tokens": 86868185.0,
      "step": 46595
    },
    {
      "entropy": 6.001949501037598,
      "epoch": 4.115153655951961,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034028052585524925,
      "loss": 5.21,
      "mean_token_accuracy": 0.21967718452215196,
      "num_tokens": 86877178.0,
      "step": 46600
    },
    {
      "entropy": 6.062430286407471,
      "epoch": 4.115595196043801,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003402503883004353,
      "loss": 5.2412,
      "mean_token_accuracy": 0.22892340421676635,
      "num_tokens": 86885014.0,
      "step": 46605
    },
    {
      "entropy": 6.052090501785278,
      "epoch": 4.116036736135641,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00034022024946740335,
      "loss": 5.2412,
      "mean_token_accuracy": 0.21970304697752,
      "num_tokens": 86894155.0,
      "step": 46610
    },
    {
      "entropy": 6.026910829544067,
      "epoch": 4.116478276227482,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034019010935674375,
      "loss": 5.157,
      "mean_token_accuracy": 0.22845847606658937,
      "num_tokens": 86902634.0,
      "step": 46615
    },
    {
      "entropy": 6.023450565338135,
      "epoch": 4.116919816319322,
      "grad_norm": 1.28125,
      "learning_rate": 0.000340159967969047,
      "loss": 5.2653,
      "mean_token_accuracy": 0.21539633870124816,
      "num_tokens": 86912810.0,
      "step": 46620
    },
    {
      "entropy": 6.043829870223999,
      "epoch": 4.117361356411162,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003401298253049037,
      "loss": 5.1927,
      "mean_token_accuracy": 0.22608870416879653,
      "num_tokens": 86921968.0,
      "step": 46625
    },
    {
      "entropy": 6.057281351089477,
      "epoch": 4.117802896503003,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003400996813649039,
      "loss": 5.2894,
      "mean_token_accuracy": 0.21032451689243317,
      "num_tokens": 86930901.0,
      "step": 46630
    },
    {
      "entropy": 6.073928642272949,
      "epoch": 4.118244436594843,
      "grad_norm": 1.328125,
      "learning_rate": 0.00034006953614963855,
      "loss": 5.3323,
      "mean_token_accuracy": 0.21307964920997619,
      "num_tokens": 86941288.0,
      "step": 46635
    },
    {
      "entropy": 6.094373893737793,
      "epoch": 4.118685976686683,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034003938965969794,
      "loss": 5.2827,
      "mean_token_accuracy": 0.21251602619886398,
      "num_tokens": 86950938.0,
      "step": 46640
    },
    {
      "entropy": 6.040459537506104,
      "epoch": 4.119127516778524,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003400092418956726,
      "loss": 5.0688,
      "mean_token_accuracy": 0.23528458625078202,
      "num_tokens": 86959434.0,
      "step": 46645
    },
    {
      "entropy": 5.93429741859436,
      "epoch": 4.119569056870364,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003399790928581533,
      "loss": 5.1499,
      "mean_token_accuracy": 0.2336612731218338,
      "num_tokens": 86968879.0,
      "step": 46650
    },
    {
      "entropy": 5.978073406219482,
      "epoch": 4.120010596962204,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033994894254773053,
      "loss": 5.2656,
      "mean_token_accuracy": 0.21866069436073304,
      "num_tokens": 86978989.0,
      "step": 46655
    },
    {
      "entropy": 6.024601411819458,
      "epoch": 4.120452137054045,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003399187909649948,
      "loss": 5.2319,
      "mean_token_accuracy": 0.22261372655630113,
      "num_tokens": 86987797.0,
      "step": 46660
    },
    {
      "entropy": 6.07278470993042,
      "epoch": 4.120893677145885,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033988863811053695,
      "loss": 5.2326,
      "mean_token_accuracy": 0.21970274448394775,
      "num_tokens": 86997384.0,
      "step": 46665
    },
    {
      "entropy": 6.059882926940918,
      "epoch": 4.121335217237725,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003398584839849476,
      "loss": 5.2644,
      "mean_token_accuracy": 0.22159926295280458,
      "num_tokens": 87005618.0,
      "step": 46670
    },
    {
      "entropy": 6.023841428756714,
      "epoch": 4.121776757329566,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003398283285888174,
      "loss": 5.2635,
      "mean_token_accuracy": 0.224546380341053,
      "num_tokens": 87014572.0,
      "step": 46675
    },
    {
      "entropy": 6.013115882873535,
      "epoch": 4.122218297421406,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00033979817192273716,
      "loss": 5.2127,
      "mean_token_accuracy": 0.2196841135621071,
      "num_tokens": 87023389.0,
      "step": 46680
    },
    {
      "entropy": 5.995945644378662,
      "epoch": 4.122659837513246,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003397680139872975,
      "loss": 5.2875,
      "mean_token_accuracy": 0.22204382717609406,
      "num_tokens": 87032706.0,
      "step": 46685
    },
    {
      "entropy": 6.091106843948364,
      "epoch": 4.123101377605087,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00033973785478308937,
      "loss": 5.4232,
      "mean_token_accuracy": 0.2088816225528717,
      "num_tokens": 87042465.0,
      "step": 46690
    },
    {
      "entropy": 6.122280645370483,
      "epoch": 4.123542917696927,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003397076943107034,
      "loss": 5.1839,
      "mean_token_accuracy": 0.21657899916172027,
      "num_tokens": 87052139.0,
      "step": 46695
    },
    {
      "entropy": 6.018119239807129,
      "epoch": 4.123984457788767,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003396775325707306,
      "loss": 5.1694,
      "mean_token_accuracy": 0.22804346829652786,
      "num_tokens": 87061363.0,
      "step": 46700
    },
    {
      "entropy": 5.984302425384522,
      "epoch": 4.124425997880608,
      "grad_norm": 1.328125,
      "learning_rate": 0.00033964736956376164,
      "loss": 5.1799,
      "mean_token_accuracy": 0.2285684585571289,
      "num_tokens": 87070182.0,
      "step": 46705
    },
    {
      "entropy": 5.948417615890503,
      "epoch": 4.124867537972448,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033961720529038746,
      "loss": 5.1308,
      "mean_token_accuracy": 0.23723100125789642,
      "num_tokens": 87078644.0,
      "step": 46710
    },
    {
      "entropy": 6.049412679672241,
      "epoch": 4.125309078064288,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003395870397511991,
      "loss": 5.2177,
      "mean_token_accuracy": 0.22391145229339598,
      "num_tokens": 87088200.0,
      "step": 46715
    },
    {
      "entropy": 5.997036409378052,
      "epoch": 4.125750618156129,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003395568729467872,
      "loss": 5.2126,
      "mean_token_accuracy": 0.21782225966453553,
      "num_tokens": 87097694.0,
      "step": 46720
    },
    {
      "entropy": 6.050383234024048,
      "epoch": 4.126192158247969,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003395267048777431,
      "loss": 5.2161,
      "mean_token_accuracy": 0.21817565560340882,
      "num_tokens": 87106886.0,
      "step": 46725
    },
    {
      "entropy": 6.085423183441162,
      "epoch": 4.126633698339809,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003394965355446573,
      "loss": 5.32,
      "mean_token_accuracy": 0.2168234407901764,
      "num_tokens": 87115756.0,
      "step": 46730
    },
    {
      "entropy": 6.0299351692199705,
      "epoch": 4.127075238431649,
      "grad_norm": 1.375,
      "learning_rate": 0.0003394663649481212,
      "loss": 5.1988,
      "mean_token_accuracy": 0.22482921034097672,
      "num_tokens": 87124684.0,
      "step": 46735
    },
    {
      "entropy": 6.032286357879639,
      "epoch": 4.12751677852349,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003394361930887256,
      "loss": 5.2994,
      "mean_token_accuracy": 0.21849847286939622,
      "num_tokens": 87134033.0,
      "step": 46740
    },
    {
      "entropy": 5.988918733596802,
      "epoch": 4.12795831861533,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033940601996706164,
      "loss": 5.2165,
      "mean_token_accuracy": 0.22398628294467926,
      "num_tokens": 87143076.0,
      "step": 46745
    },
    {
      "entropy": 5.985193729400635,
      "epoch": 4.12839985870717,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033937584558372044,
      "loss": 5.186,
      "mean_token_accuracy": 0.22149977684020997,
      "num_tokens": 87152775.0,
      "step": 46750
    },
    {
      "entropy": 6.045371341705322,
      "epoch": 4.128841398799011,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000339345669939293,
      "loss": 5.1786,
      "mean_token_accuracy": 0.22228763550519942,
      "num_tokens": 87160758.0,
      "step": 46755
    },
    {
      "entropy": 5.951877212524414,
      "epoch": 4.129282938890851,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033931549303437046,
      "loss": 5.2388,
      "mean_token_accuracy": 0.22556886374950408,
      "num_tokens": 87171129.0,
      "step": 46760
    },
    {
      "entropy": 6.038295459747315,
      "epoch": 4.129724478982691,
      "grad_norm": 1.25,
      "learning_rate": 0.00033928531486954395,
      "loss": 5.248,
      "mean_token_accuracy": 0.22343671321868896,
      "num_tokens": 87181094.0,
      "step": 46765
    },
    {
      "entropy": 6.076821422576904,
      "epoch": 4.130166019074532,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003392551354454048,
      "loss": 5.2261,
      "mean_token_accuracy": 0.21279594600200652,
      "num_tokens": 87189799.0,
      "step": 46770
    },
    {
      "entropy": 5.962515020370484,
      "epoch": 4.130607559166372,
      "grad_norm": 1.359375,
      "learning_rate": 0.000339224954762544,
      "loss": 5.2704,
      "mean_token_accuracy": 0.22240991443395614,
      "num_tokens": 87199283.0,
      "step": 46775
    },
    {
      "entropy": 6.025371885299682,
      "epoch": 4.131049099258212,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003391947728215529,
      "loss": 5.2358,
      "mean_token_accuracy": 0.22090729624032973,
      "num_tokens": 87207719.0,
      "step": 46780
    },
    {
      "entropy": 6.052111482620239,
      "epoch": 4.131490639350053,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003391645896230227,
      "loss": 5.2445,
      "mean_token_accuracy": 0.221365986764431,
      "num_tokens": 87216263.0,
      "step": 46785
    },
    {
      "entropy": 5.994659423828125,
      "epoch": 4.131932179441893,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033913440516754475,
      "loss": 5.2246,
      "mean_token_accuracy": 0.2250296100974083,
      "num_tokens": 87224932.0,
      "step": 46790
    },
    {
      "entropy": 6.041636800765991,
      "epoch": 4.132373719533733,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003391042194557101,
      "loss": 5.3012,
      "mean_token_accuracy": 0.2199682205915451,
      "num_tokens": 87235104.0,
      "step": 46795
    },
    {
      "entropy": 6.003725290298462,
      "epoch": 4.132815259625574,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033907403248811043,
      "loss": 5.2058,
      "mean_token_accuracy": 0.22558014541864396,
      "num_tokens": 87244539.0,
      "step": 46800
    },
    {
      "entropy": 5.993515253067017,
      "epoch": 4.133256799717414,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003390438442653368,
      "loss": 5.2169,
      "mean_token_accuracy": 0.22508571147918702,
      "num_tokens": 87253626.0,
      "step": 46805
    },
    {
      "entropy": 6.0072104930877686,
      "epoch": 4.133698339809254,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003390136547879807,
      "loss": 5.2978,
      "mean_token_accuracy": 0.2204422503709793,
      "num_tokens": 87263263.0,
      "step": 46810
    },
    {
      "entropy": 6.055772590637207,
      "epoch": 4.134139879901095,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033898346405663356,
      "loss": 5.2657,
      "mean_token_accuracy": 0.2129405304789543,
      "num_tokens": 87272931.0,
      "step": 46815
    },
    {
      "entropy": 6.028457689285278,
      "epoch": 4.134581419992935,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033895327207188674,
      "loss": 5.1846,
      "mean_token_accuracy": 0.22605043053627014,
      "num_tokens": 87282555.0,
      "step": 46820
    },
    {
      "entropy": 6.110195302963257,
      "epoch": 4.135022960084775,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003389230788343316,
      "loss": 5.2748,
      "mean_token_accuracy": 0.21556832194328307,
      "num_tokens": 87291904.0,
      "step": 46825
    },
    {
      "entropy": 6.122698783874512,
      "epoch": 4.135464500176616,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003388928843445598,
      "loss": 5.284,
      "mean_token_accuracy": 0.216746087372303,
      "num_tokens": 87300819.0,
      "step": 46830
    },
    {
      "entropy": 6.000822591781616,
      "epoch": 4.135906040268456,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003388626886031627,
      "loss": 5.1893,
      "mean_token_accuracy": 0.22031385749578475,
      "num_tokens": 87310051.0,
      "step": 46835
    },
    {
      "entropy": 5.9629762172698975,
      "epoch": 4.1363475803602965,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00033883249161073184,
      "loss": 5.218,
      "mean_token_accuracy": 0.21640852391719817,
      "num_tokens": 87319054.0,
      "step": 46840
    },
    {
      "entropy": 5.988977527618408,
      "epoch": 4.136789120452137,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033880229336785884,
      "loss": 5.2583,
      "mean_token_accuracy": 0.21908158510923387,
      "num_tokens": 87328470.0,
      "step": 46845
    },
    {
      "entropy": 6.059665775299072,
      "epoch": 4.1372306605439775,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003387720938751351,
      "loss": 5.2532,
      "mean_token_accuracy": 0.21542216837406158,
      "num_tokens": 87337924.0,
      "step": 46850
    },
    {
      "entropy": 5.999898862838745,
      "epoch": 4.1376722006358175,
      "grad_norm": 1.390625,
      "learning_rate": 0.00033874189313315233,
      "loss": 5.1869,
      "mean_token_accuracy": 0.22050259858369828,
      "num_tokens": 87346616.0,
      "step": 46855
    },
    {
      "entropy": 6.039550542831421,
      "epoch": 4.1381137407276585,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003387116911425022,
      "loss": 5.3191,
      "mean_token_accuracy": 0.2091870740056038,
      "num_tokens": 87356008.0,
      "step": 46860
    },
    {
      "entropy": 6.077323484420776,
      "epoch": 4.1385552808194985,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003386814879037762,
      "loss": 5.2259,
      "mean_token_accuracy": 0.21698865592479705,
      "num_tokens": 87364301.0,
      "step": 46865
    },
    {
      "entropy": 6.095392036437988,
      "epoch": 4.138996820911339,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003386512834175661,
      "loss": 5.2719,
      "mean_token_accuracy": 0.2260529428720474,
      "num_tokens": 87373192.0,
      "step": 46870
    },
    {
      "entropy": 6.04088659286499,
      "epoch": 4.1394383610031795,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003386210776844636,
      "loss": 5.2531,
      "mean_token_accuracy": 0.2218686819076538,
      "num_tokens": 87382604.0,
      "step": 46875
    },
    {
      "entropy": 5.991037464141845,
      "epoch": 4.13987990109502,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033859087070506035,
      "loss": 5.2575,
      "mean_token_accuracy": 0.22296355962753295,
      "num_tokens": 87391394.0,
      "step": 46880
    },
    {
      "entropy": 6.055443668365479,
      "epoch": 4.14032144118686,
      "grad_norm": 1.359375,
      "learning_rate": 0.000338560662479948,
      "loss": 5.2963,
      "mean_token_accuracy": 0.22352398037910462,
      "num_tokens": 87400715.0,
      "step": 46885
    },
    {
      "entropy": 6.120811080932617,
      "epoch": 4.1407629812787,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003385304530097185,
      "loss": 5.3156,
      "mean_token_accuracy": 0.21668565422296523,
      "num_tokens": 87410431.0,
      "step": 46890
    },
    {
      "entropy": 5.983977317810059,
      "epoch": 4.141204521370541,
      "grad_norm": 1.25,
      "learning_rate": 0.00033850024229496354,
      "loss": 5.1918,
      "mean_token_accuracy": 0.22672490477561952,
      "num_tokens": 87420259.0,
      "step": 46895
    },
    {
      "entropy": 5.997186803817749,
      "epoch": 4.141646061462381,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00033847003033627505,
      "loss": 5.2321,
      "mean_token_accuracy": 0.22040641009807588,
      "num_tokens": 87428716.0,
      "step": 46900
    },
    {
      "entropy": 6.052307605743408,
      "epoch": 4.142087601554221,
      "grad_norm": 1.25,
      "learning_rate": 0.0003384398171342446,
      "loss": 5.3157,
      "mean_token_accuracy": 0.21880534291267395,
      "num_tokens": 87438432.0,
      "step": 46905
    },
    {
      "entropy": 6.087466621398926,
      "epoch": 4.142529141646062,
      "grad_norm": 1.375,
      "learning_rate": 0.0003384096026894644,
      "loss": 5.2285,
      "mean_token_accuracy": 0.21322037130594254,
      "num_tokens": 87446924.0,
      "step": 46910
    },
    {
      "entropy": 6.047093915939331,
      "epoch": 4.142970681737902,
      "grad_norm": 1.375,
      "learning_rate": 0.000338379387002526,
      "loss": 5.2258,
      "mean_token_accuracy": 0.2226158559322357,
      "num_tokens": 87456046.0,
      "step": 46915
    },
    {
      "entropy": 6.063877391815185,
      "epoch": 4.143412221829742,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003383491700740215,
      "loss": 5.2617,
      "mean_token_accuracy": 0.21429235786199569,
      "num_tokens": 87464428.0,
      "step": 46920
    },
    {
      "entropy": 6.0151927947998045,
      "epoch": 4.143853761921583,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033831895190454276,
      "loss": 5.2452,
      "mean_token_accuracy": 0.21549198627471924,
      "num_tokens": 87473939.0,
      "step": 46925
    },
    {
      "entropy": 6.070584106445312,
      "epoch": 4.144295302013423,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003382887324946818,
      "loss": 5.2813,
      "mean_token_accuracy": 0.2142787307500839,
      "num_tokens": 87484143.0,
      "step": 46930
    },
    {
      "entropy": 6.13233060836792,
      "epoch": 4.144736842105263,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003382585118450305,
      "loss": 5.2576,
      "mean_token_accuracy": 0.21028421372175216,
      "num_tokens": 87492699.0,
      "step": 46935
    },
    {
      "entropy": 5.981027030944825,
      "epoch": 4.145178382197104,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033822828995618107,
      "loss": 5.2829,
      "mean_token_accuracy": 0.2254791960120201,
      "num_tokens": 87502272.0,
      "step": 46940
    },
    {
      "entropy": 6.0666313648223875,
      "epoch": 4.145619922288944,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033819806682872534,
      "loss": 5.2825,
      "mean_token_accuracy": 0.2144685849547386,
      "num_tokens": 87512267.0,
      "step": 46945
    },
    {
      "entropy": 6.020414686203003,
      "epoch": 4.146061462380784,
      "grad_norm": 1.40625,
      "learning_rate": 0.00033816784246325536,
      "loss": 5.1418,
      "mean_token_accuracy": 0.22796837836503983,
      "num_tokens": 87522331.0,
      "step": 46950
    },
    {
      "entropy": 6.0675124168396,
      "epoch": 4.146503002472625,
      "grad_norm": 1.4375,
      "learning_rate": 0.00033813761686036326,
      "loss": 5.2638,
      "mean_token_accuracy": 0.21956991404294968,
      "num_tokens": 87531762.0,
      "step": 46955
    },
    {
      "entropy": 5.983492469787597,
      "epoch": 4.146944542564465,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033810739002064114,
      "loss": 5.273,
      "mean_token_accuracy": 0.21904943287372589,
      "num_tokens": 87542094.0,
      "step": 46960
    },
    {
      "entropy": 6.059145164489746,
      "epoch": 4.147386082656305,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003380771619446813,
      "loss": 5.3041,
      "mean_token_accuracy": 0.2071674183011055,
      "num_tokens": 87551141.0,
      "step": 46965
    },
    {
      "entropy": 6.010776567459106,
      "epoch": 4.147827622748146,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003380469326330756,
      "loss": 5.1671,
      "mean_token_accuracy": 0.2360615015029907,
      "num_tokens": 87560877.0,
      "step": 46970
    },
    {
      "entropy": 6.05339503288269,
      "epoch": 4.148269162839986,
      "grad_norm": 1.25,
      "learning_rate": 0.00033801670208641636,
      "loss": 5.232,
      "mean_token_accuracy": 0.2187453642487526,
      "num_tokens": 87569654.0,
      "step": 46975
    },
    {
      "entropy": 6.028052568435669,
      "epoch": 4.148710702931826,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033798647030529577,
      "loss": 5.2224,
      "mean_token_accuracy": 0.21929402202367781,
      "num_tokens": 87578043.0,
      "step": 46980
    },
    {
      "entropy": 6.034542369842529,
      "epoch": 4.149152243023667,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003379562372903061,
      "loss": 5.2241,
      "mean_token_accuracy": 0.2207119956612587,
      "num_tokens": 87586471.0,
      "step": 46985
    },
    {
      "entropy": 6.068107652664184,
      "epoch": 4.149593783115507,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003379260030420395,
      "loss": 5.2867,
      "mean_token_accuracy": 0.21563827097415925,
      "num_tokens": 87595389.0,
      "step": 46990
    },
    {
      "entropy": 6.08360481262207,
      "epoch": 4.150035323207347,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003378957675610883,
      "loss": 5.2511,
      "mean_token_accuracy": 0.2145577609539032,
      "num_tokens": 87604190.0,
      "step": 46995
    },
    {
      "entropy": 6.057465267181397,
      "epoch": 4.150476863299188,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033786553084804485,
      "loss": 5.3386,
      "mean_token_accuracy": 0.20908631831407548,
      "num_tokens": 87614356.0,
      "step": 47000
    },
    {
      "entropy": 6.043131256103516,
      "epoch": 4.150918403391028,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003378352929035013,
      "loss": 5.2249,
      "mean_token_accuracy": 0.22774074524641036,
      "num_tokens": 87622894.0,
      "step": 47005
    },
    {
      "entropy": 6.000787448883057,
      "epoch": 4.151359943482868,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003378050537280501,
      "loss": 5.178,
      "mean_token_accuracy": 0.22009398341178893,
      "num_tokens": 87632030.0,
      "step": 47010
    },
    {
      "entropy": 6.05826587677002,
      "epoch": 4.151801483574708,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003377748133222837,
      "loss": 5.287,
      "mean_token_accuracy": 0.21893486380577087,
      "num_tokens": 87641230.0,
      "step": 47015
    },
    {
      "entropy": 6.022808361053467,
      "epoch": 4.152243023666549,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003377445716867943,
      "loss": 5.2169,
      "mean_token_accuracy": 0.22654193192720412,
      "num_tokens": 87650720.0,
      "step": 47020
    },
    {
      "entropy": 6.042055082321167,
      "epoch": 4.152684563758389,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033771432882217443,
      "loss": 5.2113,
      "mean_token_accuracy": 0.22637775242328645,
      "num_tokens": 87659959.0,
      "step": 47025
    },
    {
      "entropy": 6.077091979980469,
      "epoch": 4.153126103850229,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003376840847290167,
      "loss": 5.3238,
      "mean_token_accuracy": 0.2153591051697731,
      "num_tokens": 87669342.0,
      "step": 47030
    },
    {
      "entropy": 5.978739786148071,
      "epoch": 4.15356764394207,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003376538394079133,
      "loss": 5.2048,
      "mean_token_accuracy": 0.22953422665596007,
      "num_tokens": 87678738.0,
      "step": 47035
    },
    {
      "entropy": 6.048980569839477,
      "epoch": 4.15400918403391,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003376235928594568,
      "loss": 5.2821,
      "mean_token_accuracy": 0.2175842195749283,
      "num_tokens": 87688830.0,
      "step": 47040
    },
    {
      "entropy": 6.058171367645263,
      "epoch": 4.15445072412575,
      "grad_norm": 1.375,
      "learning_rate": 0.0003375933450842397,
      "loss": 5.2037,
      "mean_token_accuracy": 0.2223471000790596,
      "num_tokens": 87697338.0,
      "step": 47045
    },
    {
      "entropy": 6.0084045886993405,
      "epoch": 4.154892264217591,
      "grad_norm": 1.375,
      "learning_rate": 0.00033756309608285465,
      "loss": 5.1873,
      "mean_token_accuracy": 0.2244575947523117,
      "num_tokens": 87706069.0,
      "step": 47050
    },
    {
      "entropy": 6.055674600601196,
      "epoch": 4.155333804309431,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000337532845855894,
      "loss": 5.2608,
      "mean_token_accuracy": 0.21798104494810105,
      "num_tokens": 87715838.0,
      "step": 47055
    },
    {
      "entropy": 6.085498571395874,
      "epoch": 4.155775344401271,
      "grad_norm": 1.328125,
      "learning_rate": 0.00033750259440395053,
      "loss": 5.2508,
      "mean_token_accuracy": 0.22246902137994767,
      "num_tokens": 87725070.0,
      "step": 47060
    },
    {
      "entropy": 6.036937808990478,
      "epoch": 4.156216884493112,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003374723417276169,
      "loss": 5.2073,
      "mean_token_accuracy": 0.21700123697519302,
      "num_tokens": 87733422.0,
      "step": 47065
    },
    {
      "entropy": 6.076855802536011,
      "epoch": 4.156658424584952,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003374420878274855,
      "loss": 5.3331,
      "mean_token_accuracy": 0.21820258647203444,
      "num_tokens": 87743351.0,
      "step": 47070
    },
    {
      "entropy": 6.009808588027954,
      "epoch": 4.157099964676792,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003374118327041491,
      "loss": 5.2382,
      "mean_token_accuracy": 0.22016116231679916,
      "num_tokens": 87752347.0,
      "step": 47075
    },
    {
      "entropy": 6.039243698120117,
      "epoch": 4.157541504768633,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003373815763582005,
      "loss": 5.256,
      "mean_token_accuracy": 0.2162863940000534,
      "num_tokens": 87761321.0,
      "step": 47080
    },
    {
      "entropy": 6.077903604507446,
      "epoch": 4.157983044860473,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003373513187902321,
      "loss": 5.2618,
      "mean_token_accuracy": 0.22183031141757964,
      "num_tokens": 87770305.0,
      "step": 47085
    },
    {
      "entropy": 6.0221141338348385,
      "epoch": 4.158424584952313,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000337321060000837,
      "loss": 5.2377,
      "mean_token_accuracy": 0.22673409283161164,
      "num_tokens": 87779724.0,
      "step": 47090
    },
    {
      "entropy": 5.949604082107544,
      "epoch": 4.158866125044154,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033729079999060774,
      "loss": 5.1323,
      "mean_token_accuracy": 0.22739380300045015,
      "num_tokens": 87788770.0,
      "step": 47095
    },
    {
      "entropy": 6.075643920898438,
      "epoch": 4.159307665135994,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003372605387601371,
      "loss": 5.3219,
      "mean_token_accuracy": 0.21133777350187302,
      "num_tokens": 87798573.0,
      "step": 47100
    },
    {
      "entropy": 6.0713695049285885,
      "epoch": 4.159749205227834,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033723027631001794,
      "loss": 5.2775,
      "mean_token_accuracy": 0.2133309468626976,
      "num_tokens": 87807714.0,
      "step": 47105
    },
    {
      "entropy": 6.101322841644287,
      "epoch": 4.160190745319675,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003372000126408431,
      "loss": 5.3369,
      "mean_token_accuracy": 0.2167438954114914,
      "num_tokens": 87817573.0,
      "step": 47110
    },
    {
      "entropy": 6.120215559005738,
      "epoch": 4.160632285411515,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00033716974775320534,
      "loss": 5.2975,
      "mean_token_accuracy": 0.21877691000699998,
      "num_tokens": 87827582.0,
      "step": 47115
    },
    {
      "entropy": 6.035949945449829,
      "epoch": 4.1610738255033555,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003371394816476976,
      "loss": 5.2878,
      "mean_token_accuracy": 0.21465533077716828,
      "num_tokens": 87836976.0,
      "step": 47120
    },
    {
      "entropy": 6.106387948989868,
      "epoch": 4.161515365595196,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003371092143249128,
      "loss": 5.2822,
      "mean_token_accuracy": 0.21397949308156966,
      "num_tokens": 87845638.0,
      "step": 47125
    },
    {
      "entropy": 6.044862604141235,
      "epoch": 4.1619569056870365,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003370789457854437,
      "loss": 5.204,
      "mean_token_accuracy": 0.22825151681900024,
      "num_tokens": 87855130.0,
      "step": 47130
    },
    {
      "entropy": 5.998914813995361,
      "epoch": 4.1623984457788765,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003370486760298834,
      "loss": 5.1317,
      "mean_token_accuracy": 0.233625927567482,
      "num_tokens": 87863057.0,
      "step": 47135
    },
    {
      "entropy": 5.95467095375061,
      "epoch": 4.1628399858707175,
      "grad_norm": 1.375,
      "learning_rate": 0.0003370184050588248,
      "loss": 5.2251,
      "mean_token_accuracy": 0.22452993094921112,
      "num_tokens": 87872235.0,
      "step": 47140
    },
    {
      "entropy": 5.975277996063232,
      "epoch": 4.1632815259625575,
      "grad_norm": 1.203125,
      "learning_rate": 0.000336988132872861,
      "loss": 5.2158,
      "mean_token_accuracy": 0.2267138034105301,
      "num_tokens": 87882328.0,
      "step": 47145
    },
    {
      "entropy": 6.036565065383911,
      "epoch": 4.163723066054398,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003369578594725849,
      "loss": 5.3021,
      "mean_token_accuracy": 0.21689352691173552,
      "num_tokens": 87892125.0,
      "step": 47150
    },
    {
      "entropy": 6.086338996887207,
      "epoch": 4.1641646061462385,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003369275848585895,
      "loss": 5.2084,
      "mean_token_accuracy": 0.21255279034376146,
      "num_tokens": 87901062.0,
      "step": 47155
    },
    {
      "entropy": 6.055480051040649,
      "epoch": 4.164606146238079,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000336897309031468,
      "loss": 5.2205,
      "mean_token_accuracy": 0.2217582121491432,
      "num_tokens": 87910902.0,
      "step": 47160
    },
    {
      "entropy": 6.149861812591553,
      "epoch": 4.165047686329919,
      "grad_norm": 1.25,
      "learning_rate": 0.00033686703199181347,
      "loss": 5.345,
      "mean_token_accuracy": 0.20471003502607346,
      "num_tokens": 87921135.0,
      "step": 47165
    },
    {
      "entropy": 6.023531770706176,
      "epoch": 4.165489226421759,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033683675374021887,
      "loss": 5.2415,
      "mean_token_accuracy": 0.2136955589056015,
      "num_tokens": 87929795.0,
      "step": 47170
    },
    {
      "entropy": 5.9227972507476805,
      "epoch": 4.1659307665136,
      "grad_norm": 2.0625,
      "learning_rate": 0.0003368064742772775,
      "loss": 5.1027,
      "mean_token_accuracy": 0.2326638400554657,
      "num_tokens": 87940144.0,
      "step": 47175
    },
    {
      "entropy": 6.0094566822052,
      "epoch": 4.16637230660544,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00033677619360358244,
      "loss": 5.2094,
      "mean_token_accuracy": 0.22292864322662354,
      "num_tokens": 87949428.0,
      "step": 47180
    },
    {
      "entropy": 6.057882785797119,
      "epoch": 4.16681384669728,
      "grad_norm": 1.25,
      "learning_rate": 0.00033674591171972684,
      "loss": 5.306,
      "mean_token_accuracy": 0.21570100784301757,
      "num_tokens": 87958138.0,
      "step": 47185
    },
    {
      "entropy": 6.158877182006836,
      "epoch": 4.167255386789121,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033671562862630397,
      "loss": 5.4123,
      "mean_token_accuracy": 0.2099520429968834,
      "num_tokens": 87969242.0,
      "step": 47190
    },
    {
      "entropy": 6.032397508621216,
      "epoch": 4.167696926880961,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00033668534432390704,
      "loss": 5.1323,
      "mean_token_accuracy": 0.22990034222602845,
      "num_tokens": 87978122.0,
      "step": 47195
    },
    {
      "entropy": 5.966269302368164,
      "epoch": 4.168138466972801,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003366550588131293,
      "loss": 5.1142,
      "mean_token_accuracy": 0.23266829997301103,
      "num_tokens": 87987420.0,
      "step": 47200
    },
    {
      "entropy": 5.992639350891113,
      "epoch": 4.168580007064642,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000336624772094564,
      "loss": 5.2047,
      "mean_token_accuracy": 0.22011700123548508,
      "num_tokens": 87996572.0,
      "step": 47205
    },
    {
      "entropy": 6.042632102966309,
      "epoch": 4.169021547156482,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033659448416880445,
      "loss": 5.2686,
      "mean_token_accuracy": 0.21654469817876815,
      "num_tokens": 88007281.0,
      "step": 47210
    },
    {
      "entropy": 6.003139209747315,
      "epoch": 4.169463087248322,
      "grad_norm": 1.25,
      "learning_rate": 0.00033656419503644405,
      "loss": 5.2028,
      "mean_token_accuracy": 0.22166722863912583,
      "num_tokens": 88016734.0,
      "step": 47215
    },
    {
      "entropy": 5.980227708816528,
      "epoch": 4.169904627340163,
      "grad_norm": 1.40625,
      "learning_rate": 0.000336533904698076,
      "loss": 5.2089,
      "mean_token_accuracy": 0.22228860408067702,
      "num_tokens": 88025489.0,
      "step": 47220
    },
    {
      "entropy": 6.064529943466186,
      "epoch": 4.170346167432003,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003365036131542938,
      "loss": 5.3378,
      "mean_token_accuracy": 0.2102780431509018,
      "num_tokens": 88035605.0,
      "step": 47225
    },
    {
      "entropy": 6.148168468475342,
      "epoch": 4.170787707523843,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003364733204056908,
      "loss": 5.2491,
      "mean_token_accuracy": 0.217270328104496,
      "num_tokens": 88044118.0,
      "step": 47230
    },
    {
      "entropy": 6.110960054397583,
      "epoch": 4.171229247615684,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003364430264528605,
      "loss": 5.2557,
      "mean_token_accuracy": 0.22130610942840576,
      "num_tokens": 88052986.0,
      "step": 47235
    },
    {
      "entropy": 5.98006649017334,
      "epoch": 4.171670787707524,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003364127312963962,
      "loss": 5.2558,
      "mean_token_accuracy": 0.2188846006989479,
      "num_tokens": 88061670.0,
      "step": 47240
    },
    {
      "entropy": 6.0376640319824215,
      "epoch": 4.172112327799364,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003363824349368913,
      "loss": 5.352,
      "mean_token_accuracy": 0.20411040037870407,
      "num_tokens": 88072394.0,
      "step": 47245
    },
    {
      "entropy": 6.10016393661499,
      "epoch": 4.172553867891205,
      "grad_norm": 1.375,
      "learning_rate": 0.00033635213737493954,
      "loss": 5.2877,
      "mean_token_accuracy": 0.21027246564626695,
      "num_tokens": 88081962.0,
      "step": 47250
    },
    {
      "entropy": 6.033314418792725,
      "epoch": 4.172995407983045,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003363218386111342,
      "loss": 5.1389,
      "mean_token_accuracy": 0.22313956320285797,
      "num_tokens": 88091200.0,
      "step": 47255
    },
    {
      "entropy": 6.01614933013916,
      "epoch": 4.173436948074885,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000336291538646069,
      "loss": 5.232,
      "mean_token_accuracy": 0.22100694328546525,
      "num_tokens": 88100663.0,
      "step": 47260
    },
    {
      "entropy": 6.0339076042175295,
      "epoch": 4.173878488166726,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003362612374803374,
      "loss": 5.2231,
      "mean_token_accuracy": 0.2214755743741989,
      "num_tokens": 88109972.0,
      "step": 47265
    },
    {
      "entropy": 6.046849060058594,
      "epoch": 4.174320028258566,
      "grad_norm": 1.171875,
      "learning_rate": 0.000336230935114533,
      "loss": 5.2785,
      "mean_token_accuracy": 0.21183830946683885,
      "num_tokens": 88119894.0,
      "step": 47270
    },
    {
      "entropy": 6.076026964187622,
      "epoch": 4.174761568350406,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033620063154924935,
      "loss": 5.3146,
      "mean_token_accuracy": 0.21558740437030793,
      "num_tokens": 88130362.0,
      "step": 47275
    },
    {
      "entropy": 6.106430625915527,
      "epoch": 4.175203108442247,
      "grad_norm": 1.421875,
      "learning_rate": 0.00033617032678508017,
      "loss": 5.2557,
      "mean_token_accuracy": 0.21997860074043274,
      "num_tokens": 88139412.0,
      "step": 47280
    },
    {
      "entropy": 6.070490121841431,
      "epoch": 4.175644648534087,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003361400208226191,
      "loss": 5.3521,
      "mean_token_accuracy": 0.20935109853744507,
      "num_tokens": 88148466.0,
      "step": 47285
    },
    {
      "entropy": 6.044375610351563,
      "epoch": 4.176086188625927,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00033610971366245977,
      "loss": 5.2436,
      "mean_token_accuracy": 0.21733300685882567,
      "num_tokens": 88157432.0,
      "step": 47290
    },
    {
      "entropy": 6.076900672912598,
      "epoch": 4.176527728717767,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003360794053051959,
      "loss": 5.2852,
      "mean_token_accuracy": 0.21430927515029907,
      "num_tokens": 88166761.0,
      "step": 47295
    },
    {
      "entropy": 6.053371334075928,
      "epoch": 4.176969268809608,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033604909575142127,
      "loss": 5.2946,
      "mean_token_accuracy": 0.214069926738739,
      "num_tokens": 88176691.0,
      "step": 47300
    },
    {
      "entropy": 5.988549137115479,
      "epoch": 4.177410808901448,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00033601878500172963,
      "loss": 5.1232,
      "mean_token_accuracy": 0.2379852667450905,
      "num_tokens": 88185097.0,
      "step": 47305
    },
    {
      "entropy": 5.987457180023194,
      "epoch": 4.177852348993288,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003359884730567145,
      "loss": 5.2481,
      "mean_token_accuracy": 0.22586676627397537,
      "num_tokens": 88195909.0,
      "step": 47310
    },
    {
      "entropy": 6.081286811828614,
      "epoch": 4.178293889085129,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00033595815991697004,
      "loss": 5.2755,
      "mean_token_accuracy": 0.21670358628034592,
      "num_tokens": 88204642.0,
      "step": 47315
    },
    {
      "entropy": 6.080613231658935,
      "epoch": 4.178735429176969,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003359278455830898,
      "loss": 5.2915,
      "mean_token_accuracy": 0.21774663180112838,
      "num_tokens": 88213575.0,
      "step": 47320
    },
    {
      "entropy": 6.00371732711792,
      "epoch": 4.179176969268809,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033589753005566775,
      "loss": 5.2623,
      "mean_token_accuracy": 0.2218436196446419,
      "num_tokens": 88223702.0,
      "step": 47325
    },
    {
      "entropy": 6.072292041778565,
      "epoch": 4.17961850936065,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033586721333529767,
      "loss": 5.3049,
      "mean_token_accuracy": 0.21453950852155684,
      "num_tokens": 88233686.0,
      "step": 47330
    },
    {
      "entropy": 6.052868795394898,
      "epoch": 4.18006004945249,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00033583689542257355,
      "loss": 5.2334,
      "mean_token_accuracy": 0.2233275964856148,
      "num_tokens": 88242681.0,
      "step": 47335
    },
    {
      "entropy": 5.981100845336914,
      "epoch": 4.18050158954433,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00033580657631808924,
      "loss": 5.2504,
      "mean_token_accuracy": 0.22369497567415236,
      "num_tokens": 88251553.0,
      "step": 47340
    },
    {
      "entropy": 5.961998987197876,
      "epoch": 4.180943129636171,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033577625602243874,
      "loss": 5.265,
      "mean_token_accuracy": 0.21623309403657914,
      "num_tokens": 88260735.0,
      "step": 47345
    },
    {
      "entropy": 6.0703246116638185,
      "epoch": 4.181384669728011,
      "grad_norm": 1.25,
      "learning_rate": 0.0003357459345362158,
      "loss": 5.2783,
      "mean_token_accuracy": 0.21683321744203568,
      "num_tokens": 88271384.0,
      "step": 47350
    },
    {
      "entropy": 6.09986515045166,
      "epoch": 4.181826209819851,
      "grad_norm": 1.40625,
      "learning_rate": 0.00033571561186001474,
      "loss": 5.285,
      "mean_token_accuracy": 0.2120923489332199,
      "num_tokens": 88280195.0,
      "step": 47355
    },
    {
      "entropy": 5.996657085418701,
      "epoch": 4.182267749911692,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033568528799442926,
      "loss": 5.2407,
      "mean_token_accuracy": 0.22063531428575517,
      "num_tokens": 88288977.0,
      "step": 47360
    },
    {
      "entropy": 6.121722459793091,
      "epoch": 4.182709290003532,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033565496294005355,
      "loss": 5.4065,
      "mean_token_accuracy": 0.20400693118572236,
      "num_tokens": 88298200.0,
      "step": 47365
    },
    {
      "entropy": 5.979743003845215,
      "epoch": 4.183150830095372,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003356246366974816,
      "loss": 5.1585,
      "mean_token_accuracy": 0.2260609358549118,
      "num_tokens": 88307866.0,
      "step": 47370
    },
    {
      "entropy": 6.1359764575958256,
      "epoch": 4.183592370187213,
      "grad_norm": 1.359375,
      "learning_rate": 0.00033559430926730744,
      "loss": 5.3512,
      "mean_token_accuracy": 0.21837151497602464,
      "num_tokens": 88316785.0,
      "step": 47375
    },
    {
      "entropy": 6.072799253463745,
      "epoch": 4.184033910279053,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033556398065012527,
      "loss": 5.3133,
      "mean_token_accuracy": 0.2095215603709221,
      "num_tokens": 88326232.0,
      "step": 47380
    },
    {
      "entropy": 6.043931913375855,
      "epoch": 4.184475450370893,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003355336508465291,
      "loss": 5.2644,
      "mean_token_accuracy": 0.21853403896093368,
      "num_tokens": 88336246.0,
      "step": 47385
    },
    {
      "entropy": 6.03804087638855,
      "epoch": 4.184916990462734,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003355033198571133,
      "loss": 5.2483,
      "mean_token_accuracy": 0.21153480857610701,
      "num_tokens": 88345388.0,
      "step": 47390
    },
    {
      "entropy": 6.057508134841919,
      "epoch": 4.185358530554574,
      "grad_norm": 1.25,
      "learning_rate": 0.00033547298768247174,
      "loss": 5.267,
      "mean_token_accuracy": 0.2165695920586586,
      "num_tokens": 88355197.0,
      "step": 47395
    },
    {
      "entropy": 6.118351173400879,
      "epoch": 4.1858000706464145,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033544265432319874,
      "loss": 5.3241,
      "mean_token_accuracy": 0.2108940914273262,
      "num_tokens": 88364595.0,
      "step": 47400
    },
    {
      "entropy": 6.046520566940307,
      "epoch": 4.186241610738255,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033541231977988854,
      "loss": 5.1869,
      "mean_token_accuracy": 0.2257544830441475,
      "num_tokens": 88373911.0,
      "step": 47405
    },
    {
      "entropy": 6.060771036148071,
      "epoch": 4.1866831508300955,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033538198405313536,
      "loss": 5.2145,
      "mean_token_accuracy": 0.22190752923488616,
      "num_tokens": 88383839.0,
      "step": 47410
    },
    {
      "entropy": 5.9499107837677006,
      "epoch": 4.1871246909219355,
      "grad_norm": 1.34375,
      "learning_rate": 0.00033535164714353343,
      "loss": 5.1731,
      "mean_token_accuracy": 0.22875013798475266,
      "num_tokens": 88392709.0,
      "step": 47415
    },
    {
      "entropy": 6.106231451034546,
      "epoch": 4.1875662310137765,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003353213090516771,
      "loss": 5.3788,
      "mean_token_accuracy": 0.20923480540513992,
      "num_tokens": 88402817.0,
      "step": 47420
    },
    {
      "entropy": 6.130472803115845,
      "epoch": 4.1880077711056165,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033529096977816063,
      "loss": 5.36,
      "mean_token_accuracy": 0.2087341472506523,
      "num_tokens": 88412689.0,
      "step": 47425
    },
    {
      "entropy": 6.095037412643433,
      "epoch": 4.188449311197457,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033526062932357837,
      "loss": 5.3135,
      "mean_token_accuracy": 0.21486147344112397,
      "num_tokens": 88422166.0,
      "step": 47430
    },
    {
      "entropy": 6.060799312591553,
      "epoch": 4.1888908512892975,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033523028768852464,
      "loss": 5.2433,
      "mean_token_accuracy": 0.2224324405193329,
      "num_tokens": 88431446.0,
      "step": 47435
    },
    {
      "entropy": 5.913412857055664,
      "epoch": 4.1893323913811376,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003351999448735938,
      "loss": 5.1603,
      "mean_token_accuracy": 0.22893820703029633,
      "num_tokens": 88440419.0,
      "step": 47440
    },
    {
      "entropy": 6.02681245803833,
      "epoch": 4.189773931472978,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033516960087938034,
      "loss": 5.284,
      "mean_token_accuracy": 0.2138472467660904,
      "num_tokens": 88449245.0,
      "step": 47445
    },
    {
      "entropy": 6.012096118927002,
      "epoch": 4.190215471564818,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033513925570647854,
      "loss": 5.2628,
      "mean_token_accuracy": 0.22599547058343888,
      "num_tokens": 88461335.0,
      "step": 47450
    },
    {
      "entropy": 6.06014494895935,
      "epoch": 4.190657011656659,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003351089093554829,
      "loss": 5.182,
      "mean_token_accuracy": 0.22284157127141951,
      "num_tokens": 88469408.0,
      "step": 47455
    },
    {
      "entropy": 6.024022436141967,
      "epoch": 4.191098551748499,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033507856182698803,
      "loss": 5.2538,
      "mean_token_accuracy": 0.22370481491088867,
      "num_tokens": 88478266.0,
      "step": 47460
    },
    {
      "entropy": 6.042216205596924,
      "epoch": 4.191540091840339,
      "grad_norm": 1.421875,
      "learning_rate": 0.00033504821312158815,
      "loss": 5.2569,
      "mean_token_accuracy": 0.22341417968273164,
      "num_tokens": 88487515.0,
      "step": 47465
    },
    {
      "entropy": 6.085191822052002,
      "epoch": 4.19198163193218,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000335017863239878,
      "loss": 5.3891,
      "mean_token_accuracy": 0.2108893424272537,
      "num_tokens": 88497218.0,
      "step": 47470
    },
    {
      "entropy": 6.155262851715088,
      "epoch": 4.19242317202402,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003349875121824521,
      "loss": 5.294,
      "mean_token_accuracy": 0.21449677646160126,
      "num_tokens": 88506501.0,
      "step": 47475
    },
    {
      "entropy": 5.944053745269775,
      "epoch": 4.19286471211586,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003349571599499049,
      "loss": 5.1717,
      "mean_token_accuracy": 0.23076231628656388,
      "num_tokens": 88515941.0,
      "step": 47480
    },
    {
      "entropy": 6.081558799743652,
      "epoch": 4.193306252207701,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003349268065428311,
      "loss": 5.2695,
      "mean_token_accuracy": 0.21257572770118713,
      "num_tokens": 88524541.0,
      "step": 47485
    },
    {
      "entropy": 6.045158386230469,
      "epoch": 4.193747792299541,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00033489645196182517,
      "loss": 5.1805,
      "mean_token_accuracy": 0.23287620395421982,
      "num_tokens": 88533197.0,
      "step": 47490
    },
    {
      "entropy": 5.962162637710572,
      "epoch": 4.194189332391381,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033486609620748183,
      "loss": 5.1463,
      "mean_token_accuracy": 0.2229946494102478,
      "num_tokens": 88541792.0,
      "step": 47495
    },
    {
      "entropy": 5.9725486755371096,
      "epoch": 4.194630872483222,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033483573928039576,
      "loss": 5.2305,
      "mean_token_accuracy": 0.21906874775886537,
      "num_tokens": 88551751.0,
      "step": 47500
    },
    {
      "entropy": 5.973042678833008,
      "epoch": 4.195072412575062,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033480538118116143,
      "loss": 5.2348,
      "mean_token_accuracy": 0.22089468836784362,
      "num_tokens": 88561315.0,
      "step": 47505
    },
    {
      "entropy": 6.065127992630005,
      "epoch": 4.195513952666902,
      "grad_norm": 1.125,
      "learning_rate": 0.0003347750219103739,
      "loss": 5.3483,
      "mean_token_accuracy": 0.20935450196266175,
      "num_tokens": 88571977.0,
      "step": 47510
    },
    {
      "entropy": 6.095681428909302,
      "epoch": 4.195955492758743,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033474466146862753,
      "loss": 5.2866,
      "mean_token_accuracy": 0.21020708680152894,
      "num_tokens": 88582953.0,
      "step": 47515
    },
    {
      "entropy": 6.091692209243774,
      "epoch": 4.196397032850583,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033471429985651734,
      "loss": 5.2803,
      "mean_token_accuracy": 0.217060424387455,
      "num_tokens": 88591536.0,
      "step": 47520
    },
    {
      "entropy": 6.059651565551758,
      "epoch": 4.196838572942423,
      "grad_norm": 1.25,
      "learning_rate": 0.0003346839370746379,
      "loss": 5.3057,
      "mean_token_accuracy": 0.2217210575938225,
      "num_tokens": 88601638.0,
      "step": 47525
    },
    {
      "entropy": 6.132203435897827,
      "epoch": 4.197280113034264,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003346535731235841,
      "loss": 5.3253,
      "mean_token_accuracy": 0.20729554295539857,
      "num_tokens": 88610092.0,
      "step": 47530
    },
    {
      "entropy": 6.112583160400391,
      "epoch": 4.197721653126104,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033462320800395064,
      "loss": 5.2073,
      "mean_token_accuracy": 0.2239359512925148,
      "num_tokens": 88620071.0,
      "step": 47535
    },
    {
      "entropy": 6.016206312179565,
      "epoch": 4.198163193217944,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003345928417163325,
      "loss": 5.2537,
      "mean_token_accuracy": 0.21238514333963393,
      "num_tokens": 88628939.0,
      "step": 47540
    },
    {
      "entropy": 5.989529037475586,
      "epoch": 4.198604733309785,
      "grad_norm": 1.375,
      "learning_rate": 0.0003345624742613246,
      "loss": 5.2692,
      "mean_token_accuracy": 0.21908591091632842,
      "num_tokens": 88638873.0,
      "step": 47545
    },
    {
      "entropy": 6.081983280181885,
      "epoch": 4.199046273401625,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033453210563952156,
      "loss": 5.2854,
      "mean_token_accuracy": 0.22018653601408006,
      "num_tokens": 88648362.0,
      "step": 47550
    },
    {
      "entropy": 6.07322325706482,
      "epoch": 4.199487813493465,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033450173585151857,
      "loss": 5.2525,
      "mean_token_accuracy": 0.21837648153305053,
      "num_tokens": 88657991.0,
      "step": 47555
    },
    {
      "entropy": 6.05176568031311,
      "epoch": 4.199929353585306,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033447136489791026,
      "loss": 5.2517,
      "mean_token_accuracy": 0.22002002894878386,
      "num_tokens": 88667127.0,
      "step": 47560
    },
    {
      "entropy": 6.03548412322998,
      "epoch": 4.200370893677146,
      "grad_norm": 1.421875,
      "learning_rate": 0.00033444099277929186,
      "loss": 5.2179,
      "mean_token_accuracy": 0.220260389149189,
      "num_tokens": 88675761.0,
      "step": 47565
    },
    {
      "entropy": 5.966809368133545,
      "epoch": 4.200812433768986,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003344106194962581,
      "loss": 5.2167,
      "mean_token_accuracy": 0.22127882838249208,
      "num_tokens": 88684775.0,
      "step": 47570
    },
    {
      "entropy": 6.060572957992553,
      "epoch": 4.201253973860826,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033438024504940417,
      "loss": 5.236,
      "mean_token_accuracy": 0.2151380717754364,
      "num_tokens": 88693162.0,
      "step": 47575
    },
    {
      "entropy": 6.056325674057007,
      "epoch": 4.201695513952667,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000334349869439325,
      "loss": 5.2885,
      "mean_token_accuracy": 0.21571300625801088,
      "num_tokens": 88702382.0,
      "step": 47580
    },
    {
      "entropy": 6.044991874694825,
      "epoch": 4.202137054044507,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033431949266661564,
      "loss": 5.2012,
      "mean_token_accuracy": 0.2126961275935173,
      "num_tokens": 88712219.0,
      "step": 47585
    },
    {
      "entropy": 6.078792428970337,
      "epoch": 4.202578594136347,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003342891147318711,
      "loss": 5.2263,
      "mean_token_accuracy": 0.2296944096684456,
      "num_tokens": 88720236.0,
      "step": 47590
    },
    {
      "entropy": 6.03631100654602,
      "epoch": 4.203020134228188,
      "grad_norm": 1.375,
      "learning_rate": 0.00033425873563568657,
      "loss": 5.2429,
      "mean_token_accuracy": 0.22039631009101868,
      "num_tokens": 88729917.0,
      "step": 47595
    },
    {
      "entropy": 6.015591669082641,
      "epoch": 4.203461674320028,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003342283553786572,
      "loss": 5.1963,
      "mean_token_accuracy": 0.22409854382276534,
      "num_tokens": 88739667.0,
      "step": 47600
    },
    {
      "entropy": 5.981193399429321,
      "epoch": 4.203903214411868,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003341979739613779,
      "loss": 5.2887,
      "mean_token_accuracy": 0.21758417934179305,
      "num_tokens": 88750377.0,
      "step": 47605
    },
    {
      "entropy": 6.043108034133911,
      "epoch": 4.204344754503709,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033416759138444403,
      "loss": 5.2741,
      "mean_token_accuracy": 0.22396718561649323,
      "num_tokens": 88759319.0,
      "step": 47610
    },
    {
      "entropy": 6.091644382476806,
      "epoch": 4.204786294595549,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033413720764845066,
      "loss": 5.3067,
      "mean_token_accuracy": 0.21612222641706466,
      "num_tokens": 88768520.0,
      "step": 47615
    },
    {
      "entropy": 6.0687755107879635,
      "epoch": 4.205227834687389,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033410682275399306,
      "loss": 5.2422,
      "mean_token_accuracy": 0.21796325892210006,
      "num_tokens": 88777090.0,
      "step": 47620
    },
    {
      "entropy": 6.074193811416626,
      "epoch": 4.20566937477923,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033407643670166636,
      "loss": 5.3044,
      "mean_token_accuracy": 0.21629733741283416,
      "num_tokens": 88787025.0,
      "step": 47625
    },
    {
      "entropy": 6.0095597267150875,
      "epoch": 4.20611091487107,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003340460494920659,
      "loss": 5.1912,
      "mean_token_accuracy": 0.23319206535816192,
      "num_tokens": 88795876.0,
      "step": 47630
    },
    {
      "entropy": 6.0719715595245365,
      "epoch": 4.20655245496291,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000334015661125787,
      "loss": 5.295,
      "mean_token_accuracy": 0.21363986581563948,
      "num_tokens": 88805248.0,
      "step": 47635
    },
    {
      "entropy": 6.027077484130859,
      "epoch": 4.206993995054751,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003339852716034247,
      "loss": 5.2055,
      "mean_token_accuracy": 0.2238255724310875,
      "num_tokens": 88815797.0,
      "step": 47640
    },
    {
      "entropy": 5.994087648391724,
      "epoch": 4.207435535146591,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003339548809255746,
      "loss": 5.1492,
      "mean_token_accuracy": 0.22490921020507812,
      "num_tokens": 88824205.0,
      "step": 47645
    },
    {
      "entropy": 6.019708347320557,
      "epoch": 4.207877075238431,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003339244890928318,
      "loss": 5.2481,
      "mean_token_accuracy": 0.22310028970241547,
      "num_tokens": 88833451.0,
      "step": 47650
    },
    {
      "entropy": 6.077857494354248,
      "epoch": 4.208318615330272,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033389409610579183,
      "loss": 5.3381,
      "mean_token_accuracy": 0.21186116486787795,
      "num_tokens": 88843922.0,
      "step": 47655
    },
    {
      "entropy": 6.008256721496582,
      "epoch": 4.208760155422112,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033386370196505,
      "loss": 5.1781,
      "mean_token_accuracy": 0.23141681551933288,
      "num_tokens": 88852913.0,
      "step": 47660
    },
    {
      "entropy": 6.0210682392120365,
      "epoch": 4.209201695513952,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033383330667120174,
      "loss": 5.2043,
      "mean_token_accuracy": 0.22199542671442032,
      "num_tokens": 88861779.0,
      "step": 47665
    },
    {
      "entropy": 5.991804409027099,
      "epoch": 4.209643235605793,
      "grad_norm": 1.25,
      "learning_rate": 0.0003338029102248425,
      "loss": 5.2247,
      "mean_token_accuracy": 0.2151365578174591,
      "num_tokens": 88871553.0,
      "step": 47670
    },
    {
      "entropy": 6.125069904327392,
      "epoch": 4.210084775697633,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00033377251262656763,
      "loss": 5.3241,
      "mean_token_accuracy": 0.21180660873651505,
      "num_tokens": 88881066.0,
      "step": 47675
    },
    {
      "entropy": 5.971172142028808,
      "epoch": 4.2105263157894735,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003337421138769727,
      "loss": 5.0243,
      "mean_token_accuracy": 0.24856549501419067,
      "num_tokens": 88889831.0,
      "step": 47680
    },
    {
      "entropy": 6.033763074874878,
      "epoch": 4.210967855881314,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003337117139766532,
      "loss": 5.1532,
      "mean_token_accuracy": 0.2261385902762413,
      "num_tokens": 88898657.0,
      "step": 47685
    },
    {
      "entropy": 6.031268167495727,
      "epoch": 4.2114093959731544,
      "grad_norm": 1.359375,
      "learning_rate": 0.00033368131292620454,
      "loss": 5.2295,
      "mean_token_accuracy": 0.21536605507135392,
      "num_tokens": 88907773.0,
      "step": 47690
    },
    {
      "entropy": 6.085370349884033,
      "epoch": 4.2118509360649945,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003336509107262224,
      "loss": 5.2973,
      "mean_token_accuracy": 0.21648527979850768,
      "num_tokens": 88916816.0,
      "step": 47695
    },
    {
      "entropy": 6.0957444190979,
      "epoch": 4.212292476156835,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003336205073773022,
      "loss": 5.3007,
      "mean_token_accuracy": 0.21315903663635255,
      "num_tokens": 88926665.0,
      "step": 47700
    },
    {
      "entropy": 6.032852506637573,
      "epoch": 4.2127340162486755,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033359010288003965,
      "loss": 5.244,
      "mean_token_accuracy": 0.22594610899686812,
      "num_tokens": 88936627.0,
      "step": 47705
    },
    {
      "entropy": 6.032251358032227,
      "epoch": 4.2131755563405155,
      "grad_norm": 1.25,
      "learning_rate": 0.0003335596972350303,
      "loss": 5.1998,
      "mean_token_accuracy": 0.22362436801195146,
      "num_tokens": 88945594.0,
      "step": 47710
    },
    {
      "entropy": 6.081712245941162,
      "epoch": 4.2136170964323565,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033352929044286973,
      "loss": 5.2539,
      "mean_token_accuracy": 0.214470537006855,
      "num_tokens": 88954919.0,
      "step": 47715
    },
    {
      "entropy": 6.021346139907837,
      "epoch": 4.2140586365241965,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00033349888250415374,
      "loss": 5.2677,
      "mean_token_accuracy": 0.21294835954904556,
      "num_tokens": 88963981.0,
      "step": 47720
    },
    {
      "entropy": 6.048584938049316,
      "epoch": 4.214500176616037,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00033346847341947784,
      "loss": 5.3005,
      "mean_token_accuracy": 0.219223228096962,
      "num_tokens": 88973210.0,
      "step": 47725
    },
    {
      "entropy": 6.022293901443481,
      "epoch": 4.214941716707877,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003334380631894378,
      "loss": 5.2669,
      "mean_token_accuracy": 0.22158104926347733,
      "num_tokens": 88982327.0,
      "step": 47730
    },
    {
      "entropy": 6.048601150512695,
      "epoch": 4.215383256799718,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00033340765181462937,
      "loss": 5.2565,
      "mean_token_accuracy": 0.21785263419151307,
      "num_tokens": 88992759.0,
      "step": 47735
    },
    {
      "entropy": 6.098325920104981,
      "epoch": 4.215824796891558,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003333772392956483,
      "loss": 5.1688,
      "mean_token_accuracy": 0.2182217374444008,
      "num_tokens": 89000863.0,
      "step": 47740
    },
    {
      "entropy": 6.013726758956909,
      "epoch": 4.216266336983398,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033334682563309026,
      "loss": 5.1938,
      "mean_token_accuracy": 0.2345752611756325,
      "num_tokens": 89010526.0,
      "step": 47745
    },
    {
      "entropy": 6.0396136283874515,
      "epoch": 4.216707877075239,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00033331641082755113,
      "loss": 5.2666,
      "mean_token_accuracy": 0.21714781671762468,
      "num_tokens": 89019942.0,
      "step": 47750
    },
    {
      "entropy": 6.003597354888916,
      "epoch": 4.217149417167079,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033328599487962666,
      "loss": 5.1419,
      "mean_token_accuracy": 0.23423126786947251,
      "num_tokens": 89028686.0,
      "step": 47755
    },
    {
      "entropy": 5.956752490997315,
      "epoch": 4.217590957258919,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033325557778991267,
      "loss": 5.1236,
      "mean_token_accuracy": 0.23130838125944136,
      "num_tokens": 89037384.0,
      "step": 47760
    },
    {
      "entropy": 5.978174877166748,
      "epoch": 4.21803249735076,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000333225159559005,
      "loss": 5.1684,
      "mean_token_accuracy": 0.23355485498905182,
      "num_tokens": 89046678.0,
      "step": 47765
    },
    {
      "entropy": 5.939715957641601,
      "epoch": 4.2184740374426,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033319474018749965,
      "loss": 5.1313,
      "mean_token_accuracy": 0.23195002228021622,
      "num_tokens": 89055066.0,
      "step": 47770
    },
    {
      "entropy": 5.999209642410278,
      "epoch": 4.21891557753444,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00033316431967599243,
      "loss": 5.2916,
      "mean_token_accuracy": 0.2203432247042656,
      "num_tokens": 89064876.0,
      "step": 47775
    },
    {
      "entropy": 6.000910520553589,
      "epoch": 4.219357117626281,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003331338980250792,
      "loss": 5.1697,
      "mean_token_accuracy": 0.23096369951963425,
      "num_tokens": 89073764.0,
      "step": 47780
    },
    {
      "entropy": 6.007117652893067,
      "epoch": 4.219798657718121,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033310347523535607,
      "loss": 5.1919,
      "mean_token_accuracy": 0.22505367547273636,
      "num_tokens": 89082863.0,
      "step": 47785
    },
    {
      "entropy": 6.014075183868409,
      "epoch": 4.220240197809961,
      "grad_norm": 1.359375,
      "learning_rate": 0.00033307305130741884,
      "loss": 5.2083,
      "mean_token_accuracy": 0.22857119888067245,
      "num_tokens": 89091461.0,
      "step": 47790
    },
    {
      "entropy": 6.116459512710572,
      "epoch": 4.220681737901802,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003330426262418636,
      "loss": 5.2818,
      "mean_token_accuracy": 0.21913449466228485,
      "num_tokens": 89101122.0,
      "step": 47795
    },
    {
      "entropy": 6.021703290939331,
      "epoch": 4.221123277993642,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003330122000392863,
      "loss": 5.1619,
      "mean_token_accuracy": 0.229524601995945,
      "num_tokens": 89110534.0,
      "step": 47800
    },
    {
      "entropy": 5.864205169677734,
      "epoch": 4.221564818085482,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033298177270028303,
      "loss": 5.1049,
      "mean_token_accuracy": 0.23514376133680343,
      "num_tokens": 89119064.0,
      "step": 47805
    },
    {
      "entropy": 6.02984356880188,
      "epoch": 4.222006358177323,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003329513442254497,
      "loss": 5.2034,
      "mean_token_accuracy": 0.21494588553905486,
      "num_tokens": 89127979.0,
      "step": 47810
    },
    {
      "entropy": 6.0121783256530765,
      "epoch": 4.222447898269163,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003329209146153827,
      "loss": 5.2429,
      "mean_token_accuracy": 0.22214448153972627,
      "num_tokens": 89136824.0,
      "step": 47815
    },
    {
      "entropy": 6.099581575393676,
      "epoch": 4.222889438361003,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033289048387067765,
      "loss": 5.2773,
      "mean_token_accuracy": 0.213793283700943,
      "num_tokens": 89146852.0,
      "step": 47820
    },
    {
      "entropy": 6.060708999633789,
      "epoch": 4.223330978452844,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00033286005199193114,
      "loss": 5.219,
      "mean_token_accuracy": 0.217137148976326,
      "num_tokens": 89155843.0,
      "step": 47825
    },
    {
      "entropy": 6.010730457305908,
      "epoch": 4.223772518544684,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003328296189797391,
      "loss": 5.2307,
      "mean_token_accuracy": 0.21734233051538468,
      "num_tokens": 89164278.0,
      "step": 47830
    },
    {
      "entropy": 6.1043730735778805,
      "epoch": 4.224214058636524,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003327991848346977,
      "loss": 5.2784,
      "mean_token_accuracy": 0.22333239763975143,
      "num_tokens": 89174036.0,
      "step": 47835
    },
    {
      "entropy": 6.111544036865235,
      "epoch": 4.224655598728365,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033276874955740307,
      "loss": 5.2807,
      "mean_token_accuracy": 0.21039090156555176,
      "num_tokens": 89183665.0,
      "step": 47840
    },
    {
      "entropy": 6.088496923446655,
      "epoch": 4.225097138820205,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003327383131484515,
      "loss": 5.2886,
      "mean_token_accuracy": 0.2157603070139885,
      "num_tokens": 89192877.0,
      "step": 47845
    },
    {
      "entropy": 6.041431283950805,
      "epoch": 4.225538678912045,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003327078756084393,
      "loss": 5.2444,
      "mean_token_accuracy": 0.22515609711408616,
      "num_tokens": 89201783.0,
      "step": 47850
    },
    {
      "entropy": 5.961481046676636,
      "epoch": 4.225980219003885,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003326774369379626,
      "loss": 5.1402,
      "mean_token_accuracy": 0.22476355582475663,
      "num_tokens": 89211214.0,
      "step": 47855
    },
    {
      "entropy": 6.048535633087158,
      "epoch": 4.226421759095726,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003326469971376178,
      "loss": 5.2475,
      "mean_token_accuracy": 0.22246539294719697,
      "num_tokens": 89220134.0,
      "step": 47860
    },
    {
      "entropy": 6.0497579097747805,
      "epoch": 4.226863299187566,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003326165562080009,
      "loss": 5.2371,
      "mean_token_accuracy": 0.22055517435073851,
      "num_tokens": 89228953.0,
      "step": 47865
    },
    {
      "entropy": 6.055494976043701,
      "epoch": 4.227304839279406,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033258611414970857,
      "loss": 5.2622,
      "mean_token_accuracy": 0.21057970821857452,
      "num_tokens": 89238888.0,
      "step": 47870
    },
    {
      "entropy": 6.118341445922852,
      "epoch": 4.227746379371247,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000332555670963337,
      "loss": 5.2811,
      "mean_token_accuracy": 0.21619823127985,
      "num_tokens": 89248724.0,
      "step": 47875
    },
    {
      "entropy": 6.02356219291687,
      "epoch": 4.228187919463087,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003325252266494825,
      "loss": 5.2263,
      "mean_token_accuracy": 0.2176676720380783,
      "num_tokens": 89257994.0,
      "step": 47880
    },
    {
      "entropy": 5.995910739898681,
      "epoch": 4.228629459554927,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00033249478120874165,
      "loss": 5.2025,
      "mean_token_accuracy": 0.22389666885137557,
      "num_tokens": 89266893.0,
      "step": 47885
    },
    {
      "entropy": 6.090566110610962,
      "epoch": 4.229070999646768,
      "grad_norm": 1.25,
      "learning_rate": 0.0003324643346417107,
      "loss": 5.3313,
      "mean_token_accuracy": 0.21484037041664122,
      "num_tokens": 89277144.0,
      "step": 47890
    },
    {
      "entropy": 6.062097024917603,
      "epoch": 4.229512539738608,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003324338869489861,
      "loss": 5.2448,
      "mean_token_accuracy": 0.2232649639248848,
      "num_tokens": 89286683.0,
      "step": 47895
    },
    {
      "entropy": 6.120598411560058,
      "epoch": 4.229954079830448,
      "grad_norm": 1.375,
      "learning_rate": 0.0003324034381311643,
      "loss": 5.3165,
      "mean_token_accuracy": 0.21577018052339553,
      "num_tokens": 89295728.0,
      "step": 47900
    },
    {
      "entropy": 6.055761623382568,
      "epoch": 4.230395619922289,
      "grad_norm": 1.328125,
      "learning_rate": 0.00033237298818884184,
      "loss": 5.2764,
      "mean_token_accuracy": 0.22431657165288926,
      "num_tokens": 89304866.0,
      "step": 47905
    },
    {
      "entropy": 6.041124439239502,
      "epoch": 4.230837160014129,
      "grad_norm": 1.390625,
      "learning_rate": 0.000332342537122615,
      "loss": 5.1944,
      "mean_token_accuracy": 0.22340403646230697,
      "num_tokens": 89313964.0,
      "step": 47910
    },
    {
      "entropy": 6.069287395477295,
      "epoch": 4.231278700105969,
      "grad_norm": 1.453125,
      "learning_rate": 0.00033231208493308067,
      "loss": 5.2804,
      "mean_token_accuracy": 0.22313058376312256,
      "num_tokens": 89322687.0,
      "step": 47915
    },
    {
      "entropy": 6.000819396972656,
      "epoch": 4.23172024019781,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033228163162083514,
      "loss": 5.2649,
      "mean_token_accuracy": 0.2268328532576561,
      "num_tokens": 89331362.0,
      "step": 47920
    },
    {
      "entropy": 6.02420744895935,
      "epoch": 4.23216178028965,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00033225117718647506,
      "loss": 5.2293,
      "mean_token_accuracy": 0.2254909947514534,
      "num_tokens": 89340889.0,
      "step": 47925
    },
    {
      "entropy": 6.072338390350342,
      "epoch": 4.23260332038149,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003322207216305969,
      "loss": 5.277,
      "mean_token_accuracy": 0.21556888222694398,
      "num_tokens": 89350017.0,
      "step": 47930
    },
    {
      "entropy": 6.036608123779297,
      "epoch": 4.233044860473331,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003321902649537974,
      "loss": 5.2386,
      "mean_token_accuracy": 0.2190897583961487,
      "num_tokens": 89359641.0,
      "step": 47935
    },
    {
      "entropy": 6.0756364345550535,
      "epoch": 4.233486400565171,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033215980715667304,
      "loss": 5.2115,
      "mean_token_accuracy": 0.2177795648574829,
      "num_tokens": 89368601.0,
      "step": 47940
    },
    {
      "entropy": 6.071453809738159,
      "epoch": 4.233927940657011,
      "grad_norm": 1.359375,
      "learning_rate": 0.00033212934823982073,
      "loss": 5.2338,
      "mean_token_accuracy": 0.21937206089496614,
      "num_tokens": 89377647.0,
      "step": 47945
    },
    {
      "entropy": 6.033850765228271,
      "epoch": 4.234369480748852,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033209888820383683,
      "loss": 5.2668,
      "mean_token_accuracy": 0.21496698409318923,
      "num_tokens": 89387048.0,
      "step": 47950
    },
    {
      "entropy": 5.97652153968811,
      "epoch": 4.234811020840692,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003320684270493183,
      "loss": 5.2191,
      "mean_token_accuracy": 0.22280665040016173,
      "num_tokens": 89395702.0,
      "step": 47955
    },
    {
      "entropy": 6.05482587814331,
      "epoch": 4.235252560932532,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003320379647768617,
      "loss": 5.2515,
      "mean_token_accuracy": 0.2185358837246895,
      "num_tokens": 89404731.0,
      "step": 47960
    },
    {
      "entropy": 6.046424007415771,
      "epoch": 4.235694101024373,
      "grad_norm": 1.40625,
      "learning_rate": 0.00033200750138706375,
      "loss": 5.2631,
      "mean_token_accuracy": 0.21987704038619996,
      "num_tokens": 89413242.0,
      "step": 47965
    },
    {
      "entropy": 5.998749828338623,
      "epoch": 4.236135641116213,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003319770368805213,
      "loss": 5.2361,
      "mean_token_accuracy": 0.23262175023555756,
      "num_tokens": 89423172.0,
      "step": 47970
    },
    {
      "entropy": 6.032984685897827,
      "epoch": 4.2365771812080535,
      "grad_norm": 1.34375,
      "learning_rate": 0.00033194657125783114,
      "loss": 5.2675,
      "mean_token_accuracy": 0.2168906271457672,
      "num_tokens": 89432459.0,
      "step": 47975
    },
    {
      "entropy": 6.016002941131592,
      "epoch": 4.237018721299894,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033191610451959,
      "loss": 5.1758,
      "mean_token_accuracy": 0.22421194314956666,
      "num_tokens": 89441373.0,
      "step": 47980
    },
    {
      "entropy": 6.056577157974243,
      "epoch": 4.2374602613917345,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003318856366663947,
      "loss": 5.208,
      "mean_token_accuracy": 0.22246990948915482,
      "num_tokens": 89450083.0,
      "step": 47985
    },
    {
      "entropy": 5.979786682128906,
      "epoch": 4.2379018014835745,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003318551676988422,
      "loss": 5.1586,
      "mean_token_accuracy": 0.22731091678142548,
      "num_tokens": 89459638.0,
      "step": 47990
    },
    {
      "entropy": 6.040704298019409,
      "epoch": 4.2383433415754155,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003318246976175293,
      "loss": 5.2745,
      "mean_token_accuracy": 0.21432694047689438,
      "num_tokens": 89468568.0,
      "step": 47995
    },
    {
      "entropy": 6.004697275161743,
      "epoch": 4.2387848816672555,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003317942264230529,
      "loss": 5.2468,
      "mean_token_accuracy": 0.2169607788324356,
      "num_tokens": 89478045.0,
      "step": 48000
    },
    {
      "epoch": 4.2387848816672555,
      "eval_entropy": 5.764756697895573,
      "eval_loss": 5.31890344619751,
      "eval_mean_token_accuracy": 0.22472320127460182,
      "eval_num_tokens": 89478045.0,
      "eval_runtime": 20.1049,
      "eval_samples_per_second": 1756.539,
      "eval_steps_per_second": 219.598,
      "step": 48000
    },
    {
      "entropy": 6.1221513748168945,
      "epoch": 4.239226421759096,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003317637541160098,
      "loss": 5.2996,
      "mean_token_accuracy": 0.20991499572992325,
      "num_tokens": 89488208.0,
      "step": 48005
    },
    {
      "entropy": 6.098778104782104,
      "epoch": 4.239667961850936,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003317332806969972,
      "loss": 5.2731,
      "mean_token_accuracy": 0.21720168143510818,
      "num_tokens": 89497844.0,
      "step": 48010
    },
    {
      "entropy": 5.97242226600647,
      "epoch": 4.240109501942777,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033170280616661184,
      "loss": 5.2184,
      "mean_token_accuracy": 0.2246686190366745,
      "num_tokens": 89507146.0,
      "step": 48015
    },
    {
      "entropy": 6.039331436157227,
      "epoch": 4.240551042034617,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003316723305254507,
      "loss": 5.2091,
      "mean_token_accuracy": 0.218903087079525,
      "num_tokens": 89516653.0,
      "step": 48020
    },
    {
      "entropy": 6.074346303939819,
      "epoch": 4.240992582126457,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003316418537741109,
      "loss": 5.1931,
      "mean_token_accuracy": 0.2280202627182007,
      "num_tokens": 89526109.0,
      "step": 48025
    },
    {
      "entropy": 6.048233222961426,
      "epoch": 4.241434122218298,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033161137591318935,
      "loss": 5.3223,
      "mean_token_accuracy": 0.2158820539712906,
      "num_tokens": 89535771.0,
      "step": 48030
    },
    {
      "entropy": 6.056128025054932,
      "epoch": 4.241875662310138,
      "grad_norm": 1.375,
      "learning_rate": 0.00033158089694328315,
      "loss": 5.2847,
      "mean_token_accuracy": 0.2141755700111389,
      "num_tokens": 89545159.0,
      "step": 48035
    },
    {
      "entropy": 6.0754505634307865,
      "epoch": 4.242317202401978,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003315504168649893,
      "loss": 5.3512,
      "mean_token_accuracy": 0.21641218811273574,
      "num_tokens": 89554477.0,
      "step": 48040
    },
    {
      "entropy": 6.045628929138184,
      "epoch": 4.242758742493819,
      "grad_norm": 1.25,
      "learning_rate": 0.00033151993567890513,
      "loss": 5.2411,
      "mean_token_accuracy": 0.21969176977872848,
      "num_tokens": 89563147.0,
      "step": 48045
    },
    {
      "entropy": 6.0383611679077145,
      "epoch": 4.243200282585659,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003314894533856274,
      "loss": 5.2695,
      "mean_token_accuracy": 0.22078268080949784,
      "num_tokens": 89573327.0,
      "step": 48050
    },
    {
      "entropy": 5.939865207672119,
      "epoch": 4.243641822677499,
      "grad_norm": 1.375,
      "learning_rate": 0.00033145896998575347,
      "loss": 5.1494,
      "mean_token_accuracy": 0.23299940824508666,
      "num_tokens": 89581663.0,
      "step": 48055
    },
    {
      "entropy": 6.079080200195312,
      "epoch": 4.24408336276934,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003314284854798804,
      "loss": 5.2943,
      "mean_token_accuracy": 0.21639022827148438,
      "num_tokens": 89592060.0,
      "step": 48060
    },
    {
      "entropy": 5.988911294937134,
      "epoch": 4.24452490286118,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033139799986860543,
      "loss": 5.2472,
      "mean_token_accuracy": 0.21652452498674393,
      "num_tokens": 89601897.0,
      "step": 48065
    },
    {
      "entropy": 6.0750233173370365,
      "epoch": 4.24496644295302,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003313675131525257,
      "loss": 5.1713,
      "mean_token_accuracy": 0.22853025645017624,
      "num_tokens": 89610291.0,
      "step": 48070
    },
    {
      "entropy": 6.017316627502441,
      "epoch": 4.245407983044861,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033133702533223844,
      "loss": 5.2582,
      "mean_token_accuracy": 0.21901663690805434,
      "num_tokens": 89620525.0,
      "step": 48075
    },
    {
      "entropy": 5.9577436447143555,
      "epoch": 4.245849523136701,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003313065364083409,
      "loss": 5.117,
      "mean_token_accuracy": 0.23551142811775208,
      "num_tokens": 89629666.0,
      "step": 48080
    },
    {
      "entropy": 6.006594896316528,
      "epoch": 4.246291063228541,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003312760463814303,
      "loss": 5.1796,
      "mean_token_accuracy": 0.232965324819088,
      "num_tokens": 89638232.0,
      "step": 48085
    },
    {
      "entropy": 5.975753211975098,
      "epoch": 4.246732603320382,
      "grad_norm": 1.40625,
      "learning_rate": 0.000331245555252104,
      "loss": 5.1892,
      "mean_token_accuracy": 0.22097235918045044,
      "num_tokens": 89646574.0,
      "step": 48090
    },
    {
      "entropy": 6.0261815071105955,
      "epoch": 4.247174143412222,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033121506302095927,
      "loss": 5.2623,
      "mean_token_accuracy": 0.21839620023965836,
      "num_tokens": 89655924.0,
      "step": 48095
    },
    {
      "entropy": 6.011132335662841,
      "epoch": 4.247615683504062,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003311845696885934,
      "loss": 5.1759,
      "mean_token_accuracy": 0.23070118874311446,
      "num_tokens": 89664661.0,
      "step": 48100
    },
    {
      "entropy": 6.0017174243927,
      "epoch": 4.248057223595903,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003311540752556037,
      "loss": 5.2565,
      "mean_token_accuracy": 0.21408119052648544,
      "num_tokens": 89673712.0,
      "step": 48105
    },
    {
      "entropy": 6.11645188331604,
      "epoch": 4.248498763687743,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003311235797225877,
      "loss": 5.3806,
      "mean_token_accuracy": 0.2098835289478302,
      "num_tokens": 89684884.0,
      "step": 48110
    },
    {
      "entropy": 6.074457788467408,
      "epoch": 4.248940303779583,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033109308309014265,
      "loss": 5.2807,
      "mean_token_accuracy": 0.22000558525323868,
      "num_tokens": 89694656.0,
      "step": 48115
    },
    {
      "entropy": 6.099162578582764,
      "epoch": 4.249381843871424,
      "grad_norm": 1.515625,
      "learning_rate": 0.00033106258535886606,
      "loss": 5.2443,
      "mean_token_accuracy": 0.22582942396402358,
      "num_tokens": 89705071.0,
      "step": 48120
    },
    {
      "entropy": 6.050571537017822,
      "epoch": 4.249823383963264,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033103208652935517,
      "loss": 5.276,
      "mean_token_accuracy": 0.21725829988718032,
      "num_tokens": 89714287.0,
      "step": 48125
    },
    {
      "entropy": 6.009078311920166,
      "epoch": 4.250264924055104,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00033100158660220776,
      "loss": 5.2591,
      "mean_token_accuracy": 0.21912360489368438,
      "num_tokens": 89723067.0,
      "step": 48130
    },
    {
      "entropy": 6.072261953353882,
      "epoch": 4.250706464146944,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033097108557802096,
      "loss": 5.2711,
      "mean_token_accuracy": 0.21825580596923827,
      "num_tokens": 89732225.0,
      "step": 48135
    },
    {
      "entropy": 5.982676076889038,
      "epoch": 4.251148004238785,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033094058345739253,
      "loss": 5.1993,
      "mean_token_accuracy": 0.2275601491332054,
      "num_tokens": 89742853.0,
      "step": 48140
    },
    {
      "entropy": 6.069232177734375,
      "epoch": 4.251589544330625,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003309100802409199,
      "loss": 5.3057,
      "mean_token_accuracy": 0.21098411828279495,
      "num_tokens": 89754168.0,
      "step": 48145
    },
    {
      "entropy": 6.069590759277344,
      "epoch": 4.252031084422466,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003308795759292005,
      "loss": 5.202,
      "mean_token_accuracy": 0.22488936334848403,
      "num_tokens": 89763014.0,
      "step": 48150
    },
    {
      "entropy": 5.991413021087647,
      "epoch": 4.252472624514306,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033084907052283196,
      "loss": 5.2519,
      "mean_token_accuracy": 0.21844721138477324,
      "num_tokens": 89772761.0,
      "step": 48155
    },
    {
      "entropy": 6.108750534057617,
      "epoch": 4.252914164606146,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00033081856402241203,
      "loss": 5.3633,
      "mean_token_accuracy": 0.21120227128267288,
      "num_tokens": 89781600.0,
      "step": 48160
    },
    {
      "entropy": 6.050910091400146,
      "epoch": 4.253355704697986,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00033078805642853805,
      "loss": 5.2468,
      "mean_token_accuracy": 0.21763068735599517,
      "num_tokens": 89791027.0,
      "step": 48165
    },
    {
      "entropy": 6.1081384181976315,
      "epoch": 4.253797244789827,
      "grad_norm": 1.34375,
      "learning_rate": 0.00033075754774180785,
      "loss": 5.2061,
      "mean_token_accuracy": 0.2195226475596428,
      "num_tokens": 89799325.0,
      "step": 48170
    },
    {
      "entropy": 6.019128227233887,
      "epoch": 4.254238784881667,
      "grad_norm": 1.265625,
      "learning_rate": 0.000330727037962819,
      "loss": 5.191,
      "mean_token_accuracy": 0.22295982539653778,
      "num_tokens": 89807887.0,
      "step": 48175
    },
    {
      "entropy": 6.016168022155762,
      "epoch": 4.254680324973507,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003306965270921691,
      "loss": 5.2681,
      "mean_token_accuracy": 0.217441426217556,
      "num_tokens": 89816462.0,
      "step": 48180
    },
    {
      "entropy": 5.988714694976807,
      "epoch": 4.255121865065348,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00033066601513045605,
      "loss": 5.2115,
      "mean_token_accuracy": 0.2260665312409401,
      "num_tokens": 89825059.0,
      "step": 48185
    },
    {
      "entropy": 5.961385297775268,
      "epoch": 4.255563405157188,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00033063550207827725,
      "loss": 5.155,
      "mean_token_accuracy": 0.23265168070793152,
      "num_tokens": 89834535.0,
      "step": 48190
    },
    {
      "entropy": 6.041427707672119,
      "epoch": 4.256004945249028,
      "grad_norm": 1.328125,
      "learning_rate": 0.00033060498793623073,
      "loss": 5.2539,
      "mean_token_accuracy": 0.2250961735844612,
      "num_tokens": 89844081.0,
      "step": 48195
    },
    {
      "entropy": 5.957102966308594,
      "epoch": 4.256446485340869,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033057447270491415,
      "loss": 5.1766,
      "mean_token_accuracy": 0.22710752189159394,
      "num_tokens": 89853929.0,
      "step": 48200
    },
    {
      "entropy": 5.984032487869262,
      "epoch": 4.256888025432709,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003305439563849251,
      "loss": 5.2177,
      "mean_token_accuracy": 0.2267147719860077,
      "num_tokens": 89862959.0,
      "step": 48205
    },
    {
      "entropy": 6.043762111663819,
      "epoch": 4.257329565524549,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003305134389768617,
      "loss": 5.27,
      "mean_token_accuracy": 0.21360482275485992,
      "num_tokens": 89872623.0,
      "step": 48210
    },
    {
      "entropy": 6.041316413879395,
      "epoch": 4.25777110561639,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003304829204813215,
      "loss": 5.2512,
      "mean_token_accuracy": 0.2195517361164093,
      "num_tokens": 89880819.0,
      "step": 48215
    },
    {
      "entropy": 6.0173086643219,
      "epoch": 4.25821264570823,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033045240089890236,
      "loss": 5.1773,
      "mean_token_accuracy": 0.22935352027416228,
      "num_tokens": 89889868.0,
      "step": 48220
    },
    {
      "entropy": 6.068049955368042,
      "epoch": 4.25865418580007,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003304218802302024,
      "loss": 5.2502,
      "mean_token_accuracy": 0.22426638156175613,
      "num_tokens": 89899006.0,
      "step": 48225
    },
    {
      "entropy": 5.913289785385132,
      "epoch": 4.259095725891911,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033039135847581926,
      "loss": 5.072,
      "mean_token_accuracy": 0.23704774230718612,
      "num_tokens": 89909157.0,
      "step": 48230
    },
    {
      "entropy": 6.0359186172485355,
      "epoch": 4.259537265983751,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003303608356363509,
      "loss": 5.2536,
      "mean_token_accuracy": 0.21757179349660874,
      "num_tokens": 89919621.0,
      "step": 48235
    },
    {
      "entropy": 6.073157024383545,
      "epoch": 4.259978806075591,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033033031171239525,
      "loss": 5.3697,
      "mean_token_accuracy": 0.21036461889743804,
      "num_tokens": 89929025.0,
      "step": 48240
    },
    {
      "entropy": 6.077151584625244,
      "epoch": 4.260420346167432,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003302997867045503,
      "loss": 5.3461,
      "mean_token_accuracy": 0.2094286024570465,
      "num_tokens": 89939712.0,
      "step": 48245
    },
    {
      "entropy": 6.047710990905761,
      "epoch": 4.260861886259272,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033026926061341394,
      "loss": 5.1999,
      "mean_token_accuracy": 0.22235646098852158,
      "num_tokens": 89949326.0,
      "step": 48250
    },
    {
      "entropy": 6.202179861068726,
      "epoch": 4.2613034263511125,
      "grad_norm": 1.40625,
      "learning_rate": 0.00033023873343958414,
      "loss": 5.3594,
      "mean_token_accuracy": 0.21370190978050232,
      "num_tokens": 89957968.0,
      "step": 48255
    },
    {
      "entropy": 6.049559259414673,
      "epoch": 4.261744966442953,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003302082051836591,
      "loss": 5.2575,
      "mean_token_accuracy": 0.21926505118608475,
      "num_tokens": 89967495.0,
      "step": 48260
    },
    {
      "entropy": 5.979890394210815,
      "epoch": 4.2621865065347935,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003301776758462367,
      "loss": 5.1316,
      "mean_token_accuracy": 0.22604858875274658,
      "num_tokens": 89976406.0,
      "step": 48265
    },
    {
      "entropy": 6.014916944503784,
      "epoch": 4.2626280466266335,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003301471454279149,
      "loss": 5.2723,
      "mean_token_accuracy": 0.21133521497249602,
      "num_tokens": 89986553.0,
      "step": 48270
    },
    {
      "entropy": 6.057165908813476,
      "epoch": 4.2630695867184745,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00033011661392929204,
      "loss": 5.2192,
      "mean_token_accuracy": 0.22491634339094163,
      "num_tokens": 89995889.0,
      "step": 48275
    },
    {
      "entropy": 6.094903802871704,
      "epoch": 4.2635111268103145,
      "grad_norm": 1.34375,
      "learning_rate": 0.00033008608135096595,
      "loss": 5.3184,
      "mean_token_accuracy": 0.21126517206430434,
      "num_tokens": 90004853.0,
      "step": 48280
    },
    {
      "entropy": 6.060566520690918,
      "epoch": 4.263952666902155,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033005554769353484,
      "loss": 5.3319,
      "mean_token_accuracy": 0.2111051708459854,
      "num_tokens": 90015076.0,
      "step": 48285
    },
    {
      "entropy": 6.0042595863342285,
      "epoch": 4.264394206993995,
      "grad_norm": 1.359375,
      "learning_rate": 0.00033002501295759687,
      "loss": 5.2886,
      "mean_token_accuracy": 0.22093878090381622,
      "num_tokens": 90024180.0,
      "step": 48290
    },
    {
      "entropy": 6.089037561416626,
      "epoch": 4.264835747085836,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00032999447714375025,
      "loss": 5.2851,
      "mean_token_accuracy": 0.21526019424200057,
      "num_tokens": 90033693.0,
      "step": 48295
    },
    {
      "entropy": 6.001379060745239,
      "epoch": 4.265277287177676,
      "grad_norm": 1.375,
      "learning_rate": 0.0003299639402525931,
      "loss": 5.1744,
      "mean_token_accuracy": 0.22537991404533386,
      "num_tokens": 90043107.0,
      "step": 48300
    },
    {
      "entropy": 6.07304539680481,
      "epoch": 4.265718827269516,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032993340228472357,
      "loss": 5.31,
      "mean_token_accuracy": 0.21674174517393113,
      "num_tokens": 90052756.0,
      "step": 48305
    },
    {
      "entropy": 5.9394035816192625,
      "epoch": 4.266160367361357,
      "grad_norm": 1.25,
      "learning_rate": 0.00032990286324073993,
      "loss": 5.1135,
      "mean_token_accuracy": 0.23233645409345627,
      "num_tokens": 90061099.0,
      "step": 48310
    },
    {
      "entropy": 5.926008653640747,
      "epoch": 4.266601907453197,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003298723231212404,
      "loss": 5.1149,
      "mean_token_accuracy": 0.23261500597000123,
      "num_tokens": 90070005.0,
      "step": 48315
    },
    {
      "entropy": 6.051883792877197,
      "epoch": 4.267043447545037,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003298417819268234,
      "loss": 5.2121,
      "mean_token_accuracy": 0.21732547283172607,
      "num_tokens": 90078371.0,
      "step": 48320
    },
    {
      "entropy": 5.992064189910889,
      "epoch": 4.267484987636878,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032981123965808695,
      "loss": 5.1676,
      "mean_token_accuracy": 0.22390372902154923,
      "num_tokens": 90087262.0,
      "step": 48325
    },
    {
      "entropy": 6.051122808456421,
      "epoch": 4.267926527728718,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032978069631562956,
      "loss": 5.2161,
      "mean_token_accuracy": 0.21980882585048675,
      "num_tokens": 90096770.0,
      "step": 48330
    },
    {
      "entropy": 5.999482774734497,
      "epoch": 4.268368067820558,
      "grad_norm": 1.40625,
      "learning_rate": 0.00032975015190004947,
      "loss": 5.248,
      "mean_token_accuracy": 0.2271779000759125,
      "num_tokens": 90105933.0,
      "step": 48335
    },
    {
      "entropy": 6.077881669998169,
      "epoch": 4.268809607912399,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00032971960641194504,
      "loss": 5.2406,
      "mean_token_accuracy": 0.21955400258302687,
      "num_tokens": 90115542.0,
      "step": 48340
    },
    {
      "entropy": 5.947074747085571,
      "epoch": 4.269251148004239,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003296890598519146,
      "loss": 5.1645,
      "mean_token_accuracy": 0.22816011011600495,
      "num_tokens": 90124617.0,
      "step": 48345
    },
    {
      "entropy": 6.03322205543518,
      "epoch": 4.269692688096079,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032965851222055664,
      "loss": 5.3064,
      "mean_token_accuracy": 0.21606263369321824,
      "num_tokens": 90134256.0,
      "step": 48350
    },
    {
      "entropy": 6.083021640777588,
      "epoch": 4.27013422818792,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003296279635184694,
      "loss": 5.2388,
      "mean_token_accuracy": 0.21368051022291185,
      "num_tokens": 90143674.0,
      "step": 48355
    },
    {
      "entropy": 5.960357046127319,
      "epoch": 4.27057576827976,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00032959741374625156,
      "loss": 5.1292,
      "mean_token_accuracy": 0.2303976058959961,
      "num_tokens": 90153212.0,
      "step": 48360
    },
    {
      "entropy": 6.0282454013824465,
      "epoch": 4.2710173083716,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003295668629045014,
      "loss": 5.3364,
      "mean_token_accuracy": 0.21092424392700196,
      "num_tokens": 90163770.0,
      "step": 48365
    },
    {
      "entropy": 6.033769941329956,
      "epoch": 4.271458848463441,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003295363109938174,
      "loss": 5.2598,
      "mean_token_accuracy": 0.2180229038000107,
      "num_tokens": 90172633.0,
      "step": 48370
    },
    {
      "entropy": 6.054709577560425,
      "epoch": 4.271900388555281,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003295057580147981,
      "loss": 5.2826,
      "mean_token_accuracy": 0.21957656741142273,
      "num_tokens": 90182959.0,
      "step": 48375
    },
    {
      "entropy": 6.052374935150146,
      "epoch": 4.272341928647121,
      "grad_norm": 1.359375,
      "learning_rate": 0.000329475203968042,
      "loss": 5.2458,
      "mean_token_accuracy": 0.2179470181465149,
      "num_tokens": 90191717.0,
      "step": 48380
    },
    {
      "entropy": 6.032466030120849,
      "epoch": 4.272783468738962,
      "grad_norm": 1.40625,
      "learning_rate": 0.00032944464885414763,
      "loss": 5.2072,
      "mean_token_accuracy": 0.2194373205304146,
      "num_tokens": 90201267.0,
      "step": 48385
    },
    {
      "entropy": 6.011102819442749,
      "epoch": 4.273225008830802,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003294140926737136,
      "loss": 5.1841,
      "mean_token_accuracy": 0.2273390531539917,
      "num_tokens": 90210155.0,
      "step": 48390
    },
    {
      "entropy": 6.019736242294312,
      "epoch": 4.273666548922642,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003293835354273384,
      "loss": 5.2237,
      "mean_token_accuracy": 0.22180443555116652,
      "num_tokens": 90220209.0,
      "step": 48395
    },
    {
      "entropy": 6.0351386070251465,
      "epoch": 4.274108089014483,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003293529771156206,
      "loss": 5.2896,
      "mean_token_accuracy": 0.2176115557551384,
      "num_tokens": 90229486.0,
      "step": 48400
    },
    {
      "entropy": 5.993690490722656,
      "epoch": 4.274549629106323,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003293224177391591,
      "loss": 5.2254,
      "mean_token_accuracy": 0.23017251789569854,
      "num_tokens": 90239195.0,
      "step": 48405
    },
    {
      "entropy": 6.050132894515992,
      "epoch": 4.274991169198163,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003292918572985521,
      "loss": 5.23,
      "mean_token_accuracy": 0.22343695312738418,
      "num_tokens": 90248485.0,
      "step": 48410
    },
    {
      "entropy": 5.96379714012146,
      "epoch": 4.275432709290003,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032926129579439854,
      "loss": 5.2169,
      "mean_token_accuracy": 0.22131282538175584,
      "num_tokens": 90258080.0,
      "step": 48415
    },
    {
      "entropy": 6.010011005401611,
      "epoch": 4.275874249381844,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003292307332272971,
      "loss": 5.2488,
      "mean_token_accuracy": 0.21610781997442247,
      "num_tokens": 90267663.0,
      "step": 48420
    },
    {
      "entropy": 5.994490432739258,
      "epoch": 4.276315789473684,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00032920016959784643,
      "loss": 5.2011,
      "mean_token_accuracy": 0.2231466591358185,
      "num_tokens": 90276435.0,
      "step": 48425
    },
    {
      "entropy": 5.93054690361023,
      "epoch": 4.276757329565525,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00032916960490664526,
      "loss": 5.1036,
      "mean_token_accuracy": 0.23267133682966232,
      "num_tokens": 90285974.0,
      "step": 48430
    },
    {
      "entropy": 6.01851224899292,
      "epoch": 4.277198869657365,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003291390391542924,
      "loss": 5.2213,
      "mean_token_accuracy": 0.2167743444442749,
      "num_tokens": 90294854.0,
      "step": 48435
    },
    {
      "entropy": 6.019901371002197,
      "epoch": 4.277640409749205,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032910847234138654,
      "loss": 5.2738,
      "mean_token_accuracy": 0.21938747316598892,
      "num_tokens": 90304897.0,
      "step": 48440
    },
    {
      "entropy": 6.040235280990601,
      "epoch": 4.278081949841045,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003290779044685264,
      "loss": 5.2417,
      "mean_token_accuracy": 0.22122110575437545,
      "num_tokens": 90313641.0,
      "step": 48445
    },
    {
      "entropy": 6.062701654434204,
      "epoch": 4.278523489932886,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032904733553631093,
      "loss": 5.3751,
      "mean_token_accuracy": 0.20962461978197097,
      "num_tokens": 90324133.0,
      "step": 48450
    },
    {
      "entropy": 6.10500693321228,
      "epoch": 4.278965030024726,
      "grad_norm": 1.34375,
      "learning_rate": 0.00032901676554533883,
      "loss": 5.2596,
      "mean_token_accuracy": 0.21509893834590912,
      "num_tokens": 90333180.0,
      "step": 48455
    },
    {
      "entropy": 6.011363077163696,
      "epoch": 4.279406570116566,
      "grad_norm": 1.25,
      "learning_rate": 0.0003289861944962091,
      "loss": 5.2762,
      "mean_token_accuracy": 0.21776246279478073,
      "num_tokens": 90342934.0,
      "step": 48460
    },
    {
      "entropy": 6.046363258361817,
      "epoch": 4.279848110208407,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003289556223895205,
      "loss": 5.2543,
      "mean_token_accuracy": 0.22121317088603973,
      "num_tokens": 90352724.0,
      "step": 48465
    },
    {
      "entropy": 6.136553955078125,
      "epoch": 4.280289650300247,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003289250492258719,
      "loss": 5.4281,
      "mean_token_accuracy": 0.2079589247703552,
      "num_tokens": 90362066.0,
      "step": 48470
    },
    {
      "entropy": 6.118198299407959,
      "epoch": 4.280731190392087,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003288944750058622,
      "loss": 5.2985,
      "mean_token_accuracy": 0.2192852720618248,
      "num_tokens": 90370733.0,
      "step": 48475
    },
    {
      "entropy": 6.012101793289185,
      "epoch": 4.281172730483928,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003288638997300905,
      "loss": 5.2161,
      "mean_token_accuracy": 0.22105011641979216,
      "num_tokens": 90380761.0,
      "step": 48480
    },
    {
      "entropy": 5.900947618484497,
      "epoch": 4.281614270575768,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003288333233991556,
      "loss": 5.1082,
      "mean_token_accuracy": 0.23666189312934877,
      "num_tokens": 90389881.0,
      "step": 48485
    },
    {
      "entropy": 6.02625412940979,
      "epoch": 4.282055810667608,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032880274601365656,
      "loss": 5.2161,
      "mean_token_accuracy": 0.21984918415546417,
      "num_tokens": 90398632.0,
      "step": 48490
    },
    {
      "entropy": 6.072874307632446,
      "epoch": 4.282497350759449,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003287721675741922,
      "loss": 5.3494,
      "mean_token_accuracy": 0.21087322384119034,
      "num_tokens": 90407551.0,
      "step": 48495
    },
    {
      "entropy": 6.017870140075684,
      "epoch": 4.282938890851289,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003287415880813618,
      "loss": 5.188,
      "mean_token_accuracy": 0.22431463301181792,
      "num_tokens": 90416307.0,
      "step": 48500
    },
    {
      "entropy": 6.057468700408935,
      "epoch": 4.283380430943129,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003287110075357642,
      "loss": 5.2486,
      "mean_token_accuracy": 0.22444708347320558,
      "num_tokens": 90424891.0,
      "step": 48505
    },
    {
      "entropy": 6.060033369064331,
      "epoch": 4.28382197103497,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032868042593799844,
      "loss": 5.3589,
      "mean_token_accuracy": 0.21631466299295427,
      "num_tokens": 90435630.0,
      "step": 48510
    },
    {
      "entropy": 6.021633291244507,
      "epoch": 4.28426351112681,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003286498432886638,
      "loss": 5.1918,
      "mean_token_accuracy": 0.22362103313207626,
      "num_tokens": 90445381.0,
      "step": 48515
    },
    {
      "entropy": 6.009352493286133,
      "epoch": 4.28470505121865,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00032861925958835916,
      "loss": 5.2068,
      "mean_token_accuracy": 0.22613186240196229,
      "num_tokens": 90454529.0,
      "step": 48520
    },
    {
      "entropy": 5.9852683544158936,
      "epoch": 4.285146591310491,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003285886748376837,
      "loss": 5.2507,
      "mean_token_accuracy": 0.21710582673549653,
      "num_tokens": 90464729.0,
      "step": 48525
    },
    {
      "entropy": 5.994723749160767,
      "epoch": 4.285588131402331,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003285580890372367,
      "loss": 5.1965,
      "mean_token_accuracy": 0.22106048762798308,
      "num_tokens": 90475133.0,
      "step": 48530
    },
    {
      "entropy": 6.01643705368042,
      "epoch": 4.2860296714941715,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003285275021876171,
      "loss": 5.2415,
      "mean_token_accuracy": 0.21554090678691865,
      "num_tokens": 90484184.0,
      "step": 48535
    },
    {
      "entropy": 6.057605123519897,
      "epoch": 4.286471211586012,
      "grad_norm": 1.375,
      "learning_rate": 0.0003284969142894241,
      "loss": 5.2871,
      "mean_token_accuracy": 0.22082473635673522,
      "num_tokens": 90493364.0,
      "step": 48540
    },
    {
      "entropy": 6.098090648651123,
      "epoch": 4.2869127516778525,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000328466325343257,
      "loss": 5.3292,
      "mean_token_accuracy": 0.21285105496644974,
      "num_tokens": 90503010.0,
      "step": 48545
    },
    {
      "entropy": 6.048043203353882,
      "epoch": 4.2873542917696925,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032843573534971506,
      "loss": 5.2863,
      "mean_token_accuracy": 0.21738003045320511,
      "num_tokens": 90512333.0,
      "step": 48550
    },
    {
      "entropy": 6.072185039520264,
      "epoch": 4.2877958318615335,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003284051443093975,
      "loss": 5.2451,
      "mean_token_accuracy": 0.21766841560602188,
      "num_tokens": 90521590.0,
      "step": 48555
    },
    {
      "entropy": 6.028392171859741,
      "epoch": 4.2882373719533735,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003283745522229035,
      "loss": 5.2168,
      "mean_token_accuracy": 0.22060517966747284,
      "num_tokens": 90530725.0,
      "step": 48560
    },
    {
      "entropy": 6.029927539825439,
      "epoch": 4.288678912045214,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003283439590908323,
      "loss": 5.2555,
      "mean_token_accuracy": 0.2224256083369255,
      "num_tokens": 90541177.0,
      "step": 48565
    },
    {
      "entropy": 6.0641576766967775,
      "epoch": 4.289120452137054,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003283133649137834,
      "loss": 5.2229,
      "mean_token_accuracy": 0.21594574600458144,
      "num_tokens": 90551273.0,
      "step": 48570
    },
    {
      "entropy": 6.040344715118408,
      "epoch": 4.289561992228895,
      "grad_norm": 1.34375,
      "learning_rate": 0.00032828276969235586,
      "loss": 5.2077,
      "mean_token_accuracy": 0.22273287773132325,
      "num_tokens": 90560624.0,
      "step": 48575
    },
    {
      "entropy": 6.071804428100586,
      "epoch": 4.290003532320735,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003282521734271493,
      "loss": 5.2667,
      "mean_token_accuracy": 0.21035293787717818,
      "num_tokens": 90569752.0,
      "step": 48580
    },
    {
      "entropy": 6.1164937019348145,
      "epoch": 4.290445072412575,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003282215761187629,
      "loss": 5.3133,
      "mean_token_accuracy": 0.21204934865236283,
      "num_tokens": 90579716.0,
      "step": 48585
    },
    {
      "entropy": 6.059239292144776,
      "epoch": 4.290886612504416,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032819097776779605,
      "loss": 5.2742,
      "mean_token_accuracy": 0.22308889776468277,
      "num_tokens": 90589865.0,
      "step": 48590
    },
    {
      "entropy": 6.062047004699707,
      "epoch": 4.291328152596256,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00032816037837484825,
      "loss": 5.1854,
      "mean_token_accuracy": 0.22947670072317122,
      "num_tokens": 90598779.0,
      "step": 48595
    },
    {
      "entropy": 5.998826313018799,
      "epoch": 4.291769692688096,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032812977794051884,
      "loss": 5.1872,
      "mean_token_accuracy": 0.22761383652687073,
      "num_tokens": 90608494.0,
      "step": 48600
    },
    {
      "entropy": 6.056554222106934,
      "epoch": 4.292211232779937,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003280991764654073,
      "loss": 5.2214,
      "mean_token_accuracy": 0.2208447724580765,
      "num_tokens": 90619088.0,
      "step": 48605
    },
    {
      "entropy": 6.00095419883728,
      "epoch": 4.292652772871777,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032806857395011307,
      "loss": 5.247,
      "mean_token_accuracy": 0.2183338850736618,
      "num_tokens": 90628385.0,
      "step": 48610
    },
    {
      "entropy": 6.0051664352417,
      "epoch": 4.293094312963617,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003280379703952357,
      "loss": 5.1458,
      "mean_token_accuracy": 0.23243424594402312,
      "num_tokens": 90637826.0,
      "step": 48615
    },
    {
      "entropy": 6.038212585449219,
      "epoch": 4.293535853055458,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003280073658013747,
      "loss": 5.2562,
      "mean_token_accuracy": 0.21810517460107803,
      "num_tokens": 90647929.0,
      "step": 48620
    },
    {
      "entropy": 6.033287572860718,
      "epoch": 4.293977393147298,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032797676016912947,
      "loss": 5.238,
      "mean_token_accuracy": 0.21800464689731597,
      "num_tokens": 90657715.0,
      "step": 48625
    },
    {
      "entropy": 6.021102046966552,
      "epoch": 4.294418933239138,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003279461534990997,
      "loss": 5.2195,
      "mean_token_accuracy": 0.22854317873716354,
      "num_tokens": 90667087.0,
      "step": 48630
    },
    {
      "entropy": 6.032281732559204,
      "epoch": 4.294860473330979,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003279155457918848,
      "loss": 5.206,
      "mean_token_accuracy": 0.22697585076093674,
      "num_tokens": 90676040.0,
      "step": 48635
    },
    {
      "entropy": 6.092827701568604,
      "epoch": 4.295302013422819,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032788493704808455,
      "loss": 5.3559,
      "mean_token_accuracy": 0.21399894654750823,
      "num_tokens": 90685892.0,
      "step": 48640
    },
    {
      "entropy": 6.084671211242676,
      "epoch": 4.295743553514659,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032785432726829843,
      "loss": 5.2374,
      "mean_token_accuracy": 0.22420214861631393,
      "num_tokens": 90695258.0,
      "step": 48645
    },
    {
      "entropy": 6.035666561126709,
      "epoch": 4.2961850936065,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003278237164531261,
      "loss": 5.3009,
      "mean_token_accuracy": 0.21453643441200257,
      "num_tokens": 90705846.0,
      "step": 48650
    },
    {
      "entropy": 5.978988599777222,
      "epoch": 4.29662663369834,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003277931046031672,
      "loss": 5.1904,
      "mean_token_accuracy": 0.2329735741019249,
      "num_tokens": 90714708.0,
      "step": 48655
    },
    {
      "entropy": 5.984758996963501,
      "epoch": 4.29706817379018,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003277624917190214,
      "loss": 5.1226,
      "mean_token_accuracy": 0.23651546239852905,
      "num_tokens": 90723497.0,
      "step": 48660
    },
    {
      "entropy": 6.015388679504395,
      "epoch": 4.297509713882021,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003277318778012883,
      "loss": 5.2807,
      "mean_token_accuracy": 0.21689581722021103,
      "num_tokens": 90732345.0,
      "step": 48665
    },
    {
      "entropy": 6.008590412139893,
      "epoch": 4.297951253973861,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003277012628505678,
      "loss": 5.2908,
      "mean_token_accuracy": 0.21713664531707763,
      "num_tokens": 90742181.0,
      "step": 48670
    },
    {
      "entropy": 6.11655559539795,
      "epoch": 4.298392794065701,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00032767064686745957,
      "loss": 5.328,
      "mean_token_accuracy": 0.2221625939011574,
      "num_tokens": 90751889.0,
      "step": 48675
    },
    {
      "entropy": 6.013984632492066,
      "epoch": 4.298834334157542,
      "grad_norm": 1.375,
      "learning_rate": 0.0003276400298525633,
      "loss": 5.2891,
      "mean_token_accuracy": 0.21915048360824585,
      "num_tokens": 90761694.0,
      "step": 48680
    },
    {
      "entropy": 5.9604576587677,
      "epoch": 4.299275874249382,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00032760941180647876,
      "loss": 5.1456,
      "mean_token_accuracy": 0.2284931793808937,
      "num_tokens": 90770772.0,
      "step": 48685
    },
    {
      "entropy": 6.076942491531372,
      "epoch": 4.299717414341222,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003275787927298058,
      "loss": 5.332,
      "mean_token_accuracy": 0.21282655596733094,
      "num_tokens": 90780060.0,
      "step": 48690
    },
    {
      "entropy": 5.973370504379273,
      "epoch": 4.300158954433062,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003275481726231442,
      "loss": 5.1484,
      "mean_token_accuracy": 0.2292996898293495,
      "num_tokens": 90788842.0,
      "step": 48695
    },
    {
      "entropy": 6.0988061904907225,
      "epoch": 4.300600494524903,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003275175514870937,
      "loss": 5.2872,
      "mean_token_accuracy": 0.2173446699976921,
      "num_tokens": 90797411.0,
      "step": 48700
    },
    {
      "entropy": 6.063946104049682,
      "epoch": 4.301042034616743,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032748692932225434,
      "loss": 5.3395,
      "mean_token_accuracy": 0.21424626111984252,
      "num_tokens": 90806980.0,
      "step": 48705
    },
    {
      "entropy": 6.018826627731324,
      "epoch": 4.301483574708584,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00032745630612922576,
      "loss": 5.287,
      "mean_token_accuracy": 0.21724585592746734,
      "num_tokens": 90816870.0,
      "step": 48710
    },
    {
      "entropy": 6.074564981460571,
      "epoch": 4.301925114800424,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003274256819086081,
      "loss": 5.2934,
      "mean_token_accuracy": 0.2134709507226944,
      "num_tokens": 90826092.0,
      "step": 48715
    },
    {
      "entropy": 5.995055723190307,
      "epoch": 4.302366654892264,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003273950566610011,
      "loss": 5.124,
      "mean_token_accuracy": 0.23162743747234343,
      "num_tokens": 90835098.0,
      "step": 48720
    },
    {
      "entropy": 5.956968069076538,
      "epoch": 4.302808194984104,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003273644303870047,
      "loss": 5.1826,
      "mean_token_accuracy": 0.23411647975444794,
      "num_tokens": 90845075.0,
      "step": 48725
    },
    {
      "entropy": 6.0551109790802,
      "epoch": 4.303249735075945,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003273338030872189,
      "loss": 5.2648,
      "mean_token_accuracy": 0.21719565093517304,
      "num_tokens": 90855037.0,
      "step": 48730
    },
    {
      "entropy": 5.977140331268311,
      "epoch": 4.303691275167785,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003273031747622437,
      "loss": 5.1791,
      "mean_token_accuracy": 0.2286064013838768,
      "num_tokens": 90863739.0,
      "step": 48735
    },
    {
      "entropy": 6.0447922706604,
      "epoch": 4.304132815259625,
      "grad_norm": 1.359375,
      "learning_rate": 0.00032727254541267903,
      "loss": 5.2546,
      "mean_token_accuracy": 0.21261887699365617,
      "num_tokens": 90873503.0,
      "step": 48740
    },
    {
      "entropy": 6.0031496524810795,
      "epoch": 4.304574355351466,
      "grad_norm": 1.34375,
      "learning_rate": 0.00032724191503912496,
      "loss": 5.1515,
      "mean_token_accuracy": 0.2272445812821388,
      "num_tokens": 90882823.0,
      "step": 48745
    },
    {
      "entropy": 6.0427501678466795,
      "epoch": 4.305015895443306,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00032721128364218146,
      "loss": 5.2593,
      "mean_token_accuracy": 0.20964036136865616,
      "num_tokens": 90891922.0,
      "step": 48750
    },
    {
      "entropy": 6.025617837905884,
      "epoch": 4.305457435535146,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032718065122244863,
      "loss": 5.2943,
      "mean_token_accuracy": 0.2215282589197159,
      "num_tokens": 90902167.0,
      "step": 48755
    },
    {
      "entropy": 5.98639702796936,
      "epoch": 4.305898975626987,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003271500177805265,
      "loss": 5.1662,
      "mean_token_accuracy": 0.23021049201488494,
      "num_tokens": 90912282.0,
      "step": 48760
    },
    {
      "entropy": 6.003446483612061,
      "epoch": 4.306340515718827,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003271193833170152,
      "loss": 5.1526,
      "mean_token_accuracy": 0.23294443488121033,
      "num_tokens": 90921645.0,
      "step": 48765
    },
    {
      "entropy": 5.962237691879272,
      "epoch": 4.306782055810667,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003270887478325148,
      "loss": 5.2215,
      "mean_token_accuracy": 0.2267893448472023,
      "num_tokens": 90931659.0,
      "step": 48770
    },
    {
      "entropy": 6.068163967132568,
      "epoch": 4.307223595902508,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003270581113276254,
      "loss": 5.2973,
      "mean_token_accuracy": 0.21335649788379668,
      "num_tokens": 90940487.0,
      "step": 48775
    },
    {
      "entropy": 6.051117420196533,
      "epoch": 4.307665135994348,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032702747380294726,
      "loss": 5.2312,
      "mean_token_accuracy": 0.22862996757030488,
      "num_tokens": 90950710.0,
      "step": 48780
    },
    {
      "entropy": 6.116088247299194,
      "epoch": 4.308106676086188,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00032699683525908057,
      "loss": 5.362,
      "mean_token_accuracy": 0.20862789154052735,
      "num_tokens": 90960007.0,
      "step": 48785
    },
    {
      "entropy": 6.0306220054626465,
      "epoch": 4.308548216178029,
      "grad_norm": 1.359375,
      "learning_rate": 0.00032696619569662534,
      "loss": 5.2539,
      "mean_token_accuracy": 0.2257630541920662,
      "num_tokens": 90968375.0,
      "step": 48790
    },
    {
      "entropy": 6.0434363842010494,
      "epoch": 4.308989756269869,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003269355551161819,
      "loss": 5.1964,
      "mean_token_accuracy": 0.2268943503499031,
      "num_tokens": 90977591.0,
      "step": 48795
    },
    {
      "entropy": 5.933133888244629,
      "epoch": 4.309431296361709,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003269049135183505,
      "loss": 5.1098,
      "mean_token_accuracy": 0.23001569509506226,
      "num_tokens": 90986106.0,
      "step": 48800
    },
    {
      "entropy": 5.945161294937134,
      "epoch": 4.30987283645355,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032687427090373136,
      "loss": 5.1996,
      "mean_token_accuracy": 0.2280767872929573,
      "num_tokens": 90995826.0,
      "step": 48805
    },
    {
      "entropy": 6.05102162361145,
      "epoch": 4.31031437654539,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032684362727292485,
      "loss": 5.2703,
      "mean_token_accuracy": 0.21940853148698808,
      "num_tokens": 91004941.0,
      "step": 48810
    },
    {
      "entropy": 6.126395225524902,
      "epoch": 4.3107559166372305,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000326812982626531,
      "loss": 5.3226,
      "mean_token_accuracy": 0.2118456929922104,
      "num_tokens": 91014147.0,
      "step": 48815
    },
    {
      "entropy": 6.059711265563965,
      "epoch": 4.311197456729071,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003267823369651504,
      "loss": 5.2303,
      "mean_token_accuracy": 0.21952050179243088,
      "num_tokens": 91024143.0,
      "step": 48820
    },
    {
      "entropy": 5.951238441467285,
      "epoch": 4.3116389968209115,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032675169028938323,
      "loss": 5.1215,
      "mean_token_accuracy": 0.23792582303285598,
      "num_tokens": 91033368.0,
      "step": 48825
    },
    {
      "entropy": 6.056608104705811,
      "epoch": 4.3120805369127515,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003267210425998299,
      "loss": 5.2509,
      "mean_token_accuracy": 0.21936701089143754,
      "num_tokens": 91042502.0,
      "step": 48830
    },
    {
      "entropy": 6.027118396759033,
      "epoch": 4.3125220770045924,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003266903938970907,
      "loss": 5.2904,
      "mean_token_accuracy": 0.22321909815073013,
      "num_tokens": 91051798.0,
      "step": 48835
    },
    {
      "entropy": 6.077975797653198,
      "epoch": 4.3129636170964325,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00032665974418176615,
      "loss": 5.2957,
      "mean_token_accuracy": 0.20833895206451417,
      "num_tokens": 91060503.0,
      "step": 48840
    },
    {
      "entropy": 6.009342861175537,
      "epoch": 4.3134051571882726,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032662909345445657,
      "loss": 5.1364,
      "mean_token_accuracy": 0.23120843172073363,
      "num_tokens": 91069231.0,
      "step": 48845
    },
    {
      "entropy": 6.049452066421509,
      "epoch": 4.313846697280113,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032659844171576243,
      "loss": 5.3069,
      "mean_token_accuracy": 0.21233791261911392,
      "num_tokens": 91079723.0,
      "step": 48850
    },
    {
      "entropy": 6.041365337371826,
      "epoch": 4.3142882373719535,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032656778896628406,
      "loss": 5.3099,
      "mean_token_accuracy": 0.21461548656225204,
      "num_tokens": 91089507.0,
      "step": 48855
    },
    {
      "entropy": 6.110560035705566,
      "epoch": 4.314729777463794,
      "grad_norm": 1.34375,
      "learning_rate": 0.00032653713520662216,
      "loss": 5.2613,
      "mean_token_accuracy": 0.21748730093240737,
      "num_tokens": 91099072.0,
      "step": 48860
    },
    {
      "entropy": 6.101798486709595,
      "epoch": 4.315171317555634,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00032650648043737706,
      "loss": 5.2606,
      "mean_token_accuracy": 0.21562084406614304,
      "num_tokens": 91107929.0,
      "step": 48865
    },
    {
      "entropy": 6.010149097442627,
      "epoch": 4.315612857647475,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00032647582465914924,
      "loss": 5.2483,
      "mean_token_accuracy": 0.22500555217266083,
      "num_tokens": 91116923.0,
      "step": 48870
    },
    {
      "entropy": 5.961456823348999,
      "epoch": 4.316054397739315,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003264451678725394,
      "loss": 5.1508,
      "mean_token_accuracy": 0.2327345222234726,
      "num_tokens": 91126198.0,
      "step": 48875
    },
    {
      "entropy": 6.062096738815308,
      "epoch": 4.316495937831155,
      "grad_norm": 1.40625,
      "learning_rate": 0.00032641451007814793,
      "loss": 5.262,
      "mean_token_accuracy": 0.21526004523038864,
      "num_tokens": 91135141.0,
      "step": 48880
    },
    {
      "entropy": 6.021204280853271,
      "epoch": 4.316937477922996,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032638385127657546,
      "loss": 5.1341,
      "mean_token_accuracy": 0.22232139408588408,
      "num_tokens": 91144052.0,
      "step": 48885
    },
    {
      "entropy": 5.982494306564331,
      "epoch": 4.317379018014836,
      "grad_norm": 1.359375,
      "learning_rate": 0.00032635319146842254,
      "loss": 5.2201,
      "mean_token_accuracy": 0.22378459870815276,
      "num_tokens": 91153289.0,
      "step": 48890
    },
    {
      "entropy": 6.03976583480835,
      "epoch": 4.317820558106676,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003263225306542898,
      "loss": 5.2287,
      "mean_token_accuracy": 0.21994154155254364,
      "num_tokens": 91162981.0,
      "step": 48895
    },
    {
      "entropy": 6.022217798233032,
      "epoch": 4.318262098198517,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00032629186883477787,
      "loss": 5.2263,
      "mean_token_accuracy": 0.2147899478673935,
      "num_tokens": 91171593.0,
      "step": 48900
    },
    {
      "entropy": 5.993072509765625,
      "epoch": 4.318703638290357,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003262612060104874,
      "loss": 5.2672,
      "mean_token_accuracy": 0.2211688682436943,
      "num_tokens": 91180847.0,
      "step": 48905
    },
    {
      "entropy": 6.0314168453216555,
      "epoch": 4.319145178382197,
      "grad_norm": 1.34375,
      "learning_rate": 0.00032623054218201906,
      "loss": 5.2741,
      "mean_token_accuracy": 0.2202255055308342,
      "num_tokens": 91190572.0,
      "step": 48910
    },
    {
      "entropy": 6.117891073226929,
      "epoch": 4.319586718474038,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003261998773499736,
      "loss": 5.2893,
      "mean_token_accuracy": 0.2179439753293991,
      "num_tokens": 91199352.0,
      "step": 48915
    },
    {
      "entropy": 6.126251459121704,
      "epoch": 4.320028258565878,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003261692115149516,
      "loss": 5.2609,
      "mean_token_accuracy": 0.2156662568449974,
      "num_tokens": 91208230.0,
      "step": 48920
    },
    {
      "entropy": 6.069905233383179,
      "epoch": 4.320469798657718,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032613854467755383,
      "loss": 5.3272,
      "mean_token_accuracy": 0.2085884392261505,
      "num_tokens": 91217225.0,
      "step": 48925
    },
    {
      "entropy": 6.042582273483276,
      "epoch": 4.320911338749559,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003261078768383811,
      "loss": 5.2436,
      "mean_token_accuracy": 0.21820773482322692,
      "num_tokens": 91226611.0,
      "step": 48930
    },
    {
      "entropy": 6.056371784210205,
      "epoch": 4.321352878841399,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003260772079980341,
      "loss": 5.2684,
      "mean_token_accuracy": 0.21238099336624144,
      "num_tokens": 91236464.0,
      "step": 48935
    },
    {
      "entropy": 6.023149919509888,
      "epoch": 4.321794418933239,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003260465381571137,
      "loss": 5.2226,
      "mean_token_accuracy": 0.217990542948246,
      "num_tokens": 91245829.0,
      "step": 48940
    },
    {
      "entropy": 5.9848579406738285,
      "epoch": 4.32223595902508,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003260158673162205,
      "loss": 5.2255,
      "mean_token_accuracy": 0.2231573760509491,
      "num_tokens": 91254654.0,
      "step": 48945
    },
    {
      "entropy": 6.023188877105713,
      "epoch": 4.32267749911692,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032598519547595567,
      "loss": 5.283,
      "mean_token_accuracy": 0.2183112308382988,
      "num_tokens": 91264031.0,
      "step": 48950
    },
    {
      "entropy": 6.0304347515106205,
      "epoch": 4.32311903920876,
      "grad_norm": 1.25,
      "learning_rate": 0.00032595452263691973,
      "loss": 5.235,
      "mean_token_accuracy": 0.2190317004919052,
      "num_tokens": 91273638.0,
      "step": 48955
    },
    {
      "entropy": 6.061360454559326,
      "epoch": 4.323560579300601,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003259238487997138,
      "loss": 5.3086,
      "mean_token_accuracy": 0.2183957055211067,
      "num_tokens": 91282644.0,
      "step": 48960
    },
    {
      "entropy": 6.062176179885864,
      "epoch": 4.324002119392441,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00032589317396493843,
      "loss": 5.2386,
      "mean_token_accuracy": 0.22195110768079757,
      "num_tokens": 91292249.0,
      "step": 48965
    },
    {
      "entropy": 5.991433811187744,
      "epoch": 4.324443659484281,
      "grad_norm": 1.34375,
      "learning_rate": 0.00032586249813319485,
      "loss": 5.2424,
      "mean_token_accuracy": 0.2200953483581543,
      "num_tokens": 91301302.0,
      "step": 48970
    },
    {
      "entropy": 5.961963653564453,
      "epoch": 4.324885199576121,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003258318213050839,
      "loss": 5.2157,
      "mean_token_accuracy": 0.22161336243152618,
      "num_tokens": 91310486.0,
      "step": 48975
    },
    {
      "entropy": 6.104259586334228,
      "epoch": 4.325326739667962,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003258011434812064,
      "loss": 5.3262,
      "mean_token_accuracy": 0.21518594920635223,
      "num_tokens": 91320770.0,
      "step": 48980
    },
    {
      "entropy": 6.02338809967041,
      "epoch": 4.325768279759802,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003257704646621634,
      "loss": 5.198,
      "mean_token_accuracy": 0.2259119302034378,
      "num_tokens": 91329416.0,
      "step": 48985
    },
    {
      "entropy": 6.012403345108032,
      "epoch": 4.326209819851643,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00032573978484855585,
      "loss": 5.1729,
      "mean_token_accuracy": 0.2235917016863823,
      "num_tokens": 91338416.0,
      "step": 48990
    },
    {
      "entropy": 6.023179960250855,
      "epoch": 4.326651359943483,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032570910404098485,
      "loss": 5.2827,
      "mean_token_accuracy": 0.21301438212394713,
      "num_tokens": 91348179.0,
      "step": 48995
    },
    {
      "entropy": 5.933147716522217,
      "epoch": 4.327092900035323,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003256784222400513,
      "loss": 5.1407,
      "mean_token_accuracy": 0.23222050219774246,
      "num_tokens": 91356450.0,
      "step": 49000
    },
    {
      "entropy": 5.987435626983642,
      "epoch": 4.327534440127163,
      "grad_norm": 1.25,
      "learning_rate": 0.00032564773944635634,
      "loss": 5.162,
      "mean_token_accuracy": 0.22307463884353637,
      "num_tokens": 91365271.0,
      "step": 49005
    },
    {
      "entropy": 6.026652431488037,
      "epoch": 4.327975980219004,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032561705566050084,
      "loss": 5.2292,
      "mean_token_accuracy": 0.2203722447156906,
      "num_tokens": 91374958.0,
      "step": 49010
    },
    {
      "entropy": 6.0285365104675295,
      "epoch": 4.328417520310844,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003255863708830861,
      "loss": 5.2352,
      "mean_token_accuracy": 0.2191449746489525,
      "num_tokens": 91384892.0,
      "step": 49015
    },
    {
      "entropy": 6.099302053451538,
      "epoch": 4.328859060402684,
      "grad_norm": 1.375,
      "learning_rate": 0.00032555568511471306,
      "loss": 5.3429,
      "mean_token_accuracy": 0.20819327235221863,
      "num_tokens": 91394452.0,
      "step": 49020
    },
    {
      "entropy": 6.0405601978302,
      "epoch": 4.329300600494525,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000325524998355983,
      "loss": 5.2231,
      "mean_token_accuracy": 0.2162109926342964,
      "num_tokens": 91403833.0,
      "step": 49025
    },
    {
      "entropy": 6.05055341720581,
      "epoch": 4.329742140586365,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032549431060749693,
      "loss": 5.2377,
      "mean_token_accuracy": 0.21840407103300094,
      "num_tokens": 91413029.0,
      "step": 49030
    },
    {
      "entropy": 6.032879114151001,
      "epoch": 4.330183680678205,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00032546362186985597,
      "loss": 5.2603,
      "mean_token_accuracy": 0.21907900869846345,
      "num_tokens": 91421673.0,
      "step": 49035
    },
    {
      "entropy": 6.067804384231567,
      "epoch": 4.330625220770046,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003254329321436614,
      "loss": 5.2354,
      "mean_token_accuracy": 0.21990272849798204,
      "num_tokens": 91430806.0,
      "step": 49040
    },
    {
      "entropy": 6.022507095336914,
      "epoch": 4.331066760861886,
      "grad_norm": 1.375,
      "learning_rate": 0.00032540224142951443,
      "loss": 5.1773,
      "mean_token_accuracy": 0.23083896636962892,
      "num_tokens": 91439557.0,
      "step": 49045
    },
    {
      "entropy": 5.953165864944458,
      "epoch": 4.331508300953726,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032537154972801606,
      "loss": 5.2112,
      "mean_token_accuracy": 0.2287064880132675,
      "num_tokens": 91449267.0,
      "step": 49050
    },
    {
      "entropy": 6.01601505279541,
      "epoch": 4.331949841045567,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00032534085703976774,
      "loss": 5.2592,
      "mean_token_accuracy": 0.21000748127698898,
      "num_tokens": 91458545.0,
      "step": 49055
    },
    {
      "entropy": 5.956929492950439,
      "epoch": 4.332391381137407,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032531016336537075,
      "loss": 5.1711,
      "mean_token_accuracy": 0.22775033116340637,
      "num_tokens": 91467436.0,
      "step": 49060
    },
    {
      "entropy": 6.005564451217651,
      "epoch": 4.332832921229247,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003252794687054262,
      "loss": 5.0788,
      "mean_token_accuracy": 0.23276206701993943,
      "num_tokens": 91475721.0,
      "step": 49065
    },
    {
      "entropy": 6.048842144012451,
      "epoch": 4.333274461321088,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032524877306053544,
      "loss": 5.2697,
      "mean_token_accuracy": 0.21813629269599916,
      "num_tokens": 91484943.0,
      "step": 49070
    },
    {
      "entropy": 5.914062452316284,
      "epoch": 4.333716001412928,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003252180764312999,
      "loss": 5.0639,
      "mean_token_accuracy": 0.2384849727153778,
      "num_tokens": 91494263.0,
      "step": 49075
    },
    {
      "entropy": 6.0119898319244385,
      "epoch": 4.334157541504768,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003251873788183207,
      "loss": 5.2109,
      "mean_token_accuracy": 0.22550834715366364,
      "num_tokens": 91503246.0,
      "step": 49080
    },
    {
      "entropy": 6.039932060241699,
      "epoch": 4.334599081596609,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032515668022219926,
      "loss": 5.3163,
      "mean_token_accuracy": 0.2148416221141815,
      "num_tokens": 91513086.0,
      "step": 49085
    },
    {
      "entropy": 6.070546674728393,
      "epoch": 4.335040621688449,
      "grad_norm": 1.265625,
      "learning_rate": 0.000325125980643537,
      "loss": 5.2913,
      "mean_token_accuracy": 0.2169956848025322,
      "num_tokens": 91522579.0,
      "step": 49090
    },
    {
      "entropy": 6.018959808349609,
      "epoch": 4.3354821617802894,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003250952800829354,
      "loss": 5.1793,
      "mean_token_accuracy": 0.23319248706102372,
      "num_tokens": 91531633.0,
      "step": 49095
    },
    {
      "entropy": 5.982614898681641,
      "epoch": 4.33592370187213,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00032506457854099567,
      "loss": 5.1536,
      "mean_token_accuracy": 0.22701314836740494,
      "num_tokens": 91540501.0,
      "step": 49100
    },
    {
      "entropy": 6.134118318557739,
      "epoch": 4.33636524196397,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032503387601831937,
      "loss": 5.3438,
      "mean_token_accuracy": 0.20739265531301498,
      "num_tokens": 91549811.0,
      "step": 49105
    },
    {
      "entropy": 6.111767911911011,
      "epoch": 4.3368067820558105,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00032500317251550777,
      "loss": 5.3119,
      "mean_token_accuracy": 0.21853795796632766,
      "num_tokens": 91558584.0,
      "step": 49110
    },
    {
      "entropy": 5.892862796783447,
      "epoch": 4.337248322147651,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032497246803316255,
      "loss": 5.1349,
      "mean_token_accuracy": 0.2335933968424797,
      "num_tokens": 91568259.0,
      "step": 49115
    },
    {
      "entropy": 6.04819393157959,
      "epoch": 4.3376898622394915,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00032494176257188506,
      "loss": 5.2344,
      "mean_token_accuracy": 0.22140156626701354,
      "num_tokens": 91577411.0,
      "step": 49120
    },
    {
      "entropy": 6.034600687026978,
      "epoch": 4.3381314023313315,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032491105613227693,
      "loss": 5.2587,
      "mean_token_accuracy": 0.21495412141084672,
      "num_tokens": 91586851.0,
      "step": 49125
    },
    {
      "entropy": 6.06815357208252,
      "epoch": 4.338572942423172,
      "grad_norm": 1.375,
      "learning_rate": 0.00032488034871493957,
      "loss": 5.2012,
      "mean_token_accuracy": 0.22547730207443237,
      "num_tokens": 91595924.0,
      "step": 49130
    },
    {
      "entropy": 6.081540298461914,
      "epoch": 4.3390144825150125,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00032484964032047453,
      "loss": 5.2195,
      "mean_token_accuracy": 0.22366185188293458,
      "num_tokens": 91604529.0,
      "step": 49135
    },
    {
      "entropy": 5.962831258773804,
      "epoch": 4.339456022606853,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00032481893094948335,
      "loss": 5.1481,
      "mean_token_accuracy": 0.2299761012196541,
      "num_tokens": 91613000.0,
      "step": 49140
    },
    {
      "entropy": 6.058701992034912,
      "epoch": 4.339897562698693,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003247882206025677,
      "loss": 5.3654,
      "mean_token_accuracy": 0.20752007514238358,
      "num_tokens": 91623419.0,
      "step": 49145
    },
    {
      "entropy": 6.062630701065063,
      "epoch": 4.340339102790534,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032475750928032907,
      "loss": 5.2242,
      "mean_token_accuracy": 0.22428067326545714,
      "num_tokens": 91632833.0,
      "step": 49150
    },
    {
      "entropy": 6.081332063674926,
      "epoch": 4.340780642882374,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003247267969833692,
      "loss": 5.1818,
      "mean_token_accuracy": 0.2252239316701889,
      "num_tokens": 91642691.0,
      "step": 49155
    },
    {
      "entropy": 6.086390590667724,
      "epoch": 4.341222182974214,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003246960837122897,
      "loss": 5.3135,
      "mean_token_accuracy": 0.2178097039461136,
      "num_tokens": 91651910.0,
      "step": 49160
    },
    {
      "entropy": 6.021502494812012,
      "epoch": 4.341663723066055,
      "grad_norm": 1.34375,
      "learning_rate": 0.00032466536946769203,
      "loss": 5.1341,
      "mean_token_accuracy": 0.2292025476694107,
      "num_tokens": 91660698.0,
      "step": 49165
    },
    {
      "entropy": 6.109340953826904,
      "epoch": 4.342105263157895,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003246346542501781,
      "loss": 5.3062,
      "mean_token_accuracy": 0.21213509291410446,
      "num_tokens": 91669987.0,
      "step": 49170
    },
    {
      "entropy": 6.006227493286133,
      "epoch": 4.342546803249735,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003246039380603494,
      "loss": 5.2761,
      "mean_token_accuracy": 0.2232406720519066,
      "num_tokens": 91679618.0,
      "step": 49175
    },
    {
      "entropy": 6.079055213928223,
      "epoch": 4.342988343341576,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003245732208988079,
      "loss": 5.3053,
      "mean_token_accuracy": 0.2140488639473915,
      "num_tokens": 91689137.0,
      "step": 49180
    },
    {
      "entropy": 6.074229001998901,
      "epoch": 4.343429883433416,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032454250276615514,
      "loss": 5.23,
      "mean_token_accuracy": 0.22968363165855407,
      "num_tokens": 91698573.0,
      "step": 49185
    },
    {
      "entropy": 6.0211245059967045,
      "epoch": 4.343871423525256,
      "grad_norm": 1.328125,
      "learning_rate": 0.000324511783662993,
      "loss": 5.2192,
      "mean_token_accuracy": 0.22606089860200881,
      "num_tokens": 91708319.0,
      "step": 49190
    },
    {
      "entropy": 6.080251550674438,
      "epoch": 4.344312963617097,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00032448106358992315,
      "loss": 5.3325,
      "mean_token_accuracy": 0.2210884675383568,
      "num_tokens": 91717909.0,
      "step": 49195
    },
    {
      "entropy": 6.079530334472656,
      "epoch": 4.344754503708937,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032445034254754745,
      "loss": 5.2556,
      "mean_token_accuracy": 0.22103544026613237,
      "num_tokens": 91727125.0,
      "step": 49200
    },
    {
      "entropy": 6.061993598937988,
      "epoch": 4.345196043800777,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032441962053646753,
      "loss": 5.2809,
      "mean_token_accuracy": 0.21676944196224213,
      "num_tokens": 91736381.0,
      "step": 49205
    },
    {
      "entropy": 6.052382755279541,
      "epoch": 4.345637583892618,
      "grad_norm": 1.421875,
      "learning_rate": 0.00032438889755728544,
      "loss": 5.2808,
      "mean_token_accuracy": 0.2126500278711319,
      "num_tokens": 91745498.0,
      "step": 49210
    },
    {
      "entropy": 6.0297839641571045,
      "epoch": 4.346079123984458,
      "grad_norm": 1.34375,
      "learning_rate": 0.00032435817361060296,
      "loss": 5.2121,
      "mean_token_accuracy": 0.221847765147686,
      "num_tokens": 91753898.0,
      "step": 49215
    },
    {
      "entropy": 6.0735925197601315,
      "epoch": 4.346520664076298,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00032432744869702193,
      "loss": 5.2769,
      "mean_token_accuracy": 0.21420092582702638,
      "num_tokens": 91763518.0,
      "step": 49220
    },
    {
      "entropy": 6.0785314559936525,
      "epoch": 4.346962204168139,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032429672281714415,
      "loss": 5.336,
      "mean_token_accuracy": 0.21311308294534684,
      "num_tokens": 91772378.0,
      "step": 49225
    },
    {
      "entropy": 5.99944052696228,
      "epoch": 4.347403744259979,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00032426599597157167,
      "loss": 5.139,
      "mean_token_accuracy": 0.22701786011457442,
      "num_tokens": 91781263.0,
      "step": 49230
    },
    {
      "entropy": 6.030052995681762,
      "epoch": 4.347845284351819,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003242352681609063,
      "loss": 5.207,
      "mean_token_accuracy": 0.2252463921904564,
      "num_tokens": 91790328.0,
      "step": 49235
    },
    {
      "entropy": 6.052355003356934,
      "epoch": 4.34828682444366,
      "grad_norm": 1.25,
      "learning_rate": 0.00032420453938575007,
      "loss": 5.3128,
      "mean_token_accuracy": 0.21756818443536757,
      "num_tokens": 91800430.0,
      "step": 49240
    },
    {
      "entropy": 6.104255294799804,
      "epoch": 4.3487283645355,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003241738096467049,
      "loss": 5.3401,
      "mean_token_accuracy": 0.2143874943256378,
      "num_tokens": 91809457.0,
      "step": 49245
    },
    {
      "entropy": 6.050101947784424,
      "epoch": 4.34916990462734,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00032414307894437273,
      "loss": 5.2071,
      "mean_token_accuracy": 0.2239711806178093,
      "num_tokens": 91818213.0,
      "step": 49250
    },
    {
      "entropy": 6.062730121612549,
      "epoch": 4.34961144471918,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003241123472793556,
      "loss": 5.2139,
      "mean_token_accuracy": 0.22565431594848634,
      "num_tokens": 91827110.0,
      "step": 49255
    },
    {
      "entropy": 6.080104207992553,
      "epoch": 4.350052984811021,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032408161465225557,
      "loss": 5.2926,
      "mean_token_accuracy": 0.20887526273727416,
      "num_tokens": 91836136.0,
      "step": 49260
    },
    {
      "entropy": 6.02646107673645,
      "epoch": 4.350494524902861,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032405088106367464,
      "loss": 5.1914,
      "mean_token_accuracy": 0.21646656841039658,
      "num_tokens": 91845085.0,
      "step": 49265
    },
    {
      "entropy": 5.967532348632813,
      "epoch": 4.350936064994702,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032402014651421466,
      "loss": 5.0688,
      "mean_token_accuracy": 0.23502238392829894,
      "num_tokens": 91854671.0,
      "step": 49270
    },
    {
      "entropy": 6.106295585632324,
      "epoch": 4.351377605086542,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003239894110044781,
      "loss": 5.3371,
      "mean_token_accuracy": 0.2193776473402977,
      "num_tokens": 91863384.0,
      "step": 49275
    },
    {
      "entropy": 6.037597751617431,
      "epoch": 4.351819145178382,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003239586745350668,
      "loss": 5.2238,
      "mean_token_accuracy": 0.22452054917812347,
      "num_tokens": 91872339.0,
      "step": 49280
    },
    {
      "entropy": 6.053438329696656,
      "epoch": 4.352260685270222,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003239279371065829,
      "loss": 5.2366,
      "mean_token_accuracy": 0.21731108129024507,
      "num_tokens": 91880496.0,
      "step": 49285
    },
    {
      "entropy": 6.005913209915161,
      "epoch": 4.352702225362063,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032389719871962845,
      "loss": 5.2749,
      "mean_token_accuracy": 0.21946910321712493,
      "num_tokens": 91890364.0,
      "step": 49290
    },
    {
      "entropy": 6.05286340713501,
      "epoch": 4.353143765453903,
      "grad_norm": 1.25,
      "learning_rate": 0.0003238664593748058,
      "loss": 5.2529,
      "mean_token_accuracy": 0.21673347800970078,
      "num_tokens": 91899948.0,
      "step": 49295
    },
    {
      "entropy": 6.030974531173706,
      "epoch": 4.353585305545743,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003238357190727169,
      "loss": 5.2243,
      "mean_token_accuracy": 0.21844059079885483,
      "num_tokens": 91909388.0,
      "step": 49300
    },
    {
      "entropy": 6.075822257995606,
      "epoch": 4.354026845637584,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032380497781396404,
      "loss": 5.3457,
      "mean_token_accuracy": 0.2149113669991493,
      "num_tokens": 91918836.0,
      "step": 49305
    },
    {
      "entropy": 6.034208297729492,
      "epoch": 4.354468385729424,
      "grad_norm": 1.359375,
      "learning_rate": 0.00032377423559914954,
      "loss": 5.2729,
      "mean_token_accuracy": 0.22143981456756592,
      "num_tokens": 91927972.0,
      "step": 49310
    },
    {
      "entropy": 6.094947481155396,
      "epoch": 4.354909925821264,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032374349242887543,
      "loss": 5.2754,
      "mean_token_accuracy": 0.21666097790002822,
      "num_tokens": 91937159.0,
      "step": 49315
    },
    {
      "entropy": 6.089608430862427,
      "epoch": 4.355351465913105,
      "grad_norm": 1.125,
      "learning_rate": 0.000323712748303744,
      "loss": 5.2679,
      "mean_token_accuracy": 0.2237092599272728,
      "num_tokens": 91947391.0,
      "step": 49320
    },
    {
      "entropy": 5.946010780334473,
      "epoch": 4.355793006004945,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032368200322435757,
      "loss": 5.0555,
      "mean_token_accuracy": 0.2416670650243759,
      "num_tokens": 91957351.0,
      "step": 49325
    },
    {
      "entropy": 5.9505165100097654,
      "epoch": 4.356234546096785,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003236512571913184,
      "loss": 5.1607,
      "mean_token_accuracy": 0.22739427983760835,
      "num_tokens": 91965834.0,
      "step": 49330
    },
    {
      "entropy": 5.971686792373657,
      "epoch": 4.356676086188626,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032362051020522875,
      "loss": 5.1793,
      "mean_token_accuracy": 0.22418330907821654,
      "num_tokens": 91974960.0,
      "step": 49335
    },
    {
      "entropy": 6.023360776901245,
      "epoch": 4.357117626280466,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032358976226669095,
      "loss": 5.3009,
      "mean_token_accuracy": 0.21521927118301393,
      "num_tokens": 91984830.0,
      "step": 49340
    },
    {
      "entropy": 6.015460062026977,
      "epoch": 4.357559166372306,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003235590133763074,
      "loss": 5.1918,
      "mean_token_accuracy": 0.22884349077939986,
      "num_tokens": 91994491.0,
      "step": 49345
    },
    {
      "entropy": 5.967178916931152,
      "epoch": 4.358000706464147,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003235282635346803,
      "loss": 5.1841,
      "mean_token_accuracy": 0.2245100438594818,
      "num_tokens": 92003940.0,
      "step": 49350
    },
    {
      "entropy": 6.1285154819488525,
      "epoch": 4.358442246555987,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003234975127424122,
      "loss": 5.3936,
      "mean_token_accuracy": 0.21050223112106323,
      "num_tokens": 92013325.0,
      "step": 49355
    },
    {
      "entropy": 5.9745080947875975,
      "epoch": 4.358883786647827,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032346676100010533,
      "loss": 5.1632,
      "mean_token_accuracy": 0.2259178191423416,
      "num_tokens": 92022555.0,
      "step": 49360
    },
    {
      "entropy": 6.1304749011993405,
      "epoch": 4.359325326739668,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003234360083083622,
      "loss": 5.4045,
      "mean_token_accuracy": 0.20816163569688798,
      "num_tokens": 92032559.0,
      "step": 49365
    },
    {
      "entropy": 6.0320494174957275,
      "epoch": 4.359766866831508,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032340525466778526,
      "loss": 5.2309,
      "mean_token_accuracy": 0.22654195129871368,
      "num_tokens": 92041893.0,
      "step": 49370
    },
    {
      "entropy": 6.0009866714477536,
      "epoch": 4.360208406923348,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00032337450007897684,
      "loss": 5.2415,
      "mean_token_accuracy": 0.22648557126522065,
      "num_tokens": 92050556.0,
      "step": 49375
    },
    {
      "entropy": 5.961729192733765,
      "epoch": 4.360649947015189,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003233437445425395,
      "loss": 5.2872,
      "mean_token_accuracy": 0.21618181020021437,
      "num_tokens": 92058816.0,
      "step": 49380
    },
    {
      "entropy": 6.070818519592285,
      "epoch": 4.361091487107029,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032331298805907577,
      "loss": 5.3356,
      "mean_token_accuracy": 0.21531857401132584,
      "num_tokens": 92068892.0,
      "step": 49385
    },
    {
      "entropy": 6.133697366714477,
      "epoch": 4.3615330271988695,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00032328223062918793,
      "loss": 5.3065,
      "mean_token_accuracy": 0.20952605158090593,
      "num_tokens": 92077887.0,
      "step": 49390
    },
    {
      "entropy": 6.098395967483521,
      "epoch": 4.36197456729071,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003232514722534788,
      "loss": 5.2217,
      "mean_token_accuracy": 0.22253902703523637,
      "num_tokens": 92086703.0,
      "step": 49395
    },
    {
      "entropy": 6.019136953353882,
      "epoch": 4.3624161073825505,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003232207129325507,
      "loss": 5.2559,
      "mean_token_accuracy": 0.21771918535232543,
      "num_tokens": 92096669.0,
      "step": 49400
    },
    {
      "entropy": 6.001971340179443,
      "epoch": 4.3628576474743905,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032318995266700636,
      "loss": 5.1873,
      "mean_token_accuracy": 0.2260341078042984,
      "num_tokens": 92105176.0,
      "step": 49405
    },
    {
      "entropy": 6.075740909576416,
      "epoch": 4.363299187566231,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032315919145744815,
      "loss": 5.3252,
      "mean_token_accuracy": 0.21992767006158828,
      "num_tokens": 92115795.0,
      "step": 49410
    },
    {
      "entropy": 6.0883701801300045,
      "epoch": 4.3637407276580715,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003231284293044788,
      "loss": 5.2949,
      "mean_token_accuracy": 0.2173503279685974,
      "num_tokens": 92124016.0,
      "step": 49415
    },
    {
      "entropy": 6.012277126312256,
      "epoch": 4.364182267749912,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032309766620870084,
      "loss": 5.2759,
      "mean_token_accuracy": 0.21510225534439087,
      "num_tokens": 92133187.0,
      "step": 49420
    },
    {
      "entropy": 6.067049980163574,
      "epoch": 4.364623807841752,
      "grad_norm": 1.453125,
      "learning_rate": 0.00032306690217071704,
      "loss": 5.3021,
      "mean_token_accuracy": 0.2098075643181801,
      "num_tokens": 92141839.0,
      "step": 49425
    },
    {
      "entropy": 6.077973651885986,
      "epoch": 4.365065347933593,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032303613719112984,
      "loss": 5.3284,
      "mean_token_accuracy": 0.20816619396209718,
      "num_tokens": 92151288.0,
      "step": 49430
    },
    {
      "entropy": 6.0904200077056885,
      "epoch": 4.365506888025433,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003230053712705421,
      "loss": 5.2635,
      "mean_token_accuracy": 0.21751783788204193,
      "num_tokens": 92159935.0,
      "step": 49435
    },
    {
      "entropy": 6.12503833770752,
      "epoch": 4.365948428117273,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003229746044095565,
      "loss": 5.2877,
      "mean_token_accuracy": 0.22366729080677034,
      "num_tokens": 92171249.0,
      "step": 49440
    },
    {
      "entropy": 6.006118726730347,
      "epoch": 4.366389968209114,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003229438366087756,
      "loss": 5.1774,
      "mean_token_accuracy": 0.21768290251493455,
      "num_tokens": 92180105.0,
      "step": 49445
    },
    {
      "entropy": 6.02106204032898,
      "epoch": 4.366831508300954,
      "grad_norm": 1.421875,
      "learning_rate": 0.00032291306786880224,
      "loss": 5.2544,
      "mean_token_accuracy": 0.21885820478200912,
      "num_tokens": 92189475.0,
      "step": 49450
    },
    {
      "entropy": 6.003297662734985,
      "epoch": 4.367273048392794,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032288229819023913,
      "loss": 5.1799,
      "mean_token_accuracy": 0.22446053624153137,
      "num_tokens": 92198620.0,
      "step": 49455
    },
    {
      "entropy": 6.082557153701782,
      "epoch": 4.367714588484635,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003228515275736891,
      "loss": 5.2073,
      "mean_token_accuracy": 0.22101293057203292,
      "num_tokens": 92208854.0,
      "step": 49460
    },
    {
      "entropy": 6.021292066574096,
      "epoch": 4.368156128576475,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003228207560197548,
      "loss": 5.2673,
      "mean_token_accuracy": 0.21980007290840148,
      "num_tokens": 92219273.0,
      "step": 49465
    },
    {
      "entropy": 5.998938274383545,
      "epoch": 4.368597668668315,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032278998352903906,
      "loss": 5.2716,
      "mean_token_accuracy": 0.21036606580018996,
      "num_tokens": 92228360.0,
      "step": 49470
    },
    {
      "entropy": 6.026541090011596,
      "epoch": 4.369039208760156,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003227592101021447,
      "loss": 5.264,
      "mean_token_accuracy": 0.215118008852005,
      "num_tokens": 92237672.0,
      "step": 49475
    },
    {
      "entropy": 6.0332350730896,
      "epoch": 4.369480748851996,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032272843573967464,
      "loss": 5.1989,
      "mean_token_accuracy": 0.22713701128959657,
      "num_tokens": 92246616.0,
      "step": 49480
    },
    {
      "entropy": 6.032342720031738,
      "epoch": 4.369922288943836,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00032269766044223166,
      "loss": 5.2699,
      "mean_token_accuracy": 0.21569139659404754,
      "num_tokens": 92256314.0,
      "step": 49485
    },
    {
      "entropy": 6.02464599609375,
      "epoch": 4.370363829035677,
      "grad_norm": 1.34375,
      "learning_rate": 0.00032266688421041863,
      "loss": 5.2229,
      "mean_token_accuracy": 0.22272400707006454,
      "num_tokens": 92264523.0,
      "step": 49490
    },
    {
      "entropy": 6.02106671333313,
      "epoch": 4.370805369127517,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00032263610704483844,
      "loss": 5.2871,
      "mean_token_accuracy": 0.21592534333467484,
      "num_tokens": 92273167.0,
      "step": 49495
    },
    {
      "entropy": 6.03072452545166,
      "epoch": 4.371246909219357,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032260532894609395,
      "loss": 5.2473,
      "mean_token_accuracy": 0.22258199006319046,
      "num_tokens": 92283679.0,
      "step": 49500
    },
    {
      "entropy": 6.060755586624145,
      "epoch": 4.371688449311198,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00032257454991478824,
      "loss": 5.2662,
      "mean_token_accuracy": 0.21374882906675338,
      "num_tokens": 92294077.0,
      "step": 49505
    },
    {
      "entropy": 6.090130043029785,
      "epoch": 4.372129989403038,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003225437699515241,
      "loss": 5.2884,
      "mean_token_accuracy": 0.21702259331941604,
      "num_tokens": 92304174.0,
      "step": 49510
    },
    {
      "entropy": 6.063501501083374,
      "epoch": 4.372571529494878,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003225129890569045,
      "loss": 5.183,
      "mean_token_accuracy": 0.2175803542137146,
      "num_tokens": 92312916.0,
      "step": 49515
    },
    {
      "entropy": 6.029618263244629,
      "epoch": 4.373013069586719,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003224822072315326,
      "loss": 5.2308,
      "mean_token_accuracy": 0.2231118306517601,
      "num_tokens": 92322808.0,
      "step": 49520
    },
    {
      "entropy": 5.985385704040527,
      "epoch": 4.373454609678559,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032245142447601116,
      "loss": 5.2472,
      "mean_token_accuracy": 0.22241940200328827,
      "num_tokens": 92332517.0,
      "step": 49525
    },
    {
      "entropy": 6.04315447807312,
      "epoch": 4.373896149770399,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00032242064079094333,
      "loss": 5.2923,
      "mean_token_accuracy": 0.2160159796476364,
      "num_tokens": 92341751.0,
      "step": 49530
    },
    {
      "entropy": 6.130540180206299,
      "epoch": 4.374337689862239,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003223898561769321,
      "loss": 5.2854,
      "mean_token_accuracy": 0.21252651065587996,
      "num_tokens": 92352467.0,
      "step": 49535
    },
    {
      "entropy": 6.080249786376953,
      "epoch": 4.37477922995408,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003223590706345806,
      "loss": 5.2245,
      "mean_token_accuracy": 0.22813965231180192,
      "num_tokens": 92361361.0,
      "step": 49540
    },
    {
      "entropy": 6.0386213779449465,
      "epoch": 4.37522077004592,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003223282841644918,
      "loss": 5.1996,
      "mean_token_accuracy": 0.21723924726247787,
      "num_tokens": 92370583.0,
      "step": 49545
    },
    {
      "entropy": 6.06162462234497,
      "epoch": 4.375662310137761,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003222974967672688,
      "loss": 5.3161,
      "mean_token_accuracy": 0.2103470042347908,
      "num_tokens": 92380651.0,
      "step": 49550
    },
    {
      "entropy": 6.044790983200073,
      "epoch": 4.376103850229601,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003222667084435148,
      "loss": 5.2169,
      "mean_token_accuracy": 0.2276292234659195,
      "num_tokens": 92391268.0,
      "step": 49555
    },
    {
      "entropy": 6.109193944931031,
      "epoch": 4.376545390321441,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003222359191938328,
      "loss": 5.3246,
      "mean_token_accuracy": 0.2112146347761154,
      "num_tokens": 92402071.0,
      "step": 49560
    },
    {
      "entropy": 6.071422672271728,
      "epoch": 4.376986930413281,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032220512901882605,
      "loss": 5.2211,
      "mean_token_accuracy": 0.21747144162654877,
      "num_tokens": 92411569.0,
      "step": 49565
    },
    {
      "entropy": 6.068204069137574,
      "epoch": 4.377428470505122,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00032217433791909765,
      "loss": 5.2661,
      "mean_token_accuracy": 0.2181518167257309,
      "num_tokens": 92420503.0,
      "step": 49570
    },
    {
      "entropy": 6.00564136505127,
      "epoch": 4.377870010596962,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003221435458952508,
      "loss": 5.2204,
      "mean_token_accuracy": 0.22885961681604386,
      "num_tokens": 92430057.0,
      "step": 49575
    },
    {
      "entropy": 6.024640369415283,
      "epoch": 4.378311550688802,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003221127529478887,
      "loss": 5.1598,
      "mean_token_accuracy": 0.22995569556951523,
      "num_tokens": 92438364.0,
      "step": 49580
    },
    {
      "entropy": 5.988875532150269,
      "epoch": 4.378753090780643,
      "grad_norm": 1.359375,
      "learning_rate": 0.00032208195907761463,
      "loss": 5.1765,
      "mean_token_accuracy": 0.23031565994024278,
      "num_tokens": 92447492.0,
      "step": 49585
    },
    {
      "entropy": 6.075262403488159,
      "epoch": 4.379194630872483,
      "grad_norm": 1.34375,
      "learning_rate": 0.00032205116428503166,
      "loss": 5.293,
      "mean_token_accuracy": 0.21716418564319612,
      "num_tokens": 92456715.0,
      "step": 49590
    },
    {
      "entropy": 6.056954050064087,
      "epoch": 4.379636170964323,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003220203685707432,
      "loss": 5.3055,
      "mean_token_accuracy": 0.21652457416057586,
      "num_tokens": 92467021.0,
      "step": 49595
    },
    {
      "entropy": 6.130294132232666,
      "epoch": 4.380077711056164,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003219895719353524,
      "loss": 5.2734,
      "mean_token_accuracy": 0.21768767684698104,
      "num_tokens": 92475955.0,
      "step": 49600
    },
    {
      "entropy": 6.082018709182739,
      "epoch": 4.380519251148004,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003219587743794627,
      "loss": 5.2671,
      "mean_token_accuracy": 0.2118643641471863,
      "num_tokens": 92485326.0,
      "step": 49605
    },
    {
      "entropy": 6.010349273681641,
      "epoch": 4.380960791239844,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003219279759036772,
      "loss": 5.2514,
      "mean_token_accuracy": 0.22811238169670106,
      "num_tokens": 92494024.0,
      "step": 49610
    },
    {
      "entropy": 6.079763650894165,
      "epoch": 4.381402331331685,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003218971765085994,
      "loss": 5.272,
      "mean_token_accuracy": 0.21836392879486083,
      "num_tokens": 92503987.0,
      "step": 49615
    },
    {
      "entropy": 6.110907411575317,
      "epoch": 4.381843871423525,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003218663761948326,
      "loss": 5.1971,
      "mean_token_accuracy": 0.22089481502771377,
      "num_tokens": 92513047.0,
      "step": 49620
    },
    {
      "entropy": 6.096604681015014,
      "epoch": 4.382285411515365,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00032183557496298014,
      "loss": 5.2828,
      "mean_token_accuracy": 0.22320973873138428,
      "num_tokens": 92523756.0,
      "step": 49625
    },
    {
      "entropy": 6.047297573089599,
      "epoch": 4.382726951607206,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003218047728136454,
      "loss": 5.2699,
      "mean_token_accuracy": 0.220913802087307,
      "num_tokens": 92533042.0,
      "step": 49630
    },
    {
      "entropy": 6.077416610717774,
      "epoch": 4.383168491699046,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003217739697474318,
      "loss": 5.2395,
      "mean_token_accuracy": 0.22013337910175323,
      "num_tokens": 92542133.0,
      "step": 49635
    },
    {
      "entropy": 6.040828990936279,
      "epoch": 4.383610031790886,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003217431657649427,
      "loss": 5.2194,
      "mean_token_accuracy": 0.22027323693037032,
      "num_tokens": 92550991.0,
      "step": 49640
    },
    {
      "entropy": 6.031135559082031,
      "epoch": 4.384051571882727,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003217123608667816,
      "loss": 5.2839,
      "mean_token_accuracy": 0.2161594733595848,
      "num_tokens": 92560416.0,
      "step": 49645
    },
    {
      "entropy": 5.999750757217408,
      "epoch": 4.384493111974567,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003216815550535519,
      "loss": 5.214,
      "mean_token_accuracy": 0.23379204869270326,
      "num_tokens": 92569455.0,
      "step": 49650
    },
    {
      "entropy": 6.106399869918823,
      "epoch": 4.384934652066407,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003216507483258571,
      "loss": 5.2534,
      "mean_token_accuracy": 0.22258909344673156,
      "num_tokens": 92578461.0,
      "step": 49655
    },
    {
      "entropy": 5.9822286605834964,
      "epoch": 4.385376192158248,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003216199406843007,
      "loss": 5.1886,
      "mean_token_accuracy": 0.22824844270944594,
      "num_tokens": 92587850.0,
      "step": 49660
    },
    {
      "entropy": 6.023505115509034,
      "epoch": 4.385817732250088,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00032158913212948606,
      "loss": 5.1786,
      "mean_token_accuracy": 0.23123992532491683,
      "num_tokens": 92596163.0,
      "step": 49665
    },
    {
      "entropy": 6.070603132247925,
      "epoch": 4.3862592723419285,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003215583226620169,
      "loss": 5.2267,
      "mean_token_accuracy": 0.21278029680252075,
      "num_tokens": 92605467.0,
      "step": 49670
    },
    {
      "entropy": 6.0583251953125,
      "epoch": 4.386700812433769,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00032152751228249664,
      "loss": 5.2781,
      "mean_token_accuracy": 0.21591313630342485,
      "num_tokens": 92614108.0,
      "step": 49675
    },
    {
      "entropy": 6.034496784210205,
      "epoch": 4.3871423525256095,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032149670099152893,
      "loss": 5.2824,
      "mean_token_accuracy": 0.21775075942277908,
      "num_tokens": 92623480.0,
      "step": 49680
    },
    {
      "entropy": 6.109913778305054,
      "epoch": 4.3875838926174495,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003214658887897172,
      "loss": 5.3299,
      "mean_token_accuracy": 0.21792764216661453,
      "num_tokens": 92633110.0,
      "step": 49685
    },
    {
      "entropy": 6.093234348297119,
      "epoch": 4.38802543270929,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003214350756776652,
      "loss": 5.2167,
      "mean_token_accuracy": 0.21780716627836227,
      "num_tokens": 92642512.0,
      "step": 49690
    },
    {
      "entropy": 6.102414512634278,
      "epoch": 4.3884669728011305,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003214042616559765,
      "loss": 5.3569,
      "mean_token_accuracy": 0.20638346374034883,
      "num_tokens": 92652394.0,
      "step": 49695
    },
    {
      "entropy": 6.001203727722168,
      "epoch": 4.388908512892971,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003213734467252548,
      "loss": 5.2012,
      "mean_token_accuracy": 0.21549022942781448,
      "num_tokens": 92661762.0,
      "step": 49700
    },
    {
      "entropy": 6.095818901062012,
      "epoch": 4.389350052984811,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003213426308861035,
      "loss": 5.2467,
      "mean_token_accuracy": 0.2211246445775032,
      "num_tokens": 92670335.0,
      "step": 49705
    },
    {
      "entropy": 6.035959053039551,
      "epoch": 4.389791593076652,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003213118141391264,
      "loss": 5.2503,
      "mean_token_accuracy": 0.21615691781044005,
      "num_tokens": 92679433.0,
      "step": 49710
    },
    {
      "entropy": 5.983531141281128,
      "epoch": 4.390233133168492,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003212809964849273,
      "loss": 5.1756,
      "mean_token_accuracy": 0.23143292665481568,
      "num_tokens": 92688258.0,
      "step": 49715
    },
    {
      "entropy": 6.010986709594727,
      "epoch": 4.390674673260332,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003212501779241097,
      "loss": 5.2799,
      "mean_token_accuracy": 0.21861743777990342,
      "num_tokens": 92697090.0,
      "step": 49720
    },
    {
      "entropy": 5.971865701675415,
      "epoch": 4.391116213352173,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032121935845727756,
      "loss": 5.1383,
      "mean_token_accuracy": 0.2300131231546402,
      "num_tokens": 92706134.0,
      "step": 49725
    },
    {
      "entropy": 6.031006669998169,
      "epoch": 4.391557753444013,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003211885380850343,
      "loss": 5.2554,
      "mean_token_accuracy": 0.2135776922106743,
      "num_tokens": 92715057.0,
      "step": 49730
    },
    {
      "entropy": 6.072065258026123,
      "epoch": 4.391999293535853,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000321157716807984,
      "loss": 5.3077,
      "mean_token_accuracy": 0.21917413920164108,
      "num_tokens": 92724455.0,
      "step": 49735
    },
    {
      "entropy": 6.0248359680175785,
      "epoch": 4.392440833627694,
      "grad_norm": 1.25,
      "learning_rate": 0.00032112689462673015,
      "loss": 5.2055,
      "mean_token_accuracy": 0.22613064646720887,
      "num_tokens": 92733850.0,
      "step": 49740
    },
    {
      "entropy": 6.04614291191101,
      "epoch": 4.392882373719534,
      "grad_norm": 1.359375,
      "learning_rate": 0.00032109607154187674,
      "loss": 5.2393,
      "mean_token_accuracy": 0.23280955702066422,
      "num_tokens": 92743887.0,
      "step": 49745
    },
    {
      "entropy": 6.091229438781738,
      "epoch": 4.393323913811374,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003210652475540275,
      "loss": 5.2621,
      "mean_token_accuracy": 0.22369252592325212,
      "num_tokens": 92752657.0,
      "step": 49750
    },
    {
      "entropy": 6.035182809829712,
      "epoch": 4.393765453903215,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00032103442266378623,
      "loss": 5.2626,
      "mean_token_accuracy": 0.22885904610157012,
      "num_tokens": 92762182.0,
      "step": 49755
    },
    {
      "entropy": 6.082333707809449,
      "epoch": 4.394206993995055,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003210035968717569,
      "loss": 5.2953,
      "mean_token_accuracy": 0.2217119351029396,
      "num_tokens": 92771642.0,
      "step": 49760
    },
    {
      "entropy": 6.10133581161499,
      "epoch": 4.394648534086895,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032097277017854316,
      "loss": 5.3016,
      "mean_token_accuracy": 0.21918075382709504,
      "num_tokens": 92781920.0,
      "step": 49765
    },
    {
      "entropy": 6.1274001121521,
      "epoch": 4.395090074178736,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00032094194258474907,
      "loss": 5.3382,
      "mean_token_accuracy": 0.20523568689823152,
      "num_tokens": 92790892.0,
      "step": 49770
    },
    {
      "entropy": 6.049302816390991,
      "epoch": 4.395531614270576,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00032091111409097837,
      "loss": 5.1758,
      "mean_token_accuracy": 0.2207377403974533,
      "num_tokens": 92799942.0,
      "step": 49775
    },
    {
      "entropy": 5.950204133987427,
      "epoch": 4.395973154362416,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032088028469783514,
      "loss": 5.1294,
      "mean_token_accuracy": 0.23099036067724227,
      "num_tokens": 92809054.0,
      "step": 49780
    },
    {
      "entropy": 6.12145962715149,
      "epoch": 4.396414694454257,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003208494544059232,
      "loss": 5.372,
      "mean_token_accuracy": 0.20936260670423507,
      "num_tokens": 92819650.0,
      "step": 49785
    },
    {
      "entropy": 6.030419635772705,
      "epoch": 4.396856234546097,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032081862321584655,
      "loss": 5.2242,
      "mean_token_accuracy": 0.222736594080925,
      "num_tokens": 92828514.0,
      "step": 49790
    },
    {
      "entropy": 6.02163987159729,
      "epoch": 4.397297774637937,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032078779112820906,
      "loss": 5.1734,
      "mean_token_accuracy": 0.22935766577720643,
      "num_tokens": 92837977.0,
      "step": 49795
    },
    {
      "entropy": 6.047225666046143,
      "epoch": 4.397739314729778,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032075695814361485,
      "loss": 5.2128,
      "mean_token_accuracy": 0.21889323443174363,
      "num_tokens": 92847734.0,
      "step": 49800
    },
    {
      "entropy": 5.99313759803772,
      "epoch": 4.398180854821618,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003207261242626678,
      "loss": 5.2342,
      "mean_token_accuracy": 0.22170596867799758,
      "num_tokens": 92857137.0,
      "step": 49805
    },
    {
      "entropy": 6.070572423934936,
      "epoch": 4.398622394913458,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032069528948597194,
      "loss": 5.2624,
      "mean_token_accuracy": 0.22613236159086228,
      "num_tokens": 92866184.0,
      "step": 49810
    },
    {
      "entropy": 6.048924446105957,
      "epoch": 4.399063935005298,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003206644538141314,
      "loss": 5.2141,
      "mean_token_accuracy": 0.22249120324850083,
      "num_tokens": 92875021.0,
      "step": 49815
    },
    {
      "entropy": 6.0738848686218265,
      "epoch": 4.399505475097139,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003206336172477502,
      "loss": 5.2501,
      "mean_token_accuracy": 0.21595397889614104,
      "num_tokens": 92884938.0,
      "step": 49820
    },
    {
      "entropy": 6.088395357131958,
      "epoch": 4.399947015188979,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003206027797874324,
      "loss": 5.3611,
      "mean_token_accuracy": 0.20773066133260726,
      "num_tokens": 92893812.0,
      "step": 49825
    },
    {
      "entropy": 6.050552225112915,
      "epoch": 4.40038855528082,
      "grad_norm": 1.28125,
      "learning_rate": 0.000320571941433782,
      "loss": 5.2034,
      "mean_token_accuracy": 0.22526804655790328,
      "num_tokens": 92903692.0,
      "step": 49830
    },
    {
      "entropy": 6.042627286911011,
      "epoch": 4.40083009537266,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003205411021874033,
      "loss": 5.2353,
      "mean_token_accuracy": 0.2284150689840317,
      "num_tokens": 92912321.0,
      "step": 49835
    },
    {
      "entropy": 6.063397932052612,
      "epoch": 4.4012716354645,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032051026204890024,
      "loss": 5.3744,
      "mean_token_accuracy": 0.21090735048055648,
      "num_tokens": 92922560.0,
      "step": 49840
    },
    {
      "entropy": 6.085504674911499,
      "epoch": 4.40171317555634,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00032047942101887713,
      "loss": 5.2651,
      "mean_token_accuracy": 0.21916133910417557,
      "num_tokens": 92932095.0,
      "step": 49845
    },
    {
      "entropy": 6.094219779968261,
      "epoch": 4.402154715648181,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003204485790979379,
      "loss": 5.245,
      "mean_token_accuracy": 0.22079333961009978,
      "num_tokens": 92941650.0,
      "step": 49850
    },
    {
      "entropy": 6.004714632034302,
      "epoch": 4.402596255740021,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003204177362866869,
      "loss": 5.2171,
      "mean_token_accuracy": 0.22502393573522567,
      "num_tokens": 92951016.0,
      "step": 49855
    },
    {
      "entropy": 6.009087944030762,
      "epoch": 4.403037795831861,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00032038689258572823,
      "loss": 5.2761,
      "mean_token_accuracy": 0.21856490820646285,
      "num_tokens": 92960434.0,
      "step": 49860
    },
    {
      "entropy": 6.003093290328979,
      "epoch": 4.403479335923702,
      "grad_norm": 1.375,
      "learning_rate": 0.00032035604799566625,
      "loss": 5.1297,
      "mean_token_accuracy": 0.22666071653366088,
      "num_tokens": 92969332.0,
      "step": 49865
    },
    {
      "entropy": 6.066006708145141,
      "epoch": 4.403920876015542,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000320325202517105,
      "loss": 5.2226,
      "mean_token_accuracy": 0.22486210912466048,
      "num_tokens": 92977794.0,
      "step": 49870
    },
    {
      "entropy": 6.092533922195434,
      "epoch": 4.404362416107382,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00032029435615064884,
      "loss": 5.3786,
      "mean_token_accuracy": 0.21473732441663743,
      "num_tokens": 92987844.0,
      "step": 49875
    },
    {
      "entropy": 6.0882891654968265,
      "epoch": 4.404803956199223,
      "grad_norm": 1.359375,
      "learning_rate": 0.000320263508896902,
      "loss": 5.3119,
      "mean_token_accuracy": 0.20979506075382232,
      "num_tokens": 92997024.0,
      "step": 49880
    },
    {
      "entropy": 6.1146949291229244,
      "epoch": 4.405245496291063,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032023266075646874,
      "loss": 5.2681,
      "mean_token_accuracy": 0.22112446874380112,
      "num_tokens": 93005567.0,
      "step": 49885
    },
    {
      "entropy": 6.081368350982666,
      "epoch": 4.405687036382903,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032020181172995343,
      "loss": 5.2765,
      "mean_token_accuracy": 0.21722308844327926,
      "num_tokens": 93014943.0,
      "step": 49890
    },
    {
      "entropy": 6.0463316440582275,
      "epoch": 4.406128576474744,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003201709618179603,
      "loss": 5.2238,
      "mean_token_accuracy": 0.21601620316505432,
      "num_tokens": 93023972.0,
      "step": 49895
    },
    {
      "entropy": 5.981359767913818,
      "epoch": 4.406570116566584,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003201401110210938,
      "loss": 5.235,
      "mean_token_accuracy": 0.22212435454130172,
      "num_tokens": 93033469.0,
      "step": 49900
    },
    {
      "entropy": 6.039214944839477,
      "epoch": 4.407011656658424,
      "grad_norm": 1.390625,
      "learning_rate": 0.00032010925933995804,
      "loss": 5.2114,
      "mean_token_accuracy": 0.2196694552898407,
      "num_tokens": 93042993.0,
      "step": 49905
    },
    {
      "entropy": 5.9815717220306395,
      "epoch": 4.407453196750265,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003200784067751577,
      "loss": 5.1452,
      "mean_token_accuracy": 0.23256486058235168,
      "num_tokens": 93052080.0,
      "step": 49910
    },
    {
      "entropy": 5.995180368423462,
      "epoch": 4.407894736842105,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003200475533272969,
      "loss": 5.3098,
      "mean_token_accuracy": 0.2198876768350601,
      "num_tokens": 93061536.0,
      "step": 49915
    },
    {
      "entropy": 5.960009479522705,
      "epoch": 4.408336276933945,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003200166989969802,
      "loss": 5.2484,
      "mean_token_accuracy": 0.22568347603082656,
      "num_tokens": 93071134.0,
      "step": 49920
    },
    {
      "entropy": 6.084073543548584,
      "epoch": 4.408777817025786,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000319985843784812,
      "loss": 5.3006,
      "mean_token_accuracy": 0.21518827229738235,
      "num_tokens": 93080293.0,
      "step": 49925
    },
    {
      "entropy": 6.035478925704956,
      "epoch": 4.409219357117626,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003199549876913966,
      "loss": 5.2215,
      "mean_token_accuracy": 0.22189662605524063,
      "num_tokens": 93089828.0,
      "step": 49930
    },
    {
      "entropy": 6.069870233535767,
      "epoch": 4.409660897209466,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031992413071733867,
      "loss": 5.3116,
      "mean_token_accuracy": 0.21252729445695878,
      "num_tokens": 93099151.0,
      "step": 49935
    },
    {
      "entropy": 6.047096538543701,
      "epoch": 4.410102437301307,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031989327286324244,
      "loss": 5.2545,
      "mean_token_accuracy": 0.21703095138072967,
      "num_tokens": 93107841.0,
      "step": 49940
    },
    {
      "entropy": 5.998843622207642,
      "epoch": 4.410543977393147,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003198624141297127,
      "loss": 5.2496,
      "mean_token_accuracy": 0.21857085525989534,
      "num_tokens": 93117451.0,
      "step": 49945
    },
    {
      "entropy": 6.01777868270874,
      "epoch": 4.4109855174849875,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003198315545173536,
      "loss": 5.1892,
      "mean_token_accuracy": 0.2234471082687378,
      "num_tokens": 93125520.0,
      "step": 49950
    },
    {
      "entropy": 6.076767921447754,
      "epoch": 4.411427057576828,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031980069402677,
      "loss": 5.3373,
      "mean_token_accuracy": 0.21635590493679047,
      "num_tokens": 93135671.0,
      "step": 49955
    },
    {
      "entropy": 6.003994941711426,
      "epoch": 4.4118685976686685,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003197698326585662,
      "loss": 5.1379,
      "mean_token_accuracy": 0.23844470381736754,
      "num_tokens": 93145020.0,
      "step": 49960
    },
    {
      "entropy": 6.04693341255188,
      "epoch": 4.4123101377605085,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031973897041334687,
      "loss": 5.2507,
      "mean_token_accuracy": 0.22995917052030562,
      "num_tokens": 93154678.0,
      "step": 49965
    },
    {
      "entropy": 6.069682979583741,
      "epoch": 4.412751677852349,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003197081072917165,
      "loss": 5.2076,
      "mean_token_accuracy": 0.22196744829416276,
      "num_tokens": 93164253.0,
      "step": 49970
    },
    {
      "entropy": 6.042122220993042,
      "epoch": 4.4131932179441895,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003196772432942798,
      "loss": 5.3136,
      "mean_token_accuracy": 0.2179982677102089,
      "num_tokens": 93173916.0,
      "step": 49975
    },
    {
      "entropy": 6.006299257278442,
      "epoch": 4.41363475803603,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003196463784216413,
      "loss": 5.2189,
      "mean_token_accuracy": 0.21895729303359984,
      "num_tokens": 93182781.0,
      "step": 49980
    },
    {
      "entropy": 5.886580562591552,
      "epoch": 4.41407629812787,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003196155126744056,
      "loss": 5.166,
      "mean_token_accuracy": 0.23217409998178482,
      "num_tokens": 93191712.0,
      "step": 49985
    },
    {
      "entropy": 5.994838523864746,
      "epoch": 4.4145178382197106,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00031958464605317746,
      "loss": 5.1992,
      "mean_token_accuracy": 0.22687463611364364,
      "num_tokens": 93200348.0,
      "step": 49990
    },
    {
      "entropy": 5.998860311508179,
      "epoch": 4.414959378311551,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003195537785585614,
      "loss": 5.2066,
      "mean_token_accuracy": 0.22900039553642274,
      "num_tokens": 93210617.0,
      "step": 49995
    },
    {
      "entropy": 6.1080974578857425,
      "epoch": 4.415400918403391,
      "grad_norm": 1.34375,
      "learning_rate": 0.00031952291019116217,
      "loss": 5.2556,
      "mean_token_accuracy": 0.21545615941286086,
      "num_tokens": 93219538.0,
      "step": 50000
    },
    {
      "entropy": 6.094301986694336,
      "epoch": 4.415842458495232,
      "grad_norm": 1.25,
      "learning_rate": 0.0003194920409515844,
      "loss": 5.2593,
      "mean_token_accuracy": 0.2184096783399582,
      "num_tokens": 93229285.0,
      "step": 50005
    },
    {
      "entropy": 6.091911649703979,
      "epoch": 4.416283998587072,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003194611708404329,
      "loss": 5.2658,
      "mean_token_accuracy": 0.22240554839372634,
      "num_tokens": 93238269.0,
      "step": 50010
    },
    {
      "entropy": 6.067620420455933,
      "epoch": 4.416725538678912,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031943029985831233,
      "loss": 5.3631,
      "mean_token_accuracy": 0.20659298449754715,
      "num_tokens": 93248537.0,
      "step": 50015
    },
    {
      "entropy": 6.046495723724365,
      "epoch": 4.417167078770753,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003193994280058275,
      "loss": 5.2709,
      "mean_token_accuracy": 0.2238852024078369,
      "num_tokens": 93257708.0,
      "step": 50020
    },
    {
      "entropy": 6.1003273010253904,
      "epoch": 4.417608618862593,
      "grad_norm": 1.46875,
      "learning_rate": 0.00031936855528358305,
      "loss": 5.2811,
      "mean_token_accuracy": 0.2180451273918152,
      "num_tokens": 93266170.0,
      "step": 50025
    },
    {
      "entropy": 5.987220764160156,
      "epoch": 4.418050158954433,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000319337681692184,
      "loss": 5.1753,
      "mean_token_accuracy": 0.2276245579123497,
      "num_tokens": 93275252.0,
      "step": 50030
    },
    {
      "entropy": 6.017303228378296,
      "epoch": 4.418491699046274,
      "grad_norm": 1.328125,
      "learning_rate": 0.00031930680723223483,
      "loss": 5.2832,
      "mean_token_accuracy": 0.22189475297927858,
      "num_tokens": 93283980.0,
      "step": 50035
    },
    {
      "entropy": 6.05155086517334,
      "epoch": 4.418933239138114,
      "grad_norm": 1.328125,
      "learning_rate": 0.00031927593190434053,
      "loss": 5.2865,
      "mean_token_accuracy": 0.21754478365182878,
      "num_tokens": 93294395.0,
      "step": 50040
    },
    {
      "entropy": 6.096895408630371,
      "epoch": 4.419374779229954,
      "grad_norm": 1.359375,
      "learning_rate": 0.00031924505570910596,
      "loss": 5.2981,
      "mean_token_accuracy": 0.21354592591524124,
      "num_tokens": 93304268.0,
      "step": 50045
    },
    {
      "entropy": 6.075433397293091,
      "epoch": 4.419816319321795,
      "grad_norm": 1.359375,
      "learning_rate": 0.00031921417864713583,
      "loss": 5.2513,
      "mean_token_accuracy": 0.2181354805827141,
      "num_tokens": 93313445.0,
      "step": 50050
    },
    {
      "entropy": 6.049209403991699,
      "epoch": 4.420257859413635,
      "grad_norm": 1.359375,
      "learning_rate": 0.00031918330071903513,
      "loss": 5.3068,
      "mean_token_accuracy": 0.216652712225914,
      "num_tokens": 93322444.0,
      "step": 50055
    },
    {
      "entropy": 6.047723436355591,
      "epoch": 4.420699399505475,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003191524219254087,
      "loss": 5.2892,
      "mean_token_accuracy": 0.2206622764468193,
      "num_tokens": 93330854.0,
      "step": 50060
    },
    {
      "entropy": 5.987243366241455,
      "epoch": 4.421140939597316,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003191215422668614,
      "loss": 5.2474,
      "mean_token_accuracy": 0.22519634366035463,
      "num_tokens": 93340134.0,
      "step": 50065
    },
    {
      "entropy": 6.065576410293579,
      "epoch": 4.421582479689156,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003190906617439982,
      "loss": 5.2837,
      "mean_token_accuracy": 0.21077710390090942,
      "num_tokens": 93348987.0,
      "step": 50070
    },
    {
      "entropy": 6.097200632095337,
      "epoch": 4.422024019780996,
      "grad_norm": 1.375,
      "learning_rate": 0.00031905978035742403,
      "loss": 5.2864,
      "mean_token_accuracy": 0.2203537032008171,
      "num_tokens": 93358596.0,
      "step": 50075
    },
    {
      "entropy": 6.051420259475708,
      "epoch": 4.422465559872837,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003190288981077438,
      "loss": 5.2802,
      "mean_token_accuracy": 0.21885174214839936,
      "num_tokens": 93368681.0,
      "step": 50080
    },
    {
      "entropy": 6.015255546569824,
      "epoch": 4.422907099964677,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031899801499556254,
      "loss": 5.2345,
      "mean_token_accuracy": 0.23407390713691711,
      "num_tokens": 93377499.0,
      "step": 50085
    },
    {
      "entropy": 6.041907453536988,
      "epoch": 4.423348640056517,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031896713102148514,
      "loss": 5.2302,
      "mean_token_accuracy": 0.2233677938580513,
      "num_tokens": 93387159.0,
      "step": 50090
    },
    {
      "entropy": 6.050398969650269,
      "epoch": 4.423790180148357,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00031893624618611666,
      "loss": 5.259,
      "mean_token_accuracy": 0.21767404526472092,
      "num_tokens": 93396109.0,
      "step": 50095
    },
    {
      "entropy": 6.015649509429932,
      "epoch": 4.424231720240198,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00031890536049006214,
      "loss": 5.229,
      "mean_token_accuracy": 0.21933822631835936,
      "num_tokens": 93405109.0,
      "step": 50100
    },
    {
      "entropy": 5.99403977394104,
      "epoch": 4.424673260332038,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003188744739339265,
      "loss": 5.2146,
      "mean_token_accuracy": 0.22574180662631987,
      "num_tokens": 93413870.0,
      "step": 50105
    },
    {
      "entropy": 6.064032173156738,
      "epoch": 4.425114800423879,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003188435865183149,
      "loss": 5.2543,
      "mean_token_accuracy": 0.21884989589452744,
      "num_tokens": 93423432.0,
      "step": 50110
    },
    {
      "entropy": 6.009411001205445,
      "epoch": 4.425556340515719,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00031881269824383246,
      "loss": 5.2426,
      "mean_token_accuracy": 0.2193669095635414,
      "num_tokens": 93433826.0,
      "step": 50115
    },
    {
      "entropy": 6.048405647277832,
      "epoch": 4.425997880607559,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003187818091110841,
      "loss": 5.244,
      "mean_token_accuracy": 0.21886030286550523,
      "num_tokens": 93443680.0,
      "step": 50120
    },
    {
      "entropy": 6.014990663528442,
      "epoch": 4.426439420699399,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031875091912067494,
      "loss": 5.2384,
      "mean_token_accuracy": 0.2292626529932022,
      "num_tokens": 93452793.0,
      "step": 50125
    },
    {
      "entropy": 6.046043395996094,
      "epoch": 4.42688096079124,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003187200282732103,
      "loss": 5.1778,
      "mean_token_accuracy": 0.2308179959654808,
      "num_tokens": 93462062.0,
      "step": 50130
    },
    {
      "entropy": 5.9968164443969725,
      "epoch": 4.42732250088308,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00031868913656929503,
      "loss": 5.1874,
      "mean_token_accuracy": 0.22468274384737014,
      "num_tokens": 93471079.0,
      "step": 50135
    },
    {
      "entropy": 5.99547929763794,
      "epoch": 4.42776404097492,
      "grad_norm": 1.390625,
      "learning_rate": 0.00031865824400953457,
      "loss": 5.2689,
      "mean_token_accuracy": 0.21833254992961884,
      "num_tokens": 93481126.0,
      "step": 50140
    },
    {
      "entropy": 6.074772691726684,
      "epoch": 4.428205581066761,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031862735059453386,
      "loss": 5.2463,
      "mean_token_accuracy": 0.22236216366291045,
      "num_tokens": 93490382.0,
      "step": 50145
    },
    {
      "entropy": 6.059643650054932,
      "epoch": 4.428647121158601,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00031859645632489815,
      "loss": 5.3144,
      "mean_token_accuracy": 0.21317547410726548,
      "num_tokens": 93499497.0,
      "step": 50150
    },
    {
      "entropy": 6.03073787689209,
      "epoch": 4.429088661250441,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00031856556120123266,
      "loss": 5.288,
      "mean_token_accuracy": 0.22125454843044282,
      "num_tokens": 93508621.0,
      "step": 50155
    },
    {
      "entropy": 6.0147209644317625,
      "epoch": 4.429530201342282,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003185346652241426,
      "loss": 5.228,
      "mean_token_accuracy": 0.2264480099081993,
      "num_tokens": 93517244.0,
      "step": 50160
    },
    {
      "entropy": 6.0113341331481935,
      "epoch": 4.429971741434122,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003185037683942332,
      "loss": 5.1818,
      "mean_token_accuracy": 0.22496563494205474,
      "num_tokens": 93525828.0,
      "step": 50165
    },
    {
      "entropy": 6.068388032913208,
      "epoch": 4.430413281525962,
      "grad_norm": 1.34375,
      "learning_rate": 0.00031847287071210976,
      "loss": 5.3395,
      "mean_token_accuracy": 0.20802089124917983,
      "num_tokens": 93535427.0,
      "step": 50170
    },
    {
      "entropy": 6.075096607208252,
      "epoch": 4.430854821617803,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003184419721783775,
      "loss": 5.1981,
      "mean_token_accuracy": 0.22506261318922044,
      "num_tokens": 93544486.0,
      "step": 50175
    },
    {
      "entropy": 6.081974792480469,
      "epoch": 4.431296361709643,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003184110727936417,
      "loss": 5.2796,
      "mean_token_accuracy": 0.2164281576871872,
      "num_tokens": 93554451.0,
      "step": 50180
    },
    {
      "entropy": 5.985921907424927,
      "epoch": 4.431737901801483,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003183801725585077,
      "loss": 5.1885,
      "mean_token_accuracy": 0.22293712794780732,
      "num_tokens": 93563179.0,
      "step": 50185
    },
    {
      "entropy": 5.901452255249024,
      "epoch": 4.432179441893324,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003183492714735807,
      "loss": 5.1373,
      "mean_token_accuracy": 0.2364243432879448,
      "num_tokens": 93572192.0,
      "step": 50190
    },
    {
      "entropy": 6.038916873931885,
      "epoch": 4.432620981985164,
      "grad_norm": 1.4375,
      "learning_rate": 0.00031831836953946624,
      "loss": 5.2865,
      "mean_token_accuracy": 0.22048396319150926,
      "num_tokens": 93581782.0,
      "step": 50195
    },
    {
      "entropy": 6.0750038623809814,
      "epoch": 4.433062522077004,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003182874667567695,
      "loss": 5.2632,
      "mean_token_accuracy": 0.22097976207733155,
      "num_tokens": 93590876.0,
      "step": 50200
    },
    {
      "entropy": 6.031072330474854,
      "epoch": 4.433504062168845,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031825656312609594,
      "loss": 5.245,
      "mean_token_accuracy": 0.22468983829021455,
      "num_tokens": 93600647.0,
      "step": 50205
    },
    {
      "entropy": 6.01557092666626,
      "epoch": 4.433945602260685,
      "grad_norm": 1.296875,
      "learning_rate": 0.00031822565864805086,
      "loss": 5.1913,
      "mean_token_accuracy": 0.22170920222997664,
      "num_tokens": 93610097.0,
      "step": 50210
    },
    {
      "entropy": 6.0949760437011715,
      "epoch": 4.434387142352525,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003181947533232397,
      "loss": 5.2744,
      "mean_token_accuracy": 0.22017584145069122,
      "num_tokens": 93619203.0,
      "step": 50215
    },
    {
      "entropy": 5.97062201499939,
      "epoch": 4.434828682444366,
      "grad_norm": 1.359375,
      "learning_rate": 0.00031816384715226793,
      "loss": 5.1413,
      "mean_token_accuracy": 0.22437983602285386,
      "num_tokens": 93628235.0,
      "step": 50220
    },
    {
      "entropy": 6.095989990234375,
      "epoch": 4.435270222536206,
      "grad_norm": 1.34375,
      "learning_rate": 0.00031813294013574095,
      "loss": 5.3044,
      "mean_token_accuracy": 0.22050278186798095,
      "num_tokens": 93637337.0,
      "step": 50225
    },
    {
      "entropy": 6.057680988311768,
      "epoch": 4.4357117626280465,
      "grad_norm": 1.375,
      "learning_rate": 0.0003181020322742642,
      "loss": 5.3095,
      "mean_token_accuracy": 0.21804822236299515,
      "num_tokens": 93647079.0,
      "step": 50230
    },
    {
      "entropy": 6.085684776306152,
      "epoch": 4.436153302719887,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003180711235684432,
      "loss": 5.2537,
      "mean_token_accuracy": 0.21688681542873384,
      "num_tokens": 93656531.0,
      "step": 50235
    },
    {
      "entropy": 6.09502329826355,
      "epoch": 4.4365948428117274,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031804021401888336,
      "loss": 5.3536,
      "mean_token_accuracy": 0.21472365856170655,
      "num_tokens": 93666884.0,
      "step": 50240
    },
    {
      "entropy": 6.0358418941497805,
      "epoch": 4.4370363829035675,
      "grad_norm": 1.328125,
      "learning_rate": 0.00031800930362619006,
      "loss": 5.3131,
      "mean_token_accuracy": 0.2113628402352333,
      "num_tokens": 93675840.0,
      "step": 50245
    },
    {
      "entropy": 5.994051265716553,
      "epoch": 4.4374779229954076,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003179783923909691,
      "loss": 5.2121,
      "mean_token_accuracy": 0.22653213739395142,
      "num_tokens": 93684744.0,
      "step": 50250
    },
    {
      "entropy": 6.067167663574219,
      "epoch": 4.4379194630872485,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00031794748031382583,
      "loss": 5.3353,
      "mean_token_accuracy": 0.20827613919973373,
      "num_tokens": 93693964.0,
      "step": 50255
    },
    {
      "entropy": 6.1561203956604,
      "epoch": 4.4383610031790885,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003179165673953659,
      "loss": 5.3249,
      "mean_token_accuracy": 0.22026443630456924,
      "num_tokens": 93703126.0,
      "step": 50260
    },
    {
      "entropy": 6.050548267364502,
      "epoch": 4.438802543270929,
      "grad_norm": 1.421875,
      "learning_rate": 0.00031788565363619477,
      "loss": 5.2736,
      "mean_token_accuracy": 0.22146120816469192,
      "num_tokens": 93711947.0,
      "step": 50265
    },
    {
      "entropy": 6.0079552173614506,
      "epoch": 4.4392440833627695,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003178547390369181,
      "loss": 5.2348,
      "mean_token_accuracy": 0.22431884706020355,
      "num_tokens": 93721405.0,
      "step": 50270
    },
    {
      "entropy": 6.028184270858764,
      "epoch": 4.43968562345461,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031782382359814136,
      "loss": 5.2871,
      "mean_token_accuracy": 0.21905129551887512,
      "num_tokens": 93730955.0,
      "step": 50275
    },
    {
      "entropy": 6.046392488479614,
      "epoch": 4.44012716354645,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031779290732047045,
      "loss": 5.257,
      "mean_token_accuracy": 0.21507690697908402,
      "num_tokens": 93740929.0,
      "step": 50280
    },
    {
      "entropy": 6.05502347946167,
      "epoch": 4.440568703638291,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00031776199020451074,
      "loss": 5.2209,
      "mean_token_accuracy": 0.22488847374916077,
      "num_tokens": 93750610.0,
      "step": 50285
    },
    {
      "entropy": 6.0019690990448,
      "epoch": 4.441010243730131,
      "grad_norm": 1.34375,
      "learning_rate": 0.00031773107225086797,
      "loss": 5.2029,
      "mean_token_accuracy": 0.2319316104054451,
      "num_tokens": 93760236.0,
      "step": 50290
    },
    {
      "entropy": 5.975342845916748,
      "epoch": 4.441451783821971,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00031770015346014777,
      "loss": 5.2194,
      "mean_token_accuracy": 0.22545394748449327,
      "num_tokens": 93769346.0,
      "step": 50295
    },
    {
      "entropy": 6.079979372024536,
      "epoch": 4.441893323913812,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00031766923383295595,
      "loss": 5.2868,
      "mean_token_accuracy": 0.2176870658993721,
      "num_tokens": 93777913.0,
      "step": 50300
    },
    {
      "entropy": 6.06778473854065,
      "epoch": 4.442334864005652,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031763831336989803,
      "loss": 5.2309,
      "mean_token_accuracy": 0.2163509175181389,
      "num_tokens": 93787598.0,
      "step": 50305
    },
    {
      "entropy": 6.057104921340942,
      "epoch": 4.442776404097492,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031760739207157976,
      "loss": 5.2552,
      "mean_token_accuracy": 0.2150893032550812,
      "num_tokens": 93797743.0,
      "step": 50310
    },
    {
      "entropy": 6.084583950042725,
      "epoch": 4.443217944189333,
      "grad_norm": 1.34375,
      "learning_rate": 0.00031757646993860693,
      "loss": 5.2648,
      "mean_token_accuracy": 0.221408674120903,
      "num_tokens": 93806132.0,
      "step": 50315
    },
    {
      "entropy": 6.009333515167237,
      "epoch": 4.443659484281173,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003175455469715853,
      "loss": 5.1826,
      "mean_token_accuracy": 0.22293027192354203,
      "num_tokens": 93814835.0,
      "step": 50320
    },
    {
      "entropy": 6.070720672607422,
      "epoch": 4.444101024373013,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003175146231711206,
      "loss": 5.3302,
      "mean_token_accuracy": 0.21400492936372756,
      "num_tokens": 93824790.0,
      "step": 50325
    },
    {
      "entropy": 5.975285387039184,
      "epoch": 4.444542564464854,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00031748369853781854,
      "loss": 5.1479,
      "mean_token_accuracy": 0.2316602125763893,
      "num_tokens": 93833401.0,
      "step": 50330
    },
    {
      "entropy": 5.960549020767212,
      "epoch": 4.444984104556694,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003174527730722851,
      "loss": 5.1869,
      "mean_token_accuracy": 0.2300284907221794,
      "num_tokens": 93841932.0,
      "step": 50335
    },
    {
      "entropy": 5.9801758289337155,
      "epoch": 4.445425644648534,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003174218467751259,
      "loss": 5.145,
      "mean_token_accuracy": 0.23554494827985764,
      "num_tokens": 93850256.0,
      "step": 50340
    },
    {
      "entropy": 6.029037094116211,
      "epoch": 4.445867184740375,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003173909196469469,
      "loss": 5.2473,
      "mean_token_accuracy": 0.22288139760494233,
      "num_tokens": 93860093.0,
      "step": 50345
    },
    {
      "entropy": 6.01021728515625,
      "epoch": 4.446308724832215,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003173599916883539,
      "loss": 5.2674,
      "mean_token_accuracy": 0.21707877814769744,
      "num_tokens": 93869063.0,
      "step": 50350
    },
    {
      "entropy": 6.038885927200317,
      "epoch": 4.446750264924055,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00031732906289995265,
      "loss": 5.2989,
      "mean_token_accuracy": 0.21969693005084992,
      "num_tokens": 93878704.0,
      "step": 50355
    },
    {
      "entropy": 6.076512384414673,
      "epoch": 4.447191805015896,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003172981332823493,
      "loss": 5.2226,
      "mean_token_accuracy": 0.22685422599315644,
      "num_tokens": 93887804.0,
      "step": 50360
    },
    {
      "entropy": 6.0320127487182615,
      "epoch": 4.447633345107736,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003172672028361494,
      "loss": 5.304,
      "mean_token_accuracy": 0.215040223300457,
      "num_tokens": 93896990.0,
      "step": 50365
    },
    {
      "entropy": 6.000066328048706,
      "epoch": 4.448074885199576,
      "grad_norm": 1.515625,
      "learning_rate": 0.00031723627156195915,
      "loss": 5.1786,
      "mean_token_accuracy": 0.23029719591140746,
      "num_tokens": 93905516.0,
      "step": 50370
    },
    {
      "entropy": 6.038322257995605,
      "epoch": 4.448516425291416,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003172053394603843,
      "loss": 5.2747,
      "mean_token_accuracy": 0.21996768116950988,
      "num_tokens": 93915086.0,
      "step": 50375
    },
    {
      "entropy": 6.111239767074585,
      "epoch": 4.448957965383257,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003171744065320309,
      "loss": 5.2829,
      "mean_token_accuracy": 0.2161855950951576,
      "num_tokens": 93924834.0,
      "step": 50380
    },
    {
      "entropy": 6.144980764389038,
      "epoch": 4.449399505475097,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00031714347277750484,
      "loss": 5.3402,
      "mean_token_accuracy": 0.21903062760829925,
      "num_tokens": 93934770.0,
      "step": 50385
    },
    {
      "entropy": 6.003443956375122,
      "epoch": 4.449841045566938,
      "grad_norm": 1.25,
      "learning_rate": 0.0003171125381974121,
      "loss": 5.1447,
      "mean_token_accuracy": 0.2311984419822693,
      "num_tokens": 93945198.0,
      "step": 50390
    },
    {
      "entropy": 6.085482454299926,
      "epoch": 4.450282585658778,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003170816027923587,
      "loss": 5.3458,
      "mean_token_accuracy": 0.21228087395429612,
      "num_tokens": 93955587.0,
      "step": 50395
    },
    {
      "entropy": 5.936852073669433,
      "epoch": 4.450724125750618,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003170506665629506,
      "loss": 5.1193,
      "mean_token_accuracy": 0.23416538536548615,
      "num_tokens": 93964360.0,
      "step": 50400
    },
    {
      "entropy": 6.008877611160278,
      "epoch": 4.451165665842458,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003170197295097939,
      "loss": 5.1338,
      "mean_token_accuracy": 0.23739766478538513,
      "num_tokens": 93973358.0,
      "step": 50405
    },
    {
      "entropy": 5.966026306152344,
      "epoch": 4.451607205934299,
      "grad_norm": 1.453125,
      "learning_rate": 0.00031698879163349463,
      "loss": 5.1264,
      "mean_token_accuracy": 0.23881710767745973,
      "num_tokens": 93982083.0,
      "step": 50410
    },
    {
      "entropy": 5.999241161346435,
      "epoch": 4.452048746026139,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003169578529346588,
      "loss": 5.2423,
      "mean_token_accuracy": 0.22311515659093856,
      "num_tokens": 93991086.0,
      "step": 50415
    },
    {
      "entropy": 5.938956785202026,
      "epoch": 4.452490286117979,
      "grad_norm": 1.296875,
      "learning_rate": 0.00031692691341389246,
      "loss": 5.1528,
      "mean_token_accuracy": 0.22686525881290437,
      "num_tokens": 93999966.0,
      "step": 50420
    },
    {
      "entropy": 6.148738527297974,
      "epoch": 4.45293182620982,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00031689597307180177,
      "loss": 5.3662,
      "mean_token_accuracy": 0.21392159759998322,
      "num_tokens": 94010098.0,
      "step": 50425
    },
    {
      "entropy": 5.922574520111084,
      "epoch": 4.45337336630166,
      "grad_norm": 1.390625,
      "learning_rate": 0.00031686503190899275,
      "loss": 5.1397,
      "mean_token_accuracy": 0.22563256323337555,
      "num_tokens": 94019345.0,
      "step": 50430
    },
    {
      "entropy": 6.047852993011475,
      "epoch": 4.4538149063935,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003168340899260716,
      "loss": 5.2339,
      "mean_token_accuracy": 0.2174492210149765,
      "num_tokens": 94028459.0,
      "step": 50435
    },
    {
      "entropy": 6.054863595962525,
      "epoch": 4.454256446485341,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031680314712364437,
      "loss": 5.2283,
      "mean_token_accuracy": 0.2198246344923973,
      "num_tokens": 94037246.0,
      "step": 50440
    },
    {
      "entropy": 6.006714725494385,
      "epoch": 4.454697986577181,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031677220350231726,
      "loss": 5.1974,
      "mean_token_accuracy": 0.22346411049365997,
      "num_tokens": 94046313.0,
      "step": 50445
    },
    {
      "entropy": 6.011773681640625,
      "epoch": 4.455139526669021,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003167412590626965,
      "loss": 5.2225,
      "mean_token_accuracy": 0.22466596066951752,
      "num_tokens": 94056102.0,
      "step": 50450
    },
    {
      "entropy": 5.955696630477905,
      "epoch": 4.455581066760862,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003167103138053882,
      "loss": 5.1194,
      "mean_token_accuracy": 0.22876229286193847,
      "num_tokens": 94065353.0,
      "step": 50455
    },
    {
      "entropy": 5.954233026504516,
      "epoch": 4.456022606852702,
      "grad_norm": 1.390625,
      "learning_rate": 0.00031667936773099853,
      "loss": 5.1542,
      "mean_token_accuracy": 0.22870899587869645,
      "num_tokens": 94074098.0,
      "step": 50460
    },
    {
      "entropy": 5.963583517074585,
      "epoch": 4.456464146944542,
      "grad_norm": 1.375,
      "learning_rate": 0.0003166484208401338,
      "loss": 5.2567,
      "mean_token_accuracy": 0.22128936946392058,
      "num_tokens": 94083061.0,
      "step": 50465
    },
    {
      "entropy": 5.9978312969207765,
      "epoch": 4.456905687036383,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003166174731334002,
      "loss": 5.2042,
      "mean_token_accuracy": 0.2292369171977043,
      "num_tokens": 94091599.0,
      "step": 50470
    },
    {
      "entropy": 6.10058183670044,
      "epoch": 4.457347227128223,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00031658652461140395,
      "loss": 5.2322,
      "mean_token_accuracy": 0.22134115546941757,
      "num_tokens": 94100242.0,
      "step": 50475
    },
    {
      "entropy": 6.100496864318847,
      "epoch": 4.457788767220063,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003165555752747514,
      "loss": 5.2897,
      "mean_token_accuracy": 0.2119052991271019,
      "num_tokens": 94111273.0,
      "step": 50480
    },
    {
      "entropy": 5.995230007171631,
      "epoch": 4.458230307311904,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003165246251240486,
      "loss": 5.1808,
      "mean_token_accuracy": 0.21752292215824126,
      "num_tokens": 94120187.0,
      "step": 50485
    },
    {
      "entropy": 6.026410293579102,
      "epoch": 4.458671847403744,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031649367415990216,
      "loss": 5.3233,
      "mean_token_accuracy": 0.21195076704025267,
      "num_tokens": 94130897.0,
      "step": 50490
    },
    {
      "entropy": 6.076645946502685,
      "epoch": 4.459113387495584,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003164627223829182,
      "loss": 5.3415,
      "mean_token_accuracy": 0.21732379496097565,
      "num_tokens": 94140447.0,
      "step": 50495
    },
    {
      "entropy": 6.004798221588135,
      "epoch": 4.459554927587425,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000316431769793703,
      "loss": 5.1807,
      "mean_token_accuracy": 0.22236859053373337,
      "num_tokens": 94150409.0,
      "step": 50500
    },
    {
      "entropy": 6.017670679092407,
      "epoch": 4.459996467679265,
      "grad_norm": 1.375,
      "learning_rate": 0.0003164008163928632,
      "loss": 5.2012,
      "mean_token_accuracy": 0.23025232404470444,
      "num_tokens": 94159494.0,
      "step": 50505
    },
    {
      "entropy": 6.066024923324585,
      "epoch": 4.460438007771105,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003163698621810047,
      "loss": 5.2463,
      "mean_token_accuracy": 0.22467995584011077,
      "num_tokens": 94168640.0,
      "step": 50510
    },
    {
      "entropy": 6.021030521392822,
      "epoch": 4.460879547862946,
      "grad_norm": 1.328125,
      "learning_rate": 0.00031633890715873425,
      "loss": 5.2284,
      "mean_token_accuracy": 0.22431578934192659,
      "num_tokens": 94178039.0,
      "step": 50515
    },
    {
      "entropy": 6.01981372833252,
      "epoch": 4.461321087954786,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003163079513266582,
      "loss": 5.23,
      "mean_token_accuracy": 0.22065378725528717,
      "num_tokens": 94186916.0,
      "step": 50520
    },
    {
      "entropy": 5.961128091812133,
      "epoch": 4.4617626280466265,
      "grad_norm": 1.546875,
      "learning_rate": 0.0003162769946853827,
      "loss": 5.1541,
      "mean_token_accuracy": 0.23105049431324004,
      "num_tokens": 94196056.0,
      "step": 50525
    },
    {
      "entropy": 6.055885648727417,
      "epoch": 4.4622041681384665,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003162460372355145,
      "loss": 5.3544,
      "mean_token_accuracy": 0.2164454936981201,
      "num_tokens": 94205410.0,
      "step": 50530
    },
    {
      "entropy": 5.994051456451416,
      "epoch": 4.4626457082303075,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003162150789776598,
      "loss": 5.2194,
      "mean_token_accuracy": 0.22778845131397246,
      "num_tokens": 94214604.0,
      "step": 50535
    },
    {
      "entropy": 6.019695711135864,
      "epoch": 4.4630872483221475,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00031618411991242527,
      "loss": 5.1895,
      "mean_token_accuracy": 0.22326425164937974,
      "num_tokens": 94223447.0,
      "step": 50540
    },
    {
      "entropy": 6.104844808578491,
      "epoch": 4.463528788413988,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003161531600404172,
      "loss": 5.2716,
      "mean_token_accuracy": 0.21674920320510865,
      "num_tokens": 94231936.0,
      "step": 50545
    },
    {
      "entropy": 6.08623423576355,
      "epoch": 4.4639703285058285,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000316122199362242,
      "loss": 5.2933,
      "mean_token_accuracy": 0.2110053688287735,
      "num_tokens": 94242084.0,
      "step": 50550
    },
    {
      "entropy": 6.001146936416626,
      "epoch": 4.464411868597669,
      "grad_norm": 1.265625,
      "learning_rate": 0.00031609123787850645,
      "loss": 5.2544,
      "mean_token_accuracy": 0.2252486139535904,
      "num_tokens": 94252754.0,
      "step": 50555
    },
    {
      "entropy": 6.06340856552124,
      "epoch": 4.464853408689509,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003160602755898168,
      "loss": 5.2982,
      "mean_token_accuracy": 0.21778938174247742,
      "num_tokens": 94262103.0,
      "step": 50560
    },
    {
      "entropy": 6.041942167282104,
      "epoch": 4.46529494878135,
      "grad_norm": 1.25,
      "learning_rate": 0.00031602931249677976,
      "loss": 5.2674,
      "mean_token_accuracy": 0.2184070810675621,
      "num_tokens": 94272281.0,
      "step": 50565
    },
    {
      "entropy": 6.12665376663208,
      "epoch": 4.46573648887319,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031599834860000184,
      "loss": 5.3678,
      "mean_token_accuracy": 0.21168912947177887,
      "num_tokens": 94281878.0,
      "step": 50570
    },
    {
      "entropy": 6.098860216140747,
      "epoch": 4.46617802896503,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003159673839000896,
      "loss": 5.3088,
      "mean_token_accuracy": 0.21561656296253204,
      "num_tokens": 94290792.0,
      "step": 50575
    },
    {
      "entropy": 6.002708864212036,
      "epoch": 4.466619569056871,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003159364183976496,
      "loss": 5.204,
      "mean_token_accuracy": 0.22654528319835662,
      "num_tokens": 94299167.0,
      "step": 50580
    },
    {
      "entropy": 6.059002161026001,
      "epoch": 4.467061109148711,
      "grad_norm": 1.40625,
      "learning_rate": 0.00031590545209328843,
      "loss": 5.3148,
      "mean_token_accuracy": 0.2146853044629097,
      "num_tokens": 94308285.0,
      "step": 50585
    },
    {
      "entropy": 5.946333026885986,
      "epoch": 4.467502649240551,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031587448498761274,
      "loss": 5.2173,
      "mean_token_accuracy": 0.22475280463695527,
      "num_tokens": 94318974.0,
      "step": 50590
    },
    {
      "entropy": 6.040516424179077,
      "epoch": 4.467944189332392,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000315843517081229,
      "loss": 5.2588,
      "mean_token_accuracy": 0.22166985720396043,
      "num_tokens": 94328841.0,
      "step": 50595
    },
    {
      "entropy": 6.010335636138916,
      "epoch": 4.468385729424232,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003158125483747442,
      "loss": 5.2049,
      "mean_token_accuracy": 0.22787337154150009,
      "num_tokens": 94337510.0,
      "step": 50600
    },
    {
      "entropy": 6.07722315788269,
      "epoch": 4.468827269516072,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031578157886876467,
      "loss": 5.2633,
      "mean_token_accuracy": 0.22052531391382219,
      "num_tokens": 94347120.0,
      "step": 50605
    },
    {
      "entropy": 6.056341981887817,
      "epoch": 4.469268809607913,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003157506085638973,
      "loss": 5.2565,
      "mean_token_accuracy": 0.21326775401830672,
      "num_tokens": 94356506.0,
      "step": 50610
    },
    {
      "entropy": 6.000523948669434,
      "epoch": 4.469710349699753,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003157196374607486,
      "loss": 5.2022,
      "mean_token_accuracy": 0.22537857443094253,
      "num_tokens": 94366145.0,
      "step": 50615
    },
    {
      "entropy": 6.064871072769165,
      "epoch": 4.470151889791593,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00031568866555992537,
      "loss": 5.2623,
      "mean_token_accuracy": 0.21675931811332702,
      "num_tokens": 94374856.0,
      "step": 50620
    },
    {
      "entropy": 6.017176914215088,
      "epoch": 4.470593429883434,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00031565769286203435,
      "loss": 5.2645,
      "mean_token_accuracy": 0.2238522872328758,
      "num_tokens": 94383354.0,
      "step": 50625
    },
    {
      "entropy": 6.022355318069458,
      "epoch": 4.471034969975274,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003156267193676821,
      "loss": 5.2333,
      "mean_token_accuracy": 0.2238097682595253,
      "num_tokens": 94392465.0,
      "step": 50630
    },
    {
      "entropy": 5.984247159957886,
      "epoch": 4.471476510067114,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003155957450774757,
      "loss": 5.1791,
      "mean_token_accuracy": 0.23077252358198166,
      "num_tokens": 94401714.0,
      "step": 50635
    },
    {
      "entropy": 6.020223331451416,
      "epoch": 4.471918050158955,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003155647699920217,
      "loss": 5.2437,
      "mean_token_accuracy": 0.22114349156618118,
      "num_tokens": 94411286.0,
      "step": 50640
    },
    {
      "entropy": 5.9178701877594,
      "epoch": 4.472359590250795,
      "grad_norm": 1.25,
      "learning_rate": 0.0003155337941119269,
      "loss": 5.1568,
      "mean_token_accuracy": 0.2241077482700348,
      "num_tokens": 94421191.0,
      "step": 50645
    },
    {
      "entropy": 6.093020677566528,
      "epoch": 4.472801130342635,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003155028174377981,
      "loss": 5.3178,
      "mean_token_accuracy": 0.2127208650112152,
      "num_tokens": 94430666.0,
      "step": 50650
    },
    {
      "entropy": 6.082799243927002,
      "epoch": 4.473242670434475,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003154718399702422,
      "loss": 5.3035,
      "mean_token_accuracy": 0.22121178060770036,
      "num_tokens": 94440453.0,
      "step": 50655
    },
    {
      "entropy": 6.063337659835815,
      "epoch": 4.473684210526316,
      "grad_norm": 1.265625,
      "learning_rate": 0.000315440861709866,
      "loss": 5.2139,
      "mean_token_accuracy": 0.22434523701667786,
      "num_tokens": 94449344.0,
      "step": 50660
    },
    {
      "entropy": 6.013115692138672,
      "epoch": 4.474125750618156,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003154098826572763,
      "loss": 5.1683,
      "mean_token_accuracy": 0.2229877457022667,
      "num_tokens": 94458157.0,
      "step": 50665
    },
    {
      "entropy": 6.064225912094116,
      "epoch": 4.474567290709997,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031537890281307994,
      "loss": 5.2725,
      "mean_token_accuracy": 0.21907473504543304,
      "num_tokens": 94467646.0,
      "step": 50670
    },
    {
      "entropy": 5.967296552658081,
      "epoch": 4.475008830801837,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031534792217788386,
      "loss": 5.2674,
      "mean_token_accuracy": 0.21921859234571456,
      "num_tokens": 94476904.0,
      "step": 50675
    },
    {
      "entropy": 6.018720102310181,
      "epoch": 4.475450370893677,
      "grad_norm": 1.25,
      "learning_rate": 0.00031531694075229497,
      "loss": 5.1988,
      "mean_token_accuracy": 0.22042347490787506,
      "num_tokens": 94485960.0,
      "step": 50680
    },
    {
      "entropy": 6.0983212947845455,
      "epoch": 4.475891910985517,
      "grad_norm": 1.25,
      "learning_rate": 0.0003152859585369201,
      "loss": 5.2866,
      "mean_token_accuracy": 0.2161903500556946,
      "num_tokens": 94495274.0,
      "step": 50685
    },
    {
      "entropy": 6.030444478988647,
      "epoch": 4.476333451077358,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031525497553236615,
      "loss": 5.2636,
      "mean_token_accuracy": 0.217272025346756,
      "num_tokens": 94504952.0,
      "step": 50690
    },
    {
      "entropy": 6.093646669387818,
      "epoch": 4.476774991169198,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003152239917392402,
      "loss": 5.2339,
      "mean_token_accuracy": 0.2233195796608925,
      "num_tokens": 94513795.0,
      "step": 50695
    },
    {
      "entropy": 5.9655442237854,
      "epoch": 4.477216531261038,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003151930071581491,
      "loss": 5.2329,
      "mean_token_accuracy": 0.2206583574414253,
      "num_tokens": 94522872.0,
      "step": 50700
    },
    {
      "entropy": 5.996033716201782,
      "epoch": 4.477658071352879,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003151620217896999,
      "loss": 5.2249,
      "mean_token_accuracy": 0.22795525789260865,
      "num_tokens": 94532510.0,
      "step": 50705
    },
    {
      "entropy": 6.064037799835205,
      "epoch": 4.478099611444719,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003151310356344995,
      "loss": 5.2698,
      "mean_token_accuracy": 0.21695705503225327,
      "num_tokens": 94541056.0,
      "step": 50710
    },
    {
      "entropy": 6.094281244277954,
      "epoch": 4.478541151536559,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00031510004869315494,
      "loss": 5.2598,
      "mean_token_accuracy": 0.22055668234825135,
      "num_tokens": 94550388.0,
      "step": 50715
    },
    {
      "entropy": 6.0654685497283936,
      "epoch": 4.4789826916284,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003150690609662732,
      "loss": 5.2838,
      "mean_token_accuracy": 0.21969703435897828,
      "num_tokens": 94559659.0,
      "step": 50720
    },
    {
      "entropy": 5.860719442367554,
      "epoch": 4.47942423172024,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003150380724544614,
      "loss": 5.0738,
      "mean_token_accuracy": 0.24388800859451293,
      "num_tokens": 94568381.0,
      "step": 50725
    },
    {
      "entropy": 6.0275835514068605,
      "epoch": 4.47986577181208,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003150070831583265,
      "loss": 5.2958,
      "mean_token_accuracy": 0.2188476115465164,
      "num_tokens": 94578224.0,
      "step": 50730
    },
    {
      "entropy": 6.0705150127410885,
      "epoch": 4.480307311903921,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00031497609307847555,
      "loss": 5.276,
      "mean_token_accuracy": 0.21553726941347123,
      "num_tokens": 94587594.0,
      "step": 50735
    },
    {
      "entropy": 5.95890965461731,
      "epoch": 4.480748851995761,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031494510221551574,
      "loss": 5.1676,
      "mean_token_accuracy": 0.2295295774936676,
      "num_tokens": 94597831.0,
      "step": 50740
    },
    {
      "entropy": 6.065330553054809,
      "epoch": 4.481190392087601,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031491411057005405,
      "loss": 5.3144,
      "mean_token_accuracy": 0.2175091028213501,
      "num_tokens": 94607867.0,
      "step": 50745
    },
    {
      "entropy": 6.086604309082031,
      "epoch": 4.481631932179442,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031488311814269773,
      "loss": 5.2546,
      "mean_token_accuracy": 0.21918573677539827,
      "num_tokens": 94617051.0,
      "step": 50750
    },
    {
      "entropy": 5.955223274230957,
      "epoch": 4.482073472271282,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003148521249340537,
      "loss": 5.0868,
      "mean_token_accuracy": 0.2437513366341591,
      "num_tokens": 94625462.0,
      "step": 50755
    },
    {
      "entropy": 5.917551326751709,
      "epoch": 4.482515012363122,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003148211309447292,
      "loss": 5.1736,
      "mean_token_accuracy": 0.2266245037317276,
      "num_tokens": 94634350.0,
      "step": 50760
    },
    {
      "entropy": 6.015249919891358,
      "epoch": 4.482956552454963,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00031479013617533144,
      "loss": 5.2599,
      "mean_token_accuracy": 0.22278062999248505,
      "num_tokens": 94643229.0,
      "step": 50765
    },
    {
      "entropy": 6.036093378067017,
      "epoch": 4.483398092546803,
      "grad_norm": 1.375,
      "learning_rate": 0.00031475914062646757,
      "loss": 5.2294,
      "mean_token_accuracy": 0.22646304965019226,
      "num_tokens": 94652833.0,
      "step": 50770
    },
    {
      "entropy": 6.039127779006958,
      "epoch": 4.483839632638643,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003147281442987447,
      "loss": 5.2492,
      "mean_token_accuracy": 0.2210848033428192,
      "num_tokens": 94661406.0,
      "step": 50775
    },
    {
      "entropy": 5.970139837265014,
      "epoch": 4.484281172730484,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003146971471927701,
      "loss": 5.2097,
      "mean_token_accuracy": 0.22183725684881211,
      "num_tokens": 94670434.0,
      "step": 50780
    },
    {
      "entropy": 6.096702957153321,
      "epoch": 4.484722712822324,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00031466614930915094,
      "loss": 5.3163,
      "mean_token_accuracy": 0.21243428587913513,
      "num_tokens": 94679182.0,
      "step": 50785
    },
    {
      "entropy": 6.10097370147705,
      "epoch": 4.485164252914164,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003146351506484945,
      "loss": 5.3152,
      "mean_token_accuracy": 0.21494316458702087,
      "num_tokens": 94688670.0,
      "step": 50790
    },
    {
      "entropy": 5.9924358367919925,
      "epoch": 4.485605793006005,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000314604151211408,
      "loss": 5.1794,
      "mean_token_accuracy": 0.23399427384138108,
      "num_tokens": 94697488.0,
      "step": 50795
    },
    {
      "entropy": 6.01255693435669,
      "epoch": 4.486047333097845,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031457315099849866,
      "loss": 5.2436,
      "mean_token_accuracy": 0.21621133685111998,
      "num_tokens": 94706925.0,
      "step": 50800
    },
    {
      "entropy": 6.095000886917115,
      "epoch": 4.4864888731896855,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031454215001037387,
      "loss": 5.3828,
      "mean_token_accuracy": 0.21078777611255645,
      "num_tokens": 94716639.0,
      "step": 50805
    },
    {
      "entropy": 6.031996583938598,
      "epoch": 4.4869304132815255,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00031451114824764085,
      "loss": 5.208,
      "mean_token_accuracy": 0.22277006208896638,
      "num_tokens": 94726302.0,
      "step": 50810
    },
    {
      "entropy": 6.061512756347656,
      "epoch": 4.4873719533733665,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00031448014571090684,
      "loss": 5.291,
      "mean_token_accuracy": 0.21828037947416307,
      "num_tokens": 94735997.0,
      "step": 50815
    },
    {
      "entropy": 6.038099002838135,
      "epoch": 4.4878134934652065,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003144491424007793,
      "loss": 5.2296,
      "mean_token_accuracy": 0.2312270388007164,
      "num_tokens": 94745533.0,
      "step": 50820
    },
    {
      "entropy": 6.094707584381103,
      "epoch": 4.488255033557047,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003144181383178654,
      "loss": 5.2755,
      "mean_token_accuracy": 0.22113850265741347,
      "num_tokens": 94755905.0,
      "step": 50825
    },
    {
      "entropy": 6.011229801177978,
      "epoch": 4.4886965736488875,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003143871334627726,
      "loss": 5.2997,
      "mean_token_accuracy": 0.21772476881742478,
      "num_tokens": 94765179.0,
      "step": 50830
    },
    {
      "entropy": 6.037772083282471,
      "epoch": 4.489138113740728,
      "grad_norm": 1.265625,
      "learning_rate": 0.00031435612783610835,
      "loss": 5.21,
      "mean_token_accuracy": 0.22218168377876282,
      "num_tokens": 94775559.0,
      "step": 50835
    },
    {
      "entropy": 6.035936546325684,
      "epoch": 4.489579653832568,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003143251214384799,
      "loss": 5.2279,
      "mean_token_accuracy": 0.22272408306598662,
      "num_tokens": 94784260.0,
      "step": 50840
    },
    {
      "entropy": 6.0337248802185055,
      "epoch": 4.490021193924409,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031429411427049466,
      "loss": 5.2501,
      "mean_token_accuracy": 0.2192383289337158,
      "num_tokens": 94794621.0,
      "step": 50845
    },
    {
      "entropy": 6.067106151580811,
      "epoch": 4.490462734016249,
      "grad_norm": 1.421875,
      "learning_rate": 0.00031426310633276004,
      "loss": 5.2537,
      "mean_token_accuracy": 0.21425219774246215,
      "num_tokens": 94802959.0,
      "step": 50850
    },
    {
      "entropy": 6.077314233779907,
      "epoch": 4.490904274108089,
      "grad_norm": 1.328125,
      "learning_rate": 0.00031423209762588357,
      "loss": 5.2871,
      "mean_token_accuracy": 0.21675409376621246,
      "num_tokens": 94811952.0,
      "step": 50855
    },
    {
      "entropy": 6.0525470733642575,
      "epoch": 4.49134581419993,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031420108815047266,
      "loss": 5.2957,
      "mean_token_accuracy": 0.21031274348497392,
      "num_tokens": 94821048.0,
      "step": 50860
    },
    {
      "entropy": 6.074713087081909,
      "epoch": 4.49178735429177,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003141700779071346,
      "loss": 5.2975,
      "mean_token_accuracy": 0.21916881054639817,
      "num_tokens": 94830491.0,
      "step": 50865
    },
    {
      "entropy": 6.01215238571167,
      "epoch": 4.49222889438361,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003141390668964771,
      "loss": 5.1776,
      "mean_token_accuracy": 0.23557818233966826,
      "num_tokens": 94839459.0,
      "step": 50870
    },
    {
      "entropy": 5.963986825942993,
      "epoch": 4.492670434475451,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003141080551191075,
      "loss": 5.2009,
      "mean_token_accuracy": 0.22314488291740417,
      "num_tokens": 94848287.0,
      "step": 50875
    },
    {
      "entropy": 6.0138777732849125,
      "epoch": 4.493111974567291,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003140770425756333,
      "loss": 5.2171,
      "mean_token_accuracy": 0.21825242787599564,
      "num_tokens": 94858100.0,
      "step": 50880
    },
    {
      "entropy": 6.081034421920776,
      "epoch": 4.493553514659131,
      "grad_norm": 1.21875,
      "learning_rate": 0.000314046029266662,
      "loss": 5.3064,
      "mean_token_accuracy": 0.21735843420028686,
      "num_tokens": 94868026.0,
      "step": 50885
    },
    {
      "entropy": 6.119735479354858,
      "epoch": 4.493995054750972,
      "grad_norm": 1.328125,
      "learning_rate": 0.00031401501519280134,
      "loss": 5.3259,
      "mean_token_accuracy": 0.21191925555467606,
      "num_tokens": 94876783.0,
      "step": 50890
    },
    {
      "entropy": 5.998760795593261,
      "epoch": 4.494436594842812,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031398400035465864,
      "loss": 5.2256,
      "mean_token_accuracy": 0.2237299770116806,
      "num_tokens": 94887234.0,
      "step": 50895
    },
    {
      "entropy": 6.048548889160156,
      "epoch": 4.494878134934652,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031395298475284155,
      "loss": 5.2608,
      "mean_token_accuracy": 0.22104394882917405,
      "num_tokens": 94896699.0,
      "step": 50900
    },
    {
      "entropy": 6.009073257446289,
      "epoch": 4.495319675026493,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003139219683879577,
      "loss": 5.3008,
      "mean_token_accuracy": 0.2181059777736664,
      "num_tokens": 94905974.0,
      "step": 50905
    },
    {
      "entropy": 6.051841974258423,
      "epoch": 4.495761215118333,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031389095126061454,
      "loss": 5.22,
      "mean_token_accuracy": 0.22317493408918382,
      "num_tokens": 94914890.0,
      "step": 50910
    },
    {
      "entropy": 6.059266757965088,
      "epoch": 4.496202755210173,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00031385993337141986,
      "loss": 5.2576,
      "mean_token_accuracy": 0.2169677272439003,
      "num_tokens": 94923798.0,
      "step": 50915
    },
    {
      "entropy": 6.053400754928589,
      "epoch": 4.496644295302014,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00031382891472098113,
      "loss": 5.256,
      "mean_token_accuracy": 0.22038267105817794,
      "num_tokens": 94931823.0,
      "step": 50920
    },
    {
      "entropy": 6.051079845428466,
      "epoch": 4.497085835393854,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003137978953099061,
      "loss": 5.2838,
      "mean_token_accuracy": 0.22126393914222717,
      "num_tokens": 94941017.0,
      "step": 50925
    },
    {
      "entropy": 6.038876152038574,
      "epoch": 4.497527375485694,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003137668751388023,
      "loss": 5.2353,
      "mean_token_accuracy": 0.2230350822210312,
      "num_tokens": 94950402.0,
      "step": 50930
    },
    {
      "entropy": 6.054045534133911,
      "epoch": 4.497968915577534,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003137358542082775,
      "loss": 5.2352,
      "mean_token_accuracy": 0.21310080140829085,
      "num_tokens": 94959623.0,
      "step": 50935
    },
    {
      "entropy": 6.027423954010009,
      "epoch": 4.498410455669375,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003137048325189393,
      "loss": 5.2471,
      "mean_token_accuracy": 0.219474758207798,
      "num_tokens": 94969284.0,
      "step": 50940
    },
    {
      "entropy": 5.997856330871582,
      "epoch": 4.498851995761215,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003136738100713954,
      "loss": 5.2013,
      "mean_token_accuracy": 0.22403197288513182,
      "num_tokens": 94978200.0,
      "step": 50945
    },
    {
      "entropy": 6.055574607849121,
      "epoch": 4.499293535853056,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003136427868662537,
      "loss": 5.253,
      "mean_token_accuracy": 0.21939591318368912,
      "num_tokens": 94987704.0,
      "step": 50950
    },
    {
      "entropy": 6.051585340499878,
      "epoch": 4.499735075944896,
      "grad_norm": 1.25,
      "learning_rate": 0.0003136117629041217,
      "loss": 5.1981,
      "mean_token_accuracy": 0.22255864441394807,
      "num_tokens": 94996858.0,
      "step": 50955
    },
    {
      "entropy": 5.971195840835572,
      "epoch": 4.500176616036736,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031358073818560725,
      "loss": 5.1892,
      "mean_token_accuracy": 0.2362303078174591,
      "num_tokens": 95005458.0,
      "step": 50960
    },
    {
      "entropy": 6.0468357563018795,
      "epoch": 4.500618156128576,
      "grad_norm": 1.453125,
      "learning_rate": 0.000313549712711318,
      "loss": 5.2159,
      "mean_token_accuracy": 0.2280054807662964,
      "num_tokens": 95014363.0,
      "step": 50965
    },
    {
      "entropy": 6.010324001312256,
      "epoch": 4.501059696220417,
      "grad_norm": 1.421875,
      "learning_rate": 0.00031351868648186196,
      "loss": 5.2064,
      "mean_token_accuracy": 0.22691629081964493,
      "num_tokens": 95023501.0,
      "step": 50970
    },
    {
      "entropy": 5.949630928039551,
      "epoch": 4.501501236312257,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003134876594978466,
      "loss": 5.1711,
      "mean_token_accuracy": 0.2282029166817665,
      "num_tokens": 95033274.0,
      "step": 50975
    },
    {
      "entropy": 6.02598934173584,
      "epoch": 4.501942776404097,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031345663175987997,
      "loss": 5.1636,
      "mean_token_accuracy": 0.22969516068696977,
      "num_tokens": 95042963.0,
      "step": 50980
    },
    {
      "entropy": 6.022670316696167,
      "epoch": 4.502384316495938,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003134256032685698,
      "loss": 5.1624,
      "mean_token_accuracy": 0.2265353813767433,
      "num_tokens": 95051468.0,
      "step": 50985
    },
    {
      "entropy": 5.943066596984863,
      "epoch": 4.502825856587778,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031339457402452387,
      "loss": 5.1389,
      "mean_token_accuracy": 0.2264108330011368,
      "num_tokens": 95060388.0,
      "step": 50990
    },
    {
      "entropy": 6.046057891845703,
      "epoch": 4.503267396679618,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003133635440283501,
      "loss": 5.2319,
      "mean_token_accuracy": 0.22257317453622819,
      "num_tokens": 95070134.0,
      "step": 50995
    },
    {
      "entropy": 6.017930221557617,
      "epoch": 4.503708936771459,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003133325132806563,
      "loss": 5.2246,
      "mean_token_accuracy": 0.22713582813739777,
      "num_tokens": 95080160.0,
      "step": 51000
    },
    {
      "epoch": 4.503708936771459,
      "eval_entropy": 5.7126907095363695,
      "eval_loss": 5.314966678619385,
      "eval_mean_token_accuracy": 0.2258868337116317,
      "eval_num_tokens": 95080160.0,
      "eval_runtime": 20.2147,
      "eval_samples_per_second": 1746.995,
      "eval_steps_per_second": 218.405,
      "step": 51000
    },
    {
      "entropy": 6.03386812210083,
      "epoch": 4.504150476863299,
      "grad_norm": 1.375,
      "learning_rate": 0.0003133014817820504,
      "loss": 5.2423,
      "mean_token_accuracy": 0.21878099143505098,
      "num_tokens": 95089210.0,
      "step": 51005
    },
    {
      "entropy": 5.988959932327271,
      "epoch": 4.504592016955139,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031327044953314014,
      "loss": 5.2487,
      "mean_token_accuracy": 0.22018161565065383,
      "num_tokens": 95098865.0,
      "step": 51010
    },
    {
      "entropy": 6.0722212314605715,
      "epoch": 4.50503355704698,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003132394165345337,
      "loss": 5.3,
      "mean_token_accuracy": 0.20994994640350342,
      "num_tokens": 95108083.0,
      "step": 51015
    },
    {
      "entropy": 6.114263200759888,
      "epoch": 4.50547509713882,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003132083827868387,
      "loss": 5.234,
      "mean_token_accuracy": 0.22347587198019028,
      "num_tokens": 95117088.0,
      "step": 51020
    },
    {
      "entropy": 6.061977243423462,
      "epoch": 4.50591663723066,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003131773482906633,
      "loss": 5.3215,
      "mean_token_accuracy": 0.2146270290017128,
      "num_tokens": 95126019.0,
      "step": 51025
    },
    {
      "entropy": 6.075354528427124,
      "epoch": 4.506358177322501,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031314631304661536,
      "loss": 5.2497,
      "mean_token_accuracy": 0.22158991992473603,
      "num_tokens": 95135950.0,
      "step": 51030
    },
    {
      "entropy": 6.012694263458252,
      "epoch": 4.506799717414341,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003131152770553028,
      "loss": 5.2681,
      "mean_token_accuracy": 0.22254655808210372,
      "num_tokens": 95145580.0,
      "step": 51035
    },
    {
      "entropy": 6.01888747215271,
      "epoch": 4.507241257506181,
      "grad_norm": 1.546875,
      "learning_rate": 0.0003130842403173337,
      "loss": 5.2096,
      "mean_token_accuracy": 0.22368322908878327,
      "num_tokens": 95154522.0,
      "step": 51040
    },
    {
      "entropy": 5.94607982635498,
      "epoch": 4.507682797598022,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031305320283331594,
      "loss": 5.0848,
      "mean_token_accuracy": 0.24273648262023925,
      "num_tokens": 95163485.0,
      "step": 51045
    },
    {
      "entropy": 6.03512134552002,
      "epoch": 4.508124337689862,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031302216460385755,
      "loss": 5.3002,
      "mean_token_accuracy": 0.22461497634649277,
      "num_tokens": 95173425.0,
      "step": 51050
    },
    {
      "entropy": 6.014825439453125,
      "epoch": 4.508565877781702,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003129911256295667,
      "loss": 5.2967,
      "mean_token_accuracy": 0.2147938147187233,
      "num_tokens": 95182291.0,
      "step": 51055
    },
    {
      "entropy": 6.0000364780426025,
      "epoch": 4.509007417873542,
      "grad_norm": 1.25,
      "learning_rate": 0.0003129600859110512,
      "loss": 5.172,
      "mean_token_accuracy": 0.22129350900650024,
      "num_tokens": 95190999.0,
      "step": 51060
    },
    {
      "entropy": 6.042505121231079,
      "epoch": 4.509448957965383,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003129290454489192,
      "loss": 5.3074,
      "mean_token_accuracy": 0.22426247596740723,
      "num_tokens": 95200317.0,
      "step": 51065
    },
    {
      "entropy": 6.057543563842773,
      "epoch": 4.509890498057223,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003128980042437788,
      "loss": 5.3189,
      "mean_token_accuracy": 0.21663225144147874,
      "num_tokens": 95210739.0,
      "step": 51070
    },
    {
      "entropy": 6.024671936035157,
      "epoch": 4.510332038149064,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031286696229623805,
      "loss": 5.1787,
      "mean_token_accuracy": 0.22878261804580688,
      "num_tokens": 95220574.0,
      "step": 51075
    },
    {
      "entropy": 6.0493348121643065,
      "epoch": 4.510773578240904,
      "grad_norm": 1.25,
      "learning_rate": 0.0003128359196069051,
      "loss": 5.2353,
      "mean_token_accuracy": 0.2249183788895607,
      "num_tokens": 95229725.0,
      "step": 51080
    },
    {
      "entropy": 6.025968837738037,
      "epoch": 4.5112151183327445,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003128048761763879,
      "loss": 5.2091,
      "mean_token_accuracy": 0.2225511610507965,
      "num_tokens": 95239602.0,
      "step": 51085
    },
    {
      "entropy": 6.028149843215942,
      "epoch": 4.5116566584245845,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00031277383200529476,
      "loss": 5.157,
      "mean_token_accuracy": 0.2241255447268486,
      "num_tokens": 95248074.0,
      "step": 51090
    },
    {
      "entropy": 6.010294818878174,
      "epoch": 4.5120981985164255,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00031274278709423367,
      "loss": 5.2664,
      "mean_token_accuracy": 0.22165100425481796,
      "num_tokens": 95257173.0,
      "step": 51095
    },
    {
      "entropy": 6.1331761360168455,
      "epoch": 4.5125397386082655,
      "grad_norm": 1.328125,
      "learning_rate": 0.00031271174144381294,
      "loss": 5.3642,
      "mean_token_accuracy": 0.20805806815624237,
      "num_tokens": 95266822.0,
      "step": 51100
    },
    {
      "entropy": 6.100857925415039,
      "epoch": 4.5129812787001065,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00031268069505464055,
      "loss": 5.3046,
      "mean_token_accuracy": 0.2193200945854187,
      "num_tokens": 95276508.0,
      "step": 51105
    },
    {
      "entropy": 6.086715650558472,
      "epoch": 4.5134228187919465,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003126496479273248,
      "loss": 5.2668,
      "mean_token_accuracy": 0.22423662543296813,
      "num_tokens": 95286018.0,
      "step": 51110
    },
    {
      "entropy": 6.043356084823609,
      "epoch": 4.513864358883787,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031261860006247385,
      "loss": 5.2687,
      "mean_token_accuracy": 0.2170005112886429,
      "num_tokens": 95295058.0,
      "step": 51115
    },
    {
      "entropy": 5.918076705932617,
      "epoch": 4.514305898975627,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000312587551460696,
      "loss": 5.0465,
      "mean_token_accuracy": 0.24919629693031312,
      "num_tokens": 95303931.0,
      "step": 51120
    },
    {
      "entropy": 5.955110120773315,
      "epoch": 4.514747439067468,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003125565021225992,
      "loss": 5.1476,
      "mean_token_accuracy": 0.2261263072490692,
      "num_tokens": 95313448.0,
      "step": 51125
    },
    {
      "entropy": 6.022297620773315,
      "epoch": 4.515188979159308,
      "grad_norm": 1.25,
      "learning_rate": 0.0003125254520487921,
      "loss": 5.264,
      "mean_token_accuracy": 0.21666880697011948,
      "num_tokens": 95322548.0,
      "step": 51130
    },
    {
      "entropy": 6.0618644714355465,
      "epoch": 4.515630519251148,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00031249440123988265,
      "loss": 5.2774,
      "mean_token_accuracy": 0.22221512347459793,
      "num_tokens": 95331833.0,
      "step": 51135
    },
    {
      "entropy": 6.0693717956542965,
      "epoch": 4.516072059342989,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00031246334969647915,
      "loss": 5.2778,
      "mean_token_accuracy": 0.22347034215927125,
      "num_tokens": 95341247.0,
      "step": 51140
    },
    {
      "entropy": 6.096156978607178,
      "epoch": 4.516513599434829,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031243229741919,
      "loss": 5.3848,
      "mean_token_accuracy": 0.20926984697580336,
      "num_tokens": 95352157.0,
      "step": 51145
    },
    {
      "entropy": 6.051904821395874,
      "epoch": 4.516955139526669,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003124012444086234,
      "loss": 5.2823,
      "mean_token_accuracy": 0.2237311363220215,
      "num_tokens": 95361523.0,
      "step": 51150
    },
    {
      "entropy": 6.056984281539917,
      "epoch": 4.51739667961851,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031237019066538774,
      "loss": 5.2743,
      "mean_token_accuracy": 0.22359003573656083,
      "num_tokens": 95371179.0,
      "step": 51155
    },
    {
      "entropy": 5.993194150924682,
      "epoch": 4.51783821971035,
      "grad_norm": 1.421875,
      "learning_rate": 0.00031233913619009123,
      "loss": 5.2084,
      "mean_token_accuracy": 0.22302078753709792,
      "num_tokens": 95380382.0,
      "step": 51160
    },
    {
      "entropy": 5.9980583667755125,
      "epoch": 4.51827975980219,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003123080809833424,
      "loss": 5.19,
      "mean_token_accuracy": 0.22509008049964904,
      "num_tokens": 95390116.0,
      "step": 51165
    },
    {
      "entropy": 6.031252098083496,
      "epoch": 4.518721299894031,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003122770250457494,
      "loss": 5.1883,
      "mean_token_accuracy": 0.2280188739299774,
      "num_tokens": 95399481.0,
      "step": 51170
    },
    {
      "entropy": 6.106078624725342,
      "epoch": 4.519162839985871,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003122459683779206,
      "loss": 5.3392,
      "mean_token_accuracy": 0.2111035019159317,
      "num_tokens": 95409150.0,
      "step": 51175
    },
    {
      "entropy": 5.98978009223938,
      "epoch": 4.519604380077711,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003122149109804646,
      "loss": 5.1644,
      "mean_token_accuracy": 0.2255371704697609,
      "num_tokens": 95418804.0,
      "step": 51180
    },
    {
      "entropy": 6.061920022964477,
      "epoch": 4.520045920169552,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00031218385285398964,
      "loss": 5.2533,
      "mean_token_accuracy": 0.21818395853042602,
      "num_tokens": 95428061.0,
      "step": 51185
    },
    {
      "entropy": 6.134359407424927,
      "epoch": 4.520487460261392,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00031215279399910404,
      "loss": 5.323,
      "mean_token_accuracy": 0.21382232010364532,
      "num_tokens": 95437141.0,
      "step": 51190
    },
    {
      "entropy": 6.081946611404419,
      "epoch": 4.520929000353232,
      "grad_norm": 1.34375,
      "learning_rate": 0.00031212173441641646,
      "loss": 5.2844,
      "mean_token_accuracy": 0.22017906606197357,
      "num_tokens": 95446455.0,
      "step": 51195
    },
    {
      "entropy": 6.075165557861328,
      "epoch": 4.521370540445073,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003120906741065352,
      "loss": 5.3152,
      "mean_token_accuracy": 0.21878361701965332,
      "num_tokens": 95456695.0,
      "step": 51200
    },
    {
      "entropy": 5.97211651802063,
      "epoch": 4.521812080536913,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031205961307006856,
      "loss": 5.2592,
      "mean_token_accuracy": 0.2178822472691536,
      "num_tokens": 95466061.0,
      "step": 51205
    },
    {
      "entropy": 6.021544027328491,
      "epoch": 4.522253620628753,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031202855130762537,
      "loss": 5.2115,
      "mean_token_accuracy": 0.21503111273050307,
      "num_tokens": 95474821.0,
      "step": 51210
    },
    {
      "entropy": 5.967187023162841,
      "epoch": 4.522695160720593,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031199748881981384,
      "loss": 5.1976,
      "mean_token_accuracy": 0.23246073275804519,
      "num_tokens": 95484931.0,
      "step": 51215
    },
    {
      "entropy": 6.003154373168945,
      "epoch": 4.523136700812434,
      "grad_norm": 1.40625,
      "learning_rate": 0.00031196642560724265,
      "loss": 5.2209,
      "mean_token_accuracy": 0.23777785897254944,
      "num_tokens": 95493167.0,
      "step": 51220
    },
    {
      "entropy": 6.071313190460205,
      "epoch": 4.523578240904274,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00031193536167052003,
      "loss": 5.2638,
      "mean_token_accuracy": 0.22077415585517884,
      "num_tokens": 95502593.0,
      "step": 51225
    },
    {
      "entropy": 5.956775712966919,
      "epoch": 4.524019780996115,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003119042970102549,
      "loss": 5.0876,
      "mean_token_accuracy": 0.24114348441362382,
      "num_tokens": 95511317.0,
      "step": 51230
    },
    {
      "entropy": 6.020994186401367,
      "epoch": 4.524461321087955,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003118732316270553,
      "loss": 5.2316,
      "mean_token_accuracy": 0.21932052522897721,
      "num_tokens": 95520242.0,
      "step": 51235
    },
    {
      "entropy": 6.064249181747437,
      "epoch": 4.524902861179795,
      "grad_norm": 1.375,
      "learning_rate": 0.0003118421655215303,
      "loss": 5.2872,
      "mean_token_accuracy": 0.2158590540289879,
      "num_tokens": 95529415.0,
      "step": 51240
    },
    {
      "entropy": 6.0222780227661135,
      "epoch": 4.525344401271635,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003118110986942881,
      "loss": 5.2218,
      "mean_token_accuracy": 0.2245274230837822,
      "num_tokens": 95538856.0,
      "step": 51245
    },
    {
      "entropy": 6.146059989929199,
      "epoch": 4.525785941363476,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003117800311459375,
      "loss": 5.3893,
      "mean_token_accuracy": 0.20983501225709916,
      "num_tokens": 95548553.0,
      "step": 51250
    },
    {
      "entropy": 6.140642213821411,
      "epoch": 4.526227481455316,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003117489628770869,
      "loss": 5.2689,
      "mean_token_accuracy": 0.21419315189123153,
      "num_tokens": 95558495.0,
      "step": 51255
    },
    {
      "entropy": 6.048066425323486,
      "epoch": 4.526669021547156,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003117178938883452,
      "loss": 5.2411,
      "mean_token_accuracy": 0.2187207281589508,
      "num_tokens": 95567747.0,
      "step": 51260
    },
    {
      "entropy": 6.039315843582154,
      "epoch": 4.527110561638997,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003116868241803207,
      "loss": 5.2469,
      "mean_token_accuracy": 0.22519977241754532,
      "num_tokens": 95576658.0,
      "step": 51265
    },
    {
      "entropy": 6.053203582763672,
      "epoch": 4.527552101730837,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031165575375362216,
      "loss": 5.1888,
      "mean_token_accuracy": 0.21926985383033754,
      "num_tokens": 95585589.0,
      "step": 51270
    },
    {
      "entropy": 6.092959356307984,
      "epoch": 4.527993641822677,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003116246826088583,
      "loss": 5.2942,
      "mean_token_accuracy": 0.21417630165815355,
      "num_tokens": 95594460.0,
      "step": 51275
    },
    {
      "entropy": 6.037305498123169,
      "epoch": 4.528435181914518,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031159361074663765,
      "loss": 5.2145,
      "mean_token_accuracy": 0.22339971661567687,
      "num_tokens": 95603343.0,
      "step": 51280
    },
    {
      "entropy": 5.940203285217285,
      "epoch": 4.528876722006358,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00031156253816756914,
      "loss": 5.1826,
      "mean_token_accuracy": 0.23202226161956788,
      "num_tokens": 95613218.0,
      "step": 51285
    },
    {
      "entropy": 5.972400236129761,
      "epoch": 4.529318262098198,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003115314648722612,
      "loss": 5.1708,
      "mean_token_accuracy": 0.2268904522061348,
      "num_tokens": 95622287.0,
      "step": 51290
    },
    {
      "entropy": 6.0720690250396725,
      "epoch": 4.529759802190039,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031150039086132273,
      "loss": 5.336,
      "mean_token_accuracy": 0.20936258733272553,
      "num_tokens": 95632286.0,
      "step": 51295
    },
    {
      "entropy": 6.060075998306274,
      "epoch": 4.530201342281879,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00031146931613536223,
      "loss": 5.2572,
      "mean_token_accuracy": 0.2244809940457344,
      "num_tokens": 95642577.0,
      "step": 51300
    },
    {
      "entropy": 6.042535495758057,
      "epoch": 4.530642882373719,
      "grad_norm": 1.359375,
      "learning_rate": 0.00031143824069498866,
      "loss": 5.2518,
      "mean_token_accuracy": 0.21618098765611649,
      "num_tokens": 95652658.0,
      "step": 51305
    },
    {
      "entropy": 5.950520038604736,
      "epoch": 4.53108442246556,
      "grad_norm": 1.265625,
      "learning_rate": 0.00031140716454081065,
      "loss": 5.1277,
      "mean_token_accuracy": 0.22808357775211335,
      "num_tokens": 95662710.0,
      "step": 51310
    },
    {
      "entropy": 6.057318115234375,
      "epoch": 4.5315259625574,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000311376087673437,
      "loss": 5.2395,
      "mean_token_accuracy": 0.2208109438419342,
      "num_tokens": 95673571.0,
      "step": 51315
    },
    {
      "entropy": 6.047490453720092,
      "epoch": 4.53196750264924,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003113450100934765,
      "loss": 5.2453,
      "mean_token_accuracy": 0.21838372200727463,
      "num_tokens": 95682581.0,
      "step": 51320
    },
    {
      "entropy": 6.0615462303161625,
      "epoch": 4.532409042741081,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003113139318015379,
      "loss": 5.326,
      "mean_token_accuracy": 0.21115211993455887,
      "num_tokens": 95692697.0,
      "step": 51325
    },
    {
      "entropy": 6.040004396438599,
      "epoch": 4.532850582832921,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031128285279823,
      "loss": 5.239,
      "mean_token_accuracy": 0.22195335179567338,
      "num_tokens": 95701951.0,
      "step": 51330
    },
    {
      "entropy": 6.093921232223511,
      "epoch": 4.533292122924761,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00031125177308416165,
      "loss": 5.2732,
      "mean_token_accuracy": 0.21829540580511092,
      "num_tokens": 95711297.0,
      "step": 51335
    },
    {
      "entropy": 6.017175626754761,
      "epoch": 4.533733663016601,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003112206926599417,
      "loss": 5.2355,
      "mean_token_accuracy": 0.22334630489349366,
      "num_tokens": 95720627.0,
      "step": 51340
    },
    {
      "entropy": 6.001351928710937,
      "epoch": 4.534175203108442,
      "grad_norm": 1.3125,
      "learning_rate": 0.000311189611526179,
      "loss": 5.2065,
      "mean_token_accuracy": 0.2276992216706276,
      "num_tokens": 95729401.0,
      "step": 51345
    },
    {
      "entropy": 6.073207902908325,
      "epoch": 4.534616743200282,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031115852968348237,
      "loss": 5.2426,
      "mean_token_accuracy": 0.2187807857990265,
      "num_tokens": 95739146.0,
      "step": 51350
    },
    {
      "entropy": 6.073341178894043,
      "epoch": 4.535058283292123,
      "grad_norm": 1.359375,
      "learning_rate": 0.00031112744713246074,
      "loss": 5.3116,
      "mean_token_accuracy": 0.22028181254863738,
      "num_tokens": 95749279.0,
      "step": 51355
    },
    {
      "entropy": 6.035161066055298,
      "epoch": 4.535499823383963,
      "grad_norm": 1.5,
      "learning_rate": 0.00031109636387372295,
      "loss": 5.1342,
      "mean_token_accuracy": 0.2362685337662697,
      "num_tokens": 95758277.0,
      "step": 51360
    },
    {
      "entropy": 6.012306213378906,
      "epoch": 4.5359413634758035,
      "grad_norm": 1.34375,
      "learning_rate": 0.00031106527990787787,
      "loss": 5.2692,
      "mean_token_accuracy": 0.21636745929718018,
      "num_tokens": 95767914.0,
      "step": 51365
    },
    {
      "entropy": 5.924245023727417,
      "epoch": 4.5363829035676435,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00031103419523553457,
      "loss": 5.1669,
      "mean_token_accuracy": 0.23140785247087478,
      "num_tokens": 95777253.0,
      "step": 51370
    },
    {
      "entropy": 6.087069892883301,
      "epoch": 4.5368244436594845,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0003110031098573017,
      "loss": 5.3066,
      "mean_token_accuracy": 0.21800709515810013,
      "num_tokens": 95787533.0,
      "step": 51375
    },
    {
      "entropy": 6.065097713470459,
      "epoch": 4.5372659837513245,
      "grad_norm": 1.328125,
      "learning_rate": 0.00031097202377378856,
      "loss": 5.2443,
      "mean_token_accuracy": 0.21866924464702606,
      "num_tokens": 95796533.0,
      "step": 51380
    },
    {
      "entropy": 6.084074401855469,
      "epoch": 4.5377075238431654,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003109409369856039,
      "loss": 5.2738,
      "mean_token_accuracy": 0.22273004204034805,
      "num_tokens": 95806902.0,
      "step": 51385
    },
    {
      "entropy": 6.005260896682739,
      "epoch": 4.5381490639350055,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003109098494933568,
      "loss": 5.2404,
      "mean_token_accuracy": 0.21948911994695663,
      "num_tokens": 95816386.0,
      "step": 51390
    },
    {
      "entropy": 6.078106212615967,
      "epoch": 4.5385906040268456,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031087876129765605,
      "loss": 5.2008,
      "mean_token_accuracy": 0.22602997124195098,
      "num_tokens": 95825503.0,
      "step": 51395
    },
    {
      "entropy": 6.006036615371704,
      "epoch": 4.539032144118686,
      "grad_norm": 1.328125,
      "learning_rate": 0.00031084767239911073,
      "loss": 5.243,
      "mean_token_accuracy": 0.2244933009147644,
      "num_tokens": 95834429.0,
      "step": 51400
    },
    {
      "entropy": 5.940882587432862,
      "epoch": 4.5394736842105265,
      "grad_norm": 1.25,
      "learning_rate": 0.00031081658279833,
      "loss": 5.1537,
      "mean_token_accuracy": 0.24001335799694062,
      "num_tokens": 95844766.0,
      "step": 51405
    },
    {
      "entropy": 6.060229396820068,
      "epoch": 4.539915224302367,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003107854924959227,
      "loss": 5.2028,
      "mean_token_accuracy": 0.2196443870663643,
      "num_tokens": 95854335.0,
      "step": 51410
    },
    {
      "entropy": 6.030467081069946,
      "epoch": 4.540356764394207,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000310754401492498,
      "loss": 5.1929,
      "mean_token_accuracy": 0.2215151533484459,
      "num_tokens": 95863424.0,
      "step": 51415
    },
    {
      "entropy": 5.967924880981445,
      "epoch": 4.540798304486048,
      "grad_norm": 1.375,
      "learning_rate": 0.0003107233097886649,
      "loss": 5.1553,
      "mean_token_accuracy": 0.22176275104284288,
      "num_tokens": 95871945.0,
      "step": 51420
    },
    {
      "entropy": 6.035799169540406,
      "epoch": 4.541239844577888,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003106922173850325,
      "loss": 5.2274,
      "mean_token_accuracy": 0.22363280206918718,
      "num_tokens": 95881000.0,
      "step": 51425
    },
    {
      "entropy": 6.062141418457031,
      "epoch": 4.541681384669728,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00031066112428220985,
      "loss": 5.3202,
      "mean_token_accuracy": 0.21699631214141846,
      "num_tokens": 95890286.0,
      "step": 51430
    },
    {
      "entropy": 5.987793254852295,
      "epoch": 4.542122924761569,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003106300304808061,
      "loss": 5.1191,
      "mean_token_accuracy": 0.2342773973941803,
      "num_tokens": 95900008.0,
      "step": 51435
    },
    {
      "entropy": 6.00024824142456,
      "epoch": 4.542564464853409,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031059893598143026,
      "loss": 5.1812,
      "mean_token_accuracy": 0.2282746210694313,
      "num_tokens": 95909389.0,
      "step": 51440
    },
    {
      "entropy": 5.979001045227051,
      "epoch": 4.543006004945249,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003105678407846915,
      "loss": 5.1411,
      "mean_token_accuracy": 0.23343109488487243,
      "num_tokens": 95919035.0,
      "step": 51445
    },
    {
      "entropy": 6.025582647323608,
      "epoch": 4.54344754503709,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000310536744891199,
      "loss": 5.179,
      "mean_token_accuracy": 0.2208606034517288,
      "num_tokens": 95928114.0,
      "step": 51450
    },
    {
      "entropy": 5.984578561782837,
      "epoch": 4.54388908512893,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003105056483015619,
      "loss": 5.2103,
      "mean_token_accuracy": 0.2267756164073944,
      "num_tokens": 95936904.0,
      "step": 51455
    },
    {
      "entropy": 5.997446918487549,
      "epoch": 4.54433062522077,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003104745510163893,
      "loss": 5.1988,
      "mean_token_accuracy": 0.21989749372005463,
      "num_tokens": 95945823.0,
      "step": 51460
    },
    {
      "entropy": 6.029147434234619,
      "epoch": 4.544772165312611,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003104434530362903,
      "loss": 5.2052,
      "mean_token_accuracy": 0.22215196639299392,
      "num_tokens": 95954772.0,
      "step": 51465
    },
    {
      "entropy": 6.046457004547119,
      "epoch": 4.545213705404451,
      "grad_norm": 1.375,
      "learning_rate": 0.0003104123543618744,
      "loss": 5.3091,
      "mean_token_accuracy": 0.2141388550400734,
      "num_tokens": 95963057.0,
      "step": 51470
    },
    {
      "entropy": 6.114950323104859,
      "epoch": 4.545655245496291,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003103812549937506,
      "loss": 5.2752,
      "mean_token_accuracy": 0.21934564262628556,
      "num_tokens": 95972161.0,
      "step": 51475
    },
    {
      "entropy": 6.096015930175781,
      "epoch": 4.546096785588132,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003103501549325281,
      "loss": 5.2901,
      "mean_token_accuracy": 0.21314697712659836,
      "num_tokens": 95982428.0,
      "step": 51480
    },
    {
      "entropy": 6.021536016464234,
      "epoch": 4.546538325679972,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003103190541788162,
      "loss": 5.2682,
      "mean_token_accuracy": 0.2280149281024933,
      "num_tokens": 95991964.0,
      "step": 51485
    },
    {
      "entropy": 6.048902940750122,
      "epoch": 4.546979865771812,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003102879527332241,
      "loss": 5.1577,
      "mean_token_accuracy": 0.22652845531702043,
      "num_tokens": 96000714.0,
      "step": 51490
    },
    {
      "entropy": 6.0202240467071535,
      "epoch": 4.547421405863652,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000310256850596361,
      "loss": 5.297,
      "mean_token_accuracy": 0.21061520725488664,
      "num_tokens": 96010256.0,
      "step": 51495
    },
    {
      "entropy": 6.0804849624633786,
      "epoch": 4.547862945955493,
      "grad_norm": 1.375,
      "learning_rate": 0.00031022574776883634,
      "loss": 5.2655,
      "mean_token_accuracy": 0.21589360535144805,
      "num_tokens": 96019655.0,
      "step": 51500
    },
    {
      "entropy": 6.040716314315796,
      "epoch": 4.548304486047333,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003101946442512593,
      "loss": 5.2139,
      "mean_token_accuracy": 0.23274083584547042,
      "num_tokens": 96028896.0,
      "step": 51505
    },
    {
      "entropy": 6.0301892280578615,
      "epoch": 4.548746026139174,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031016354004423916,
      "loss": 5.2359,
      "mean_token_accuracy": 0.22368062287569046,
      "num_tokens": 96039760.0,
      "step": 51510
    },
    {
      "entropy": 6.042342710494995,
      "epoch": 4.549187566231014,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003101324351483853,
      "loss": 5.2069,
      "mean_token_accuracy": 0.23064843118190764,
      "num_tokens": 96048240.0,
      "step": 51515
    },
    {
      "entropy": 6.051011943817139,
      "epoch": 4.549629106322854,
      "grad_norm": 1.375,
      "learning_rate": 0.0003101013295643069,
      "loss": 5.2205,
      "mean_token_accuracy": 0.22837756872177123,
      "num_tokens": 96057004.0,
      "step": 51520
    },
    {
      "entropy": 5.979472827911377,
      "epoch": 4.550070646414694,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003100702232926136,
      "loss": 5.2633,
      "mean_token_accuracy": 0.22446771562099457,
      "num_tokens": 96066679.0,
      "step": 51525
    },
    {
      "entropy": 5.981230020523071,
      "epoch": 4.550512186506535,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00031003911633391447,
      "loss": 5.1227,
      "mean_token_accuracy": 0.23293748795986174,
      "num_tokens": 96076207.0,
      "step": 51530
    },
    {
      "entropy": 6.085381698608399,
      "epoch": 4.550953726598375,
      "grad_norm": 1.34375,
      "learning_rate": 0.000310008008688819,
      "loss": 5.373,
      "mean_token_accuracy": 0.20220860987901687,
      "num_tokens": 96085237.0,
      "step": 51535
    },
    {
      "entropy": 6.100530958175659,
      "epoch": 4.551395266690215,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030997690035793654,
      "loss": 5.2913,
      "mean_token_accuracy": 0.2192571997642517,
      "num_tokens": 96095758.0,
      "step": 51540
    },
    {
      "entropy": 6.082007980346679,
      "epoch": 4.551836806782056,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003099457913418766,
      "loss": 5.2675,
      "mean_token_accuracy": 0.22168381363153458,
      "num_tokens": 96104510.0,
      "step": 51545
    },
    {
      "entropy": 6.113527774810791,
      "epoch": 4.552278346873896,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030991468164124833,
      "loss": 5.2642,
      "mean_token_accuracy": 0.22165673673152925,
      "num_tokens": 96115208.0,
      "step": 51550
    },
    {
      "entropy": 6.063199520111084,
      "epoch": 4.552719886965736,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003098835712566614,
      "loss": 5.326,
      "mean_token_accuracy": 0.21610725820064544,
      "num_tokens": 96125114.0,
      "step": 51555
    },
    {
      "entropy": 6.00048599243164,
      "epoch": 4.553161427057577,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030985246018872507,
      "loss": 5.2777,
      "mean_token_accuracy": 0.22516521215438842,
      "num_tokens": 96134389.0,
      "step": 51560
    },
    {
      "entropy": 6.033515357971192,
      "epoch": 4.553602967149417,
      "grad_norm": 1.375,
      "learning_rate": 0.0003098213484380489,
      "loss": 5.2536,
      "mean_token_accuracy": 0.21874288022518157,
      "num_tokens": 96143252.0,
      "step": 51565
    },
    {
      "entropy": 6.0230693340301515,
      "epoch": 4.554044507241257,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00030979023600524237,
      "loss": 5.1739,
      "mean_token_accuracy": 0.23311991840600968,
      "num_tokens": 96151969.0,
      "step": 51570
    },
    {
      "entropy": 6.045546388626098,
      "epoch": 4.554486047333098,
      "grad_norm": 1.25,
      "learning_rate": 0.00030975912289091484,
      "loss": 5.2707,
      "mean_token_accuracy": 0.22247494459152223,
      "num_tokens": 96162778.0,
      "step": 51575
    },
    {
      "entropy": 6.021856117248535,
      "epoch": 4.554927587424938,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003097280090956758,
      "loss": 5.2669,
      "mean_token_accuracy": 0.2206105411052704,
      "num_tokens": 96171589.0,
      "step": 51580
    },
    {
      "entropy": 6.060900020599365,
      "epoch": 4.555369127516778,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0003096968946201349,
      "loss": 5.2116,
      "mean_token_accuracy": 0.22943131029605865,
      "num_tokens": 96181410.0,
      "step": 51585
    },
    {
      "entropy": 6.118806505203247,
      "epoch": 4.555810667608619,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003096657794649016,
      "loss": 5.3549,
      "mean_token_accuracy": 0.20632321536540985,
      "num_tokens": 96191546.0,
      "step": 51590
    },
    {
      "entropy": 6.02675232887268,
      "epoch": 4.556252207700459,
      "grad_norm": 1.4375,
      "learning_rate": 0.00030963466363058536,
      "loss": 5.2074,
      "mean_token_accuracy": 0.23311295062303544,
      "num_tokens": 96200585.0,
      "step": 51595
    },
    {
      "entropy": 6.03455171585083,
      "epoch": 4.556693747792299,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030960354711779577,
      "loss": 5.2751,
      "mean_token_accuracy": 0.21558839082717896,
      "num_tokens": 96209714.0,
      "step": 51600
    },
    {
      "entropy": 6.028346586227417,
      "epoch": 4.55713528788414,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003095724299271423,
      "loss": 5.1834,
      "mean_token_accuracy": 0.2270253285765648,
      "num_tokens": 96218382.0,
      "step": 51605
    },
    {
      "entropy": 6.120908641815186,
      "epoch": 4.55757682797598,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003095413120592347,
      "loss": 5.3649,
      "mean_token_accuracy": 0.21066599935293198,
      "num_tokens": 96228582.0,
      "step": 51610
    },
    {
      "entropy": 6.0848828792572025,
      "epoch": 4.55801836806782,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030951019351468234,
      "loss": 5.24,
      "mean_token_accuracy": 0.21738502532243728,
      "num_tokens": 96238142.0,
      "step": 51615
    },
    {
      "entropy": 6.0689950466156,
      "epoch": 4.55845990815966,
      "grad_norm": 1.203125,
      "learning_rate": 0.000309479074294095,
      "loss": 5.2891,
      "mean_token_accuracy": 0.21761548519134521,
      "num_tokens": 96247805.0,
      "step": 51620
    },
    {
      "entropy": 6.0773460388183596,
      "epoch": 4.558901448251501,
      "grad_norm": 1.328125,
      "learning_rate": 0.00030944795439808223,
      "loss": 5.2478,
      "mean_token_accuracy": 0.21894153505563735,
      "num_tokens": 96256844.0,
      "step": 51625
    },
    {
      "entropy": 6.022658348083496,
      "epoch": 4.559342988343341,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003094168338272535,
      "loss": 5.1983,
      "mean_token_accuracy": 0.22325184792280198,
      "num_tokens": 96266977.0,
      "step": 51630
    },
    {
      "entropy": 5.992998695373535,
      "epoch": 4.559784528435182,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00030938571258221873,
      "loss": 5.2571,
      "mean_token_accuracy": 0.22187526077032088,
      "num_tokens": 96275813.0,
      "step": 51635
    },
    {
      "entropy": 6.040212488174438,
      "epoch": 4.560226068527022,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030935459066358723,
      "loss": 5.2916,
      "mean_token_accuracy": 0.21816359609365463,
      "num_tokens": 96286039.0,
      "step": 51640
    },
    {
      "entropy": 5.980687379837036,
      "epoch": 4.5606676086188624,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00030932346807196885,
      "loss": 5.175,
      "mean_token_accuracy": 0.22249783724546432,
      "num_tokens": 96295647.0,
      "step": 51645
    },
    {
      "entropy": 6.052506637573242,
      "epoch": 4.5611091487107025,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003092923448079733,
      "loss": 5.2459,
      "mean_token_accuracy": 0.21735197007656099,
      "num_tokens": 96305298.0,
      "step": 51650
    },
    {
      "entropy": 6.017192411422729,
      "epoch": 4.561550688802543,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030926122087221024,
      "loss": 5.2538,
      "mean_token_accuracy": 0.21475619375705718,
      "num_tokens": 96315318.0,
      "step": 51655
    },
    {
      "entropy": 6.082444381713867,
      "epoch": 4.5619922288943835,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003092300962652894,
      "loss": 5.2496,
      "mean_token_accuracy": 0.2217893436551094,
      "num_tokens": 96325316.0,
      "step": 51660
    },
    {
      "entropy": 6.006418752670288,
      "epoch": 4.562433768986224,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00030919897098782037,
      "loss": 5.2107,
      "mean_token_accuracy": 0.2216939941048622,
      "num_tokens": 96334240.0,
      "step": 51665
    },
    {
      "entropy": 6.0237720012664795,
      "epoch": 4.5628753090780645,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000309167845040413,
      "loss": 5.2161,
      "mean_token_accuracy": 0.2260952264070511,
      "num_tokens": 96343215.0,
      "step": 51670
    },
    {
      "entropy": 6.019490766525268,
      "epoch": 4.5633168491699045,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003091367184236769,
      "loss": 5.2555,
      "mean_token_accuracy": 0.22198258340358734,
      "num_tokens": 96352610.0,
      "step": 51675
    },
    {
      "entropy": 6.008432292938233,
      "epoch": 4.563758389261745,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00030910559113822203,
      "loss": 5.2618,
      "mean_token_accuracy": 0.22513247281312943,
      "num_tokens": 96362183.0,
      "step": 51680
    },
    {
      "entropy": 6.089629364013672,
      "epoch": 4.5641999293535855,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030907446318465784,
      "loss": 5.301,
      "mean_token_accuracy": 0.21311096847057343,
      "num_tokens": 96372431.0,
      "step": 51685
    },
    {
      "entropy": 6.131595849990845,
      "epoch": 4.564641469445426,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00030904333456359444,
      "loss": 5.3196,
      "mean_token_accuracy": 0.20356470197439194,
      "num_tokens": 96380606.0,
      "step": 51690
    },
    {
      "entropy": 6.002891111373901,
      "epoch": 4.565083009537266,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003090122052756414,
      "loss": 5.2605,
      "mean_token_accuracy": 0.21774163991212844,
      "num_tokens": 96390257.0,
      "step": 51695
    },
    {
      "entropy": 5.993734741210938,
      "epoch": 4.565524549629107,
      "grad_norm": 1.375,
      "learning_rate": 0.00030898107532140867,
      "loss": 5.2093,
      "mean_token_accuracy": 0.22797070443630219,
      "num_tokens": 96399248.0,
      "step": 51700
    },
    {
      "entropy": 5.99481692314148,
      "epoch": 4.565966089720947,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000308949944701506,
      "loss": 5.2225,
      "mean_token_accuracy": 0.2180814877152443,
      "num_tokens": 96408223.0,
      "step": 51705
    },
    {
      "entropy": 6.004155063629151,
      "epoch": 4.566407629812787,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003089188134165431,
      "loss": 5.211,
      "mean_token_accuracy": 0.21953134685754777,
      "num_tokens": 96417308.0,
      "step": 51710
    },
    {
      "entropy": 6.076648092269897,
      "epoch": 4.566849169904628,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00030888768146713005,
      "loss": 5.315,
      "mean_token_accuracy": 0.21123477071523666,
      "num_tokens": 96426524.0,
      "step": 51715
    },
    {
      "entropy": 6.021949720382691,
      "epoch": 4.567290709996468,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030885654885387656,
      "loss": 5.2865,
      "mean_token_accuracy": 0.21475039720535277,
      "num_tokens": 96435508.0,
      "step": 51720
    },
    {
      "entropy": 6.062174129486084,
      "epoch": 4.567732250088308,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00030882541557739247,
      "loss": 5.2299,
      "mean_token_accuracy": 0.22776298820972443,
      "num_tokens": 96444927.0,
      "step": 51725
    },
    {
      "entropy": 6.0103644847869875,
      "epoch": 4.568173790180149,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030879428163828787,
      "loss": 5.2682,
      "mean_token_accuracy": 0.22534780204296112,
      "num_tokens": 96454515.0,
      "step": 51730
    },
    {
      "entropy": 5.967941427230835,
      "epoch": 4.568615330271989,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003087631470371724,
      "loss": 5.2165,
      "mean_token_accuracy": 0.22726544588804246,
      "num_tokens": 96463287.0,
      "step": 51735
    },
    {
      "entropy": 6.143257713317871,
      "epoch": 4.569056870363829,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003087320117746561,
      "loss": 5.2465,
      "mean_token_accuracy": 0.22034126818180083,
      "num_tokens": 96472190.0,
      "step": 51740
    },
    {
      "entropy": 6.043005800247192,
      "epoch": 4.56949841045567,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030870087585134893,
      "loss": 5.1434,
      "mean_token_accuracy": 0.2232836365699768,
      "num_tokens": 96481252.0,
      "step": 51745
    },
    {
      "entropy": 6.009641218185425,
      "epoch": 4.56993995054751,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030866973926786076,
      "loss": 5.1713,
      "mean_token_accuracy": 0.22823173701763153,
      "num_tokens": 96490873.0,
      "step": 51750
    },
    {
      "entropy": 6.017526912689209,
      "epoch": 4.57038149063935,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003086386020248015,
      "loss": 5.248,
      "mean_token_accuracy": 0.22428581565618516,
      "num_tokens": 96499500.0,
      "step": 51755
    },
    {
      "entropy": 6.062466049194336,
      "epoch": 4.570823030731191,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003086074641227812,
      "loss": 5.2607,
      "mean_token_accuracy": 0.2175396203994751,
      "num_tokens": 96508602.0,
      "step": 51760
    },
    {
      "entropy": 6.014065933227539,
      "epoch": 4.571264570823031,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003085763255624098,
      "loss": 5.1405,
      "mean_token_accuracy": 0.227251298725605,
      "num_tokens": 96517699.0,
      "step": 51765
    },
    {
      "entropy": 6.033297967910767,
      "epoch": 4.571706110914871,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00030854518634429716,
      "loss": 5.2921,
      "mean_token_accuracy": 0.21957348883152009,
      "num_tokens": 96526578.0,
      "step": 51770
    },
    {
      "entropy": 6.02634744644165,
      "epoch": 4.572147651006711,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003085140464690535,
      "loss": 5.229,
      "mean_token_accuracy": 0.22128651887178422,
      "num_tokens": 96536012.0,
      "step": 51775
    },
    {
      "entropy": 6.003471088409424,
      "epoch": 4.572589191098552,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003084829059372887,
      "loss": 5.2019,
      "mean_token_accuracy": 0.22282925099134446,
      "num_tokens": 96544433.0,
      "step": 51780
    },
    {
      "entropy": 5.978329801559449,
      "epoch": 4.573030731190392,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030845176474961276,
      "loss": 5.1747,
      "mean_token_accuracy": 0.23451523035764693,
      "num_tokens": 96553676.0,
      "step": 51785
    },
    {
      "entropy": 6.075777244567871,
      "epoch": 4.573472271282233,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003084206229066359,
      "loss": 5.243,
      "mean_token_accuracy": 0.2250689223408699,
      "num_tokens": 96563111.0,
      "step": 51790
    },
    {
      "entropy": 5.995978832244873,
      "epoch": 4.573913811374073,
      "grad_norm": 1.25,
      "learning_rate": 0.000308389480408968,
      "loss": 5.1902,
      "mean_token_accuracy": 0.2277138575911522,
      "num_tokens": 96572283.0,
      "step": 51795
    },
    {
      "entropy": 6.022646379470825,
      "epoch": 4.574355351465913,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030835833725721917,
      "loss": 5.2599,
      "mean_token_accuracy": 0.22003717273473739,
      "num_tokens": 96581128.0,
      "step": 51800
    },
    {
      "entropy": 6.076579761505127,
      "epoch": 4.574796891557753,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00030832719345199943,
      "loss": 5.2666,
      "mean_token_accuracy": 0.2118977874517441,
      "num_tokens": 96590986.0,
      "step": 51805
    },
    {
      "entropy": 6.007929801940918,
      "epoch": 4.575238431649594,
      "grad_norm": 1.375,
      "learning_rate": 0.00030829604899391906,
      "loss": 5.1904,
      "mean_token_accuracy": 0.22396421581506729,
      "num_tokens": 96600420.0,
      "step": 51810
    },
    {
      "entropy": 6.057980155944824,
      "epoch": 4.575679971741434,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00030826490388358787,
      "loss": 5.3457,
      "mean_token_accuracy": 0.21096092760562896,
      "num_tokens": 96610006.0,
      "step": 51815
    },
    {
      "entropy": 6.007547330856323,
      "epoch": 4.576121511833274,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030823375812161625,
      "loss": 5.2155,
      "mean_token_accuracy": 0.22919034957885742,
      "num_tokens": 96618859.0,
      "step": 51820
    },
    {
      "entropy": 6.053813791275024,
      "epoch": 4.576563051925115,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003082026117086141,
      "loss": 5.3047,
      "mean_token_accuracy": 0.22289955914020537,
      "num_tokens": 96627931.0,
      "step": 51825
    },
    {
      "entropy": 6.028058242797852,
      "epoch": 4.577004592016955,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00030817146464519167,
      "loss": 5.2311,
      "mean_token_accuracy": 0.2212057039141655,
      "num_tokens": 96636707.0,
      "step": 51830
    },
    {
      "entropy": 6.081181955337525,
      "epoch": 4.577446132108795,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00030814031693195923,
      "loss": 5.2884,
      "mean_token_accuracy": 0.2161325916647911,
      "num_tokens": 96646399.0,
      "step": 51835
    },
    {
      "entropy": 6.071191358566284,
      "epoch": 4.577887672200636,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003081091685695267,
      "loss": 5.306,
      "mean_token_accuracy": 0.21472116708755493,
      "num_tokens": 96656299.0,
      "step": 51840
    },
    {
      "entropy": 6.043561840057373,
      "epoch": 4.578329212292476,
      "grad_norm": 1.328125,
      "learning_rate": 0.00030807801955850443,
      "loss": 5.2638,
      "mean_token_accuracy": 0.22690692096948623,
      "num_tokens": 96665049.0,
      "step": 51845
    },
    {
      "entropy": 5.9194488525390625,
      "epoch": 4.578770752384316,
      "grad_norm": 1.5,
      "learning_rate": 0.00030804686989950266,
      "loss": 5.1036,
      "mean_token_accuracy": 0.22665663212537765,
      "num_tokens": 96673583.0,
      "step": 51850
    },
    {
      "entropy": 6.061968231201172,
      "epoch": 4.579212292476157,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003080157195931314,
      "loss": 5.27,
      "mean_token_accuracy": 0.21782780289649964,
      "num_tokens": 96683361.0,
      "step": 51855
    },
    {
      "entropy": 6.0328864574432375,
      "epoch": 4.579653832567997,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000307984568640001,
      "loss": 5.254,
      "mean_token_accuracy": 0.2236607328057289,
      "num_tokens": 96693122.0,
      "step": 51860
    },
    {
      "entropy": 6.084800767898559,
      "epoch": 4.580095372659837,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003079534170407217,
      "loss": 5.231,
      "mean_token_accuracy": 0.221280600130558,
      "num_tokens": 96701591.0,
      "step": 51865
    },
    {
      "entropy": 6.0484272003173825,
      "epoch": 4.580536912751678,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003079222647959037,
      "loss": 5.3497,
      "mean_token_accuracy": 0.2058253511786461,
      "num_tokens": 96710945.0,
      "step": 51870
    },
    {
      "entropy": 6.017931938171387,
      "epoch": 4.580978452843518,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003078911119061572,
      "loss": 5.2171,
      "mean_token_accuracy": 0.21913393586874008,
      "num_tokens": 96720689.0,
      "step": 51875
    },
    {
      "entropy": 6.039345550537109,
      "epoch": 4.581419992935358,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003078599583720925,
      "loss": 5.285,
      "mean_token_accuracy": 0.2194061055779457,
      "num_tokens": 96730036.0,
      "step": 51880
    },
    {
      "entropy": 6.071523380279541,
      "epoch": 4.581861533027199,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030782880419432,
      "loss": 5.3285,
      "mean_token_accuracy": 0.21112748980522156,
      "num_tokens": 96739406.0,
      "step": 51885
    },
    {
      "entropy": 6.050321912765503,
      "epoch": 4.582303073119039,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003077976493734498,
      "loss": 5.2215,
      "mean_token_accuracy": 0.22266059815883638,
      "num_tokens": 96747812.0,
      "step": 51890
    },
    {
      "entropy": 6.060449457168579,
      "epoch": 4.582744613210879,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030776649391009236,
      "loss": 5.2498,
      "mean_token_accuracy": 0.22651513218879699,
      "num_tokens": 96756294.0,
      "step": 51895
    },
    {
      "entropy": 6.056379652023315,
      "epoch": 4.583186153302719,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000307735337804858,
      "loss": 5.2711,
      "mean_token_accuracy": 0.22653734236955642,
      "num_tokens": 96765383.0,
      "step": 51900
    },
    {
      "entropy": 6.014493370056153,
      "epoch": 4.58362769339456,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00030770418105835685,
      "loss": 5.2935,
      "mean_token_accuracy": 0.22199879437685013,
      "num_tokens": 96774933.0,
      "step": 51905
    },
    {
      "entropy": 5.998779916763306,
      "epoch": 4.5840692334864,
      "grad_norm": 1.4375,
      "learning_rate": 0.00030767302367119945,
      "loss": 5.2689,
      "mean_token_accuracy": 0.21923537403345109,
      "num_tokens": 96783235.0,
      "step": 51910
    },
    {
      "entropy": 6.013357734680175,
      "epoch": 4.584510773578241,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003076418656439961,
      "loss": 5.2469,
      "mean_token_accuracy": 0.2194809526205063,
      "num_tokens": 96792457.0,
      "step": 51915
    },
    {
      "entropy": 6.037715196609497,
      "epoch": 4.584952313670081,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00030761070697735724,
      "loss": 5.2606,
      "mean_token_accuracy": 0.2265486642718315,
      "num_tokens": 96801275.0,
      "step": 51920
    },
    {
      "entropy": 6.124417543411255,
      "epoch": 4.585393853761921,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003075795476718931,
      "loss": 5.3572,
      "mean_token_accuracy": 0.20967619121074677,
      "num_tokens": 96811564.0,
      "step": 51925
    },
    {
      "entropy": 6.02652587890625,
      "epoch": 4.5858353938537615,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030754838772821416,
      "loss": 5.1837,
      "mean_token_accuracy": 0.2338342234492302,
      "num_tokens": 96821317.0,
      "step": 51930
    },
    {
      "entropy": 6.031321620941162,
      "epoch": 4.586276933945602,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003075172271469308,
      "loss": 5.2209,
      "mean_token_accuracy": 0.22201302647590637,
      "num_tokens": 96830709.0,
      "step": 51935
    },
    {
      "entropy": 6.0670592308044435,
      "epoch": 4.5867184740374425,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003074860659286535,
      "loss": 5.2434,
      "mean_token_accuracy": 0.22306993901729583,
      "num_tokens": 96839208.0,
      "step": 51940
    },
    {
      "entropy": 6.0642163276672365,
      "epoch": 4.587160014129283,
      "grad_norm": 1.375,
      "learning_rate": 0.0003074549040739926,
      "loss": 5.32,
      "mean_token_accuracy": 0.21978389769792556,
      "num_tokens": 96848681.0,
      "step": 51945
    },
    {
      "entropy": 6.039669513702393,
      "epoch": 4.5876015542211235,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030742374158355864,
      "loss": 5.2984,
      "mean_token_accuracy": 0.21660050600767136,
      "num_tokens": 96858306.0,
      "step": 51950
    },
    {
      "entropy": 6.04000506401062,
      "epoch": 4.5880430943129635,
      "grad_norm": 1.328125,
      "learning_rate": 0.000307392578457962,
      "loss": 5.2497,
      "mean_token_accuracy": 0.22669800966978074,
      "num_tokens": 96867799.0,
      "step": 51955
    },
    {
      "entropy": 6.075552749633789,
      "epoch": 4.588484634404804,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003073614146978131,
      "loss": 5.253,
      "mean_token_accuracy": 0.21844439059495926,
      "num_tokens": 96875839.0,
      "step": 51960
    },
    {
      "entropy": 5.9892792224884035,
      "epoch": 4.5889261744966445,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003073302503037226,
      "loss": 5.2441,
      "mean_token_accuracy": 0.2288251057267189,
      "num_tokens": 96884305.0,
      "step": 51965
    },
    {
      "entropy": 5.928381156921387,
      "epoch": 4.589367714588485,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003072990852763008,
      "loss": 5.1582,
      "mean_token_accuracy": 0.22104837149381637,
      "num_tokens": 96893782.0,
      "step": 51970
    },
    {
      "entropy": 6.012468528747559,
      "epoch": 4.589809254680325,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003072679196161583,
      "loss": 5.234,
      "mean_token_accuracy": 0.2177085295319557,
      "num_tokens": 96902726.0,
      "step": 51975
    },
    {
      "entropy": 6.087898635864258,
      "epoch": 4.590250794772166,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003072367533239056,
      "loss": 5.3648,
      "mean_token_accuracy": 0.21473534852266313,
      "num_tokens": 96913006.0,
      "step": 51980
    },
    {
      "entropy": 6.072362470626831,
      "epoch": 4.590692334864006,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00030720558640015327,
      "loss": 5.2065,
      "mean_token_accuracy": 0.22136864960193633,
      "num_tokens": 96922244.0,
      "step": 51985
    },
    {
      "entropy": 6.030195808410644,
      "epoch": 4.591133874955846,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030717441884551183,
      "loss": 5.1863,
      "mean_token_accuracy": 0.22908525466918944,
      "num_tokens": 96930424.0,
      "step": 51990
    },
    {
      "entropy": 5.991921138763428,
      "epoch": 4.591575415047687,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030714325066059185,
      "loss": 5.2599,
      "mean_token_accuracy": 0.21397242248058318,
      "num_tokens": 96940062.0,
      "step": 51995
    },
    {
      "entropy": 6.033878135681152,
      "epoch": 4.592016955139527,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00030711208184600383,
      "loss": 5.2763,
      "mean_token_accuracy": 0.2134414240717888,
      "num_tokens": 96949913.0,
      "step": 52000
    },
    {
      "entropy": 6.038950347900391,
      "epoch": 4.592458495231367,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003070809124023584,
      "loss": 5.2723,
      "mean_token_accuracy": 0.22017012685537338,
      "num_tokens": 96959477.0,
      "step": 52005
    },
    {
      "entropy": 6.068460464477539,
      "epoch": 4.592900035323208,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003070497423302661,
      "loss": 5.2694,
      "mean_token_accuracy": 0.22199157923460006,
      "num_tokens": 96968740.0,
      "step": 52010
    },
    {
      "entropy": 6.087869882583618,
      "epoch": 4.593341575415048,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030701857163033757,
      "loss": 5.296,
      "mean_token_accuracy": 0.21501156985759734,
      "num_tokens": 96977935.0,
      "step": 52015
    },
    {
      "entropy": 6.12738938331604,
      "epoch": 4.593783115506888,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030698740030318344,
      "loss": 5.3832,
      "mean_token_accuracy": 0.2132135108113289,
      "num_tokens": 96988370.0,
      "step": 52020
    },
    {
      "entropy": 6.119157028198242,
      "epoch": 4.594224655598729,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003069562283494144,
      "loss": 5.3006,
      "mean_token_accuracy": 0.21929920315742493,
      "num_tokens": 96998434.0,
      "step": 52025
    },
    {
      "entropy": 5.946434783935547,
      "epoch": 4.594666195690569,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00030692505576964096,
      "loss": 5.0862,
      "mean_token_accuracy": 0.2384801372885704,
      "num_tokens": 97007459.0,
      "step": 52030
    },
    {
      "entropy": 6.0063385486602785,
      "epoch": 4.595107735782409,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030689388256447385,
      "loss": 5.2147,
      "mean_token_accuracy": 0.21975667029619217,
      "num_tokens": 97016426.0,
      "step": 52035
    },
    {
      "entropy": 6.053132247924805,
      "epoch": 4.59554927587425,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003068627087345238,
      "loss": 5.2955,
      "mean_token_accuracy": 0.2167125791311264,
      "num_tokens": 97026034.0,
      "step": 52040
    },
    {
      "entropy": 6.062428092956543,
      "epoch": 4.59599081596609,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003068315342804012,
      "loss": 5.1855,
      "mean_token_accuracy": 0.22887178510427475,
      "num_tokens": 97036267.0,
      "step": 52045
    },
    {
      "entropy": 6.048151445388794,
      "epoch": 4.59643235605793,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00030680035920271713,
      "loss": 5.2852,
      "mean_token_accuracy": 0.22309403419494628,
      "num_tokens": 97044961.0,
      "step": 52050
    },
    {
      "entropy": 6.052394676208496,
      "epoch": 4.59687389614977,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030676918350208205,
      "loss": 5.2181,
      "mean_token_accuracy": 0.22820911705493926,
      "num_tokens": 97054236.0,
      "step": 52055
    },
    {
      "entropy": 5.999620246887207,
      "epoch": 4.597315436241611,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003067380071791068,
      "loss": 5.2774,
      "mean_token_accuracy": 0.21590325236320496,
      "num_tokens": 97063558.0,
      "step": 52060
    },
    {
      "entropy": 5.9684751510620115,
      "epoch": 4.597756976333451,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003067068302344019,
      "loss": 5.1652,
      "mean_token_accuracy": 0.2251383036375046,
      "num_tokens": 97072625.0,
      "step": 52065
    },
    {
      "entropy": 6.0825169563293455,
      "epoch": 4.598198516425292,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003066756526685784,
      "loss": 5.295,
      "mean_token_accuracy": 0.222673200070858,
      "num_tokens": 97082047.0,
      "step": 52070
    },
    {
      "entropy": 5.987621021270752,
      "epoch": 4.598640056517132,
      "grad_norm": 1.546875,
      "learning_rate": 0.0003066444744822467,
      "loss": 5.1726,
      "mean_token_accuracy": 0.23233262002468108,
      "num_tokens": 97090804.0,
      "step": 52075
    },
    {
      "entropy": 6.0564549446105955,
      "epoch": 4.599081596608972,
      "grad_norm": 1.25,
      "learning_rate": 0.00030661329567601786,
      "loss": 5.3267,
      "mean_token_accuracy": 0.2228786304593086,
      "num_tokens": 97100682.0,
      "step": 52080
    },
    {
      "entropy": 6.045048618316651,
      "epoch": 4.599523136700812,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030658211625050246,
      "loss": 5.3508,
      "mean_token_accuracy": 0.2170161098241806,
      "num_tokens": 97111031.0,
      "step": 52085
    },
    {
      "entropy": 6.082386112213134,
      "epoch": 4.599964676792653,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003065509362063114,
      "loss": 5.3006,
      "mean_token_accuracy": 0.2131194144487381,
      "num_tokens": 97119988.0,
      "step": 52090
    },
    {
      "entropy": 6.056108903884888,
      "epoch": 4.600406216884493,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030651975554405543,
      "loss": 5.2568,
      "mean_token_accuracy": 0.22124978452920913,
      "num_tokens": 97129947.0,
      "step": 52095
    },
    {
      "entropy": 6.03003420829773,
      "epoch": 4.600847756976333,
      "grad_norm": 1.375,
      "learning_rate": 0.00030648857426434544,
      "loss": 5.1949,
      "mean_token_accuracy": 0.22694693356752396,
      "num_tokens": 97140677.0,
      "step": 52100
    },
    {
      "entropy": 6.139239120483398,
      "epoch": 4.601289297068174,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003064573923677921,
      "loss": 5.3565,
      "mean_token_accuracy": 0.21230851113796234,
      "num_tokens": 97150402.0,
      "step": 52105
    },
    {
      "entropy": 6.077597188949585,
      "epoch": 4.601730837160014,
      "grad_norm": 1.25,
      "learning_rate": 0.0003064262098550064,
      "loss": 5.2486,
      "mean_token_accuracy": 0.22292913347482682,
      "num_tokens": 97159539.0,
      "step": 52110
    },
    {
      "entropy": 6.0668117046356205,
      "epoch": 4.602172377251854,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003063950267265992,
      "loss": 5.2078,
      "mean_token_accuracy": 0.22652770429849625,
      "num_tokens": 97168332.0,
      "step": 52115
    },
    {
      "entropy": 6.020342206954956,
      "epoch": 4.602613917343695,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00030636384298318107,
      "loss": 5.1952,
      "mean_token_accuracy": 0.22397733330726624,
      "num_tokens": 97177090.0,
      "step": 52120
    },
    {
      "entropy": 6.068851470947266,
      "epoch": 4.603055457435535,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003063326586253633,
      "loss": 5.3754,
      "mean_token_accuracy": 0.20865686237812042,
      "num_tokens": 97186294.0,
      "step": 52125
    },
    {
      "entropy": 6.0853015899658205,
      "epoch": 4.603496997527375,
      "grad_norm": 1.46875,
      "learning_rate": 0.00030630147365375644,
      "loss": 5.2841,
      "mean_token_accuracy": 0.21545587182044984,
      "num_tokens": 97195189.0,
      "step": 52130
    },
    {
      "entropy": 6.021648502349853,
      "epoch": 4.603938537619216,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030627028806897154,
      "loss": 5.2548,
      "mean_token_accuracy": 0.22289353758096694,
      "num_tokens": 97203918.0,
      "step": 52135
    },
    {
      "entropy": 5.983114099502563,
      "epoch": 4.604380077711056,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00030623910187161946,
      "loss": 5.1299,
      "mean_token_accuracy": 0.23456103205680848,
      "num_tokens": 97212550.0,
      "step": 52140
    },
    {
      "entropy": 6.096062898635864,
      "epoch": 4.604821617802896,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003062079150623112,
      "loss": 5.2971,
      "mean_token_accuracy": 0.21376026570796966,
      "num_tokens": 97221793.0,
      "step": 52145
    },
    {
      "entropy": 6.115578603744507,
      "epoch": 4.605263157894737,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003061767276416576,
      "loss": 5.341,
      "mean_token_accuracy": 0.21013542860746384,
      "num_tokens": 97230884.0,
      "step": 52150
    },
    {
      "entropy": 6.000761699676514,
      "epoch": 4.605704697986577,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003061455396102696,
      "loss": 5.2572,
      "mean_token_accuracy": 0.2179681420326233,
      "num_tokens": 97240362.0,
      "step": 52155
    },
    {
      "entropy": 6.093262720108032,
      "epoch": 4.606146238078417,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003061143509687583,
      "loss": 5.2646,
      "mean_token_accuracy": 0.21791450977325438,
      "num_tokens": 97248932.0,
      "step": 52160
    },
    {
      "entropy": 6.0849097728729244,
      "epoch": 4.606587778170258,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030608316171773444,
      "loss": 5.2063,
      "mean_token_accuracy": 0.21685246974229813,
      "num_tokens": 97258213.0,
      "step": 52165
    },
    {
      "entropy": 6.025115919113159,
      "epoch": 4.607029318262098,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003060519718578092,
      "loss": 5.2186,
      "mean_token_accuracy": 0.2243146553635597,
      "num_tokens": 97267472.0,
      "step": 52170
    },
    {
      "entropy": 5.976731061935425,
      "epoch": 4.607470858353938,
      "grad_norm": 1.4375,
      "learning_rate": 0.00030602078138959344,
      "loss": 5.2142,
      "mean_token_accuracy": 0.22266520261764527,
      "num_tokens": 97276109.0,
      "step": 52175
    },
    {
      "entropy": 6.037236928939819,
      "epoch": 4.607912398445778,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00030598959031369826,
      "loss": 5.3227,
      "mean_token_accuracy": 0.21415407955646515,
      "num_tokens": 97284438.0,
      "step": 52180
    },
    {
      "entropy": 5.968881702423095,
      "epoch": 4.608353938537619,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003059583986307346,
      "loss": 5.2591,
      "mean_token_accuracy": 0.2275417998433113,
      "num_tokens": 97293861.0,
      "step": 52185
    },
    {
      "entropy": 5.9843668937683105,
      "epoch": 4.608795478629459,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030592720634131356,
      "loss": 5.2279,
      "mean_token_accuracy": 0.22622978836297988,
      "num_tokens": 97303418.0,
      "step": 52190
    },
    {
      "entropy": 6.0624871253967285,
      "epoch": 4.6092370187213,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003058960134460461,
      "loss": 5.2209,
      "mean_token_accuracy": 0.22362909466028214,
      "num_tokens": 97312108.0,
      "step": 52195
    },
    {
      "entropy": 6.043094444274902,
      "epoch": 4.60967855881314,
      "grad_norm": 1.375,
      "learning_rate": 0.0003058648199455433,
      "loss": 5.2436,
      "mean_token_accuracy": 0.21614499092102052,
      "num_tokens": 97321553.0,
      "step": 52200
    },
    {
      "entropy": 6.050907993316651,
      "epoch": 4.61012009890498,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00030583362584041626,
      "loss": 5.2532,
      "mean_token_accuracy": 0.2208789512515068,
      "num_tokens": 97330400.0,
      "step": 52205
    },
    {
      "entropy": 6.0554827690124515,
      "epoch": 4.6105616389968205,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000305802431131276,
      "loss": 5.2196,
      "mean_token_accuracy": 0.21984711438417434,
      "num_tokens": 97338913.0,
      "step": 52210
    },
    {
      "entropy": 6.049071836471557,
      "epoch": 4.611003179088661,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003057712358187336,
      "loss": 5.2749,
      "mean_token_accuracy": 0.2172769635915756,
      "num_tokens": 97347469.0,
      "step": 52215
    },
    {
      "entropy": 5.969475126266479,
      "epoch": 4.6114447191805015,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030574003990340023,
      "loss": 5.2524,
      "mean_token_accuracy": 0.2291898772120476,
      "num_tokens": 97356919.0,
      "step": 52220
    },
    {
      "entropy": 6.097788906097412,
      "epoch": 4.611886259272342,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000305708843385887,
      "loss": 5.2125,
      "mean_token_accuracy": 0.22495101541280746,
      "num_tokens": 97366058.0,
      "step": 52225
    },
    {
      "entropy": 6.034520292282105,
      "epoch": 4.6123277993641825,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003056776462668049,
      "loss": 5.2777,
      "mean_token_accuracy": 0.21678948402404785,
      "num_tokens": 97376325.0,
      "step": 52230
    },
    {
      "entropy": 6.01314001083374,
      "epoch": 4.6127693394560225,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003056464485467653,
      "loss": 5.3234,
      "mean_token_accuracy": 0.2143211081624031,
      "num_tokens": 97385961.0,
      "step": 52235
    },
    {
      "entropy": 6.076448249816894,
      "epoch": 4.613210879547863,
      "grad_norm": 1.40625,
      "learning_rate": 0.000305615250226379,
      "loss": 5.2631,
      "mean_token_accuracy": 0.21547337919473647,
      "num_tokens": 97394522.0,
      "step": 52240
    },
    {
      "entropy": 6.062446403503418,
      "epoch": 4.6136524196397035,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030558405130625755,
      "loss": 5.2092,
      "mean_token_accuracy": 0.22233630418777467,
      "num_tokens": 97404113.0,
      "step": 52245
    },
    {
      "entropy": 6.060015869140625,
      "epoch": 4.614093959731544,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00030555285178701186,
      "loss": 5.2754,
      "mean_token_accuracy": 0.2226386234164238,
      "num_tokens": 97414424.0,
      "step": 52250
    },
    {
      "entropy": 6.007137775421143,
      "epoch": 4.614535499823384,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003055216516692532,
      "loss": 5.1678,
      "mean_token_accuracy": 0.22554033249616623,
      "num_tokens": 97422606.0,
      "step": 52255
    },
    {
      "entropy": 6.088907051086426,
      "epoch": 4.614977039915225,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00030549045095359267,
      "loss": 5.283,
      "mean_token_accuracy": 0.22382775247097014,
      "num_tokens": 97431627.0,
      "step": 52260
    },
    {
      "entropy": 6.077367877960205,
      "epoch": 4.615418580007065,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003054592496406416,
      "loss": 5.315,
      "mean_token_accuracy": 0.21881612837314607,
      "num_tokens": 97441266.0,
      "step": 52265
    },
    {
      "entropy": 5.957552194595337,
      "epoch": 4.615860120098905,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030542804773101115,
      "loss": 5.187,
      "mean_token_accuracy": 0.22646407634019852,
      "num_tokens": 97450807.0,
      "step": 52270
    },
    {
      "entropy": 6.049994897842407,
      "epoch": 4.616301660190746,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003053968452253126,
      "loss": 5.2009,
      "mean_token_accuracy": 0.22385649383068085,
      "num_tokens": 97460431.0,
      "step": 52275
    },
    {
      "entropy": 5.994827127456665,
      "epoch": 4.616743200282586,
      "grad_norm": 1.328125,
      "learning_rate": 0.000305365642124157,
      "loss": 5.1455,
      "mean_token_accuracy": 0.23176960200071334,
      "num_tokens": 97469677.0,
      "step": 52280
    },
    {
      "entropy": 5.999058961868286,
      "epoch": 4.617184740374426,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030533443842815586,
      "loss": 5.1835,
      "mean_token_accuracy": 0.2296268031001091,
      "num_tokens": 97478936.0,
      "step": 52285
    },
    {
      "entropy": 5.980969333648682,
      "epoch": 4.617626280466267,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00030530323413792025,
      "loss": 5.2331,
      "mean_token_accuracy": 0.22494005113840104,
      "num_tokens": 97488222.0,
      "step": 52290
    },
    {
      "entropy": 6.042471170425415,
      "epoch": 4.618067820558107,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030527202925406156,
      "loss": 5.247,
      "mean_token_accuracy": 0.2238452821969986,
      "num_tokens": 97497133.0,
      "step": 52295
    },
    {
      "entropy": 6.0039482593536375,
      "epoch": 4.618509360649947,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030524082377719107,
      "loss": 5.1841,
      "mean_token_accuracy": 0.22902010530233383,
      "num_tokens": 97506198.0,
      "step": 52300
    },
    {
      "entropy": 6.012518453598022,
      "epoch": 4.618950900741788,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030520961770792,
      "loss": 5.2187,
      "mean_token_accuracy": 0.2246124655008316,
      "num_tokens": 97514722.0,
      "step": 52305
    },
    {
      "entropy": 6.018968534469605,
      "epoch": 4.619392440833628,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003051784110468597,
      "loss": 5.2404,
      "mean_token_accuracy": 0.2255624860525131,
      "num_tokens": 97524418.0,
      "step": 52310
    },
    {
      "entropy": 5.935358905792237,
      "epoch": 4.619833980925468,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030514720379462145,
      "loss": 5.162,
      "mean_token_accuracy": 0.2312308669090271,
      "num_tokens": 97533984.0,
      "step": 52315
    },
    {
      "entropy": 6.065497970581054,
      "epoch": 4.620275521017309,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030511599595181674,
      "loss": 5.2855,
      "mean_token_accuracy": 0.22158818691968918,
      "num_tokens": 97542723.0,
      "step": 52320
    },
    {
      "entropy": 6.07277021408081,
      "epoch": 4.620717061109149,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00030508478751905665,
      "loss": 5.2486,
      "mean_token_accuracy": 0.2232775241136551,
      "num_tokens": 97551456.0,
      "step": 52325
    },
    {
      "entropy": 5.98573145866394,
      "epoch": 4.621158601200989,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003050535784969528,
      "loss": 5.2121,
      "mean_token_accuracy": 0.23344701081514357,
      "num_tokens": 97560631.0,
      "step": 52330
    },
    {
      "entropy": 6.077793025970459,
      "epoch": 4.621600141292829,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003050223688861164,
      "loss": 5.3519,
      "mean_token_accuracy": 0.21615810990333556,
      "num_tokens": 97570742.0,
      "step": 52335
    },
    {
      "entropy": 6.088828277587891,
      "epoch": 4.62204168138467,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00030499115868715885,
      "loss": 5.2808,
      "mean_token_accuracy": 0.2228200316429138,
      "num_tokens": 97581003.0,
      "step": 52340
    },
    {
      "entropy": 6.01220440864563,
      "epoch": 4.62248322147651,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003049599479006916,
      "loss": 5.1538,
      "mean_token_accuracy": 0.2336730495095253,
      "num_tokens": 97590397.0,
      "step": 52345
    },
    {
      "entropy": 6.111645126342774,
      "epoch": 4.622924761568351,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00030492873652732594,
      "loss": 5.321,
      "mean_token_accuracy": 0.21916348338127137,
      "num_tokens": 97600191.0,
      "step": 52350
    },
    {
      "entropy": 6.104408216476441,
      "epoch": 4.623366301660191,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00030489752456767347,
      "loss": 5.3377,
      "mean_token_accuracy": 0.213889916241169,
      "num_tokens": 97608912.0,
      "step": 52355
    },
    {
      "entropy": 6.135921621322632,
      "epoch": 4.623807841752031,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00030486631202234534,
      "loss": 5.3074,
      "mean_token_accuracy": 0.2157763957977295,
      "num_tokens": 97617631.0,
      "step": 52360
    },
    {
      "entropy": 6.05002269744873,
      "epoch": 4.624249381843871,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030483509889195327,
      "loss": 5.1723,
      "mean_token_accuracy": 0.22869362235069274,
      "num_tokens": 97626713.0,
      "step": 52365
    },
    {
      "entropy": 6.066608524322509,
      "epoch": 4.624690921935712,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003048038851771085,
      "loss": 5.2588,
      "mean_token_accuracy": 0.2171972617506981,
      "num_tokens": 97636252.0,
      "step": 52370
    },
    {
      "entropy": 6.076579666137695,
      "epoch": 4.625132462027552,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00030477267087842267,
      "loss": 5.3282,
      "mean_token_accuracy": 0.2202013149857521,
      "num_tokens": 97645347.0,
      "step": 52375
    },
    {
      "entropy": 6.0891913890838625,
      "epoch": 4.625574002119392,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000304741455996507,
      "loss": 5.2304,
      "mean_token_accuracy": 0.22067975997924805,
      "num_tokens": 97654435.0,
      "step": 52380
    },
    {
      "entropy": 6.056708765029907,
      "epoch": 4.626015542211233,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003047102405319733,
      "loss": 5.2999,
      "mean_token_accuracy": 0.2207583338022232,
      "num_tokens": 97663844.0,
      "step": 52385
    },
    {
      "entropy": 6.031840181350708,
      "epoch": 4.626457082303073,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003046790244854327,
      "loss": 5.2542,
      "mean_token_accuracy": 0.2160596251487732,
      "num_tokens": 97672951.0,
      "step": 52390
    },
    {
      "entropy": 5.991131687164307,
      "epoch": 4.626898622394913,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000304647807857497,
      "loss": 5.2093,
      "mean_token_accuracy": 0.2223599970340729,
      "num_tokens": 97681187.0,
      "step": 52395
    },
    {
      "entropy": 6.045425796508789,
      "epoch": 4.627340162486754,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003046165906487776,
      "loss": 5.3401,
      "mean_token_accuracy": 0.21415073722600936,
      "num_tokens": 97690298.0,
      "step": 52400
    },
    {
      "entropy": 5.997149658203125,
      "epoch": 4.627781702578594,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00030458537285988604,
      "loss": 5.2681,
      "mean_token_accuracy": 0.22337619364261627,
      "num_tokens": 97699676.0,
      "step": 52405
    },
    {
      "entropy": 6.098217010498047,
      "epoch": 4.628223242670434,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003045541544914338,
      "loss": 5.2649,
      "mean_token_accuracy": 0.2190857633948326,
      "num_tokens": 97709446.0,
      "step": 52410
    },
    {
      "entropy": 5.994238948822021,
      "epoch": 4.628664782762275,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003045229355440325,
      "loss": 5.1827,
      "mean_token_accuracy": 0.22648973017930984,
      "num_tokens": 97717981.0,
      "step": 52415
    },
    {
      "entropy": 5.944191884994507,
      "epoch": 4.629106322854115,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003044917160182938,
      "loss": 5.1931,
      "mean_token_accuracy": 0.2302930846810341,
      "num_tokens": 97727653.0,
      "step": 52420
    },
    {
      "entropy": 5.996671295166015,
      "epoch": 4.629547862945955,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030446049591482895,
      "loss": 5.211,
      "mean_token_accuracy": 0.2285235583782196,
      "num_tokens": 97738253.0,
      "step": 52425
    },
    {
      "entropy": 6.018295669555664,
      "epoch": 4.629989403037796,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003044292752342499,
      "loss": 5.1267,
      "mean_token_accuracy": 0.2300614222884178,
      "num_tokens": 97747661.0,
      "step": 52430
    },
    {
      "entropy": 6.0880965232849125,
      "epoch": 4.630430943129636,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00030439805397716795,
      "loss": 5.3018,
      "mean_token_accuracy": 0.21986270397901536,
      "num_tokens": 97757098.0,
      "step": 52435
    },
    {
      "entropy": 5.96878752708435,
      "epoch": 4.630872483221476,
      "grad_norm": 1.390625,
      "learning_rate": 0.00030436683214419495,
      "loss": 5.1645,
      "mean_token_accuracy": 0.23076634258031845,
      "num_tokens": 97765326.0,
      "step": 52440
    },
    {
      "entropy": 6.0122381210327145,
      "epoch": 4.631314023313317,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030433560973594243,
      "loss": 5.2038,
      "mean_token_accuracy": 0.23036014437675476,
      "num_tokens": 97775961.0,
      "step": 52445
    },
    {
      "entropy": 6.022179794311524,
      "epoch": 4.631755563405157,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030430438675302195,
      "loss": 5.3016,
      "mean_token_accuracy": 0.22006694376468658,
      "num_tokens": 97785318.0,
      "step": 52450
    },
    {
      "entropy": 5.97112398147583,
      "epoch": 4.632197103496997,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003042731631960452,
      "loss": 5.2256,
      "mean_token_accuracy": 0.2238365024328232,
      "num_tokens": 97795508.0,
      "step": 52455
    },
    {
      "entropy": 5.985183572769165,
      "epoch": 4.632638643588837,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003042419390656239,
      "loss": 5.1864,
      "mean_token_accuracy": 0.23275825530290603,
      "num_tokens": 97803932.0,
      "step": 52460
    },
    {
      "entropy": 6.118350172042847,
      "epoch": 4.633080183680678,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003042107143623695,
      "loss": 5.3496,
      "mean_token_accuracy": 0.21549306064844131,
      "num_tokens": 97814379.0,
      "step": 52465
    },
    {
      "entropy": 6.114810371398926,
      "epoch": 4.633521723772518,
      "grad_norm": 1.4375,
      "learning_rate": 0.00030417948908689394,
      "loss": 5.292,
      "mean_token_accuracy": 0.21359787583351136,
      "num_tokens": 97823710.0,
      "step": 52470
    },
    {
      "entropy": 6.041645145416259,
      "epoch": 4.633963263864359,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003041482632398087,
      "loss": 5.2271,
      "mean_token_accuracy": 0.22095587253570556,
      "num_tokens": 97834067.0,
      "step": 52475
    },
    {
      "entropy": 6.059876203536987,
      "epoch": 4.634404803956199,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003041170368217256,
      "loss": 5.3923,
      "mean_token_accuracy": 0.21599969416856765,
      "num_tokens": 97843067.0,
      "step": 52480
    },
    {
      "entropy": 6.023591804504394,
      "epoch": 4.634846344048039,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030408580983325634,
      "loss": 5.2615,
      "mean_token_accuracy": 0.21246389150619507,
      "num_tokens": 97853015.0,
      "step": 52485
    },
    {
      "entropy": 6.116689157485962,
      "epoch": 4.6352878841398795,
      "grad_norm": 1.453125,
      "learning_rate": 0.00030405458227501255,
      "loss": 5.3073,
      "mean_token_accuracy": 0.2234611302614212,
      "num_tokens": 97861882.0,
      "step": 52490
    },
    {
      "entropy": 6.049499273300171,
      "epoch": 4.63572942423172,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030402335414760604,
      "loss": 5.169,
      "mean_token_accuracy": 0.2239794075489044,
      "num_tokens": 97871231.0,
      "step": 52495
    },
    {
      "entropy": 6.068097019195557,
      "epoch": 4.6361709643235605,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003039921254516486,
      "loss": 5.281,
      "mean_token_accuracy": 0.21724320948123932,
      "num_tokens": 97880223.0,
      "step": 52500
    },
    {
      "entropy": 6.037273263931274,
      "epoch": 4.636612504415401,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003039608961877518,
      "loss": 5.2729,
      "mean_token_accuracy": 0.2242664247751236,
      "num_tokens": 97889251.0,
      "step": 52505
    },
    {
      "entropy": 6.027593898773193,
      "epoch": 4.6370540445072415,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003039296663565275,
      "loss": 5.2371,
      "mean_token_accuracy": 0.22845786660909653,
      "num_tokens": 97898245.0,
      "step": 52510
    },
    {
      "entropy": 6.073487710952759,
      "epoch": 4.6374955845990815,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003038984359585876,
      "loss": 5.2658,
      "mean_token_accuracy": 0.2233736753463745,
      "num_tokens": 97906837.0,
      "step": 52515
    },
    {
      "entropy": 6.063505744934082,
      "epoch": 4.637937124690922,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00030386720499454373,
      "loss": 5.2632,
      "mean_token_accuracy": 0.2178092211484909,
      "num_tokens": 97916077.0,
      "step": 52520
    },
    {
      "entropy": 6.01588773727417,
      "epoch": 4.6383786647827625,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00030383597346500773,
      "loss": 5.2366,
      "mean_token_accuracy": 0.22549020498991013,
      "num_tokens": 97925033.0,
      "step": 52525
    },
    {
      "entropy": 6.044680500030518,
      "epoch": 4.638820204874603,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003038047413705914,
      "loss": 5.248,
      "mean_token_accuracy": 0.22494291216135026,
      "num_tokens": 97933895.0,
      "step": 52530
    },
    {
      "entropy": 6.02127161026001,
      "epoch": 4.639261744966443,
      "grad_norm": 1.375,
      "learning_rate": 0.00030377350871190655,
      "loss": 5.1902,
      "mean_token_accuracy": 0.2240193024277687,
      "num_tokens": 97942831.0,
      "step": 52535
    },
    {
      "entropy": 6.063392400741577,
      "epoch": 4.6397032850582836,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030374227548956505,
      "loss": 5.1863,
      "mean_token_accuracy": 0.22760095000267028,
      "num_tokens": 97951900.0,
      "step": 52540
    },
    {
      "entropy": 6.052659511566162,
      "epoch": 4.640144825150124,
      "grad_norm": 1.375,
      "learning_rate": 0.0003037110417041787,
      "loss": 5.2598,
      "mean_token_accuracy": 0.2218284234404564,
      "num_tokens": 97961029.0,
      "step": 52545
    },
    {
      "entropy": 6.063736486434936,
      "epoch": 4.640586365241964,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030367980735635945,
      "loss": 5.3027,
      "mean_token_accuracy": 0.22063824981451036,
      "num_tokens": 97970275.0,
      "step": 52550
    },
    {
      "entropy": 6.091079139709473,
      "epoch": 4.641027905333805,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000303648572446719,
      "loss": 5.2845,
      "mean_token_accuracy": 0.21618800908327102,
      "num_tokens": 97979856.0,
      "step": 52555
    },
    {
      "entropy": 6.120592164993286,
      "epoch": 4.641469445425645,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00030361733697586936,
      "loss": 5.3246,
      "mean_token_accuracy": 0.21227778643369674,
      "num_tokens": 97989740.0,
      "step": 52560
    },
    {
      "entropy": 5.962530612945557,
      "epoch": 4.641910985517485,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003035861009444223,
      "loss": 5.1161,
      "mean_token_accuracy": 0.23893265873193742,
      "num_tokens": 97998894.0,
      "step": 52565
    },
    {
      "entropy": 6.071327638626099,
      "epoch": 4.642352525609326,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030355486435298983,
      "loss": 5.3355,
      "mean_token_accuracy": 0.21424668282270432,
      "num_tokens": 98008062.0,
      "step": 52570
    },
    {
      "entropy": 6.068814945220947,
      "epoch": 4.642794065701166,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003035236272021838,
      "loss": 5.1834,
      "mean_token_accuracy": 0.22335621267557143,
      "num_tokens": 98016764.0,
      "step": 52575
    },
    {
      "entropy": 6.0236729145050045,
      "epoch": 4.643235605793006,
      "grad_norm": 1.390625,
      "learning_rate": 0.00030349238949261615,
      "loss": 5.2039,
      "mean_token_accuracy": 0.22561356872320176,
      "num_tokens": 98025459.0,
      "step": 52580
    },
    {
      "entropy": 5.991366338729859,
      "epoch": 4.643677145884847,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003034611512248988,
      "loss": 5.2126,
      "mean_token_accuracy": 0.22646698653697966,
      "num_tokens": 98034790.0,
      "step": 52585
    },
    {
      "entropy": 6.064182710647583,
      "epoch": 4.644118685976687,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030342991239964366,
      "loss": 5.2379,
      "mean_token_accuracy": 0.21905699670314788,
      "num_tokens": 98044057.0,
      "step": 52590
    },
    {
      "entropy": 6.031960678100586,
      "epoch": 4.644560226068527,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003033986730174627,
      "loss": 5.1305,
      "mean_token_accuracy": 0.2343716099858284,
      "num_tokens": 98053298.0,
      "step": 52595
    },
    {
      "entropy": 6.102035760879517,
      "epoch": 4.645001766160368,
      "grad_norm": 1.25,
      "learning_rate": 0.00030336743307896784,
      "loss": 5.2973,
      "mean_token_accuracy": 0.21524155437946318,
      "num_tokens": 98062785.0,
      "step": 52600
    },
    {
      "entropy": 6.077810621261596,
      "epoch": 4.645443306252208,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030333619258477123,
      "loss": 5.2543,
      "mean_token_accuracy": 0.21927611380815507,
      "num_tokens": 98072726.0,
      "step": 52605
    },
    {
      "entropy": 5.963824653625489,
      "epoch": 4.645884846344048,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003033049515354846,
      "loss": 5.1762,
      "mean_token_accuracy": 0.23093230575323104,
      "num_tokens": 98081937.0,
      "step": 52610
    },
    {
      "entropy": 6.079023694992065,
      "epoch": 4.646326386435888,
      "grad_norm": 1.421875,
      "learning_rate": 0.00030327370993172014,
      "loss": 5.336,
      "mean_token_accuracy": 0.21441642493009566,
      "num_tokens": 98091489.0,
      "step": 52615
    },
    {
      "entropy": 6.127726125717163,
      "epoch": 4.646767926527729,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030324246777408976,
      "loss": 5.3462,
      "mean_token_accuracy": 0.2151074305176735,
      "num_tokens": 98100501.0,
      "step": 52620
    },
    {
      "entropy": 6.082920551300049,
      "epoch": 4.647209466619569,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003032112250632055,
      "loss": 5.2717,
      "mean_token_accuracy": 0.22493387907743453,
      "num_tokens": 98109650.0,
      "step": 52625
    },
    {
      "entropy": 6.069586277008057,
      "epoch": 4.64765100671141,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030317998179967936,
      "loss": 5.2954,
      "mean_token_accuracy": 0.22263315916061402,
      "num_tokens": 98120421.0,
      "step": 52630
    },
    {
      "entropy": 6.088148736953736,
      "epoch": 4.64809254680325,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00030314873798412346,
      "loss": 5.2512,
      "mean_token_accuracy": 0.22131777852773665,
      "num_tokens": 98129829.0,
      "step": 52635
    },
    {
      "entropy": 5.976587390899658,
      "epoch": 4.64853408689509,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030311749361714975,
      "loss": 5.1381,
      "mean_token_accuracy": 0.23122316598892212,
      "num_tokens": 98139229.0,
      "step": 52640
    },
    {
      "entropy": 6.000230741500855,
      "epoch": 4.64897562698693,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00030308624869937033,
      "loss": 5.2152,
      "mean_token_accuracy": 0.22305875271558762,
      "num_tokens": 98148731.0,
      "step": 52645
    },
    {
      "entropy": 6.0514076232910154,
      "epoch": 4.649417167078771,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003030550032313973,
      "loss": 5.1967,
      "mean_token_accuracy": 0.22900252789258957,
      "num_tokens": 98157886.0,
      "step": 52650
    },
    {
      "entropy": 6.077514696121216,
      "epoch": 4.649858707170611,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00030302375721384266,
      "loss": 5.3111,
      "mean_token_accuracy": 0.2099177986383438,
      "num_tokens": 98166988.0,
      "step": 52655
    },
    {
      "entropy": 6.154877519607544,
      "epoch": 4.650300247262451,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030299251064731857,
      "loss": 5.324,
      "mean_token_accuracy": 0.21150567829608918,
      "num_tokens": 98176360.0,
      "step": 52660
    },
    {
      "entropy": 6.025263881683349,
      "epoch": 4.650741787354292,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003029612635324371,
      "loss": 5.2409,
      "mean_token_accuracy": 0.22028011679649354,
      "num_tokens": 98185475.0,
      "step": 52665
    },
    {
      "entropy": 5.966798734664917,
      "epoch": 4.651183327446132,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003029300158698105,
      "loss": 5.1869,
      "mean_token_accuracy": 0.22606071382761,
      "num_tokens": 98194863.0,
      "step": 52670
    },
    {
      "entropy": 6.056330347061158,
      "epoch": 4.651624867537972,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030289876766005067,
      "loss": 5.2137,
      "mean_token_accuracy": 0.2227076604962349,
      "num_tokens": 98203387.0,
      "step": 52675
    },
    {
      "entropy": 6.082599210739136,
      "epoch": 4.652066407629813,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00030286751890376985,
      "loss": 5.1764,
      "mean_token_accuracy": 0.222381092607975,
      "num_tokens": 98212365.0,
      "step": 52680
    },
    {
      "entropy": 6.038247108459473,
      "epoch": 4.652507947721653,
      "grad_norm": 1.40625,
      "learning_rate": 0.00030283626960158017,
      "loss": 5.2222,
      "mean_token_accuracy": 0.21502559036016464,
      "num_tokens": 98220663.0,
      "step": 52685
    },
    {
      "entropy": 5.965808963775634,
      "epoch": 4.652949487813493,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003028050197540939,
      "loss": 5.1714,
      "mean_token_accuracy": 0.23141634911298753,
      "num_tokens": 98229970.0,
      "step": 52690
    },
    {
      "entropy": 6.021854209899902,
      "epoch": 4.653391027905334,
      "grad_norm": 1.328125,
      "learning_rate": 0.00030277376936192296,
      "loss": 5.2003,
      "mean_token_accuracy": 0.22551226615905762,
      "num_tokens": 98239186.0,
      "step": 52695
    },
    {
      "entropy": 5.940040349960327,
      "epoch": 4.653832567997174,
      "grad_norm": 1.609375,
      "learning_rate": 0.00030274251842567977,
      "loss": 5.1662,
      "mean_token_accuracy": 0.2290506199002266,
      "num_tokens": 98248036.0,
      "step": 52700
    },
    {
      "entropy": 5.944090175628662,
      "epoch": 4.654274108089014,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030271126694597644,
      "loss": 5.1285,
      "mean_token_accuracy": 0.2341877982020378,
      "num_tokens": 98257044.0,
      "step": 52705
    },
    {
      "entropy": 6.026203346252442,
      "epoch": 4.654715648180855,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00030268001492342507,
      "loss": 5.1888,
      "mean_token_accuracy": 0.22506840527057648,
      "num_tokens": 98265529.0,
      "step": 52710
    },
    {
      "entropy": 6.020377111434937,
      "epoch": 4.655157188272695,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030264876235863806,
      "loss": 5.2102,
      "mean_token_accuracy": 0.23089172542095185,
      "num_tokens": 98274555.0,
      "step": 52715
    },
    {
      "entropy": 6.141514682769776,
      "epoch": 4.655598728364535,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030261750925222744,
      "loss": 5.3254,
      "mean_token_accuracy": 0.2175574004650116,
      "num_tokens": 98283311.0,
      "step": 52720
    },
    {
      "entropy": 6.049583053588867,
      "epoch": 4.656040268456376,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030258625560480553,
      "loss": 5.2566,
      "mean_token_accuracy": 0.2146862581372261,
      "num_tokens": 98292809.0,
      "step": 52725
    },
    {
      "entropy": 5.986868953704834,
      "epoch": 4.656481808548216,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003025550014169845,
      "loss": 5.18,
      "mean_token_accuracy": 0.22684442549943923,
      "num_tokens": 98302335.0,
      "step": 52730
    },
    {
      "entropy": 6.020013523101807,
      "epoch": 4.656923348640056,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003025237466893767,
      "loss": 5.2441,
      "mean_token_accuracy": 0.22258022874593736,
      "num_tokens": 98312116.0,
      "step": 52735
    },
    {
      "entropy": 5.969479417800903,
      "epoch": 4.657364888731896,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003024924914225944,
      "loss": 5.1737,
      "mean_token_accuracy": 0.23580623120069505,
      "num_tokens": 98321582.0,
      "step": 52740
    },
    {
      "entropy": 6.068461608886719,
      "epoch": 4.657806428823737,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030246123561724985,
      "loss": 5.2769,
      "mean_token_accuracy": 0.22048001140356063,
      "num_tokens": 98331065.0,
      "step": 52745
    },
    {
      "entropy": 6.054486989974976,
      "epoch": 4.658247968915577,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003024299792739553,
      "loss": 5.2543,
      "mean_token_accuracy": 0.2223142519593239,
      "num_tokens": 98339838.0,
      "step": 52750
    },
    {
      "entropy": 6.019526052474975,
      "epoch": 4.658689509007418,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000302398722393323,
      "loss": 5.2606,
      "mean_token_accuracy": 0.22116465866565704,
      "num_tokens": 98349664.0,
      "step": 52755
    },
    {
      "entropy": 6.006218767166137,
      "epoch": 4.659131049099258,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00030236746497596534,
      "loss": 5.2717,
      "mean_token_accuracy": 0.2225496307015419,
      "num_tokens": 98358807.0,
      "step": 52760
    },
    {
      "entropy": 6.043183517456055,
      "epoch": 4.659572589191098,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030233620702249463,
      "loss": 5.1921,
      "mean_token_accuracy": 0.23134994506835938,
      "num_tokens": 98367179.0,
      "step": 52765
    },
    {
      "entropy": 6.042371416091919,
      "epoch": 4.6600141292829385,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003023049485335232,
      "loss": 5.2451,
      "mean_token_accuracy": 0.21778919845819472,
      "num_tokens": 98375589.0,
      "step": 52770
    },
    {
      "entropy": 6.084535455703735,
      "epoch": 4.660455669374779,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00030227368950966327,
      "loss": 5.2487,
      "mean_token_accuracy": 0.21975739002227784,
      "num_tokens": 98385656.0,
      "step": 52775
    },
    {
      "entropy": 6.158698606491089,
      "epoch": 4.6608972094666195,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003022424299515273,
      "loss": 5.3165,
      "mean_token_accuracy": 0.2157350093126297,
      "num_tokens": 98395269.0,
      "step": 52780
    },
    {
      "entropy": 6.032495594024658,
      "epoch": 4.66133874955846,
      "grad_norm": 1.25,
      "learning_rate": 0.0003022111698597276,
      "loss": 5.2214,
      "mean_token_accuracy": 0.22260996103286743,
      "num_tokens": 98405076.0,
      "step": 52785
    },
    {
      "entropy": 6.00750412940979,
      "epoch": 4.6617802896503004,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003021799092348766,
      "loss": 5.2276,
      "mean_token_accuracy": 0.21911124736070633,
      "num_tokens": 98413981.0,
      "step": 52790
    },
    {
      "entropy": 5.982139205932617,
      "epoch": 4.6622218297421405,
      "grad_norm": 1.390625,
      "learning_rate": 0.00030214864807758657,
      "loss": 5.2534,
      "mean_token_accuracy": 0.2184978261590004,
      "num_tokens": 98423213.0,
      "step": 52795
    },
    {
      "entropy": 6.023605108261108,
      "epoch": 4.6626633698339806,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030211738638847007,
      "loss": 5.1981,
      "mean_token_accuracy": 0.2218171939253807,
      "num_tokens": 98432475.0,
      "step": 52800
    },
    {
      "entropy": 5.9174353122711185,
      "epoch": 4.6631049099258215,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00030208612416813933,
      "loss": 5.1153,
      "mean_token_accuracy": 0.2310450404882431,
      "num_tokens": 98441117.0,
      "step": 52805
    },
    {
      "entropy": 6.091187953948975,
      "epoch": 4.6635464500176615,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00030205486141720683,
      "loss": 5.3008,
      "mean_token_accuracy": 0.21341426372528077,
      "num_tokens": 98451117.0,
      "step": 52810
    },
    {
      "entropy": 6.11251277923584,
      "epoch": 4.663987990109502,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030202359813628497,
      "loss": 5.272,
      "mean_token_accuracy": 0.2195122167468071,
      "num_tokens": 98460097.0,
      "step": 52815
    },
    {
      "entropy": 6.040381383895874,
      "epoch": 4.6644295302013425,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003019923343259863,
      "loss": 5.203,
      "mean_token_accuracy": 0.22865511924028398,
      "num_tokens": 98469414.0,
      "step": 52820
    },
    {
      "entropy": 6.088582611083984,
      "epoch": 4.664871070293183,
      "grad_norm": 1.4375,
      "learning_rate": 0.00030196106998692305,
      "loss": 5.2812,
      "mean_token_accuracy": 0.21958217918872833,
      "num_tokens": 98478620.0,
      "step": 52825
    },
    {
      "entropy": 5.968236637115479,
      "epoch": 4.665312610385023,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030192980511970776,
      "loss": 5.1016,
      "mean_token_accuracy": 0.23240741342306137,
      "num_tokens": 98487514.0,
      "step": 52830
    },
    {
      "entropy": 5.996429061889648,
      "epoch": 4.665754150476864,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00030189853972495303,
      "loss": 5.1955,
      "mean_token_accuracy": 0.2314342737197876,
      "num_tokens": 98495424.0,
      "step": 52835
    },
    {
      "entropy": 5.992571067810059,
      "epoch": 4.666195690568704,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000301867273803271,
      "loss": 5.2895,
      "mean_token_accuracy": 0.22314245849847794,
      "num_tokens": 98505043.0,
      "step": 52840
    },
    {
      "entropy": 6.051070833206177,
      "epoch": 4.666637230660544,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003018360073552744,
      "loss": 5.2381,
      "mean_token_accuracy": 0.22499449104070662,
      "num_tokens": 98513710.0,
      "step": 52845
    },
    {
      "entropy": 6.008578491210938,
      "epoch": 4.667078770752385,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003018047403815758,
      "loss": 5.1883,
      "mean_token_accuracy": 0.22671865224838256,
      "num_tokens": 98522657.0,
      "step": 52850
    },
    {
      "entropy": 6.101751136779785,
      "epoch": 4.667520310844225,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003017734728827875,
      "loss": 5.276,
      "mean_token_accuracy": 0.21971959322690965,
      "num_tokens": 98531739.0,
      "step": 52855
    },
    {
      "entropy": 6.026125049591064,
      "epoch": 4.667961850936065,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00030174220485952206,
      "loss": 5.1669,
      "mean_token_accuracy": 0.2321300819516182,
      "num_tokens": 98540404.0,
      "step": 52860
    },
    {
      "entropy": 6.035815572738647,
      "epoch": 4.668403391027906,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00030171093631239216,
      "loss": 5.278,
      "mean_token_accuracy": 0.22243451923131943,
      "num_tokens": 98549317.0,
      "step": 52865
    },
    {
      "entropy": 6.001697063446045,
      "epoch": 4.668844931119746,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00030167966724201013,
      "loss": 5.201,
      "mean_token_accuracy": 0.22793205380439757,
      "num_tokens": 98559860.0,
      "step": 52870
    },
    {
      "entropy": 6.050649929046631,
      "epoch": 4.669286471211586,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003016483976489886,
      "loss": 5.2682,
      "mean_token_accuracy": 0.2190545156598091,
      "num_tokens": 98568636.0,
      "step": 52875
    },
    {
      "entropy": 5.964159250259399,
      "epoch": 4.669728011303427,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003016171275339401,
      "loss": 5.1907,
      "mean_token_accuracy": 0.23244341164827348,
      "num_tokens": 98577754.0,
      "step": 52880
    },
    {
      "entropy": 5.968952417373657,
      "epoch": 4.670169551395267,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030158585689747723,
      "loss": 5.1567,
      "mean_token_accuracy": 0.22815798819065095,
      "num_tokens": 98587319.0,
      "step": 52885
    },
    {
      "entropy": 6.001821088790893,
      "epoch": 4.670611091487107,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003015545857402126,
      "loss": 5.2516,
      "mean_token_accuracy": 0.22048345506191253,
      "num_tokens": 98596334.0,
      "step": 52890
    },
    {
      "entropy": 6.070473957061767,
      "epoch": 4.671052631578947,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030152331406275865,
      "loss": 5.3017,
      "mean_token_accuracy": 0.21438712477684022,
      "num_tokens": 98605408.0,
      "step": 52895
    },
    {
      "entropy": 6.058600378036499,
      "epoch": 4.671494171670788,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003014920418657281,
      "loss": 5.257,
      "mean_token_accuracy": 0.21855923980474473,
      "num_tokens": 98614124.0,
      "step": 52900
    },
    {
      "entropy": 6.043726444244385,
      "epoch": 4.671935711762628,
      "grad_norm": 1.4375,
      "learning_rate": 0.00030146076914973346,
      "loss": 5.2299,
      "mean_token_accuracy": 0.22058036774396897,
      "num_tokens": 98623491.0,
      "step": 52905
    },
    {
      "entropy": 6.073151063919068,
      "epoch": 4.672377251854469,
      "grad_norm": 1.515625,
      "learning_rate": 0.00030142949591538745,
      "loss": 5.318,
      "mean_token_accuracy": 0.2105977639555931,
      "num_tokens": 98634064.0,
      "step": 52910
    },
    {
      "entropy": 6.058108520507813,
      "epoch": 4.672818791946309,
      "grad_norm": 1.390625,
      "learning_rate": 0.00030139822216330265,
      "loss": 5.2251,
      "mean_token_accuracy": 0.23041686564683914,
      "num_tokens": 98643629.0,
      "step": 52915
    },
    {
      "entropy": 6.067491817474365,
      "epoch": 4.673260332038149,
      "grad_norm": 1.734375,
      "learning_rate": 0.00030136694789409163,
      "loss": 5.2075,
      "mean_token_accuracy": 0.22030248939990998,
      "num_tokens": 98654203.0,
      "step": 52920
    },
    {
      "entropy": 6.077182292938232,
      "epoch": 4.673701872129989,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003013356731083672,
      "loss": 5.2778,
      "mean_token_accuracy": 0.21625804305076599,
      "num_tokens": 98663731.0,
      "step": 52925
    },
    {
      "entropy": 6.056910419464112,
      "epoch": 4.67414341222183,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00030130439780674177,
      "loss": 5.2623,
      "mean_token_accuracy": 0.2222818359732628,
      "num_tokens": 98672609.0,
      "step": 52930
    },
    {
      "entropy": 6.06661810874939,
      "epoch": 4.67458495231367,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003012731219898283,
      "loss": 5.1918,
      "mean_token_accuracy": 0.22885549813508987,
      "num_tokens": 98681734.0,
      "step": 52935
    },
    {
      "entropy": 6.000039196014404,
      "epoch": 4.67502649240551,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003012418456582393,
      "loss": 5.24,
      "mean_token_accuracy": 0.22424447238445283,
      "num_tokens": 98690174.0,
      "step": 52940
    },
    {
      "entropy": 6.01829719543457,
      "epoch": 4.675468032497351,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003012105688125874,
      "loss": 5.2867,
      "mean_token_accuracy": 0.22039142698049546,
      "num_tokens": 98698958.0,
      "step": 52945
    },
    {
      "entropy": 6.134668064117432,
      "epoch": 4.675909572589191,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003011792914534853,
      "loss": 5.3618,
      "mean_token_accuracy": 0.20871088206768035,
      "num_tokens": 98708070.0,
      "step": 52950
    },
    {
      "entropy": 6.169289922714233,
      "epoch": 4.676351112681031,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00030114801358154593,
      "loss": 5.3692,
      "mean_token_accuracy": 0.2172737091779709,
      "num_tokens": 98717486.0,
      "step": 52955
    },
    {
      "entropy": 6.078028631210327,
      "epoch": 4.676792652772872,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030111673519738174,
      "loss": 5.2121,
      "mean_token_accuracy": 0.2277482494711876,
      "num_tokens": 98726582.0,
      "step": 52960
    },
    {
      "entropy": 6.001724529266357,
      "epoch": 4.677234192864712,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003010854563016056,
      "loss": 5.216,
      "mean_token_accuracy": 0.2321178749203682,
      "num_tokens": 98736844.0,
      "step": 52965
    },
    {
      "entropy": 6.0319713115692135,
      "epoch": 4.677675732956552,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003010541768948302,
      "loss": 5.305,
      "mean_token_accuracy": 0.21488004326820373,
      "num_tokens": 98747018.0,
      "step": 52970
    },
    {
      "entropy": 6.147765827178955,
      "epoch": 4.678117273048393,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00030102289697766827,
      "loss": 5.2472,
      "mean_token_accuracy": 0.2197589322924614,
      "num_tokens": 98756304.0,
      "step": 52975
    },
    {
      "entropy": 6.115427494049072,
      "epoch": 4.678558813140233,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00030099161655073263,
      "loss": 5.3444,
      "mean_token_accuracy": 0.21722585707902908,
      "num_tokens": 98765471.0,
      "step": 52980
    },
    {
      "entropy": 6.001520347595215,
      "epoch": 4.679000353232073,
      "grad_norm": 1.390625,
      "learning_rate": 0.00030096033561463593,
      "loss": 5.2597,
      "mean_token_accuracy": 0.22197602242231368,
      "num_tokens": 98774792.0,
      "step": 52985
    },
    {
      "entropy": 6.064077234268188,
      "epoch": 4.679441893323914,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00030092905416999105,
      "loss": 5.3035,
      "mean_token_accuracy": 0.21518024504184724,
      "num_tokens": 98783875.0,
      "step": 52990
    },
    {
      "entropy": 6.001260375976562,
      "epoch": 4.679883433415754,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030089777221741084,
      "loss": 5.2896,
      "mean_token_accuracy": 0.2198464810848236,
      "num_tokens": 98793860.0,
      "step": 52995
    },
    {
      "entropy": 6.053949403762817,
      "epoch": 4.680324973507594,
      "grad_norm": 1.21875,
      "learning_rate": 0.000300866489757508,
      "loss": 5.2846,
      "mean_token_accuracy": 0.2140503153204918,
      "num_tokens": 98804730.0,
      "step": 53000
    },
    {
      "entropy": 5.957002782821656,
      "epoch": 4.680766513599435,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003008352067908952,
      "loss": 5.1703,
      "mean_token_accuracy": 0.23311106562614442,
      "num_tokens": 98813414.0,
      "step": 53005
    },
    {
      "entropy": 6.058737134933471,
      "epoch": 4.681208053691275,
      "grad_norm": 1.46875,
      "learning_rate": 0.00030080392331818546,
      "loss": 5.208,
      "mean_token_accuracy": 0.22205926328897477,
      "num_tokens": 98822112.0,
      "step": 53010
    },
    {
      "entropy": 6.058559703826904,
      "epoch": 4.681649593783115,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030077263933999147,
      "loss": 5.2626,
      "mean_token_accuracy": 0.22366718649864198,
      "num_tokens": 98831673.0,
      "step": 53015
    },
    {
      "entropy": 6.038046503067017,
      "epoch": 4.682091133874955,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003007413548569262,
      "loss": 5.3361,
      "mean_token_accuracy": 0.2154768392443657,
      "num_tokens": 98840663.0,
      "step": 53020
    },
    {
      "entropy": 6.138865900039673,
      "epoch": 4.682532673966796,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00030071006986960244,
      "loss": 5.4075,
      "mean_token_accuracy": 0.21174502819776536,
      "num_tokens": 98850411.0,
      "step": 53025
    },
    {
      "entropy": 6.022521591186523,
      "epoch": 4.682974214058636,
      "grad_norm": 1.390625,
      "learning_rate": 0.00030067878437863303,
      "loss": 5.2322,
      "mean_token_accuracy": 0.2231406658887863,
      "num_tokens": 98859385.0,
      "step": 53030
    },
    {
      "entropy": 6.02979736328125,
      "epoch": 4.683415754150477,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030064749838463084,
      "loss": 5.2797,
      "mean_token_accuracy": 0.22219710499048234,
      "num_tokens": 98868581.0,
      "step": 53035
    },
    {
      "entropy": 5.954588174819946,
      "epoch": 4.683857294242317,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003006162118882087,
      "loss": 5.178,
      "mean_token_accuracy": 0.22361732721328736,
      "num_tokens": 98877991.0,
      "step": 53040
    },
    {
      "entropy": 6.033837223052979,
      "epoch": 4.684298834334157,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003005849248899795,
      "loss": 5.2365,
      "mean_token_accuracy": 0.224464550614357,
      "num_tokens": 98887372.0,
      "step": 53045
    },
    {
      "entropy": 6.098164510726929,
      "epoch": 4.6847403744259974,
      "grad_norm": 1.375,
      "learning_rate": 0.0003005536373905563,
      "loss": 5.2015,
      "mean_token_accuracy": 0.22214835733175278,
      "num_tokens": 98896824.0,
      "step": 53050
    },
    {
      "entropy": 6.0729845523834225,
      "epoch": 4.685181914517838,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00030052234939055176,
      "loss": 5.2705,
      "mean_token_accuracy": 0.22121940106153487,
      "num_tokens": 98905355.0,
      "step": 53055
    },
    {
      "entropy": 6.054833698272705,
      "epoch": 4.685623454609678,
      "grad_norm": 1.484375,
      "learning_rate": 0.00030049106089057904,
      "loss": 5.2371,
      "mean_token_accuracy": 0.22438955307006836,
      "num_tokens": 98913540.0,
      "step": 53060
    },
    {
      "entropy": 6.009105777740478,
      "epoch": 4.686064994701519,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030045977189125083,
      "loss": 5.2689,
      "mean_token_accuracy": 0.21889499574899673,
      "num_tokens": 98923374.0,
      "step": 53065
    },
    {
      "entropy": 6.04231071472168,
      "epoch": 4.686506534793359,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003004284823931802,
      "loss": 5.263,
      "mean_token_accuracy": 0.22578988075256348,
      "num_tokens": 98932435.0,
      "step": 53070
    },
    {
      "entropy": 6.0476484298706055,
      "epoch": 4.6869480748851995,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00030039719239698007,
      "loss": 5.2367,
      "mean_token_accuracy": 0.22596161514520646,
      "num_tokens": 98941196.0,
      "step": 53075
    },
    {
      "entropy": 6.067796516418457,
      "epoch": 4.6873896149770395,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003003659019032634,
      "loss": 5.3185,
      "mean_token_accuracy": 0.21702548265457153,
      "num_tokens": 98950877.0,
      "step": 53080
    },
    {
      "entropy": 6.12905969619751,
      "epoch": 4.6878311550688805,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00030033461091264315,
      "loss": 5.2563,
      "mean_token_accuracy": 0.21822184324264526,
      "num_tokens": 98959910.0,
      "step": 53085
    },
    {
      "entropy": 6.02737979888916,
      "epoch": 4.6882726951607205,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003003033194257322,
      "loss": 5.1864,
      "mean_token_accuracy": 0.22786992639303208,
      "num_tokens": 98968932.0,
      "step": 53090
    },
    {
      "entropy": 6.045907688140869,
      "epoch": 4.688714235252561,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00030027202744314365,
      "loss": 5.2978,
      "mean_token_accuracy": 0.21866774559020996,
      "num_tokens": 98978558.0,
      "step": 53095
    },
    {
      "entropy": 6.067431497573852,
      "epoch": 4.6891557753444015,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003002407349654905,
      "loss": 5.3274,
      "mean_token_accuracy": 0.2098805919289589,
      "num_tokens": 98988348.0,
      "step": 53100
    },
    {
      "entropy": 6.0871169090271,
      "epoch": 4.689597315436242,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003002094419933856,
      "loss": 5.3238,
      "mean_token_accuracy": 0.2164008915424347,
      "num_tokens": 98997443.0,
      "step": 53105
    },
    {
      "entropy": 6.075009822845459,
      "epoch": 4.690038855528082,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00030017814852744215,
      "loss": 5.3077,
      "mean_token_accuracy": 0.21381716132164003,
      "num_tokens": 99006286.0,
      "step": 53110
    },
    {
      "entropy": 6.031685209274292,
      "epoch": 4.690480395619923,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003001468545682731,
      "loss": 5.1892,
      "mean_token_accuracy": 0.22926014214754104,
      "num_tokens": 99015516.0,
      "step": 53115
    },
    {
      "entropy": 6.0748930931091305,
      "epoch": 4.690921935711763,
      "grad_norm": 1.46875,
      "learning_rate": 0.00030011556011649143,
      "loss": 5.2734,
      "mean_token_accuracy": 0.2202693447470665,
      "num_tokens": 99024172.0,
      "step": 53120
    },
    {
      "entropy": 6.0237109661102295,
      "epoch": 4.691363475803603,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003000842651727103,
      "loss": 5.1805,
      "mean_token_accuracy": 0.2326088845729828,
      "num_tokens": 99034822.0,
      "step": 53125
    },
    {
      "entropy": 6.018715143203735,
      "epoch": 4.691805015895444,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003000529697375426,
      "loss": 5.2118,
      "mean_token_accuracy": 0.22760390490293503,
      "num_tokens": 99044323.0,
      "step": 53130
    },
    {
      "entropy": 6.068567991256714,
      "epoch": 4.692246555987284,
      "grad_norm": 1.328125,
      "learning_rate": 0.00030002167381160144,
      "loss": 5.2566,
      "mean_token_accuracy": 0.21439773887395858,
      "num_tokens": 99053465.0,
      "step": 53135
    },
    {
      "entropy": 6.050239849090576,
      "epoch": 4.692688096079124,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029999037739550003,
      "loss": 5.3548,
      "mean_token_accuracy": 0.21218154430389405,
      "num_tokens": 99063174.0,
      "step": 53140
    },
    {
      "entropy": 6.024475240707398,
      "epoch": 4.693129636170965,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029995908048985125,
      "loss": 5.2457,
      "mean_token_accuracy": 0.2231910541653633,
      "num_tokens": 99072497.0,
      "step": 53145
    },
    {
      "entropy": 6.055537176132202,
      "epoch": 4.693571176262805,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002999277830952683,
      "loss": 5.1894,
      "mean_token_accuracy": 0.23151684701442718,
      "num_tokens": 99081640.0,
      "step": 53150
    },
    {
      "entropy": 6.11285367012024,
      "epoch": 4.694012716354645,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002998964852123643,
      "loss": 5.3079,
      "mean_token_accuracy": 0.21235403418540955,
      "num_tokens": 99092103.0,
      "step": 53155
    },
    {
      "entropy": 6.088313198089599,
      "epoch": 4.694454256446486,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029986518684175226,
      "loss": 5.2214,
      "mean_token_accuracy": 0.2256855472922325,
      "num_tokens": 99101019.0,
      "step": 53160
    },
    {
      "entropy": 6.076145172119141,
      "epoch": 4.694895796538326,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002998338879840454,
      "loss": 5.2354,
      "mean_token_accuracy": 0.22264487743377687,
      "num_tokens": 99110123.0,
      "step": 53165
    },
    {
      "entropy": 6.091257953643799,
      "epoch": 4.695337336630166,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002998025886398567,
      "loss": 5.2147,
      "mean_token_accuracy": 0.22294439673423766,
      "num_tokens": 99119329.0,
      "step": 53170
    },
    {
      "entropy": 6.039666843414307,
      "epoch": 4.695778876722006,
      "grad_norm": 1.375,
      "learning_rate": 0.0002997712888097995,
      "loss": 5.2799,
      "mean_token_accuracy": 0.22295041084289552,
      "num_tokens": 99128597.0,
      "step": 53175
    },
    {
      "entropy": 5.961242055892944,
      "epoch": 4.696220416813847,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002997399884944868,
      "loss": 5.2212,
      "mean_token_accuracy": 0.22845748960971832,
      "num_tokens": 99138274.0,
      "step": 53180
    },
    {
      "entropy": 6.1067389965057375,
      "epoch": 4.696661956905687,
      "grad_norm": 1.375,
      "learning_rate": 0.00029970868769453184,
      "loss": 5.2638,
      "mean_token_accuracy": 0.22594459503889083,
      "num_tokens": 99146983.0,
      "step": 53185
    },
    {
      "entropy": 6.070300006866455,
      "epoch": 4.697103496997528,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00029967738641054764,
      "loss": 5.2198,
      "mean_token_accuracy": 0.2250143066048622,
      "num_tokens": 99156080.0,
      "step": 53190
    },
    {
      "entropy": 6.108403825759888,
      "epoch": 4.697545037089368,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00029964608464314756,
      "loss": 5.252,
      "mean_token_accuracy": 0.22187752276659012,
      "num_tokens": 99165019.0,
      "step": 53195
    },
    {
      "entropy": 6.063605737686157,
      "epoch": 4.697986577181208,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002996147823929447,
      "loss": 5.2513,
      "mean_token_accuracy": 0.2149145022034645,
      "num_tokens": 99174105.0,
      "step": 53200
    },
    {
      "entropy": 6.047848033905029,
      "epoch": 4.698428117273048,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002995834796605523,
      "loss": 5.2288,
      "mean_token_accuracy": 0.21797591000795363,
      "num_tokens": 99183742.0,
      "step": 53205
    },
    {
      "entropy": 6.164152956008911,
      "epoch": 4.698869657364889,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002995521764465834,
      "loss": 5.3777,
      "mean_token_accuracy": 0.2114158883690834,
      "num_tokens": 99192851.0,
      "step": 53210
    },
    {
      "entropy": 6.031906414031982,
      "epoch": 4.699311197456729,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029952087275165134,
      "loss": 5.2276,
      "mean_token_accuracy": 0.22122025340795518,
      "num_tokens": 99202270.0,
      "step": 53215
    },
    {
      "entropy": 6.06518325805664,
      "epoch": 4.699752737548569,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029948956857636937,
      "loss": 5.2503,
      "mean_token_accuracy": 0.22361101359128951,
      "num_tokens": 99211459.0,
      "step": 53220
    },
    {
      "entropy": 6.102450656890869,
      "epoch": 4.70019427764041,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029945826392135065,
      "loss": 5.2575,
      "mean_token_accuracy": 0.21829914003610612,
      "num_tokens": 99222200.0,
      "step": 53225
    },
    {
      "entropy": 6.0338808536529545,
      "epoch": 4.70063581773225,
      "grad_norm": 1.234375,
      "learning_rate": 0.00029942695878720847,
      "loss": 5.1849,
      "mean_token_accuracy": 0.2205914229154587,
      "num_tokens": 99232018.0,
      "step": 53230
    },
    {
      "entropy": 5.998761081695557,
      "epoch": 4.70107735782409,
      "grad_norm": 1.3125,
      "learning_rate": 0.00029939565317455604,
      "loss": 5.1724,
      "mean_token_accuracy": 0.2272698312997818,
      "num_tokens": 99241138.0,
      "step": 53235
    },
    {
      "entropy": 5.972371435165405,
      "epoch": 4.701518897915931,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00029936434708400665,
      "loss": 5.1595,
      "mean_token_accuracy": 0.2222268745303154,
      "num_tokens": 99250296.0,
      "step": 53240
    },
    {
      "entropy": 6.098923683166504,
      "epoch": 4.701960438007771,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029933304051617363,
      "loss": 5.2577,
      "mean_token_accuracy": 0.218852636218071,
      "num_tokens": 99259503.0,
      "step": 53245
    },
    {
      "entropy": 5.984187936782837,
      "epoch": 4.702401978099611,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00029930173347167015,
      "loss": 5.1911,
      "mean_token_accuracy": 0.22357518076896668,
      "num_tokens": 99268305.0,
      "step": 53250
    },
    {
      "entropy": 6.088347482681274,
      "epoch": 4.702843518191452,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002992704259511094,
      "loss": 5.286,
      "mean_token_accuracy": 0.21995923221111296,
      "num_tokens": 99278031.0,
      "step": 53255
    },
    {
      "entropy": 6.096342754364014,
      "epoch": 4.703285058283292,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00029923911795510497,
      "loss": 5.3114,
      "mean_token_accuracy": 0.21238358765840532,
      "num_tokens": 99286957.0,
      "step": 53260
    },
    {
      "entropy": 6.12192120552063,
      "epoch": 4.703726598375132,
      "grad_norm": 1.484375,
      "learning_rate": 0.00029920780948426994,
      "loss": 5.2296,
      "mean_token_accuracy": 0.2238196074962616,
      "num_tokens": 99295396.0,
      "step": 53265
    },
    {
      "entropy": 6.030779075622559,
      "epoch": 4.704168138466973,
      "grad_norm": 1.40625,
      "learning_rate": 0.00029917650053921767,
      "loss": 5.2712,
      "mean_token_accuracy": 0.22029815167188643,
      "num_tokens": 99305256.0,
      "step": 53270
    },
    {
      "entropy": 6.044755458831787,
      "epoch": 4.704609678558813,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029914519112056155,
      "loss": 5.2441,
      "mean_token_accuracy": 0.2209386095404625,
      "num_tokens": 99315971.0,
      "step": 53275
    },
    {
      "entropy": 6.033116054534912,
      "epoch": 4.705051218650653,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002991138812289149,
      "loss": 5.2487,
      "mean_token_accuracy": 0.22456149160861968,
      "num_tokens": 99326264.0,
      "step": 53280
    },
    {
      "entropy": 6.096076536178589,
      "epoch": 4.705492758742494,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000299082570864891,
      "loss": 5.2913,
      "mean_token_accuracy": 0.21670921444892882,
      "num_tokens": 99335184.0,
      "step": 53285
    },
    {
      "entropy": 5.981477880477906,
      "epoch": 4.705934298834334,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002990512600291032,
      "loss": 5.156,
      "mean_token_accuracy": 0.23624032139778137,
      "num_tokens": 99343601.0,
      "step": 53290
    },
    {
      "entropy": 6.034730005264282,
      "epoch": 4.706375838926174,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002990199487221649,
      "loss": 5.3316,
      "mean_token_accuracy": 0.21727863252162932,
      "num_tokens": 99352912.0,
      "step": 53295
    },
    {
      "entropy": 6.073510074615479,
      "epoch": 4.706817379018015,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002989886369446894,
      "loss": 5.3282,
      "mean_token_accuracy": 0.21817300617694854,
      "num_tokens": 99361978.0,
      "step": 53300
    },
    {
      "entropy": 6.0853235721588135,
      "epoch": 4.707258919109855,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029895732469729025,
      "loss": 5.2881,
      "mean_token_accuracy": 0.21660494357347487,
      "num_tokens": 99370708.0,
      "step": 53305
    },
    {
      "entropy": 6.034816837310791,
      "epoch": 4.707700459201695,
      "grad_norm": 1.375,
      "learning_rate": 0.00029892601198058073,
      "loss": 5.1998,
      "mean_token_accuracy": 0.22670699059963226,
      "num_tokens": 99378830.0,
      "step": 53310
    },
    {
      "entropy": 6.027603197097778,
      "epoch": 4.708141999293536,
      "grad_norm": 1.375,
      "learning_rate": 0.00029889469879517425,
      "loss": 5.2769,
      "mean_token_accuracy": 0.22096991986036302,
      "num_tokens": 99387565.0,
      "step": 53315
    },
    {
      "entropy": 6.003838300704956,
      "epoch": 4.708583539385376,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002988633851416841,
      "loss": 5.2246,
      "mean_token_accuracy": 0.2235969826579094,
      "num_tokens": 99397131.0,
      "step": 53320
    },
    {
      "entropy": 5.947904253005982,
      "epoch": 4.709025079477216,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029883207102072393,
      "loss": 5.1176,
      "mean_token_accuracy": 0.23215089589357377,
      "num_tokens": 99406643.0,
      "step": 53325
    },
    {
      "entropy": 6.034085369110107,
      "epoch": 4.709466619569056,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000298800756432907,
      "loss": 5.1762,
      "mean_token_accuracy": 0.2275853395462036,
      "num_tokens": 99416109.0,
      "step": 53330
    },
    {
      "entropy": 6.036314249038696,
      "epoch": 4.709908159660897,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002987694413788468,
      "loss": 5.1887,
      "mean_token_accuracy": 0.22224732637405395,
      "num_tokens": 99424236.0,
      "step": 53335
    },
    {
      "entropy": 6.014817714691162,
      "epoch": 4.710349699752737,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002987381258591567,
      "loss": 5.1286,
      "mean_token_accuracy": 0.23295253217220308,
      "num_tokens": 99432605.0,
      "step": 53340
    },
    {
      "entropy": 5.997411108016967,
      "epoch": 4.710791239844578,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002987068098744503,
      "loss": 5.2241,
      "mean_token_accuracy": 0.22645166367292405,
      "num_tokens": 99440568.0,
      "step": 53345
    },
    {
      "entropy": 5.966975831985474,
      "epoch": 4.711232779936418,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002986754934253409,
      "loss": 5.1487,
      "mean_token_accuracy": 0.23388146311044694,
      "num_tokens": 99450025.0,
      "step": 53350
    },
    {
      "entropy": 6.039076471328736,
      "epoch": 4.7116743200282585,
      "grad_norm": 1.3125,
      "learning_rate": 0.00029864417651244207,
      "loss": 5.2151,
      "mean_token_accuracy": 0.22896508872509003,
      "num_tokens": 99459105.0,
      "step": 53355
    },
    {
      "entropy": 6.087052059173584,
      "epoch": 4.7121158601200985,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029861285913636727,
      "loss": 5.2988,
      "mean_token_accuracy": 0.21156253516674042,
      "num_tokens": 99469204.0,
      "step": 53360
    },
    {
      "entropy": 6.085298490524292,
      "epoch": 4.7125574002119395,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00029858154129773,
      "loss": 5.3049,
      "mean_token_accuracy": 0.219907546043396,
      "num_tokens": 99478519.0,
      "step": 53365
    },
    {
      "entropy": 6.025917387008667,
      "epoch": 4.7129989403037795,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002985502229971438,
      "loss": 5.2436,
      "mean_token_accuracy": 0.2217080295085907,
      "num_tokens": 99487182.0,
      "step": 53370
    },
    {
      "entropy": 6.142800998687744,
      "epoch": 4.71344048039562,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000298518904235222,
      "loss": 5.3553,
      "mean_token_accuracy": 0.21176586747169496,
      "num_tokens": 99496685.0,
      "step": 53375
    },
    {
      "entropy": 6.098040437698364,
      "epoch": 4.7138820204874605,
      "grad_norm": 1.375,
      "learning_rate": 0.00029848758501257837,
      "loss": 5.3029,
      "mean_token_accuracy": 0.2140083521604538,
      "num_tokens": 99505767.0,
      "step": 53380
    },
    {
      "entropy": 6.023417663574219,
      "epoch": 4.714323560579301,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00029845626532982626,
      "loss": 5.1956,
      "mean_token_accuracy": 0.2317446142435074,
      "num_tokens": 99514920.0,
      "step": 53385
    },
    {
      "entropy": 6.03801326751709,
      "epoch": 4.714765100671141,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029842494518757923,
      "loss": 5.2651,
      "mean_token_accuracy": 0.21602197140455245,
      "num_tokens": 99524054.0,
      "step": 53390
    },
    {
      "entropy": 5.91349983215332,
      "epoch": 4.715206640762982,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002983936245864509,
      "loss": 5.1682,
      "mean_token_accuracy": 0.22937270551919936,
      "num_tokens": 99533984.0,
      "step": 53395
    },
    {
      "entropy": 6.086732864379883,
      "epoch": 4.715648180854822,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002983623035270547,
      "loss": 5.2711,
      "mean_token_accuracy": 0.21697564721107482,
      "num_tokens": 99543685.0,
      "step": 53400
    },
    {
      "entropy": 5.934382390975952,
      "epoch": 4.716089720946662,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002983309820100043,
      "loss": 5.0394,
      "mean_token_accuracy": 0.24508453756570817,
      "num_tokens": 99551659.0,
      "step": 53405
    },
    {
      "entropy": 5.993869113922119,
      "epoch": 4.716531261038503,
      "grad_norm": 1.390625,
      "learning_rate": 0.00029829966003591327,
      "loss": 5.2695,
      "mean_token_accuracy": 0.21361242085695267,
      "num_tokens": 99561161.0,
      "step": 53410
    },
    {
      "entropy": 6.012023973464966,
      "epoch": 4.716972801130343,
      "grad_norm": 1.359375,
      "learning_rate": 0.00029826833760539505,
      "loss": 5.3012,
      "mean_token_accuracy": 0.224408520758152,
      "num_tokens": 99570925.0,
      "step": 53415
    },
    {
      "entropy": 6.084084510803223,
      "epoch": 4.717414341222183,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002982370147190634,
      "loss": 5.2429,
      "mean_token_accuracy": 0.220643450319767,
      "num_tokens": 99579562.0,
      "step": 53420
    },
    {
      "entropy": 6.099534559249878,
      "epoch": 4.717855881314024,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002982056913775319,
      "loss": 5.2687,
      "mean_token_accuracy": 0.22030719965696335,
      "num_tokens": 99588804.0,
      "step": 53425
    },
    {
      "entropy": 6.049863195419311,
      "epoch": 4.718297421405864,
      "grad_norm": 1.234375,
      "learning_rate": 0.000298174367581414,
      "loss": 5.2169,
      "mean_token_accuracy": 0.2144991233944893,
      "num_tokens": 99598388.0,
      "step": 53430
    },
    {
      "entropy": 6.096501731872559,
      "epoch": 4.718738961497704,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029814304333132346,
      "loss": 5.2819,
      "mean_token_accuracy": 0.2203215941786766,
      "num_tokens": 99607834.0,
      "step": 53435
    },
    {
      "entropy": 6.138031625747681,
      "epoch": 4.719180501589545,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029811171862787383,
      "loss": 5.3583,
      "mean_token_accuracy": 0.2056211695075035,
      "num_tokens": 99617944.0,
      "step": 53440
    },
    {
      "entropy": 6.061831331253051,
      "epoch": 4.719622041681385,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002980803934716788,
      "loss": 5.2022,
      "mean_token_accuracy": 0.22303376942873002,
      "num_tokens": 99627572.0,
      "step": 53445
    },
    {
      "entropy": 6.0494050025939945,
      "epoch": 4.720063581773225,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029804906786335203,
      "loss": 5.3086,
      "mean_token_accuracy": 0.21795739233493805,
      "num_tokens": 99637905.0,
      "step": 53450
    },
    {
      "entropy": 6.041937780380249,
      "epoch": 4.720505121865065,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002980177418035071,
      "loss": 5.3144,
      "mean_token_accuracy": 0.21304962933063507,
      "num_tokens": 99647471.0,
      "step": 53455
    },
    {
      "entropy": 5.999951696395874,
      "epoch": 4.720946661956906,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002979864152927577,
      "loss": 5.1453,
      "mean_token_accuracy": 0.23506402522325515,
      "num_tokens": 99657446.0,
      "step": 53460
    },
    {
      "entropy": 6.088510274887085,
      "epoch": 4.721388202048746,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002979550883317175,
      "loss": 5.2371,
      "mean_token_accuracy": 0.22192367017269135,
      "num_tokens": 99668106.0,
      "step": 53465
    },
    {
      "entropy": 6.114775562286377,
      "epoch": 4.721829742140587,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002979237609210001,
      "loss": 5.2592,
      "mean_token_accuracy": 0.2211780682206154,
      "num_tokens": 99677663.0,
      "step": 53470
    },
    {
      "entropy": 6.087085580825805,
      "epoch": 4.722271282232427,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002978924330612194,
      "loss": 5.2902,
      "mean_token_accuracy": 0.2159252718091011,
      "num_tokens": 99685653.0,
      "step": 53475
    },
    {
      "entropy": 6.08561692237854,
      "epoch": 4.722712822324267,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029786110475298885,
      "loss": 5.2409,
      "mean_token_accuracy": 0.22273146361112595,
      "num_tokens": 99693834.0,
      "step": 53480
    },
    {
      "entropy": 6.00920524597168,
      "epoch": 4.723154362416107,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029782977599692226,
      "loss": 5.1962,
      "mean_token_accuracy": 0.22859523743391036,
      "num_tokens": 99702287.0,
      "step": 53485
    },
    {
      "entropy": 6.027482509613037,
      "epoch": 4.723595902507948,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029779844679363345,
      "loss": 5.2848,
      "mean_token_accuracy": 0.2207554116845131,
      "num_tokens": 99711377.0,
      "step": 53490
    },
    {
      "entropy": 6.0272540092468265,
      "epoch": 4.724037442599788,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002977671171437359,
      "loss": 5.3024,
      "mean_token_accuracy": 0.22084389626979828,
      "num_tokens": 99721440.0,
      "step": 53495
    },
    {
      "entropy": 6.0334044933319095,
      "epoch": 4.724478982691628,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002977357870478436,
      "loss": 5.2486,
      "mean_token_accuracy": 0.22237810641527175,
      "num_tokens": 99730571.0,
      "step": 53500
    },
    {
      "entropy": 6.064307498931885,
      "epoch": 4.724920522783469,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029770445650657,
      "loss": 5.3174,
      "mean_token_accuracy": 0.21872489005327225,
      "num_tokens": 99740014.0,
      "step": 53505
    },
    {
      "entropy": 6.056478834152221,
      "epoch": 4.725362062875309,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029767312552052917,
      "loss": 5.2711,
      "mean_token_accuracy": 0.22048529535531997,
      "num_tokens": 99750218.0,
      "step": 53510
    },
    {
      "entropy": 6.0236677646636965,
      "epoch": 4.725803602967149,
      "grad_norm": 1.765625,
      "learning_rate": 0.0002976417940903346,
      "loss": 5.1598,
      "mean_token_accuracy": 0.22182633429765702,
      "num_tokens": 99759717.0,
      "step": 53515
    },
    {
      "entropy": 6.168697452545166,
      "epoch": 4.72624514305899,
      "grad_norm": 1.3125,
      "learning_rate": 0.00029761046221660025,
      "loss": 5.3017,
      "mean_token_accuracy": 0.2150585025548935,
      "num_tokens": 99769726.0,
      "step": 53520
    },
    {
      "entropy": 6.011307859420777,
      "epoch": 4.72668668315083,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002975791298999397,
      "loss": 5.2076,
      "mean_token_accuracy": 0.22562885731458665,
      "num_tokens": 99779473.0,
      "step": 53525
    },
    {
      "entropy": 6.059014463424683,
      "epoch": 4.72712822324267,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002975477971409669,
      "loss": 5.2253,
      "mean_token_accuracy": 0.22531527876853943,
      "num_tokens": 99789502.0,
      "step": 53530
    },
    {
      "entropy": 6.023262882232666,
      "epoch": 4.727569763334511,
      "grad_norm": 1.375,
      "learning_rate": 0.0002975164639402956,
      "loss": 5.1963,
      "mean_token_accuracy": 0.2260812610387802,
      "num_tokens": 99798036.0,
      "step": 53535
    },
    {
      "entropy": 6.004972791671753,
      "epoch": 4.728011303426351,
      "grad_norm": 1.3125,
      "learning_rate": 0.00029748513029853954,
      "loss": 5.1829,
      "mean_token_accuracy": 0.2289073348045349,
      "num_tokens": 99807622.0,
      "step": 53540
    },
    {
      "entropy": 6.004963493347168,
      "epoch": 4.728452843518191,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002974537962163126,
      "loss": 5.2049,
      "mean_token_accuracy": 0.22841427922248841,
      "num_tokens": 99817220.0,
      "step": 53545
    },
    {
      "entropy": 6.018936443328857,
      "epoch": 4.728894383610032,
      "grad_norm": 1.46875,
      "learning_rate": 0.00029742246169422846,
      "loss": 5.2261,
      "mean_token_accuracy": 0.21943080723285674,
      "num_tokens": 99826899.0,
      "step": 53550
    },
    {
      "entropy": 6.02133150100708,
      "epoch": 4.729335923701872,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002973911267329011,
      "loss": 5.2343,
      "mean_token_accuracy": 0.22081039547920228,
      "num_tokens": 99836104.0,
      "step": 53555
    },
    {
      "entropy": 6.062137651443481,
      "epoch": 4.729777463793712,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002973597913329444,
      "loss": 5.2238,
      "mean_token_accuracy": 0.22286295145750046,
      "num_tokens": 99845155.0,
      "step": 53560
    },
    {
      "entropy": 5.998758554458618,
      "epoch": 4.730219003885553,
      "grad_norm": 1.3125,
      "learning_rate": 0.000297328455494972,
      "loss": 5.1701,
      "mean_token_accuracy": 0.232456012070179,
      "num_tokens": 99854090.0,
      "step": 53565
    },
    {
      "entropy": 6.097750329971314,
      "epoch": 4.730660543977393,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002972971192195979,
      "loss": 5.2701,
      "mean_token_accuracy": 0.2182851701974869,
      "num_tokens": 99864234.0,
      "step": 53570
    },
    {
      "entropy": 6.081766986846924,
      "epoch": 4.731102084069233,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002972657825074359,
      "loss": 5.2271,
      "mean_token_accuracy": 0.22095700055360795,
      "num_tokens": 99872895.0,
      "step": 53575
    },
    {
      "entropy": 5.952299785614014,
      "epoch": 4.731543624161074,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029723444535909987,
      "loss": 5.1495,
      "mean_token_accuracy": 0.22851888388395308,
      "num_tokens": 99883166.0,
      "step": 53580
    },
    {
      "entropy": 6.059799432754517,
      "epoch": 4.731985164252914,
      "grad_norm": 1.40625,
      "learning_rate": 0.00029720310777520375,
      "loss": 5.205,
      "mean_token_accuracy": 0.2220145583152771,
      "num_tokens": 99892125.0,
      "step": 53585
    },
    {
      "entropy": 6.136021518707276,
      "epoch": 4.732426704344754,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029717176975636134,
      "loss": 5.3475,
      "mean_token_accuracy": 0.2138415440917015,
      "num_tokens": 99901359.0,
      "step": 53590
    },
    {
      "entropy": 6.055138731002808,
      "epoch": 4.732868244436595,
      "grad_norm": 1.25,
      "learning_rate": 0.0002971404313031865,
      "loss": 5.2882,
      "mean_token_accuracy": 0.21640869677066804,
      "num_tokens": 99910414.0,
      "step": 53595
    },
    {
      "entropy": 6.081405687332153,
      "epoch": 4.733309784528435,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002971090924162934,
      "loss": 5.3677,
      "mean_token_accuracy": 0.21419043987989425,
      "num_tokens": 99920227.0,
      "step": 53600
    },
    {
      "entropy": 6.132946109771728,
      "epoch": 4.733751324620275,
      "grad_norm": 1.375,
      "learning_rate": 0.0002970777530962956,
      "loss": 5.2946,
      "mean_token_accuracy": 0.21720935106277467,
      "num_tokens": 99929065.0,
      "step": 53605
    },
    {
      "entropy": 6.065053272247314,
      "epoch": 4.734192864712115,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002970464133438072,
      "loss": 5.201,
      "mean_token_accuracy": 0.22371379286050797,
      "num_tokens": 99938266.0,
      "step": 53610
    },
    {
      "entropy": 6.100692892074585,
      "epoch": 4.734634404803956,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00029701507315944197,
      "loss": 5.2686,
      "mean_token_accuracy": 0.2202488139271736,
      "num_tokens": 99947306.0,
      "step": 53615
    },
    {
      "entropy": 6.089151239395141,
      "epoch": 4.735075944895796,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029698373254381425,
      "loss": 5.3106,
      "mean_token_accuracy": 0.21798589080572128,
      "num_tokens": 99957291.0,
      "step": 53620
    },
    {
      "entropy": 6.153366994857788,
      "epoch": 4.735517484987637,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002969523914975375,
      "loss": 5.2994,
      "mean_token_accuracy": 0.21365876495838165,
      "num_tokens": 99966088.0,
      "step": 53625
    },
    {
      "entropy": 5.9509196281433105,
      "epoch": 4.735959025079477,
      "grad_norm": 1.25,
      "learning_rate": 0.000296921050021226,
      "loss": 5.1359,
      "mean_token_accuracy": 0.23207794725894929,
      "num_tokens": 99975387.0,
      "step": 53630
    },
    {
      "entropy": 6.013784551620484,
      "epoch": 4.7364005651713175,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00029688970811549354,
      "loss": 5.1286,
      "mean_token_accuracy": 0.23167888969182968,
      "num_tokens": 99984725.0,
      "step": 53635
    },
    {
      "entropy": 6.057236003875732,
      "epoch": 4.7368421052631575,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00029685836578095413,
      "loss": 5.2319,
      "mean_token_accuracy": 0.21946544498205184,
      "num_tokens": 99993520.0,
      "step": 53640
    },
    {
      "entropy": 6.027833366394043,
      "epoch": 4.7372836453549985,
      "grad_norm": 1.234375,
      "learning_rate": 0.00029682702301822176,
      "loss": 5.2471,
      "mean_token_accuracy": 0.2219287171959877,
      "num_tokens": 100003602.0,
      "step": 53645
    },
    {
      "entropy": 6.130818700790405,
      "epoch": 4.7377251854468385,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002967956798279105,
      "loss": 5.2885,
      "mean_token_accuracy": 0.21092484891414642,
      "num_tokens": 100013010.0,
      "step": 53650
    },
    {
      "entropy": 6.088682460784912,
      "epoch": 4.738166725538679,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029676433621063424,
      "loss": 5.2706,
      "mean_token_accuracy": 0.21818211674690247,
      "num_tokens": 100022756.0,
      "step": 53655
    },
    {
      "entropy": 5.949555921554565,
      "epoch": 4.7386082656305195,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000296732992167007,
      "loss": 5.0702,
      "mean_token_accuracy": 0.23819994777441025,
      "num_tokens": 100031848.0,
      "step": 53660
    },
    {
      "entropy": 6.041798210144043,
      "epoch": 4.73904980572236,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002967016476976428,
      "loss": 5.3076,
      "mean_token_accuracy": 0.21729132086038588,
      "num_tokens": 100042388.0,
      "step": 53665
    },
    {
      "entropy": 6.0540547370910645,
      "epoch": 4.7394913458142,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002966703028031557,
      "loss": 5.1421,
      "mean_token_accuracy": 0.23259084671735764,
      "num_tokens": 100051927.0,
      "step": 53670
    },
    {
      "entropy": 5.973486137390137,
      "epoch": 4.739932885906041,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029663895748415967,
      "loss": 5.0717,
      "mean_token_accuracy": 0.24304133504629136,
      "num_tokens": 100061419.0,
      "step": 53675
    },
    {
      "entropy": 6.0576738834381105,
      "epoch": 4.740374425997881,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002966076117412687,
      "loss": 5.2614,
      "mean_token_accuracy": 0.2208349213004112,
      "num_tokens": 100070737.0,
      "step": 53680
    },
    {
      "entropy": 6.037282848358155,
      "epoch": 4.740815966089721,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029657626557509696,
      "loss": 5.1908,
      "mean_token_accuracy": 0.22323861867189407,
      "num_tokens": 100079553.0,
      "step": 53685
    },
    {
      "entropy": 6.058582067489624,
      "epoch": 4.741257506181562,
      "grad_norm": 1.359375,
      "learning_rate": 0.00029654491898625844,
      "loss": 5.2344,
      "mean_token_accuracy": 0.2204655349254608,
      "num_tokens": 100088932.0,
      "step": 53690
    },
    {
      "entropy": 6.035842800140381,
      "epoch": 4.741699046273402,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002965135719753673,
      "loss": 5.2231,
      "mean_token_accuracy": 0.2230399489402771,
      "num_tokens": 100097469.0,
      "step": 53695
    },
    {
      "entropy": 6.056171178817749,
      "epoch": 4.742140586365242,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029648222454303736,
      "loss": 5.2847,
      "mean_token_accuracy": 0.22282267659902572,
      "num_tokens": 100107894.0,
      "step": 53700
    },
    {
      "entropy": 5.96567850112915,
      "epoch": 4.742582126457083,
      "grad_norm": 1.328125,
      "learning_rate": 0.000296450876689883,
      "loss": 5.1438,
      "mean_token_accuracy": 0.23111927062273024,
      "num_tokens": 100117547.0,
      "step": 53705
    },
    {
      "entropy": 6.079860877990723,
      "epoch": 4.743023666548923,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002964195284165181,
      "loss": 5.2835,
      "mean_token_accuracy": 0.2203971579670906,
      "num_tokens": 100127122.0,
      "step": 53710
    },
    {
      "entropy": 6.07754693031311,
      "epoch": 4.743465206640763,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002963881797235568,
      "loss": 5.2943,
      "mean_token_accuracy": 0.21852121651172637,
      "num_tokens": 100136516.0,
      "step": 53715
    },
    {
      "entropy": 6.023001718521118,
      "epoch": 4.743906746732604,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00029635683061161335,
      "loss": 5.1804,
      "mean_token_accuracy": 0.23211707770824433,
      "num_tokens": 100146219.0,
      "step": 53720
    },
    {
      "entropy": 6.138419103622437,
      "epoch": 4.744348286824444,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002963254810813016,
      "loss": 5.3627,
      "mean_token_accuracy": 0.21029507070779802,
      "num_tokens": 100156502.0,
      "step": 53725
    },
    {
      "entropy": 5.979936790466309,
      "epoch": 4.744789826916284,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00029629413113323584,
      "loss": 5.196,
      "mean_token_accuracy": 0.2236461043357849,
      "num_tokens": 100164961.0,
      "step": 53730
    },
    {
      "entropy": 6.070103359222412,
      "epoch": 4.745231367008124,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029626278076803013,
      "loss": 5.3291,
      "mean_token_accuracy": 0.21382853537797927,
      "num_tokens": 100173850.0,
      "step": 53735
    },
    {
      "entropy": 6.1412910461425785,
      "epoch": 4.745672907099965,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002962314299862987,
      "loss": 5.2489,
      "mean_token_accuracy": 0.22272955477237702,
      "num_tokens": 100182739.0,
      "step": 53740
    },
    {
      "entropy": 6.140345335006714,
      "epoch": 4.746114447191805,
      "grad_norm": 1.4375,
      "learning_rate": 0.00029620007878865555,
      "loss": 5.2637,
      "mean_token_accuracy": 0.2158778578042984,
      "num_tokens": 100191653.0,
      "step": 53745
    },
    {
      "entropy": 6.051365709304809,
      "epoch": 4.746555987283646,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00029616872717571496,
      "loss": 5.2486,
      "mean_token_accuracy": 0.22736181765794755,
      "num_tokens": 100200137.0,
      "step": 53750
    },
    {
      "entropy": 5.959221649169922,
      "epoch": 4.746997527375486,
      "grad_norm": 1.328125,
      "learning_rate": 0.000296137375148091,
      "loss": 5.189,
      "mean_token_accuracy": 0.23099228292703627,
      "num_tokens": 100210416.0,
      "step": 53755
    },
    {
      "entropy": 6.0468991756439205,
      "epoch": 4.747439067467326,
      "grad_norm": 1.4375,
      "learning_rate": 0.00029610602270639804,
      "loss": 5.212,
      "mean_token_accuracy": 0.22765135616064072,
      "num_tokens": 100219671.0,
      "step": 53760
    },
    {
      "entropy": 5.9769844055175785,
      "epoch": 4.747880607559166,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029607466985125,
      "loss": 5.1874,
      "mean_token_accuracy": 0.2221895709633827,
      "num_tokens": 100229168.0,
      "step": 53765
    },
    {
      "entropy": 6.08186092376709,
      "epoch": 4.748322147651007,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029604331658326114,
      "loss": 5.3461,
      "mean_token_accuracy": 0.21253465563058854,
      "num_tokens": 100238624.0,
      "step": 53770
    },
    {
      "entropy": 6.017095947265625,
      "epoch": 4.748763687742847,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002960119629030456,
      "loss": 5.2087,
      "mean_token_accuracy": 0.23019490987062455,
      "num_tokens": 100247663.0,
      "step": 53775
    },
    {
      "entropy": 6.090999031066895,
      "epoch": 4.749205227834687,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002959806088112178,
      "loss": 5.2958,
      "mean_token_accuracy": 0.21570903360843657,
      "num_tokens": 100256631.0,
      "step": 53780
    },
    {
      "entropy": 5.991155624389648,
      "epoch": 4.749646767926528,
      "grad_norm": 1.375,
      "learning_rate": 0.00029594925430839183,
      "loss": 5.1025,
      "mean_token_accuracy": 0.22922645807266234,
      "num_tokens": 100265744.0,
      "step": 53785
    },
    {
      "entropy": 6.013508319854736,
      "epoch": 4.750088308018368,
      "grad_norm": 1.390625,
      "learning_rate": 0.00029591789939518174,
      "loss": 5.2201,
      "mean_token_accuracy": 0.22401549369096757,
      "num_tokens": 100275125.0,
      "step": 53790
    },
    {
      "entropy": 6.039568281173706,
      "epoch": 4.750529848110208,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029588654407220193,
      "loss": 5.1818,
      "mean_token_accuracy": 0.2305081531405449,
      "num_tokens": 100284049.0,
      "step": 53795
    },
    {
      "entropy": 6.11863431930542,
      "epoch": 4.750971388202049,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002958551883400667,
      "loss": 5.2789,
      "mean_token_accuracy": 0.2171795293688774,
      "num_tokens": 100293533.0,
      "step": 53800
    },
    {
      "entropy": 6.094614267349243,
      "epoch": 4.751412928293889,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002958238321993901,
      "loss": 5.2663,
      "mean_token_accuracy": 0.21818250119686128,
      "num_tokens": 100302907.0,
      "step": 53805
    },
    {
      "entropy": 6.041608572006226,
      "epoch": 4.751854468385729,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002957924756507864,
      "loss": 5.2152,
      "mean_token_accuracy": 0.22116616368293762,
      "num_tokens": 100311764.0,
      "step": 53810
    },
    {
      "entropy": 6.023539638519287,
      "epoch": 4.75229600847757,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002957611186948701,
      "loss": 5.2317,
      "mean_token_accuracy": 0.21935751289129257,
      "num_tokens": 100320561.0,
      "step": 53815
    },
    {
      "entropy": 6.049736452102661,
      "epoch": 4.75273754856941,
      "grad_norm": 1.25,
      "learning_rate": 0.00029572976133225516,
      "loss": 5.2535,
      "mean_token_accuracy": 0.22300448566675185,
      "num_tokens": 100329875.0,
      "step": 53820
    },
    {
      "entropy": 5.980596494674683,
      "epoch": 4.75317908866125,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002956984035635561,
      "loss": 5.1966,
      "mean_token_accuracy": 0.2236359804868698,
      "num_tokens": 100339283.0,
      "step": 53825
    },
    {
      "entropy": 5.990444612503052,
      "epoch": 4.753620628753091,
      "grad_norm": 1.390625,
      "learning_rate": 0.00029566704538938696,
      "loss": 5.1348,
      "mean_token_accuracy": 0.23094189018011094,
      "num_tokens": 100348942.0,
      "step": 53830
    },
    {
      "entropy": 6.048077154159546,
      "epoch": 4.754062168844931,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002956356868103623,
      "loss": 5.2755,
      "mean_token_accuracy": 0.22353435605764388,
      "num_tokens": 100358325.0,
      "step": 53835
    },
    {
      "entropy": 5.996874523162842,
      "epoch": 4.754503708936771,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002956043278270962,
      "loss": 5.1575,
      "mean_token_accuracy": 0.23560840785503387,
      "num_tokens": 100367549.0,
      "step": 53840
    },
    {
      "entropy": 6.031670045852661,
      "epoch": 4.754945249028612,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029557296844020305,
      "loss": 5.2613,
      "mean_token_accuracy": 0.2201864704489708,
      "num_tokens": 100376974.0,
      "step": 53845
    },
    {
      "entropy": 5.997309589385987,
      "epoch": 4.755386789120452,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002955416086502971,
      "loss": 5.2506,
      "mean_token_accuracy": 0.22604839950799943,
      "num_tokens": 100386642.0,
      "step": 53850
    },
    {
      "entropy": 6.070215940475464,
      "epoch": 4.755828329212292,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029551024845799275,
      "loss": 5.2997,
      "mean_token_accuracy": 0.2192176416516304,
      "num_tokens": 100396874.0,
      "step": 53855
    },
    {
      "entropy": 5.98539137840271,
      "epoch": 4.756269869304133,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002954788878639042,
      "loss": 5.155,
      "mean_token_accuracy": 0.23192679286003112,
      "num_tokens": 100405952.0,
      "step": 53860
    },
    {
      "entropy": 6.102971792221069,
      "epoch": 4.756711409395973,
      "grad_norm": 1.265625,
      "learning_rate": 0.000295447526868646,
      "loss": 5.3686,
      "mean_token_accuracy": 0.21607712060213088,
      "num_tokens": 100415900.0,
      "step": 53865
    },
    {
      "entropy": 6.041302299499511,
      "epoch": 4.757152949487813,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002954161654728324,
      "loss": 5.2738,
      "mean_token_accuracy": 0.2265323966741562,
      "num_tokens": 100425006.0,
      "step": 53870
    },
    {
      "entropy": 6.033534097671509,
      "epoch": 4.757594489579654,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002953848036770777,
      "loss": 5.2325,
      "mean_token_accuracy": 0.2192823112010956,
      "num_tokens": 100433686.0,
      "step": 53875
    },
    {
      "entropy": 5.954775285720825,
      "epoch": 4.758036029671494,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00029535344148199624,
      "loss": 5.0762,
      "mean_token_accuracy": 0.23619128465652467,
      "num_tokens": 100442407.0,
      "step": 53880
    },
    {
      "entropy": 5.980261707305909,
      "epoch": 4.758477569763334,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029532207888820244,
      "loss": 5.1702,
      "mean_token_accuracy": 0.2294472798705101,
      "num_tokens": 100451476.0,
      "step": 53885
    },
    {
      "entropy": 6.02455506324768,
      "epoch": 4.758919109855174,
      "grad_norm": 1.359375,
      "learning_rate": 0.00029529071589631074,
      "loss": 5.214,
      "mean_token_accuracy": 0.22434910535812377,
      "num_tokens": 100460689.0,
      "step": 53890
    },
    {
      "entropy": 6.012349796295166,
      "epoch": 4.759360649947015,
      "grad_norm": 1.359375,
      "learning_rate": 0.00029525935250693534,
      "loss": 5.1824,
      "mean_token_accuracy": 0.22075506001710893,
      "num_tokens": 100469573.0,
      "step": 53895
    },
    {
      "entropy": 5.979785919189453,
      "epoch": 4.759802190038855,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002952279887206909,
      "loss": 5.1985,
      "mean_token_accuracy": 0.22657249122858047,
      "num_tokens": 100478864.0,
      "step": 53900
    },
    {
      "entropy": 6.061070108413697,
      "epoch": 4.760243730130696,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002951966245381916,
      "loss": 5.2991,
      "mean_token_accuracy": 0.2158491402864456,
      "num_tokens": 100488147.0,
      "step": 53905
    },
    {
      "entropy": 6.062256336212158,
      "epoch": 4.760685270222536,
      "grad_norm": 1.390625,
      "learning_rate": 0.00029516525996005187,
      "loss": 5.2818,
      "mean_token_accuracy": 0.2207522824406624,
      "num_tokens": 100497255.0,
      "step": 53910
    },
    {
      "entropy": 6.086244344711304,
      "epoch": 4.7611268103143765,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029513389498688625,
      "loss": 5.2924,
      "mean_token_accuracy": 0.21401480883359908,
      "num_tokens": 100506114.0,
      "step": 53915
    },
    {
      "entropy": 6.062914419174194,
      "epoch": 4.7615683504062165,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000295102529619309,
      "loss": 5.2202,
      "mean_token_accuracy": 0.2217021256685257,
      "num_tokens": 100515352.0,
      "step": 53920
    },
    {
      "entropy": 6.0575785636901855,
      "epoch": 4.7620098904980575,
      "grad_norm": 1.390625,
      "learning_rate": 0.00029507116385793465,
      "loss": 5.1633,
      "mean_token_accuracy": 0.2305621936917305,
      "num_tokens": 100523991.0,
      "step": 53925
    },
    {
      "entropy": 6.043101119995117,
      "epoch": 4.7624514305898975,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002950397977033776,
      "loss": 5.3028,
      "mean_token_accuracy": 0.21457558870315552,
      "num_tokens": 100534460.0,
      "step": 53930
    },
    {
      "entropy": 6.048871850967407,
      "epoch": 4.762892970681738,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002950084311562524,
      "loss": 5.2238,
      "mean_token_accuracy": 0.2245767518877983,
      "num_tokens": 100543105.0,
      "step": 53935
    },
    {
      "entropy": 6.039324855804443,
      "epoch": 4.7633345107735785,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002949770642171733,
      "loss": 5.2454,
      "mean_token_accuracy": 0.21902401596307755,
      "num_tokens": 100552634.0,
      "step": 53940
    },
    {
      "entropy": 6.050445985794068,
      "epoch": 4.7637760508654186,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029494569688675486,
      "loss": 5.2665,
      "mean_token_accuracy": 0.21510395109653474,
      "num_tokens": 100562116.0,
      "step": 53945
    },
    {
      "entropy": 6.057997989654541,
      "epoch": 4.764217590957259,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00029491432916561167,
      "loss": 5.2651,
      "mean_token_accuracy": 0.22433690428733827,
      "num_tokens": 100571112.0,
      "step": 53950
    },
    {
      "entropy": 6.0962341785430905,
      "epoch": 4.7646591310490995,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00029488296105435806,
      "loss": 5.2081,
      "mean_token_accuracy": 0.22555013000965118,
      "num_tokens": 100581061.0,
      "step": 53955
    },
    {
      "entropy": 6.046308755874634,
      "epoch": 4.76510067114094,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00029485159255360856,
      "loss": 5.2465,
      "mean_token_accuracy": 0.2172827184200287,
      "num_tokens": 100589991.0,
      "step": 53960
    },
    {
      "entropy": 6.041256999969482,
      "epoch": 4.76554221123278,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002948202236639776,
      "loss": 5.2373,
      "mean_token_accuracy": 0.22485012114048003,
      "num_tokens": 100598729.0,
      "step": 53965
    },
    {
      "entropy": 6.066609716415405,
      "epoch": 4.765983751324621,
      "grad_norm": 1.28125,
      "learning_rate": 0.00029478885438607977,
      "loss": 5.231,
      "mean_token_accuracy": 0.2191901057958603,
      "num_tokens": 100607541.0,
      "step": 53970
    },
    {
      "entropy": 6.082042598724366,
      "epoch": 4.766425291416461,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002947574847205295,
      "loss": 5.2453,
      "mean_token_accuracy": 0.2183339610695839,
      "num_tokens": 100616427.0,
      "step": 53975
    },
    {
      "entropy": 6.015865421295166,
      "epoch": 4.766866831508301,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002947261146679414,
      "loss": 5.2368,
      "mean_token_accuracy": 0.22130235582590102,
      "num_tokens": 100626427.0,
      "step": 53980
    },
    {
      "entropy": 5.952264213562012,
      "epoch": 4.767308371600142,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002946947442289299,
      "loss": 5.11,
      "mean_token_accuracy": 0.23140606433153152,
      "num_tokens": 100634955.0,
      "step": 53985
    },
    {
      "entropy": 6.067717218399048,
      "epoch": 4.767749911691982,
      "grad_norm": 1.5,
      "learning_rate": 0.0002946633734041095,
      "loss": 5.3198,
      "mean_token_accuracy": 0.21638166010379792,
      "num_tokens": 100644994.0,
      "step": 53990
    },
    {
      "entropy": 6.081340885162353,
      "epoch": 4.768191451783822,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00029463200219409486,
      "loss": 5.2229,
      "mean_token_accuracy": 0.22263213545083999,
      "num_tokens": 100653842.0,
      "step": 53995
    },
    {
      "entropy": 6.076512718200684,
      "epoch": 4.768632991875663,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029460063059950047,
      "loss": 5.2233,
      "mean_token_accuracy": 0.228511743247509,
      "num_tokens": 100664177.0,
      "step": 54000
    },
    {
      "epoch": 4.768632991875663,
      "eval_entropy": 5.766466747998921,
      "eval_loss": 5.3104400634765625,
      "eval_mean_token_accuracy": 0.22655995563094505,
      "eval_num_tokens": 100664177.0,
      "eval_runtime": 20.0347,
      "eval_samples_per_second": 1762.688,
      "eval_steps_per_second": 220.367,
      "step": 54000
    },
    {
      "entropy": 6.114556884765625,
      "epoch": 4.769074531967503,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002945692586209408,
      "loss": 5.3771,
      "mean_token_accuracy": 0.21228503286838532,
      "num_tokens": 100673669.0,
      "step": 54005
    },
    {
      "entropy": 6.014226865768433,
      "epoch": 4.769516072059343,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002945378862590306,
      "loss": 5.2103,
      "mean_token_accuracy": 0.22832316607236863,
      "num_tokens": 100683001.0,
      "step": 54010
    },
    {
      "entropy": 6.075849103927612,
      "epoch": 4.769957612151183,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029450651351438427,
      "loss": 5.2625,
      "mean_token_accuracy": 0.21684684455394745,
      "num_tokens": 100692466.0,
      "step": 54015
    },
    {
      "entropy": 6.113894939422607,
      "epoch": 4.770399152243024,
      "grad_norm": 1.25,
      "learning_rate": 0.00029447514038761645,
      "loss": 5.2901,
      "mean_token_accuracy": 0.21865109652280806,
      "num_tokens": 100702092.0,
      "step": 54020
    },
    {
      "entropy": 6.047034788131714,
      "epoch": 4.770840692334864,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002944437668793417,
      "loss": 5.2483,
      "mean_token_accuracy": 0.2242418959736824,
      "num_tokens": 100712727.0,
      "step": 54025
    },
    {
      "entropy": 5.961920738220215,
      "epoch": 4.771282232426705,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002944123929901746,
      "loss": 5.1406,
      "mean_token_accuracy": 0.23077901303768159,
      "num_tokens": 100722095.0,
      "step": 54030
    },
    {
      "entropy": 6.070462799072265,
      "epoch": 4.771723772518545,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002943810187207298,
      "loss": 5.3113,
      "mean_token_accuracy": 0.21052516996860504,
      "num_tokens": 100730731.0,
      "step": 54035
    },
    {
      "entropy": 5.927001571655273,
      "epoch": 4.772165312610385,
      "grad_norm": 1.5,
      "learning_rate": 0.0002943496440716218,
      "loss": 5.1033,
      "mean_token_accuracy": 0.23262849152088166,
      "num_tokens": 100739472.0,
      "step": 54040
    },
    {
      "entropy": 6.100700044631958,
      "epoch": 4.772606852702225,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002943182690434653,
      "loss": 5.2839,
      "mean_token_accuracy": 0.21497849375009537,
      "num_tokens": 100749324.0,
      "step": 54045
    },
    {
      "entropy": 6.078263998031616,
      "epoch": 4.773048392794066,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002942868936368749,
      "loss": 5.2269,
      "mean_token_accuracy": 0.22188192307949067,
      "num_tokens": 100758276.0,
      "step": 54050
    },
    {
      "entropy": 6.047050189971924,
      "epoch": 4.773489932885906,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002942555178524652,
      "loss": 5.2203,
      "mean_token_accuracy": 0.22716485112905502,
      "num_tokens": 100766686.0,
      "step": 54055
    },
    {
      "entropy": 6.032462024688721,
      "epoch": 4.773931472977746,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00029422414169085094,
      "loss": 5.1977,
      "mean_token_accuracy": 0.2255384624004364,
      "num_tokens": 100774977.0,
      "step": 54060
    },
    {
      "entropy": 6.0215606689453125,
      "epoch": 4.774373013069587,
      "grad_norm": 1.25,
      "learning_rate": 0.00029419276515264657,
      "loss": 5.231,
      "mean_token_accuracy": 0.22050793766975402,
      "num_tokens": 100784574.0,
      "step": 54065
    },
    {
      "entropy": 6.064417362213135,
      "epoch": 4.774814553161427,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029416138823846696,
      "loss": 5.2223,
      "mean_token_accuracy": 0.21864588856697081,
      "num_tokens": 100793503.0,
      "step": 54070
    },
    {
      "entropy": 6.078106641769409,
      "epoch": 4.775256093253267,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029413001094892653,
      "loss": 5.2227,
      "mean_token_accuracy": 0.22024817764759064,
      "num_tokens": 100802745.0,
      "step": 54075
    },
    {
      "entropy": 6.0682044506073,
      "epoch": 4.775697633345108,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002940986332846402,
      "loss": 5.2213,
      "mean_token_accuracy": 0.22240860462188722,
      "num_tokens": 100811517.0,
      "step": 54080
    },
    {
      "entropy": 6.009048986434936,
      "epoch": 4.776139173436948,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002940672552462224,
      "loss": 5.2144,
      "mean_token_accuracy": 0.2301095739006996,
      "num_tokens": 100820748.0,
      "step": 54085
    },
    {
      "entropy": 6.049788188934326,
      "epoch": 4.776580713528788,
      "grad_norm": 1.3125,
      "learning_rate": 0.00029403587683428796,
      "loss": 5.222,
      "mean_token_accuracy": 0.22916426807641982,
      "num_tokens": 100830422.0,
      "step": 54090
    },
    {
      "entropy": 5.997008180618286,
      "epoch": 4.777022253620629,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002940044980494515,
      "loss": 5.2156,
      "mean_token_accuracy": 0.22336872667074203,
      "num_tokens": 100838980.0,
      "step": 54095
    },
    {
      "entropy": 6.061352062225342,
      "epoch": 4.777463793712469,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029397311889232773,
      "loss": 5.2266,
      "mean_token_accuracy": 0.21383165717124938,
      "num_tokens": 100848739.0,
      "step": 54100
    },
    {
      "entropy": 6.009482383728027,
      "epoch": 4.777905333804309,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002939417393635314,
      "loss": 5.2361,
      "mean_token_accuracy": 0.22361187338829042,
      "num_tokens": 100857518.0,
      "step": 54105
    },
    {
      "entropy": 6.101108837127685,
      "epoch": 4.77834687389615,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002939103594636771,
      "loss": 5.3273,
      "mean_token_accuracy": 0.21322769522666932,
      "num_tokens": 100867480.0,
      "step": 54110
    },
    {
      "entropy": 6.046001052856445,
      "epoch": 4.77878841398799,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002938789791933796,
      "loss": 5.1706,
      "mean_token_accuracy": 0.22833192795515062,
      "num_tokens": 100876933.0,
      "step": 54115
    },
    {
      "entropy": 6.023180389404297,
      "epoch": 4.77922995407983,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002938475985532537,
      "loss": 5.2375,
      "mean_token_accuracy": 0.22579851746559143,
      "num_tokens": 100886515.0,
      "step": 54120
    },
    {
      "entropy": 6.074264669418335,
      "epoch": 4.779671494171671,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029381621754391395,
      "loss": 5.3121,
      "mean_token_accuracy": 0.2192767322063446,
      "num_tokens": 100897104.0,
      "step": 54125
    },
    {
      "entropy": 6.085560750961304,
      "epoch": 4.780113034263511,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00029378483616597533,
      "loss": 5.3113,
      "mean_token_accuracy": 0.21234953105449678,
      "num_tokens": 100906867.0,
      "step": 54130
    },
    {
      "entropy": 6.025297737121582,
      "epoch": 4.780554574355351,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002937534544200524,
      "loss": 5.2684,
      "mean_token_accuracy": 0.22040412575006485,
      "num_tokens": 100916127.0,
      "step": 54135
    },
    {
      "entropy": 6.087685203552246,
      "epoch": 4.780996114447192,
      "grad_norm": 1.25,
      "learning_rate": 0.00029372207230676003,
      "loss": 5.3053,
      "mean_token_accuracy": 0.2132120966911316,
      "num_tokens": 100925200.0,
      "step": 54140
    },
    {
      "entropy": 6.0085694789886475,
      "epoch": 4.781437654539032,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002936906898267129,
      "loss": 5.2592,
      "mean_token_accuracy": 0.22364795058965684,
      "num_tokens": 100934084.0,
      "step": 54145
    },
    {
      "entropy": 5.996689653396606,
      "epoch": 4.781879194630872,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002936593069805257,
      "loss": 5.1081,
      "mean_token_accuracy": 0.2345293641090393,
      "num_tokens": 100943595.0,
      "step": 54150
    },
    {
      "entropy": 6.089011764526367,
      "epoch": 4.782320734722713,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002936279237688133,
      "loss": 5.2717,
      "mean_token_accuracy": 0.21670939028263092,
      "num_tokens": 100952675.0,
      "step": 54155
    },
    {
      "entropy": 6.054854154586792,
      "epoch": 4.782762274814553,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002935965401921905,
      "loss": 5.2565,
      "mean_token_accuracy": 0.2194048747420311,
      "num_tokens": 100962948.0,
      "step": 54160
    },
    {
      "entropy": 6.083700799942017,
      "epoch": 4.783203814906393,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002935651562512721,
      "loss": 5.252,
      "mean_token_accuracy": 0.22023638784885408,
      "num_tokens": 100972077.0,
      "step": 54165
    },
    {
      "entropy": 6.0814008712768555,
      "epoch": 4.783645354998233,
      "grad_norm": 1.25,
      "learning_rate": 0.0002935337719466727,
      "loss": 5.2724,
      "mean_token_accuracy": 0.22423920333385466,
      "num_tokens": 100981093.0,
      "step": 54170
    },
    {
      "entropy": 6.087555313110352,
      "epoch": 4.784086895090074,
      "grad_norm": 1.453125,
      "learning_rate": 0.00029350238727900743,
      "loss": 5.2776,
      "mean_token_accuracy": 0.21638938188552856,
      "num_tokens": 100990336.0,
      "step": 54175
    },
    {
      "entropy": 6.012143087387085,
      "epoch": 4.784528435181914,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002934710022488908,
      "loss": 5.2614,
      "mean_token_accuracy": 0.22317229807376862,
      "num_tokens": 101000068.0,
      "step": 54180
    },
    {
      "entropy": 5.972120141983032,
      "epoch": 4.784969975273755,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029343961685693776,
      "loss": 5.2449,
      "mean_token_accuracy": 0.22497146874666213,
      "num_tokens": 101009180.0,
      "step": 54185
    },
    {
      "entropy": 6.010469532012939,
      "epoch": 4.785411515365595,
      "grad_norm": 1.546875,
      "learning_rate": 0.00029340823110376313,
      "loss": 5.2413,
      "mean_token_accuracy": 0.20930409282445908,
      "num_tokens": 101018199.0,
      "step": 54190
    },
    {
      "entropy": 6.122262001037598,
      "epoch": 4.7858530554574354,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002933768449899817,
      "loss": 5.2758,
      "mean_token_accuracy": 0.21832099556922913,
      "num_tokens": 101026378.0,
      "step": 54195
    },
    {
      "entropy": 6.052288246154785,
      "epoch": 4.7862945955492755,
      "grad_norm": 1.5,
      "learning_rate": 0.00029334545851620843,
      "loss": 5.1706,
      "mean_token_accuracy": 0.2302236422896385,
      "num_tokens": 101035317.0,
      "step": 54200
    },
    {
      "entropy": 6.124141025543213,
      "epoch": 4.786736135641116,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002933140716830579,
      "loss": 5.3187,
      "mean_token_accuracy": 0.21944456696510314,
      "num_tokens": 101044859.0,
      "step": 54205
    },
    {
      "entropy": 5.9460855484008786,
      "epoch": 4.7871776757329565,
      "grad_norm": 1.359375,
      "learning_rate": 0.00029328268449114526,
      "loss": 5.1385,
      "mean_token_accuracy": 0.22493553459644317,
      "num_tokens": 101053875.0,
      "step": 54210
    },
    {
      "entropy": 6.056414556503296,
      "epoch": 4.7876192158247965,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002932512969410852,
      "loss": 5.3333,
      "mean_token_accuracy": 0.20343596041202544,
      "num_tokens": 101063228.0,
      "step": 54215
    },
    {
      "entropy": 6.076673698425293,
      "epoch": 4.7880607559166375,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002932199090334925,
      "loss": 5.3343,
      "mean_token_accuracy": 0.21728368252515792,
      "num_tokens": 101073256.0,
      "step": 54220
    },
    {
      "entropy": 6.007012844085693,
      "epoch": 4.7885022960084775,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029318852076898224,
      "loss": 5.1575,
      "mean_token_accuracy": 0.22873047292232512,
      "num_tokens": 101082689.0,
      "step": 54225
    },
    {
      "entropy": 6.117455196380615,
      "epoch": 4.788943836100318,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002931571321481692,
      "loss": 5.2749,
      "mean_token_accuracy": 0.21749770790338516,
      "num_tokens": 101092193.0,
      "step": 54230
    },
    {
      "entropy": 6.077174663543701,
      "epoch": 4.7893853761921585,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029312574317166823,
      "loss": 5.2849,
      "mean_token_accuracy": 0.2304350972175598,
      "num_tokens": 101102419.0,
      "step": 54235
    },
    {
      "entropy": 6.095309638977051,
      "epoch": 4.789826916283999,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002930943538400943,
      "loss": 5.2763,
      "mean_token_accuracy": 0.22012246549129486,
      "num_tokens": 101112258.0,
      "step": 54240
    },
    {
      "entropy": 6.015294456481934,
      "epoch": 4.790268456375839,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002930629641540623,
      "loss": 5.1489,
      "mean_token_accuracy": 0.23155426532030104,
      "num_tokens": 101121177.0,
      "step": 54245
    },
    {
      "entropy": 6.018430137634278,
      "epoch": 4.79070999646768,
      "grad_norm": 1.375,
      "learning_rate": 0.00029303157411418704,
      "loss": 5.1738,
      "mean_token_accuracy": 0.23491950035095216,
      "num_tokens": 101130153.0,
      "step": 54250
    },
    {
      "entropy": 6.044982051849365,
      "epoch": 4.79115153655952,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002930001837210835,
      "loss": 5.3031,
      "mean_token_accuracy": 0.2128067970275879,
      "num_tokens": 101139584.0,
      "step": 54255
    },
    {
      "entropy": 5.9658143520355225,
      "epoch": 4.79159307665136,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002929687929753666,
      "loss": 5.2118,
      "mean_token_accuracy": 0.2339214190840721,
      "num_tokens": 101148096.0,
      "step": 54260
    },
    {
      "entropy": 6.083902645111084,
      "epoch": 4.792034616743201,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029293740187765133,
      "loss": 5.318,
      "mean_token_accuracy": 0.21200615763664246,
      "num_tokens": 101157095.0,
      "step": 54265
    },
    {
      "entropy": 6.074271774291992,
      "epoch": 4.792476156835041,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029290601042855245,
      "loss": 5.2407,
      "mean_token_accuracy": 0.21428124308586122,
      "num_tokens": 101166843.0,
      "step": 54270
    },
    {
      "entropy": 6.076080083847046,
      "epoch": 4.792917696926881,
      "grad_norm": 1.59375,
      "learning_rate": 0.00029287461862868516,
      "loss": 5.2199,
      "mean_token_accuracy": 0.22421278059482574,
      "num_tokens": 101175553.0,
      "step": 54275
    },
    {
      "entropy": 6.033529615402221,
      "epoch": 4.793359237018722,
      "grad_norm": 1.3125,
      "learning_rate": 0.00029284322647866415,
      "loss": 5.2226,
      "mean_token_accuracy": 0.21573404222726822,
      "num_tokens": 101185405.0,
      "step": 54280
    },
    {
      "entropy": 6.082808971405029,
      "epoch": 4.793800777110562,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002928118339791045,
      "loss": 5.2885,
      "mean_token_accuracy": 0.22132809609174728,
      "num_tokens": 101194657.0,
      "step": 54285
    },
    {
      "entropy": 6.036460590362549,
      "epoch": 4.794242317202402,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029278044113062115,
      "loss": 5.1536,
      "mean_token_accuracy": 0.2267281711101532,
      "num_tokens": 101204228.0,
      "step": 54290
    },
    {
      "entropy": 6.055885601043701,
      "epoch": 4.794683857294242,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002927490479338291,
      "loss": 5.238,
      "mean_token_accuracy": 0.22514035254716874,
      "num_tokens": 101214537.0,
      "step": 54295
    },
    {
      "entropy": 6.011689281463623,
      "epoch": 4.795125397386083,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002927176543893432,
      "loss": 5.2562,
      "mean_token_accuracy": 0.21391512900590898,
      "num_tokens": 101224890.0,
      "step": 54300
    },
    {
      "entropy": 6.065608263015747,
      "epoch": 4.795566937477923,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029268626049777863,
      "loss": 5.2997,
      "mean_token_accuracy": 0.21564335525035858,
      "num_tokens": 101234331.0,
      "step": 54305
    },
    {
      "entropy": 6.04762635231018,
      "epoch": 4.796008477569764,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002926548662597502,
      "loss": 5.2311,
      "mean_token_accuracy": 0.22390355318784713,
      "num_tokens": 101243654.0,
      "step": 54310
    },
    {
      "entropy": 5.974041271209717,
      "epoch": 4.796450017661604,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000292623471675873,
      "loss": 5.1157,
      "mean_token_accuracy": 0.23437278419733049,
      "num_tokens": 101252924.0,
      "step": 54315
    },
    {
      "entropy": 6.031518125534058,
      "epoch": 4.796891557753444,
      "grad_norm": 1.515625,
      "learning_rate": 0.00029259207674676204,
      "loss": 5.3321,
      "mean_token_accuracy": 0.22125389128923417,
      "num_tokens": 101263100.0,
      "step": 54320
    },
    {
      "entropy": 6.0547287940979,
      "epoch": 4.797333097845284,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002925606814730323,
      "loss": 5.2506,
      "mean_token_accuracy": 0.22252277731895448,
      "num_tokens": 101271305.0,
      "step": 54325
    },
    {
      "entropy": 6.087602519989014,
      "epoch": 4.797774637937125,
      "grad_norm": 1.375,
      "learning_rate": 0.0002925292858552988,
      "loss": 5.2029,
      "mean_token_accuracy": 0.2231658235192299,
      "num_tokens": 101280260.0,
      "step": 54330
    },
    {
      "entropy": 6.043287944793701,
      "epoch": 4.798216178028965,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029249788989417654,
      "loss": 5.2204,
      "mean_token_accuracy": 0.22345547378063202,
      "num_tokens": 101289472.0,
      "step": 54335
    },
    {
      "entropy": 6.0725139617919925,
      "epoch": 4.798657718120805,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002924664935902806,
      "loss": 5.21,
      "mean_token_accuracy": 0.21977491974830626,
      "num_tokens": 101298503.0,
      "step": 54340
    },
    {
      "entropy": 6.04119725227356,
      "epoch": 4.799099258212646,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000292435096944226,
      "loss": 5.1774,
      "mean_token_accuracy": 0.22127342075109482,
      "num_tokens": 101309160.0,
      "step": 54345
    },
    {
      "entropy": 6.053101205825806,
      "epoch": 4.799540798304486,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029240369995662776,
      "loss": 5.2644,
      "mean_token_accuracy": 0.2163906767964363,
      "num_tokens": 101318981.0,
      "step": 54350
    },
    {
      "entropy": 6.080749797821045,
      "epoch": 4.799982338396326,
      "grad_norm": 1.28125,
      "learning_rate": 0.00029237230262810085,
      "loss": 5.2826,
      "mean_token_accuracy": 0.2093798577785492,
      "num_tokens": 101327726.0,
      "step": 54355
    },
    {
      "entropy": 6.067434978485108,
      "epoch": 4.800423878488167,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029234090495926045,
      "loss": 5.3224,
      "mean_token_accuracy": 0.22034769207239152,
      "num_tokens": 101338372.0,
      "step": 54360
    },
    {
      "entropy": 6.056603479385376,
      "epoch": 4.800865418580007,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002923095069507216,
      "loss": 5.1977,
      "mean_token_accuracy": 0.22501939684152603,
      "num_tokens": 101348198.0,
      "step": 54365
    },
    {
      "entropy": 6.044725608825684,
      "epoch": 4.801306958671847,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002922781086030993,
      "loss": 5.1942,
      "mean_token_accuracy": 0.23806336373090745,
      "num_tokens": 101357778.0,
      "step": 54370
    },
    {
      "entropy": 6.050053930282592,
      "epoch": 4.801748498763688,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002922467099170088,
      "loss": 5.264,
      "mean_token_accuracy": 0.21903471201658248,
      "num_tokens": 101367202.0,
      "step": 54375
    },
    {
      "entropy": 6.048165035247803,
      "epoch": 4.802190038855528,
      "grad_norm": 1.359375,
      "learning_rate": 0.00029221531089306495,
      "loss": 5.2578,
      "mean_token_accuracy": 0.22036244571208954,
      "num_tokens": 101377091.0,
      "step": 54380
    },
    {
      "entropy": 6.097947835922241,
      "epoch": 4.802631578947368,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002921839115318831,
      "loss": 5.2362,
      "mean_token_accuracy": 0.22243023663759232,
      "num_tokens": 101386649.0,
      "step": 54385
    },
    {
      "entropy": 6.076847982406616,
      "epoch": 4.803073119039209,
      "grad_norm": 1.375,
      "learning_rate": 0.00029215251183407805,
      "loss": 5.2886,
      "mean_token_accuracy": 0.22037257850170136,
      "num_tokens": 101395670.0,
      "step": 54390
    },
    {
      "entropy": 6.021548891067505,
      "epoch": 4.803514659131049,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002921211118002651,
      "loss": 5.2233,
      "mean_token_accuracy": 0.2233205184340477,
      "num_tokens": 101404576.0,
      "step": 54395
    },
    {
      "entropy": 6.09860200881958,
      "epoch": 4.803956199222889,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002920897114310593,
      "loss": 5.3297,
      "mean_token_accuracy": 0.21724033057689668,
      "num_tokens": 101414602.0,
      "step": 54400
    },
    {
      "entropy": 6.0142433643341064,
      "epoch": 4.80439773931473,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002920583107270758,
      "loss": 5.172,
      "mean_token_accuracy": 0.22960223853588105,
      "num_tokens": 101423878.0,
      "step": 54405
    },
    {
      "entropy": 6.03351354598999,
      "epoch": 4.80483927940657,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002920269096889296,
      "loss": 5.2464,
      "mean_token_accuracy": 0.22532508820295333,
      "num_tokens": 101432635.0,
      "step": 54410
    },
    {
      "entropy": 6.114325761795044,
      "epoch": 4.80528081949841,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029199550831723604,
      "loss": 5.3333,
      "mean_token_accuracy": 0.20931147187948226,
      "num_tokens": 101442589.0,
      "step": 54415
    },
    {
      "entropy": 6.069237232208252,
      "epoch": 4.805722359590251,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029196410661261004,
      "loss": 5.2634,
      "mean_token_accuracy": 0.21538061648607254,
      "num_tokens": 101452150.0,
      "step": 54420
    },
    {
      "entropy": 6.07257046699524,
      "epoch": 4.806163899682091,
      "grad_norm": 1.453125,
      "learning_rate": 0.00029193270457566697,
      "loss": 5.2651,
      "mean_token_accuracy": 0.22058549523353577,
      "num_tokens": 101460278.0,
      "step": 54425
    },
    {
      "entropy": 5.997336959838867,
      "epoch": 4.806605439773931,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002919013022070217,
      "loss": 5.1619,
      "mean_token_accuracy": 0.22342493385076523,
      "num_tokens": 101468668.0,
      "step": 54430
    },
    {
      "entropy": 6.039183139801025,
      "epoch": 4.807046979865772,
      "grad_norm": 1.25,
      "learning_rate": 0.0002918698995072896,
      "loss": 5.2037,
      "mean_token_accuracy": 0.22904239147901534,
      "num_tokens": 101478420.0,
      "step": 54435
    },
    {
      "entropy": 6.040695238113403,
      "epoch": 4.807488519957612,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002918384964770858,
      "loss": 5.2779,
      "mean_token_accuracy": 0.2224355459213257,
      "num_tokens": 101487978.0,
      "step": 54440
    },
    {
      "entropy": 6.065792226791382,
      "epoch": 4.807930060049452,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002918070931170253,
      "loss": 5.2741,
      "mean_token_accuracy": 0.21880536377429963,
      "num_tokens": 101497059.0,
      "step": 54445
    },
    {
      "entropy": 6.149503135681153,
      "epoch": 4.808371600141292,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00029177568942772355,
      "loss": 5.3108,
      "mean_token_accuracy": 0.21013759076595306,
      "num_tokens": 101506552.0,
      "step": 54450
    },
    {
      "entropy": 6.0609376430511475,
      "epoch": 4.808813140233133,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002917442854097955,
      "loss": 5.2355,
      "mean_token_accuracy": 0.22488498389720918,
      "num_tokens": 101516738.0,
      "step": 54455
    },
    {
      "entropy": 6.094410371780396,
      "epoch": 4.809254680324973,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029171288106385644,
      "loss": 5.3076,
      "mean_token_accuracy": 0.21705031245946885,
      "num_tokens": 101526594.0,
      "step": 54460
    },
    {
      "entropy": 6.0526079654693605,
      "epoch": 4.809696220416814,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00029168147639052156,
      "loss": 5.2394,
      "mean_token_accuracy": 0.22952306717634202,
      "num_tokens": 101536715.0,
      "step": 54465
    },
    {
      "entropy": 6.036999750137329,
      "epoch": 4.810137760508654,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002916500713904061,
      "loss": 5.1766,
      "mean_token_accuracy": 0.2288274884223938,
      "num_tokens": 101545517.0,
      "step": 54470
    },
    {
      "entropy": 6.109799718856811,
      "epoch": 4.810579300600494,
      "grad_norm": 1.25,
      "learning_rate": 0.0002916186660641251,
      "loss": 5.3529,
      "mean_token_accuracy": 0.2115970477461815,
      "num_tokens": 101555883.0,
      "step": 54475
    },
    {
      "entropy": 6.09185094833374,
      "epoch": 4.8110208406923345,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029158726041229394,
      "loss": 5.3185,
      "mean_token_accuracy": 0.2189461410045624,
      "num_tokens": 101565727.0,
      "step": 54480
    },
    {
      "entropy": 6.073488569259643,
      "epoch": 4.811462380784175,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002915558544355278,
      "loss": 5.3137,
      "mean_token_accuracy": 0.2165201336145401,
      "num_tokens": 101575733.0,
      "step": 54485
    },
    {
      "entropy": 6.192319869995117,
      "epoch": 4.8119039208760155,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029152444813444186,
      "loss": 5.3428,
      "mean_token_accuracy": 0.2071605697274208,
      "num_tokens": 101585335.0,
      "step": 54490
    },
    {
      "entropy": 6.0820427417755125,
      "epoch": 4.8123454609678555,
      "grad_norm": 1.25,
      "learning_rate": 0.0002914930415096514,
      "loss": 5.2511,
      "mean_token_accuracy": 0.22306964844465255,
      "num_tokens": 101594720.0,
      "step": 54495
    },
    {
      "entropy": 6.003778028488159,
      "epoch": 4.8127870010596965,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00029146163456177153,
      "loss": 5.2461,
      "mean_token_accuracy": 0.22322487235069274,
      "num_tokens": 101603582.0,
      "step": 54500
    },
    {
      "entropy": 6.012280797958374,
      "epoch": 4.8132285411515365,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002914302272914177,
      "loss": 5.2073,
      "mean_token_accuracy": 0.22922038435935974,
      "num_tokens": 101612389.0,
      "step": 54505
    },
    {
      "entropy": 6.000100040435791,
      "epoch": 4.813670081243377,
      "grad_norm": 1.375,
      "learning_rate": 0.00029139881969920506,
      "loss": 5.2019,
      "mean_token_accuracy": 0.22887191623449327,
      "num_tokens": 101621816.0,
      "step": 54510
    },
    {
      "entropy": 6.01985387802124,
      "epoch": 4.8141116213352175,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002913674117857489,
      "loss": 5.2497,
      "mean_token_accuracy": 0.22550371587276458,
      "num_tokens": 101631576.0,
      "step": 54515
    },
    {
      "entropy": 6.034173917770386,
      "epoch": 4.814553161427058,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002913360035516644,
      "loss": 5.2216,
      "mean_token_accuracy": 0.224354088306427,
      "num_tokens": 101641179.0,
      "step": 54520
    },
    {
      "entropy": 6.108440828323364,
      "epoch": 4.814994701518898,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029130459499756694,
      "loss": 5.2665,
      "mean_token_accuracy": 0.22154286801815032,
      "num_tokens": 101650857.0,
      "step": 54525
    },
    {
      "entropy": 6.10884656906128,
      "epoch": 4.815436241610739,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002912731861240716,
      "loss": 5.2693,
      "mean_token_accuracy": 0.21567293852567673,
      "num_tokens": 101660627.0,
      "step": 54530
    },
    {
      "entropy": 6.02791199684143,
      "epoch": 4.815877781702579,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00029124177693179394,
      "loss": 5.2144,
      "mean_token_accuracy": 0.22803260684013366,
      "num_tokens": 101670260.0,
      "step": 54535
    },
    {
      "entropy": 6.004865026473999,
      "epoch": 4.816319321794419,
      "grad_norm": 1.3125,
      "learning_rate": 0.00029121036742134907,
      "loss": 5.1578,
      "mean_token_accuracy": 0.23533227145671845,
      "num_tokens": 101679755.0,
      "step": 54540
    },
    {
      "entropy": 6.139831686019898,
      "epoch": 4.81676086188626,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002911789575933523,
      "loss": 5.3912,
      "mean_token_accuracy": 0.22139603048563003,
      "num_tokens": 101688726.0,
      "step": 54545
    },
    {
      "entropy": 6.002870798110962,
      "epoch": 4.8172024019781,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002911475474484189,
      "loss": 5.1979,
      "mean_token_accuracy": 0.22742023468017578,
      "num_tokens": 101698594.0,
      "step": 54550
    },
    {
      "entropy": 6.130014419555664,
      "epoch": 4.81764394206994,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002911161369871644,
      "loss": 5.4136,
      "mean_token_accuracy": 0.20475993156433106,
      "num_tokens": 101707908.0,
      "step": 54555
    },
    {
      "entropy": 6.0203516483306885,
      "epoch": 4.818085482161781,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002910847262102038,
      "loss": 5.0919,
      "mean_token_accuracy": 0.23481134921312333,
      "num_tokens": 101716811.0,
      "step": 54560
    },
    {
      "entropy": 6.136241340637207,
      "epoch": 4.818527022253621,
      "grad_norm": 1.4375,
      "learning_rate": 0.00029105331511815254,
      "loss": 5.3528,
      "mean_token_accuracy": 0.20991623252630234,
      "num_tokens": 101725668.0,
      "step": 54565
    },
    {
      "entropy": 6.045255470275879,
      "epoch": 4.818968562345461,
      "grad_norm": 1.25,
      "learning_rate": 0.000291021903711626,
      "loss": 5.1818,
      "mean_token_accuracy": 0.21838841140270232,
      "num_tokens": 101734959.0,
      "step": 54570
    },
    {
      "entropy": 6.084988260269165,
      "epoch": 4.819410102437301,
      "grad_norm": 1.359375,
      "learning_rate": 0.00029099049199123955,
      "loss": 5.2491,
      "mean_token_accuracy": 0.22987105399370195,
      "num_tokens": 101743615.0,
      "step": 54575
    },
    {
      "entropy": 6.04188756942749,
      "epoch": 4.819851642529142,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029095907995760844,
      "loss": 5.2038,
      "mean_token_accuracy": 0.22682141065597533,
      "num_tokens": 101752432.0,
      "step": 54580
    },
    {
      "entropy": 6.071153688430786,
      "epoch": 4.820293182620982,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029092766761134795,
      "loss": 5.3098,
      "mean_token_accuracy": 0.21933654993772506,
      "num_tokens": 101762831.0,
      "step": 54585
    },
    {
      "entropy": 6.05112624168396,
      "epoch": 4.820734722712823,
      "grad_norm": 1.375,
      "learning_rate": 0.0002908962549530736,
      "loss": 5.1679,
      "mean_token_accuracy": 0.2300880640745163,
      "num_tokens": 101770941.0,
      "step": 54590
    },
    {
      "entropy": 6.069560289382935,
      "epoch": 4.821176262804663,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002908648419834006,
      "loss": 5.3287,
      "mean_token_accuracy": 0.21463610082864762,
      "num_tokens": 101780850.0,
      "step": 54595
    },
    {
      "entropy": 6.009507417678833,
      "epoch": 4.821617802896503,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002908334287029444,
      "loss": 5.1358,
      "mean_token_accuracy": 0.2186424106359482,
      "num_tokens": 101791120.0,
      "step": 54600
    },
    {
      "entropy": 6.0680194854736325,
      "epoch": 4.822059342988343,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029080201511232025,
      "loss": 5.2625,
      "mean_token_accuracy": 0.22119388729333878,
      "num_tokens": 101800098.0,
      "step": 54605
    },
    {
      "entropy": 6.092315673828125,
      "epoch": 4.822500883080184,
      "grad_norm": 1.8125,
      "learning_rate": 0.0002907706012121437,
      "loss": 5.3199,
      "mean_token_accuracy": 0.2217171609401703,
      "num_tokens": 101809366.0,
      "step": 54610
    },
    {
      "entropy": 5.911322450637817,
      "epoch": 4.822942423172024,
      "grad_norm": 1.28125,
      "learning_rate": 0.00029073918700303,
      "loss": 5.0935,
      "mean_token_accuracy": 0.24265896081924437,
      "num_tokens": 101818378.0,
      "step": 54615
    },
    {
      "entropy": 5.996654224395752,
      "epoch": 4.823383963263864,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029070777248559466,
      "loss": 5.1753,
      "mean_token_accuracy": 0.22859304547309875,
      "num_tokens": 101827501.0,
      "step": 54620
    },
    {
      "entropy": 6.088297033309937,
      "epoch": 4.823825503355705,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002906763576604529,
      "loss": 5.3581,
      "mean_token_accuracy": 0.2116013526916504,
      "num_tokens": 101837295.0,
      "step": 54625
    },
    {
      "entropy": 6.094502878189087,
      "epoch": 4.824267043447545,
      "grad_norm": 1.25,
      "learning_rate": 0.0002906449425282202,
      "loss": 5.2667,
      "mean_token_accuracy": 0.21657532453536987,
      "num_tokens": 101846498.0,
      "step": 54630
    },
    {
      "entropy": 6.065976810455322,
      "epoch": 4.824708583539385,
      "grad_norm": 1.328125,
      "learning_rate": 0.000290613527089512,
      "loss": 5.2944,
      "mean_token_accuracy": 0.22590167820453644,
      "num_tokens": 101855308.0,
      "step": 54635
    },
    {
      "entropy": 6.0677416801452635,
      "epoch": 4.825150123631226,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002905821113449437,
      "loss": 5.2194,
      "mean_token_accuracy": 0.21778471618890763,
      "num_tokens": 101865281.0,
      "step": 54640
    },
    {
      "entropy": 6.03996639251709,
      "epoch": 4.825591663723066,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002905506952951307,
      "loss": 5.2304,
      "mean_token_accuracy": 0.2255015641450882,
      "num_tokens": 101875562.0,
      "step": 54645
    },
    {
      "entropy": 6.013488531112671,
      "epoch": 4.826033203814906,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002905192789406884,
      "loss": 5.1873,
      "mean_token_accuracy": 0.2232766330242157,
      "num_tokens": 101884180.0,
      "step": 54650
    },
    {
      "entropy": 6.027576684951782,
      "epoch": 4.826474743906747,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029048786228223235,
      "loss": 5.1989,
      "mean_token_accuracy": 0.2252906233072281,
      "num_tokens": 101894472.0,
      "step": 54655
    },
    {
      "entropy": 6.039594602584839,
      "epoch": 4.826916283998587,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029045644532037777,
      "loss": 5.2175,
      "mean_token_accuracy": 0.21880590468645095,
      "num_tokens": 101903981.0,
      "step": 54660
    },
    {
      "entropy": 6.090957307815552,
      "epoch": 4.827357824090427,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029042502805574027,
      "loss": 5.3339,
      "mean_token_accuracy": 0.21840128302574158,
      "num_tokens": 101913915.0,
      "step": 54665
    },
    {
      "entropy": 6.094878005981445,
      "epoch": 4.827799364182268,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002903936104889352,
      "loss": 5.219,
      "mean_token_accuracy": 0.21853099167346954,
      "num_tokens": 101923954.0,
      "step": 54670
    },
    {
      "entropy": 6.014292001724243,
      "epoch": 4.828240904274108,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029036219262057813,
      "loss": 5.2257,
      "mean_token_accuracy": 0.22436462193727494,
      "num_tokens": 101932876.0,
      "step": 54675
    },
    {
      "entropy": 6.011656379699707,
      "epoch": 4.828682444365948,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00029033077445128435,
      "loss": 5.214,
      "mean_token_accuracy": 0.22399693429470063,
      "num_tokens": 101942118.0,
      "step": 54680
    },
    {
      "entropy": 6.075991153717041,
      "epoch": 4.829123984457789,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00029029935598166943,
      "loss": 5.2871,
      "mean_token_accuracy": 0.21424282640218734,
      "num_tokens": 101951946.0,
      "step": 54685
    },
    {
      "entropy": 6.06880407333374,
      "epoch": 4.829565524549629,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002902679372123489,
      "loss": 5.2576,
      "mean_token_accuracy": 0.22060510367155076,
      "num_tokens": 101962056.0,
      "step": 54690
    },
    {
      "entropy": 6.08807406425476,
      "epoch": 4.830007064641469,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002902365181439381,
      "loss": 5.2427,
      "mean_token_accuracy": 0.2241177186369896,
      "num_tokens": 101971352.0,
      "step": 54695
    },
    {
      "entropy": 5.942915773391723,
      "epoch": 4.83044860473331,
      "grad_norm": 1.234375,
      "learning_rate": 0.00029020509877705265,
      "loss": 5.0933,
      "mean_token_accuracy": 0.235978427529335,
      "num_tokens": 101981057.0,
      "step": 54700
    },
    {
      "entropy": 6.0092689037323,
      "epoch": 4.83089014482515,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029017367911230785,
      "loss": 5.1939,
      "mean_token_accuracy": 0.22080948948860168,
      "num_tokens": 101989949.0,
      "step": 54705
    },
    {
      "entropy": 6.077118873596191,
      "epoch": 4.83133168491699,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002901422591503194,
      "loss": 5.2689,
      "mean_token_accuracy": 0.21862131506204605,
      "num_tokens": 101999187.0,
      "step": 54710
    },
    {
      "entropy": 6.083589553833008,
      "epoch": 4.831773225008831,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029011083889170263,
      "loss": 5.2739,
      "mean_token_accuracy": 0.2139814615249634,
      "num_tokens": 102009193.0,
      "step": 54715
    },
    {
      "entropy": 6.107234621047974,
      "epoch": 4.832214765100671,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002900794183370732,
      "loss": 5.4053,
      "mean_token_accuracy": 0.20861772298812867,
      "num_tokens": 102019605.0,
      "step": 54720
    },
    {
      "entropy": 6.133562231063843,
      "epoch": 4.832656305192511,
      "grad_norm": 1.390625,
      "learning_rate": 0.00029004799748704645,
      "loss": 5.3428,
      "mean_token_accuracy": 0.2072906419634819,
      "num_tokens": 102029255.0,
      "step": 54725
    },
    {
      "entropy": 5.960929346084595,
      "epoch": 4.833097845284351,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002900165763422381,
      "loss": 5.1029,
      "mean_token_accuracy": 0.23320896178483963,
      "num_tokens": 102038011.0,
      "step": 54730
    },
    {
      "entropy": 5.991989707946777,
      "epoch": 4.833539385376192,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002899851549032635,
      "loss": 5.1686,
      "mean_token_accuracy": 0.23095216006040573,
      "num_tokens": 102046790.0,
      "step": 54735
    },
    {
      "entropy": 6.033354187011719,
      "epoch": 4.833980925468032,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002899537331707383,
      "loss": 5.1742,
      "mean_token_accuracy": 0.22732955813407899,
      "num_tokens": 102055827.0,
      "step": 54740
    },
    {
      "entropy": 6.035234451293945,
      "epoch": 4.834422465559873,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00028992231114527786,
      "loss": 5.2856,
      "mean_token_accuracy": 0.21897264420986176,
      "num_tokens": 102064695.0,
      "step": 54745
    },
    {
      "entropy": 6.026830101013184,
      "epoch": 4.834864005651713,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002898908888274979,
      "loss": 5.1737,
      "mean_token_accuracy": 0.2264883488416672,
      "num_tokens": 102073073.0,
      "step": 54750
    },
    {
      "entropy": 6.04601411819458,
      "epoch": 4.835305545743553,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028985946621801385,
      "loss": 5.317,
      "mean_token_accuracy": 0.21584934145212173,
      "num_tokens": 102082942.0,
      "step": 54755
    },
    {
      "entropy": 6.1041131019592285,
      "epoch": 4.8357470858353935,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002898280433174413,
      "loss": 5.2467,
      "mean_token_accuracy": 0.2190461978316307,
      "num_tokens": 102092315.0,
      "step": 54760
    },
    {
      "entropy": 6.063352298736572,
      "epoch": 4.836188625927234,
      "grad_norm": 1.40625,
      "learning_rate": 0.00028979662012639586,
      "loss": 5.2101,
      "mean_token_accuracy": 0.22024068385362625,
      "num_tokens": 102101882.0,
      "step": 54765
    },
    {
      "entropy": 6.0441803455352785,
      "epoch": 4.8366301660190745,
      "grad_norm": 1.375,
      "learning_rate": 0.000289765196645493,
      "loss": 5.2148,
      "mean_token_accuracy": 0.22390421777963637,
      "num_tokens": 102110767.0,
      "step": 54770
    },
    {
      "entropy": 6.074973154067993,
      "epoch": 4.8370717061109145,
      "grad_norm": 1.375,
      "learning_rate": 0.00028973377287534844,
      "loss": 5.2672,
      "mean_token_accuracy": 0.22000281363725663,
      "num_tokens": 102119093.0,
      "step": 54775
    },
    {
      "entropy": 5.9869036197662355,
      "epoch": 4.8375132462027555,
      "grad_norm": 1.25,
      "learning_rate": 0.00028970234881657754,
      "loss": 5.1833,
      "mean_token_accuracy": 0.22217458188533784,
      "num_tokens": 102128955.0,
      "step": 54780
    },
    {
      "entropy": 6.031954193115235,
      "epoch": 4.8379547862945955,
      "grad_norm": 1.375,
      "learning_rate": 0.000289670924469796,
      "loss": 5.158,
      "mean_token_accuracy": 0.22203155159950255,
      "num_tokens": 102137955.0,
      "step": 54785
    },
    {
      "entropy": 6.019688081741333,
      "epoch": 4.838396326386436,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028963949983561946,
      "loss": 5.2565,
      "mean_token_accuracy": 0.22237527072429658,
      "num_tokens": 102148348.0,
      "step": 54790
    },
    {
      "entropy": 6.043976831436157,
      "epoch": 4.8388378664782765,
      "grad_norm": 1.375,
      "learning_rate": 0.00028960807491466346,
      "loss": 5.2892,
      "mean_token_accuracy": 0.21285985112190248,
      "num_tokens": 102157846.0,
      "step": 54795
    },
    {
      "entropy": 6.045740413665771,
      "epoch": 4.839279406570117,
      "grad_norm": 1.34375,
      "learning_rate": 0.00028957664970754356,
      "loss": 5.2231,
      "mean_token_accuracy": 0.2225173071026802,
      "num_tokens": 102166824.0,
      "step": 54800
    },
    {
      "entropy": 6.04933352470398,
      "epoch": 4.839720946661957,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002895452242148753,
      "loss": 5.2651,
      "mean_token_accuracy": 0.21909735351800919,
      "num_tokens": 102176295.0,
      "step": 54805
    },
    {
      "entropy": 6.0295408248901365,
      "epoch": 4.840162486753798,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00028951379843727444,
      "loss": 5.1977,
      "mean_token_accuracy": 0.22394683510065078,
      "num_tokens": 102184609.0,
      "step": 54810
    },
    {
      "entropy": 6.108981084823609,
      "epoch": 4.840604026845638,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002894823723753565,
      "loss": 5.3416,
      "mean_token_accuracy": 0.21039033979177474,
      "num_tokens": 102194465.0,
      "step": 54815
    },
    {
      "entropy": 6.049669694900513,
      "epoch": 4.841045566937478,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002894509460297371,
      "loss": 5.2499,
      "mean_token_accuracy": 0.21975243240594863,
      "num_tokens": 102204472.0,
      "step": 54820
    },
    {
      "entropy": 6.097823143005371,
      "epoch": 4.841487107029319,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002894195194010319,
      "loss": 5.2887,
      "mean_token_accuracy": 0.2151476711034775,
      "num_tokens": 102213636.0,
      "step": 54825
    },
    {
      "entropy": 6.131978511810303,
      "epoch": 4.841928647121159,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028938809248985655,
      "loss": 5.3037,
      "mean_token_accuracy": 0.21346478760242463,
      "num_tokens": 102223867.0,
      "step": 54830
    },
    {
      "entropy": 6.15651044845581,
      "epoch": 4.842370187212999,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028935666529682663,
      "loss": 5.2977,
      "mean_token_accuracy": 0.21986383497714995,
      "num_tokens": 102233050.0,
      "step": 54835
    },
    {
      "entropy": 5.999396514892578,
      "epoch": 4.84281172730484,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028932523782255776,
      "loss": 5.2398,
      "mean_token_accuracy": 0.22044421136379241,
      "num_tokens": 102243092.0,
      "step": 54840
    },
    {
      "entropy": 6.0104776382446286,
      "epoch": 4.84325326739668,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002892938100676657,
      "loss": 5.2072,
      "mean_token_accuracy": 0.22301664799451829,
      "num_tokens": 102252575.0,
      "step": 54845
    },
    {
      "entropy": 6.092718982696534,
      "epoch": 4.84369480748852,
      "grad_norm": 1.421875,
      "learning_rate": 0.00028926238203276603,
      "loss": 5.2668,
      "mean_token_accuracy": 0.2212194100022316,
      "num_tokens": 102261242.0,
      "step": 54850
    },
    {
      "entropy": 6.033739185333252,
      "epoch": 4.84413634758036,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002892309537184743,
      "loss": 5.142,
      "mean_token_accuracy": 0.23080601692199706,
      "num_tokens": 102269395.0,
      "step": 54855
    },
    {
      "entropy": 6.0374657154083256,
      "epoch": 4.844577887672201,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028919952512540645,
      "loss": 5.1888,
      "mean_token_accuracy": 0.2258877143263817,
      "num_tokens": 102278639.0,
      "step": 54860
    },
    {
      "entropy": 6.031810760498047,
      "epoch": 4.845019427764041,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002891680962541779,
      "loss": 5.2193,
      "mean_token_accuracy": 0.2277596712112427,
      "num_tokens": 102287645.0,
      "step": 54865
    },
    {
      "entropy": 6.030710649490357,
      "epoch": 4.845460967855882,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028913666710540434,
      "loss": 5.2302,
      "mean_token_accuracy": 0.2175803780555725,
      "num_tokens": 102296948.0,
      "step": 54870
    },
    {
      "entropy": 5.9668943881988525,
      "epoch": 4.845902507947722,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00028910523767970155,
      "loss": 5.1418,
      "mean_token_accuracy": 0.23686392307281495,
      "num_tokens": 102307581.0,
      "step": 54875
    },
    {
      "entropy": 6.057305908203125,
      "epoch": 4.846344048039562,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002890738079776852,
      "loss": 5.2692,
      "mean_token_accuracy": 0.21892789155244827,
      "num_tokens": 102318244.0,
      "step": 54880
    },
    {
      "entropy": 6.139332628250122,
      "epoch": 4.846785588131402,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002890423779999709,
      "loss": 5.317,
      "mean_token_accuracy": 0.21313853263854982,
      "num_tokens": 102327562.0,
      "step": 54885
    },
    {
      "entropy": 6.086068916320801,
      "epoch": 4.847227128223243,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002890109477471744,
      "loss": 5.285,
      "mean_token_accuracy": 0.21481698602437974,
      "num_tokens": 102337553.0,
      "step": 54890
    },
    {
      "entropy": 6.036645698547363,
      "epoch": 4.847668668315083,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002889795172199115,
      "loss": 5.2034,
      "mean_token_accuracy": 0.2314106211066246,
      "num_tokens": 102346525.0,
      "step": 54895
    },
    {
      "entropy": 6.035616111755371,
      "epoch": 4.848110208406923,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002889480864187977,
      "loss": 5.2246,
      "mean_token_accuracy": 0.22055349498987198,
      "num_tokens": 102355579.0,
      "step": 54900
    },
    {
      "entropy": 6.137354755401612,
      "epoch": 4.848551748498764,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002889166553444489,
      "loss": 5.3536,
      "mean_token_accuracy": 0.21402721405029296,
      "num_tokens": 102364483.0,
      "step": 54905
    },
    {
      "entropy": 6.131111192703247,
      "epoch": 4.848993288590604,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028888522399748073,
      "loss": 5.2457,
      "mean_token_accuracy": 0.22133916318416597,
      "num_tokens": 102374002.0,
      "step": 54910
    },
    {
      "entropy": 5.9872818946838375,
      "epoch": 4.849434828682444,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002888537923785089,
      "loss": 5.1981,
      "mean_token_accuracy": 0.23026204556226731,
      "num_tokens": 102384581.0,
      "step": 54915
    },
    {
      "entropy": 6.060829305648804,
      "epoch": 4.849876368774285,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002888223604881492,
      "loss": 5.364,
      "mean_token_accuracy": 0.21077674478292466,
      "num_tokens": 102396379.0,
      "step": 54920
    },
    {
      "entropy": 6.037704753875732,
      "epoch": 4.850317908866125,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028879092832701723,
      "loss": 5.2728,
      "mean_token_accuracy": 0.22169052213430404,
      "num_tokens": 102406191.0,
      "step": 54925
    },
    {
      "entropy": 6.0139491081237795,
      "epoch": 4.850759448957965,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002887594958957289,
      "loss": 5.1451,
      "mean_token_accuracy": 0.2350727841258049,
      "num_tokens": 102415586.0,
      "step": 54930
    },
    {
      "entropy": 6.055977010726929,
      "epoch": 4.851200989049806,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002887280631948998,
      "loss": 5.2165,
      "mean_token_accuracy": 0.22767869979143143,
      "num_tokens": 102425704.0,
      "step": 54935
    },
    {
      "entropy": 6.094387149810791,
      "epoch": 4.851642529141646,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002886966302251458,
      "loss": 5.2278,
      "mean_token_accuracy": 0.21965894252061843,
      "num_tokens": 102435626.0,
      "step": 54940
    },
    {
      "entropy": 6.050828886032105,
      "epoch": 4.852084069233486,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002886651969870825,
      "loss": 5.2021,
      "mean_token_accuracy": 0.22630850821733475,
      "num_tokens": 102445027.0,
      "step": 54945
    },
    {
      "entropy": 6.039860343933105,
      "epoch": 4.852525609325327,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002886337634813259,
      "loss": 5.186,
      "mean_token_accuracy": 0.2345767840743065,
      "num_tokens": 102453175.0,
      "step": 54950
    },
    {
      "entropy": 6.003546047210693,
      "epoch": 4.852967149417167,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002886023297084915,
      "loss": 5.1939,
      "mean_token_accuracy": 0.2252606213092804,
      "num_tokens": 102462396.0,
      "step": 54955
    },
    {
      "entropy": 6.065056896209716,
      "epoch": 4.853408689509007,
      "grad_norm": 1.40625,
      "learning_rate": 0.00028857089566919537,
      "loss": 5.2819,
      "mean_token_accuracy": 0.22171641141176224,
      "num_tokens": 102471961.0,
      "step": 54960
    },
    {
      "entropy": 6.034629535675049,
      "epoch": 4.853850229600848,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028853946136405296,
      "loss": 5.1811,
      "mean_token_accuracy": 0.22981057614088057,
      "num_tokens": 102482369.0,
      "step": 54965
    },
    {
      "entropy": 6.084611797332764,
      "epoch": 4.854291769692688,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002885080267936803,
      "loss": 5.282,
      "mean_token_accuracy": 0.21373874545097352,
      "num_tokens": 102491729.0,
      "step": 54970
    },
    {
      "entropy": 6.088733291625976,
      "epoch": 4.854733309784528,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028847659195869297,
      "loss": 5.2866,
      "mean_token_accuracy": 0.2133502796292305,
      "num_tokens": 102501539.0,
      "step": 54975
    },
    {
      "entropy": 6.0188898086547855,
      "epoch": 4.855174849876369,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000288445156859707,
      "loss": 5.2104,
      "mean_token_accuracy": 0.22608264535665512,
      "num_tokens": 102511631.0,
      "step": 54980
    },
    {
      "entropy": 6.08184289932251,
      "epoch": 4.855616389968209,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00028841372149733795,
      "loss": 5.2348,
      "mean_token_accuracy": 0.22288694977760315,
      "num_tokens": 102520979.0,
      "step": 54985
    },
    {
      "entropy": 6.035020732879639,
      "epoch": 4.856057930060049,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002883822858722018,
      "loss": 5.211,
      "mean_token_accuracy": 0.2270339086651802,
      "num_tokens": 102529970.0,
      "step": 54990
    },
    {
      "entropy": 6.078511381149292,
      "epoch": 4.85649947015189,
      "grad_norm": 1.421875,
      "learning_rate": 0.00028835084998491426,
      "loss": 5.2407,
      "mean_token_accuracy": 0.21649349331855774,
      "num_tokens": 102539502.0,
      "step": 54995
    },
    {
      "entropy": 6.080436182022095,
      "epoch": 4.85694101024373,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002883194138360911,
      "loss": 5.2687,
      "mean_token_accuracy": 0.22377052903175354,
      "num_tokens": 102548275.0,
      "step": 55000
    },
    {
      "entropy": 6.032503986358643,
      "epoch": 4.85738255033557,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002882879774263482,
      "loss": 5.1739,
      "mean_token_accuracy": 0.2331118956208229,
      "num_tokens": 102557180.0,
      "step": 55005
    },
    {
      "entropy": 6.092382478713989,
      "epoch": 4.85782409042741,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002882565407563014,
      "loss": 5.3499,
      "mean_token_accuracy": 0.21561795622110366,
      "num_tokens": 102567349.0,
      "step": 55010
    },
    {
      "entropy": 6.064191627502441,
      "epoch": 4.858265630519251,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00028822510382656657,
      "loss": 5.3069,
      "mean_token_accuracy": 0.22051185518503189,
      "num_tokens": 102576868.0,
      "step": 55015
    },
    {
      "entropy": 6.097965431213379,
      "epoch": 4.858707170611091,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028819366663775944,
      "loss": 5.2277,
      "mean_token_accuracy": 0.22290243208408356,
      "num_tokens": 102585530.0,
      "step": 55020
    },
    {
      "entropy": 6.113088512420655,
      "epoch": 4.859148710702932,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002881622291904959,
      "loss": 5.3144,
      "mean_token_accuracy": 0.21160818785429,
      "num_tokens": 102594615.0,
      "step": 55025
    },
    {
      "entropy": 6.0766054630279545,
      "epoch": 4.859590250794772,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002881307914853917,
      "loss": 5.2856,
      "mean_token_accuracy": 0.22101301401853563,
      "num_tokens": 102604600.0,
      "step": 55030
    },
    {
      "entropy": 6.055761003494263,
      "epoch": 4.860031790886612,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00028809935352306287,
      "loss": 5.2528,
      "mean_token_accuracy": 0.22433197051286696,
      "num_tokens": 102613190.0,
      "step": 55035
    },
    {
      "entropy": 6.150272607803345,
      "epoch": 4.8604733309784525,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002880679153041251,
      "loss": 5.4032,
      "mean_token_accuracy": 0.21631651520729064,
      "num_tokens": 102623708.0,
      "step": 55040
    },
    {
      "entropy": 6.055891704559326,
      "epoch": 4.860914871070293,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002880364768291943,
      "loss": 5.2024,
      "mean_token_accuracy": 0.2207249030470848,
      "num_tokens": 102633493.0,
      "step": 55045
    },
    {
      "entropy": 6.104701948165894,
      "epoch": 4.8613564111621335,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002880050380988864,
      "loss": 5.291,
      "mean_token_accuracy": 0.21427439153194427,
      "num_tokens": 102642483.0,
      "step": 55050
    },
    {
      "entropy": 6.045454597473144,
      "epoch": 4.8617979512539735,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002879735991138171,
      "loss": 5.208,
      "mean_token_accuracy": 0.22546921968460082,
      "num_tokens": 102651739.0,
      "step": 55055
    },
    {
      "entropy": 5.961918544769287,
      "epoch": 4.8622394913458145,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002879421598746024,
      "loss": 5.1689,
      "mean_token_accuracy": 0.2272886484861374,
      "num_tokens": 102661557.0,
      "step": 55060
    },
    {
      "entropy": 6.052403831481934,
      "epoch": 4.8626810314376545,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002879107203818581,
      "loss": 5.2984,
      "mean_token_accuracy": 0.2218918591737747,
      "num_tokens": 102672914.0,
      "step": 55065
    },
    {
      "entropy": 6.109771585464477,
      "epoch": 4.863122571529495,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028787928063620015,
      "loss": 5.2901,
      "mean_token_accuracy": 0.22460891157388688,
      "num_tokens": 102681711.0,
      "step": 55070
    },
    {
      "entropy": 6.0793355941772464,
      "epoch": 4.8635641116213355,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002878478406382445,
      "loss": 5.3296,
      "mean_token_accuracy": 0.21990997046232225,
      "num_tokens": 102691030.0,
      "step": 55075
    },
    {
      "entropy": 5.917256784439087,
      "epoch": 4.864005651713176,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002878164003886069,
      "loss": 5.1051,
      "mean_token_accuracy": 0.2352490931749344,
      "num_tokens": 102700322.0,
      "step": 55080
    },
    {
      "entropy": 6.136836242675781,
      "epoch": 4.864447191805016,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028778495988790334,
      "loss": 5.3895,
      "mean_token_accuracy": 0.20766876190900802,
      "num_tokens": 102709587.0,
      "step": 55085
    },
    {
      "entropy": 6.1040397644042965,
      "epoch": 4.8648887318968566,
      "grad_norm": 1.390625,
      "learning_rate": 0.00028775351913674967,
      "loss": 5.267,
      "mean_token_accuracy": 0.21732005327939988,
      "num_tokens": 102718544.0,
      "step": 55090
    },
    {
      "entropy": 6.1289643287658695,
      "epoch": 4.865330271988697,
      "grad_norm": 1.359375,
      "learning_rate": 0.00028772207813576177,
      "loss": 5.2076,
      "mean_token_accuracy": 0.2177622899413109,
      "num_tokens": 102728130.0,
      "step": 55095
    },
    {
      "entropy": 6.0326728343963625,
      "epoch": 4.865771812080537,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028769063688555564,
      "loss": 5.2445,
      "mean_token_accuracy": 0.2230493038892746,
      "num_tokens": 102737212.0,
      "step": 55100
    },
    {
      "entropy": 6.134122037887574,
      "epoch": 4.866213352172378,
      "grad_norm": 1.34375,
      "learning_rate": 0.00028765919538674715,
      "loss": 5.3263,
      "mean_token_accuracy": 0.21460118889808655,
      "num_tokens": 102745707.0,
      "step": 55105
    },
    {
      "entropy": 6.0616614818573,
      "epoch": 4.866654892264218,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028762775363995225,
      "loss": 5.2842,
      "mean_token_accuracy": 0.21800937056541442,
      "num_tokens": 102755651.0,
      "step": 55110
    },
    {
      "entropy": 6.0193239688873295,
      "epoch": 4.867096432356058,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002875963116457868,
      "loss": 5.277,
      "mean_token_accuracy": 0.21775470823049545,
      "num_tokens": 102765745.0,
      "step": 55115
    },
    {
      "entropy": 6.075144338607788,
      "epoch": 4.867537972447899,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002875648694048668,
      "loss": 5.2165,
      "mean_token_accuracy": 0.22258132547140122,
      "num_tokens": 102774455.0,
      "step": 55120
    },
    {
      "entropy": 6.095785617828369,
      "epoch": 4.867979512539739,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028753342691780815,
      "loss": 5.2734,
      "mean_token_accuracy": 0.2143101528286934,
      "num_tokens": 102784793.0,
      "step": 55125
    },
    {
      "entropy": 6.058955812454224,
      "epoch": 4.868421052631579,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002875019841852267,
      "loss": 5.2534,
      "mean_token_accuracy": 0.2176300913095474,
      "num_tokens": 102793588.0,
      "step": 55130
    },
    {
      "entropy": 5.9880530834198,
      "epoch": 4.868862592723419,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002874705412077385,
      "loss": 5.1561,
      "mean_token_accuracy": 0.2396731898188591,
      "num_tokens": 102802558.0,
      "step": 55135
    },
    {
      "entropy": 5.932499122619629,
      "epoch": 4.86930413281526,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002874390979859596,
      "loss": 5.1173,
      "mean_token_accuracy": 0.23650178760290147,
      "num_tokens": 102812838.0,
      "step": 55140
    },
    {
      "entropy": 6.076902532577515,
      "epoch": 4.8697456729071,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002874076545205058,
      "loss": 5.2857,
      "mean_token_accuracy": 0.21683234274387359,
      "num_tokens": 102822218.0,
      "step": 55145
    },
    {
      "entropy": 6.100531339645386,
      "epoch": 4.870187212998941,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002873762108119931,
      "loss": 5.3452,
      "mean_token_accuracy": 0.21649006009101868,
      "num_tokens": 102831739.0,
      "step": 55150
    },
    {
      "entropy": 6.008925867080689,
      "epoch": 4.870628753090781,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028734476686103743,
      "loss": 5.2175,
      "mean_token_accuracy": 0.22554219365119935,
      "num_tokens": 102840657.0,
      "step": 55155
    },
    {
      "entropy": 6.040749788284302,
      "epoch": 4.871070293182621,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002873133226682549,
      "loss": 5.2001,
      "mean_token_accuracy": 0.22837148755788803,
      "num_tokens": 102848846.0,
      "step": 55160
    },
    {
      "entropy": 6.054190397262573,
      "epoch": 4.871511833274461,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028728187823426126,
      "loss": 5.1718,
      "mean_token_accuracy": 0.23247093856334686,
      "num_tokens": 102858757.0,
      "step": 55165
    },
    {
      "entropy": 6.082274770736694,
      "epoch": 4.871953373366302,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002872504335596727,
      "loss": 5.2609,
      "mean_token_accuracy": 0.2164839595556259,
      "num_tokens": 102867329.0,
      "step": 55170
    },
    {
      "entropy": 5.984916114807129,
      "epoch": 4.872394913458142,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00028721898864510507,
      "loss": 5.1515,
      "mean_token_accuracy": 0.22636647671461105,
      "num_tokens": 102875894.0,
      "step": 55175
    },
    {
      "entropy": 6.0257645606994625,
      "epoch": 4.872836453549982,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002871875434911745,
      "loss": 5.2838,
      "mean_token_accuracy": 0.21880505979061127,
      "num_tokens": 102885165.0,
      "step": 55180
    },
    {
      "entropy": 5.982967138290405,
      "epoch": 4.873277993641823,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0002871560980984968,
      "loss": 5.1514,
      "mean_token_accuracy": 0.22957687973976135,
      "num_tokens": 102895427.0,
      "step": 55185
    },
    {
      "entropy": 6.0043713569641115,
      "epoch": 4.873719533733663,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028712465246768804,
      "loss": 5.1992,
      "mean_token_accuracy": 0.22959301322698594,
      "num_tokens": 102905999.0,
      "step": 55190
    },
    {
      "entropy": 6.05212664604187,
      "epoch": 4.874161073825503,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028709320659936423,
      "loss": 5.2618,
      "mean_token_accuracy": 0.21500236243009568,
      "num_tokens": 102916160.0,
      "step": 55195
    },
    {
      "entropy": 6.0378186225891115,
      "epoch": 4.874602613917344,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028706176049414137,
      "loss": 5.143,
      "mean_token_accuracy": 0.22668862640857695,
      "num_tokens": 102924502.0,
      "step": 55200
    },
    {
      "entropy": 6.0256147384643555,
      "epoch": 4.875044154009184,
      "grad_norm": 1.4375,
      "learning_rate": 0.00028703031415263546,
      "loss": 5.2461,
      "mean_token_accuracy": 0.22819055020809173,
      "num_tokens": 102933414.0,
      "step": 55205
    },
    {
      "entropy": 6.032432794570923,
      "epoch": 4.875485694101024,
      "grad_norm": 1.453125,
      "learning_rate": 0.00028699886757546265,
      "loss": 5.2949,
      "mean_token_accuracy": 0.22085026502609253,
      "num_tokens": 102941950.0,
      "step": 55210
    },
    {
      "entropy": 6.062118768692017,
      "epoch": 4.875927234192865,
      "grad_norm": 1.328125,
      "learning_rate": 0.00028696742076323876,
      "loss": 5.258,
      "mean_token_accuracy": 0.21889220774173737,
      "num_tokens": 102950891.0,
      "step": 55215
    },
    {
      "entropy": 6.113462209701538,
      "epoch": 4.876368774284705,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002869359737165799,
      "loss": 5.3015,
      "mean_token_accuracy": 0.2187534049153328,
      "num_tokens": 102960610.0,
      "step": 55220
    },
    {
      "entropy": 6.009484148025512,
      "epoch": 4.876810314376545,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002869045264361022,
      "loss": 5.198,
      "mean_token_accuracy": 0.22834363877773284,
      "num_tokens": 102970523.0,
      "step": 55225
    },
    {
      "entropy": 6.0934913635253904,
      "epoch": 4.877251854468386,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002868730789224215,
      "loss": 5.2407,
      "mean_token_accuracy": 0.21671107560396194,
      "num_tokens": 102979889.0,
      "step": 55230
    },
    {
      "entropy": 6.05963659286499,
      "epoch": 4.877693394560226,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00028684163117615396,
      "loss": 5.3126,
      "mean_token_accuracy": 0.21657199412584305,
      "num_tokens": 102989590.0,
      "step": 55235
    },
    {
      "entropy": 5.996762371063232,
      "epoch": 4.878134934652066,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028681018319791563,
      "loss": 5.151,
      "mean_token_accuracy": 0.23033954799175263,
      "num_tokens": 102999203.0,
      "step": 55240
    },
    {
      "entropy": 6.089912939071655,
      "epoch": 4.878576474743907,
      "grad_norm": 1.375,
      "learning_rate": 0.0002867787349883225,
      "loss": 5.3028,
      "mean_token_accuracy": 0.22441934496164323,
      "num_tokens": 103009269.0,
      "step": 55245
    },
    {
      "entropy": 6.151309776306152,
      "epoch": 4.879018014835747,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00028674728654799064,
      "loss": 5.3442,
      "mean_token_accuracy": 0.21088075041770935,
      "num_tokens": 103018874.0,
      "step": 55250
    },
    {
      "entropy": 6.007666301727295,
      "epoch": 4.879459554927587,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002867158378775362,
      "loss": 5.2084,
      "mean_token_accuracy": 0.22514523416757584,
      "num_tokens": 103029221.0,
      "step": 55255
    },
    {
      "entropy": 6.029514408111572,
      "epoch": 4.879901095019428,
      "grad_norm": 1.328125,
      "learning_rate": 0.000286684388977575,
      "loss": 5.2405,
      "mean_token_accuracy": 0.2137842670083046,
      "num_tokens": 103038451.0,
      "step": 55260
    },
    {
      "entropy": 6.03513240814209,
      "epoch": 4.880342635111268,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00028665293984872335,
      "loss": 5.1913,
      "mean_token_accuracy": 0.22335971593856813,
      "num_tokens": 103047759.0,
      "step": 55265
    },
    {
      "entropy": 6.092176628112793,
      "epoch": 4.880784175203108,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00028662149049159724,
      "loss": 5.2838,
      "mean_token_accuracy": 0.21614395529031755,
      "num_tokens": 103057438.0,
      "step": 55270
    },
    {
      "entropy": 6.044149398803711,
      "epoch": 4.881225715294949,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002865900409068128,
      "loss": 5.2469,
      "mean_token_accuracy": 0.22037434726953506,
      "num_tokens": 103067012.0,
      "step": 55275
    },
    {
      "entropy": 6.097521209716797,
      "epoch": 4.881667255386789,
      "grad_norm": 1.359375,
      "learning_rate": 0.000286558591094986,
      "loss": 5.3158,
      "mean_token_accuracy": 0.21698811501264573,
      "num_tokens": 103076217.0,
      "step": 55280
    },
    {
      "entropy": 6.034359931945801,
      "epoch": 4.882108795478629,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028652714105673293,
      "loss": 5.1358,
      "mean_token_accuracy": 0.22933807075023652,
      "num_tokens": 103085417.0,
      "step": 55285
    },
    {
      "entropy": 6.0255554676055905,
      "epoch": 4.882550335570469,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002864956907926698,
      "loss": 5.1631,
      "mean_token_accuracy": 0.2271426498889923,
      "num_tokens": 103095024.0,
      "step": 55290
    },
    {
      "entropy": 6.040237808227539,
      "epoch": 4.88299187566231,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002864642403034126,
      "loss": 5.1949,
      "mean_token_accuracy": 0.22237641513347625,
      "num_tokens": 103104546.0,
      "step": 55295
    },
    {
      "entropy": 6.045297431945801,
      "epoch": 4.88343341575415,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002864327895895774,
      "loss": 5.2618,
      "mean_token_accuracy": 0.21994535326957704,
      "num_tokens": 103114285.0,
      "step": 55300
    },
    {
      "entropy": 6.1253515720367435,
      "epoch": 4.883874955845991,
      "grad_norm": 1.25,
      "learning_rate": 0.0002864013386517804,
      "loss": 5.4431,
      "mean_token_accuracy": 0.203119133412838,
      "num_tokens": 103124315.0,
      "step": 55305
    },
    {
      "entropy": 6.077098369598389,
      "epoch": 4.884316495937831,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002863698874906377,
      "loss": 5.1792,
      "mean_token_accuracy": 0.22859359979629518,
      "num_tokens": 103133963.0,
      "step": 55310
    },
    {
      "entropy": 6.0322709560394285,
      "epoch": 4.884758036029671,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028633843610676525,
      "loss": 5.2121,
      "mean_token_accuracy": 0.22513076514005662,
      "num_tokens": 103142398.0,
      "step": 55315
    },
    {
      "entropy": 6.055155801773071,
      "epoch": 4.8851995761215115,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00028630698450077935,
      "loss": 5.2024,
      "mean_token_accuracy": 0.2258098840713501,
      "num_tokens": 103151833.0,
      "step": 55320
    },
    {
      "entropy": 6.100893926620484,
      "epoch": 4.885641116213352,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028627553267329605,
      "loss": 5.3066,
      "mean_token_accuracy": 0.21443851441144943,
      "num_tokens": 103161777.0,
      "step": 55325
    },
    {
      "entropy": 6.081901121139526,
      "epoch": 4.8860826563051925,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002862440806249315,
      "loss": 5.2291,
      "mean_token_accuracy": 0.22426914274692536,
      "num_tokens": 103170776.0,
      "step": 55330
    },
    {
      "entropy": 6.058515501022339,
      "epoch": 4.8865241963970325,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028621262835630177,
      "loss": 5.2523,
      "mean_token_accuracy": 0.21976661831140518,
      "num_tokens": 103180241.0,
      "step": 55335
    },
    {
      "entropy": 6.003627109527588,
      "epoch": 4.8869657364888734,
      "grad_norm": 1.46875,
      "learning_rate": 0.00028618117586802304,
      "loss": 5.1911,
      "mean_token_accuracy": 0.22579171657562255,
      "num_tokens": 103188525.0,
      "step": 55340
    },
    {
      "entropy": 6.06125054359436,
      "epoch": 4.8874072765807135,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002861497231607115,
      "loss": 5.2827,
      "mean_token_accuracy": 0.21253552585840224,
      "num_tokens": 103197216.0,
      "step": 55345
    },
    {
      "entropy": 6.0707415580749515,
      "epoch": 4.8878488166725536,
      "grad_norm": 1.375,
      "learning_rate": 0.0002861182702349832,
      "loss": 5.2846,
      "mean_token_accuracy": 0.2216160461306572,
      "num_tokens": 103206219.0,
      "step": 55350
    },
    {
      "entropy": 6.025346136093139,
      "epoch": 4.8882903567643945,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028608681709145424,
      "loss": 5.1864,
      "mean_token_accuracy": 0.22885316610336304,
      "num_tokens": 103215141.0,
      "step": 55355
    },
    {
      "entropy": 6.037259674072265,
      "epoch": 4.8887318968562345,
      "grad_norm": 1.453125,
      "learning_rate": 0.00028605536373074085,
      "loss": 5.2232,
      "mean_token_accuracy": 0.22761396765708924,
      "num_tokens": 103224742.0,
      "step": 55360
    },
    {
      "entropy": 5.96082878112793,
      "epoch": 4.889173436948075,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028602391015345927,
      "loss": 5.1949,
      "mean_token_accuracy": 0.23527868688106537,
      "num_tokens": 103232654.0,
      "step": 55365
    },
    {
      "entropy": 6.0081888198852536,
      "epoch": 4.8896149770399155,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028599245636022543,
      "loss": 5.1361,
      "mean_token_accuracy": 0.22559056878089906,
      "num_tokens": 103241964.0,
      "step": 55370
    },
    {
      "entropy": 6.0014325141906735,
      "epoch": 4.890056517131756,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002859610023516557,
      "loss": 5.2015,
      "mean_token_accuracy": 0.22353656142950057,
      "num_tokens": 103250679.0,
      "step": 55375
    },
    {
      "entropy": 6.1061930656433105,
      "epoch": 4.890498057223596,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002859295481283662,
      "loss": 5.2696,
      "mean_token_accuracy": 0.2165765270590782,
      "num_tokens": 103260035.0,
      "step": 55380
    },
    {
      "entropy": 6.035006666183472,
      "epoch": 4.890939597315437,
      "grad_norm": 1.34375,
      "learning_rate": 0.000285898093690973,
      "loss": 5.2918,
      "mean_token_accuracy": 0.21846067160367966,
      "num_tokens": 103269422.0,
      "step": 55385
    },
    {
      "entropy": 6.12196455001831,
      "epoch": 4.891381137407277,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028586663904009234,
      "loss": 5.265,
      "mean_token_accuracy": 0.22245350629091262,
      "num_tokens": 103279318.0,
      "step": 55390
    },
    {
      "entropy": 6.0557506561279295,
      "epoch": 4.891822677499117,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002858351841763404,
      "loss": 5.2864,
      "mean_token_accuracy": 0.22177371382713318,
      "num_tokens": 103288295.0,
      "step": 55395
    },
    {
      "entropy": 5.976905250549317,
      "epoch": 4.892264217590958,
      "grad_norm": 1.34375,
      "learning_rate": 0.00028580372910033344,
      "loss": 5.1851,
      "mean_token_accuracy": 0.23066844046115875,
      "num_tokens": 103296858.0,
      "step": 55400
    },
    {
      "entropy": 6.010330867767334,
      "epoch": 4.892705757682798,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002857722738126874,
      "loss": 5.123,
      "mean_token_accuracy": 0.23234065175056456,
      "num_tokens": 103306417.0,
      "step": 55405
    },
    {
      "entropy": 6.086170864105225,
      "epoch": 4.893147297774638,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002857408183140188,
      "loss": 5.2526,
      "mean_token_accuracy": 0.2207666367292404,
      "num_tokens": 103316501.0,
      "step": 55410
    },
    {
      "entropy": 6.0141034603118895,
      "epoch": 4.893588837866478,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002857093626049436,
      "loss": 5.2657,
      "mean_token_accuracy": 0.21780641824007035,
      "num_tokens": 103326031.0,
      "step": 55415
    },
    {
      "entropy": 6.090510129928589,
      "epoch": 4.894030377958319,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002856779066860782,
      "loss": 5.3134,
      "mean_token_accuracy": 0.21419512629508972,
      "num_tokens": 103335204.0,
      "step": 55420
    },
    {
      "entropy": 6.086677169799804,
      "epoch": 4.894471918050159,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002856464505580386,
      "loss": 5.2738,
      "mean_token_accuracy": 0.22024219781160354,
      "num_tokens": 103345742.0,
      "step": 55425
    },
    {
      "entropy": 6.086591482162476,
      "epoch": 4.894913458142,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028561499422144113,
      "loss": 5.2167,
      "mean_token_accuracy": 0.22310238033533097,
      "num_tokens": 103354788.0,
      "step": 55430
    },
    {
      "entropy": 5.996408367156983,
      "epoch": 4.89535499823384,
      "grad_norm": 1.375,
      "learning_rate": 0.00028558353767690187,
      "loss": 5.2636,
      "mean_token_accuracy": 0.22284620106220246,
      "num_tokens": 103363949.0,
      "step": 55435
    },
    {
      "entropy": 6.020040893554688,
      "epoch": 4.89579653832568,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028555208092503713,
      "loss": 5.2357,
      "mean_token_accuracy": 0.23017463386058806,
      "num_tokens": 103372543.0,
      "step": 55440
    },
    {
      "entropy": 6.026459121704102,
      "epoch": 4.89623807841752,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002855206239664632,
      "loss": 5.2433,
      "mean_token_accuracy": 0.2247896149754524,
      "num_tokens": 103381373.0,
      "step": 55445
    },
    {
      "entropy": 6.074599981307983,
      "epoch": 4.896679618509361,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00028548916680179616,
      "loss": 5.2503,
      "mean_token_accuracy": 0.22323601245880126,
      "num_tokens": 103390179.0,
      "step": 55450
    },
    {
      "entropy": 6.047282457351685,
      "epoch": 4.897121158601201,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00028545770943165234,
      "loss": 5.2791,
      "mean_token_accuracy": 0.22155072987079621,
      "num_tokens": 103400014.0,
      "step": 55455
    },
    {
      "entropy": 5.973917436599732,
      "epoch": 4.897562698693041,
      "grad_norm": 1.375,
      "learning_rate": 0.0002854262518566479,
      "loss": 5.1357,
      "mean_token_accuracy": 0.2424969866871834,
      "num_tokens": 103408990.0,
      "step": 55460
    },
    {
      "entropy": 5.997885751724243,
      "epoch": 4.898004238784882,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002853947940773993,
      "loss": 5.2075,
      "mean_token_accuracy": 0.22221532911062242,
      "num_tokens": 103419368.0,
      "step": 55465
    },
    {
      "entropy": 6.055965375900269,
      "epoch": 4.898445778876722,
      "grad_norm": 1.34375,
      "learning_rate": 0.00028536333609452234,
      "loss": 5.3304,
      "mean_token_accuracy": 0.21386135071516038,
      "num_tokens": 103428964.0,
      "step": 55470
    },
    {
      "entropy": 6.0124828815460205,
      "epoch": 4.898887318968562,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002853318779086337,
      "loss": 5.1348,
      "mean_token_accuracy": 0.2320249393582344,
      "num_tokens": 103438074.0,
      "step": 55475
    },
    {
      "entropy": 6.056667900085449,
      "epoch": 4.899328859060403,
      "grad_norm": 1.375,
      "learning_rate": 0.00028530041952034934,
      "loss": 5.2435,
      "mean_token_accuracy": 0.21516669690608978,
      "num_tokens": 103447395.0,
      "step": 55480
    },
    {
      "entropy": 6.034773063659668,
      "epoch": 4.899770399152243,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002852689609302857,
      "loss": 5.194,
      "mean_token_accuracy": 0.22695554345846175,
      "num_tokens": 103457012.0,
      "step": 55485
    },
    {
      "entropy": 6.111459970474243,
      "epoch": 4.900211939244083,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00028523750213905894,
      "loss": 5.3308,
      "mean_token_accuracy": 0.21702015548944473,
      "num_tokens": 103467729.0,
      "step": 55490
    },
    {
      "entropy": 6.132442760467529,
      "epoch": 4.900653479335924,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002852060431472853,
      "loss": 5.3073,
      "mean_token_accuracy": 0.2089808315038681,
      "num_tokens": 103477857.0,
      "step": 55495
    },
    {
      "entropy": 6.0900187492370605,
      "epoch": 4.901095019427764,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00028517458395558107,
      "loss": 5.1582,
      "mean_token_accuracy": 0.22851658314466478,
      "num_tokens": 103486895.0,
      "step": 55500
    },
    {
      "entropy": 6.041134691238403,
      "epoch": 4.901536559519604,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002851431245645626,
      "loss": 5.2811,
      "mean_token_accuracy": 0.21841103732585906,
      "num_tokens": 103496293.0,
      "step": 55505
    },
    {
      "entropy": 5.982244825363159,
      "epoch": 4.901978099611445,
      "grad_norm": 1.34375,
      "learning_rate": 0.00028511166497484597,
      "loss": 5.1527,
      "mean_token_accuracy": 0.2290387973189354,
      "num_tokens": 103506719.0,
      "step": 55510
    },
    {
      "entropy": 6.0560462951660154,
      "epoch": 4.902419639703285,
      "grad_norm": 1.40625,
      "learning_rate": 0.00028508020518704764,
      "loss": 5.2658,
      "mean_token_accuracy": 0.22269216775894166,
      "num_tokens": 103516450.0,
      "step": 55515
    },
    {
      "entropy": 6.057534503936767,
      "epoch": 4.902861179795125,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002850487452017839,
      "loss": 5.3013,
      "mean_token_accuracy": 0.22059336602687835,
      "num_tokens": 103526325.0,
      "step": 55520
    },
    {
      "entropy": 6.118698930740356,
      "epoch": 4.903302719886966,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002850172850196708,
      "loss": 5.3077,
      "mean_token_accuracy": 0.21650459468364716,
      "num_tokens": 103535632.0,
      "step": 55525
    },
    {
      "entropy": 6.028482294082641,
      "epoch": 4.903744259978806,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028498582464132487,
      "loss": 5.1911,
      "mean_token_accuracy": 0.2275482937693596,
      "num_tokens": 103545592.0,
      "step": 55530
    },
    {
      "entropy": 6.039769411087036,
      "epoch": 4.904185800070646,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002849543640673624,
      "loss": 5.1781,
      "mean_token_accuracy": 0.23050940930843353,
      "num_tokens": 103554538.0,
      "step": 55535
    },
    {
      "entropy": 6.036166334152222,
      "epoch": 4.904627340162487,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002849229032983994,
      "loss": 5.1806,
      "mean_token_accuracy": 0.22937033474445342,
      "num_tokens": 103564343.0,
      "step": 55540
    },
    {
      "entropy": 6.138940286636353,
      "epoch": 4.905068880254327,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002848914423350525,
      "loss": 5.2995,
      "mean_token_accuracy": 0.21899387538433074,
      "num_tokens": 103573917.0,
      "step": 55545
    },
    {
      "entropy": 6.018236255645752,
      "epoch": 4.905510420346167,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002848599811779379,
      "loss": 5.1923,
      "mean_token_accuracy": 0.22492597252130508,
      "num_tokens": 103582992.0,
      "step": 55550
    },
    {
      "entropy": 6.071676111221313,
      "epoch": 4.905951960438008,
      "grad_norm": 1.34375,
      "learning_rate": 0.00028482851982767184,
      "loss": 5.2867,
      "mean_token_accuracy": 0.21726544052362443,
      "num_tokens": 103592597.0,
      "step": 55555
    },
    {
      "entropy": 6.117518949508667,
      "epoch": 4.906393500529848,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028479705828487063,
      "loss": 5.3494,
      "mean_token_accuracy": 0.2147160768508911,
      "num_tokens": 103602456.0,
      "step": 55560
    },
    {
      "entropy": 6.066470670700073,
      "epoch": 4.906835040621688,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002847655965501506,
      "loss": 5.1932,
      "mean_token_accuracy": 0.2263448193669319,
      "num_tokens": 103611155.0,
      "step": 55565
    },
    {
      "entropy": 6.042045640945434,
      "epoch": 4.907276580713528,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028473413462412817,
      "loss": 5.2496,
      "mean_token_accuracy": 0.22653857320547105,
      "num_tokens": 103619676.0,
      "step": 55570
    },
    {
      "entropy": 6.015290403366089,
      "epoch": 4.907718120805369,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028470267250741956,
      "loss": 5.2483,
      "mean_token_accuracy": 0.2231249451637268,
      "num_tokens": 103628772.0,
      "step": 55575
    },
    {
      "entropy": 5.998124361038208,
      "epoch": 4.908159660897209,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00028467121020064103,
      "loss": 5.2186,
      "mean_token_accuracy": 0.22566382437944413,
      "num_tokens": 103638041.0,
      "step": 55580
    },
    {
      "entropy": 6.045027256011963,
      "epoch": 4.90860120098905,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00028463974770440907,
      "loss": 5.2385,
      "mean_token_accuracy": 0.22079157680273057,
      "num_tokens": 103646932.0,
      "step": 55585
    },
    {
      "entropy": 6.009152030944824,
      "epoch": 4.90904274108089,
      "grad_norm": 1.375,
      "learning_rate": 0.0002846082850193399,
      "loss": 5.2311,
      "mean_token_accuracy": 0.22235515862703323,
      "num_tokens": 103655902.0,
      "step": 55590
    },
    {
      "entropy": 5.9542622566223145,
      "epoch": 4.90948428117273,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00028457682214604995,
      "loss": 5.1906,
      "mean_token_accuracy": 0.23415677845478058,
      "num_tokens": 103665102.0,
      "step": 55595
    },
    {
      "entropy": 6.004604864120483,
      "epoch": 4.9099258212645704,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002845453590851554,
      "loss": 5.119,
      "mean_token_accuracy": 0.23347872495651245,
      "num_tokens": 103674018.0,
      "step": 55600
    },
    {
      "entropy": 6.053213119506836,
      "epoch": 4.910367361356411,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028451389583727276,
      "loss": 5.2394,
      "mean_token_accuracy": 0.2282985433936119,
      "num_tokens": 103684790.0,
      "step": 55605
    },
    {
      "entropy": 5.967590951919556,
      "epoch": 4.910808901448251,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002844824324030183,
      "loss": 5.1513,
      "mean_token_accuracy": 0.2243446946144104,
      "num_tokens": 103693990.0,
      "step": 55610
    },
    {
      "entropy": 6.044803142547607,
      "epoch": 4.9112504415400915,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002844509687830084,
      "loss": 5.1907,
      "mean_token_accuracy": 0.22616310566663742,
      "num_tokens": 103703099.0,
      "step": 55615
    },
    {
      "entropy": 5.995152187347412,
      "epoch": 4.911691981631932,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002844195049778594,
      "loss": 5.1597,
      "mean_token_accuracy": 0.22758469581604004,
      "num_tokens": 103711746.0,
      "step": 55620
    },
    {
      "entropy": 6.00442271232605,
      "epoch": 4.9121335217237725,
      "grad_norm": 1.328125,
      "learning_rate": 0.00028438804098818767,
      "loss": 5.1684,
      "mean_token_accuracy": 0.2236298829317093,
      "num_tokens": 103720492.0,
      "step": 55625
    },
    {
      "entropy": 6.139531946182251,
      "epoch": 4.9125750618156125,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002843565768146095,
      "loss": 5.3237,
      "mean_token_accuracy": 0.21206597685813905,
      "num_tokens": 103729866.0,
      "step": 55630
    },
    {
      "entropy": 6.012110900878906,
      "epoch": 4.9130166019074535,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002843251124577414,
      "loss": 5.1948,
      "mean_token_accuracy": 0.23077826499938964,
      "num_tokens": 103738447.0,
      "step": 55635
    },
    {
      "entropy": 6.030248546600342,
      "epoch": 4.9134581419992935,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002842936479181996,
      "loss": 5.2153,
      "mean_token_accuracy": 0.22399666160345078,
      "num_tokens": 103747214.0,
      "step": 55640
    },
    {
      "entropy": 5.925600957870484,
      "epoch": 4.913899682091134,
      "grad_norm": 1.390625,
      "learning_rate": 0.00028426218319660044,
      "loss": 5.1446,
      "mean_token_accuracy": 0.2304267853498459,
      "num_tokens": 103756526.0,
      "step": 55645
    },
    {
      "entropy": 6.021153736114502,
      "epoch": 4.9143412221829745,
      "grad_norm": 1.484375,
      "learning_rate": 0.00028423071829356044,
      "loss": 5.2938,
      "mean_token_accuracy": 0.21574838161468507,
      "num_tokens": 103765626.0,
      "step": 55650
    },
    {
      "entropy": 6.054450464248657,
      "epoch": 4.914782762274815,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002841992532096959,
      "loss": 5.2551,
      "mean_token_accuracy": 0.2264857307076454,
      "num_tokens": 103775444.0,
      "step": 55655
    },
    {
      "entropy": 6.046548700332641,
      "epoch": 4.915224302366655,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002841677879456233,
      "loss": 5.233,
      "mean_token_accuracy": 0.23121203035116195,
      "num_tokens": 103784668.0,
      "step": 55660
    },
    {
      "entropy": 6.036210680007935,
      "epoch": 4.915665842458496,
      "grad_norm": 1.390625,
      "learning_rate": 0.00028413632250195894,
      "loss": 5.2152,
      "mean_token_accuracy": 0.21934465318918228,
      "num_tokens": 103793895.0,
      "step": 55665
    },
    {
      "entropy": 6.010747480392456,
      "epoch": 4.916107382550336,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00028410485687931916,
      "loss": 5.1629,
      "mean_token_accuracy": 0.2276706576347351,
      "num_tokens": 103802895.0,
      "step": 55670
    },
    {
      "entropy": 6.051443958282471,
      "epoch": 4.916548922642176,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028407339107832036,
      "loss": 5.2121,
      "mean_token_accuracy": 0.23547389954328538,
      "num_tokens": 103811978.0,
      "step": 55675
    },
    {
      "entropy": 6.059592437744141,
      "epoch": 4.916990462734017,
      "grad_norm": 1.390625,
      "learning_rate": 0.00028404192509957915,
      "loss": 5.2824,
      "mean_token_accuracy": 0.22179624438285828,
      "num_tokens": 103820928.0,
      "step": 55680
    },
    {
      "entropy": 6.035628986358643,
      "epoch": 4.917432002825857,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028401045894371166,
      "loss": 5.2556,
      "mean_token_accuracy": 0.21345915645360947,
      "num_tokens": 103830253.0,
      "step": 55685
    },
    {
      "entropy": 5.955310106277466,
      "epoch": 4.917873542917697,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00028397899261133443,
      "loss": 5.0928,
      "mean_token_accuracy": 0.23586656004190446,
      "num_tokens": 103839665.0,
      "step": 55690
    },
    {
      "entropy": 6.07917013168335,
      "epoch": 4.918315083009537,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002839475261030638,
      "loss": 5.237,
      "mean_token_accuracy": 0.222864505648613,
      "num_tokens": 103848673.0,
      "step": 55695
    },
    {
      "entropy": 6.069603395462036,
      "epoch": 4.918756623101378,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00028391605941951623,
      "loss": 5.2766,
      "mean_token_accuracy": 0.22318734377622604,
      "num_tokens": 103857891.0,
      "step": 55700
    },
    {
      "entropy": 6.03624587059021,
      "epoch": 4.919198163193218,
      "grad_norm": 1.34375,
      "learning_rate": 0.00028388459256130816,
      "loss": 5.2375,
      "mean_token_accuracy": 0.22710915058851242,
      "num_tokens": 103867197.0,
      "step": 55705
    },
    {
      "entropy": 6.052307224273681,
      "epoch": 4.919639703285059,
      "grad_norm": 1.328125,
      "learning_rate": 0.00028385312552905593,
      "loss": 5.2067,
      "mean_token_accuracy": 0.2264177158474922,
      "num_tokens": 103876540.0,
      "step": 55710
    },
    {
      "entropy": 6.10645866394043,
      "epoch": 4.920081243376899,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002838216583233761,
      "loss": 5.2856,
      "mean_token_accuracy": 0.21887914538383485,
      "num_tokens": 103885638.0,
      "step": 55715
    },
    {
      "entropy": 5.985408449172974,
      "epoch": 4.920522783468739,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002837901909448849,
      "loss": 5.2019,
      "mean_token_accuracy": 0.22118840366601944,
      "num_tokens": 103895312.0,
      "step": 55720
    },
    {
      "entropy": 6.043701457977295,
      "epoch": 4.920964323560579,
      "grad_norm": 1.328125,
      "learning_rate": 0.00028375872339419893,
      "loss": 5.2548,
      "mean_token_accuracy": 0.22208064496517183,
      "num_tokens": 103904057.0,
      "step": 55725
    },
    {
      "entropy": 6.050293731689453,
      "epoch": 4.92140586365242,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002837272556719345,
      "loss": 5.361,
      "mean_token_accuracy": 0.2141176789999008,
      "num_tokens": 103914264.0,
      "step": 55730
    },
    {
      "entropy": 6.011467933654785,
      "epoch": 4.92184740374426,
      "grad_norm": 1.421875,
      "learning_rate": 0.00028369578777870814,
      "loss": 5.1523,
      "mean_token_accuracy": 0.22326601296663284,
      "num_tokens": 103923186.0,
      "step": 55735
    },
    {
      "entropy": 6.0697096824646,
      "epoch": 4.9222889438361,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002836643197151362,
      "loss": 5.3139,
      "mean_token_accuracy": 0.21876053512096405,
      "num_tokens": 103932319.0,
      "step": 55740
    },
    {
      "entropy": 6.0560449123382565,
      "epoch": 4.922730483927941,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002836328514818352,
      "loss": 5.2713,
      "mean_token_accuracy": 0.22212502658367156,
      "num_tokens": 103941804.0,
      "step": 55745
    },
    {
      "entropy": 6.01679573059082,
      "epoch": 4.923172024019781,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002836013830794215,
      "loss": 5.202,
      "mean_token_accuracy": 0.22427988648414612,
      "num_tokens": 103950893.0,
      "step": 55750
    },
    {
      "entropy": 6.088992691040039,
      "epoch": 4.923613564111621,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002835699145085117,
      "loss": 5.3093,
      "mean_token_accuracy": 0.2163187623023987,
      "num_tokens": 103960705.0,
      "step": 55755
    },
    {
      "entropy": 6.07949800491333,
      "epoch": 4.924055104203462,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028353844576972203,
      "loss": 5.275,
      "mean_token_accuracy": 0.21306130141019822,
      "num_tokens": 103968925.0,
      "step": 55760
    },
    {
      "entropy": 5.982359933853149,
      "epoch": 4.924496644295302,
      "grad_norm": 1.328125,
      "learning_rate": 0.00028350697686366916,
      "loss": 5.2247,
      "mean_token_accuracy": 0.22196399867534639,
      "num_tokens": 103978371.0,
      "step": 55765
    },
    {
      "entropy": 6.083266401290894,
      "epoch": 4.924938184387142,
      "grad_norm": 1.4375,
      "learning_rate": 0.00028347550779096937,
      "loss": 5.3106,
      "mean_token_accuracy": 0.21362218260765076,
      "num_tokens": 103987152.0,
      "step": 55770
    },
    {
      "entropy": 6.1246131420135494,
      "epoch": 4.925379724478983,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028344403855223914,
      "loss": 5.276,
      "mean_token_accuracy": 0.21898669749498367,
      "num_tokens": 103995808.0,
      "step": 55775
    },
    {
      "entropy": 6.078204250335693,
      "epoch": 4.925821264570823,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00028341256914809517,
      "loss": 5.2877,
      "mean_token_accuracy": 0.21330465525388717,
      "num_tokens": 104006095.0,
      "step": 55780
    },
    {
      "entropy": 5.98389744758606,
      "epoch": 4.926262804662663,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00028338109957915366,
      "loss": 5.1536,
      "mean_token_accuracy": 0.22825625836849212,
      "num_tokens": 104015070.0,
      "step": 55785
    },
    {
      "entropy": 5.993691492080688,
      "epoch": 4.926704344754504,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028334962984603125,
      "loss": 5.1083,
      "mean_token_accuracy": 0.23400361090898514,
      "num_tokens": 104024227.0,
      "step": 55790
    },
    {
      "entropy": 6.006072521209717,
      "epoch": 4.927145884846344,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028331815994934426,
      "loss": 5.1535,
      "mean_token_accuracy": 0.23300192654132842,
      "num_tokens": 104033958.0,
      "step": 55795
    },
    {
      "entropy": 6.080094575881958,
      "epoch": 4.927587424938184,
      "grad_norm": 1.5,
      "learning_rate": 0.0002832866898897092,
      "loss": 5.312,
      "mean_token_accuracy": 0.22391662895679473,
      "num_tokens": 104042739.0,
      "step": 55800
    },
    {
      "entropy": 6.035082817077637,
      "epoch": 4.928028965030025,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00028325521966774276,
      "loss": 5.2504,
      "mean_token_accuracy": 0.22041285037994385,
      "num_tokens": 104052121.0,
      "step": 55805
    },
    {
      "entropy": 6.020398950576782,
      "epoch": 4.928470505121865,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002832237492840612,
      "loss": 5.2529,
      "mean_token_accuracy": 0.22564626187086106,
      "num_tokens": 104061616.0,
      "step": 55810
    },
    {
      "entropy": 5.987677764892578,
      "epoch": 4.928912045213705,
      "grad_norm": 1.453125,
      "learning_rate": 0.000283192278739281,
      "loss": 5.1027,
      "mean_token_accuracy": 0.23666051179170608,
      "num_tokens": 104069863.0,
      "step": 55815
    },
    {
      "entropy": 6.048363780975341,
      "epoch": 4.929353585305546,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002831608080340188,
      "loss": 5.248,
      "mean_token_accuracy": 0.21578313112258912,
      "num_tokens": 104078657.0,
      "step": 55820
    },
    {
      "entropy": 6.0152486801147464,
      "epoch": 4.929795125397386,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002831293371688909,
      "loss": 5.1667,
      "mean_token_accuracy": 0.23054830282926558,
      "num_tokens": 104087484.0,
      "step": 55825
    },
    {
      "entropy": 6.065231037139893,
      "epoch": 4.930236665489226,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028309786614451397,
      "loss": 5.2383,
      "mean_token_accuracy": 0.22688756436109542,
      "num_tokens": 104096757.0,
      "step": 55830
    },
    {
      "entropy": 6.058841896057129,
      "epoch": 4.930678205581067,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002830663949615044,
      "loss": 5.2687,
      "mean_token_accuracy": 0.22586465179920195,
      "num_tokens": 104105826.0,
      "step": 55835
    },
    {
      "entropy": 6.09197850227356,
      "epoch": 4.931119745672907,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028303492362047886,
      "loss": 5.3339,
      "mean_token_accuracy": 0.21731724590063095,
      "num_tokens": 104115776.0,
      "step": 55840
    },
    {
      "entropy": 6.123762130737305,
      "epoch": 4.931561285764747,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002830034521220537,
      "loss": 5.276,
      "mean_token_accuracy": 0.21963244378566743,
      "num_tokens": 104125584.0,
      "step": 55845
    },
    {
      "entropy": 6.081539297103882,
      "epoch": 4.932002825856587,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028297198046684543,
      "loss": 5.3206,
      "mean_token_accuracy": 0.2173621192574501,
      "num_tokens": 104134954.0,
      "step": 55850
    },
    {
      "entropy": 6.077397871017456,
      "epoch": 4.932444365948428,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028294050865547063,
      "loss": 5.2983,
      "mean_token_accuracy": 0.21182382851839066,
      "num_tokens": 104143860.0,
      "step": 55855
    },
    {
      "entropy": 6.057211685180664,
      "epoch": 4.932885906040268,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002829090366885457,
      "loss": 5.2912,
      "mean_token_accuracy": 0.21816257536411285,
      "num_tokens": 104153530.0,
      "step": 55860
    },
    {
      "entropy": 6.044768381118774,
      "epoch": 4.933327446132109,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002828775645666873,
      "loss": 5.2756,
      "mean_token_accuracy": 0.22401908040046692,
      "num_tokens": 104164268.0,
      "step": 55865
    },
    {
      "entropy": 6.0240888595581055,
      "epoch": 4.933768986223949,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002828460922905119,
      "loss": 5.2302,
      "mean_token_accuracy": 0.22719531804323195,
      "num_tokens": 104173848.0,
      "step": 55870
    },
    {
      "entropy": 6.01424732208252,
      "epoch": 4.934210526315789,
      "grad_norm": 1.34375,
      "learning_rate": 0.000282814619860636,
      "loss": 5.1678,
      "mean_token_accuracy": 0.2323947563767433,
      "num_tokens": 104182942.0,
      "step": 55875
    },
    {
      "entropy": 5.977741813659668,
      "epoch": 4.934652066407629,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028278314727767616,
      "loss": 5.1838,
      "mean_token_accuracy": 0.22958560436964034,
      "num_tokens": 104193011.0,
      "step": 55880
    },
    {
      "entropy": 6.1471953868865965,
      "epoch": 4.93509360649947,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002827516745422488,
      "loss": 5.3453,
      "mean_token_accuracy": 0.21348896771669387,
      "num_tokens": 104202961.0,
      "step": 55885
    },
    {
      "entropy": 6.141497039794922,
      "epoch": 4.93553514659131,
      "grad_norm": 1.421875,
      "learning_rate": 0.00028272020165497064,
      "loss": 5.1836,
      "mean_token_accuracy": 0.2319609582424164,
      "num_tokens": 104211538.0,
      "step": 55890
    },
    {
      "entropy": 6.069897651672363,
      "epoch": 4.9359766866831505,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002826887286164581,
      "loss": 5.2652,
      "mean_token_accuracy": 0.2209451675415039,
      "num_tokens": 104219894.0,
      "step": 55895
    },
    {
      "entropy": 6.024998331069947,
      "epoch": 4.936418226774991,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028265725542732774,
      "loss": 5.2365,
      "mean_token_accuracy": 0.21749506443738936,
      "num_tokens": 104228916.0,
      "step": 55900
    },
    {
      "entropy": 5.956444835662841,
      "epoch": 4.9368597668668315,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002826257820881961,
      "loss": 5.1382,
      "mean_token_accuracy": 0.2310498982667923,
      "num_tokens": 104238183.0,
      "step": 55905
    },
    {
      "entropy": 6.05078444480896,
      "epoch": 4.9373013069586715,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002825943085996797,
      "loss": 5.2505,
      "mean_token_accuracy": 0.21634772419929504,
      "num_tokens": 104247214.0,
      "step": 55910
    },
    {
      "entropy": 6.068523645401001,
      "epoch": 4.9377428470505125,
      "grad_norm": 1.46875,
      "learning_rate": 0.00028256283496239516,
      "loss": 5.2621,
      "mean_token_accuracy": 0.21697497814893724,
      "num_tokens": 104256477.0,
      "step": 55915
    },
    {
      "entropy": 6.0846678733825685,
      "epoch": 4.9381843871423525,
      "grad_norm": 1.34375,
      "learning_rate": 0.000282531361176959,
      "loss": 5.2983,
      "mean_token_accuracy": 0.21471430957317353,
      "num_tokens": 104265683.0,
      "step": 55920
    },
    {
      "entropy": 6.134388446807861,
      "epoch": 4.938625927234193,
      "grad_norm": 1.359375,
      "learning_rate": 0.00028249988724398766,
      "loss": 5.3053,
      "mean_token_accuracy": 0.21735720336437225,
      "num_tokens": 104275497.0,
      "step": 55925
    },
    {
      "entropy": 6.120714092254639,
      "epoch": 4.9390674673260335,
      "grad_norm": 1.34375,
      "learning_rate": 0.00028246841316409795,
      "loss": 5.2295,
      "mean_token_accuracy": 0.22059171944856643,
      "num_tokens": 104284445.0,
      "step": 55930
    },
    {
      "entropy": 6.044152021408081,
      "epoch": 4.939509007417874,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002824369389379061,
      "loss": 5.24,
      "mean_token_accuracy": 0.22086252719163896,
      "num_tokens": 104293364.0,
      "step": 55935
    },
    {
      "entropy": 6.042601919174194,
      "epoch": 4.939950547509714,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028240546456602893,
      "loss": 5.2559,
      "mean_token_accuracy": 0.22209815979003905,
      "num_tokens": 104304085.0,
      "step": 55940
    },
    {
      "entropy": 6.125866508483886,
      "epoch": 4.940392087601555,
      "grad_norm": 1.359375,
      "learning_rate": 0.00028237399004908296,
      "loss": 5.3502,
      "mean_token_accuracy": 0.21283169686794282,
      "num_tokens": 104313915.0,
      "step": 55945
    },
    {
      "entropy": 6.071991205215454,
      "epoch": 4.940833627693395,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002823425153876846,
      "loss": 5.2688,
      "mean_token_accuracy": 0.22332937866449357,
      "num_tokens": 104323834.0,
      "step": 55950
    },
    {
      "entropy": 6.128975439071655,
      "epoch": 4.941275167785235,
      "grad_norm": 1.4375,
      "learning_rate": 0.00028231104058245064,
      "loss": 5.3077,
      "mean_token_accuracy": 0.21575336903333664,
      "num_tokens": 104333372.0,
      "step": 55955
    },
    {
      "entropy": 6.013987445831299,
      "epoch": 4.941716707877076,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028227956563399746,
      "loss": 5.1981,
      "mean_token_accuracy": 0.22788852602243423,
      "num_tokens": 104342936.0,
      "step": 55960
    },
    {
      "entropy": 5.991474628448486,
      "epoch": 4.942158247968916,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002822480905429418,
      "loss": 5.1209,
      "mean_token_accuracy": 0.2309712991118431,
      "num_tokens": 104351330.0,
      "step": 55965
    },
    {
      "entropy": 6.066426229476929,
      "epoch": 4.942599788060756,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002822166153099002,
      "loss": 5.2511,
      "mean_token_accuracy": 0.21917422264814376,
      "num_tokens": 104361209.0,
      "step": 55970
    },
    {
      "entropy": 6.091613245010376,
      "epoch": 4.943041328152596,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00028218513993548915,
      "loss": 5.2902,
      "mean_token_accuracy": 0.215957573056221,
      "num_tokens": 104370487.0,
      "step": 55975
    },
    {
      "entropy": 6.041250419616699,
      "epoch": 4.943482868244437,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002821536644203253,
      "loss": 5.2444,
      "mean_token_accuracy": 0.2265189066529274,
      "num_tokens": 104379609.0,
      "step": 55980
    },
    {
      "entropy": 6.024241209030151,
      "epoch": 4.943924408336277,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002821221887650253,
      "loss": 5.2783,
      "mean_token_accuracy": 0.22090939581394195,
      "num_tokens": 104388999.0,
      "step": 55985
    },
    {
      "entropy": 6.143504190444946,
      "epoch": 4.944365948428118,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002820907129702056,
      "loss": 5.3035,
      "mean_token_accuracy": 0.21467583179473876,
      "num_tokens": 104398588.0,
      "step": 55990
    },
    {
      "entropy": 6.0826108932495115,
      "epoch": 4.944807488519958,
      "grad_norm": 1.390625,
      "learning_rate": 0.000282059237036483,
      "loss": 5.1893,
      "mean_token_accuracy": 0.23435863107442856,
      "num_tokens": 104407903.0,
      "step": 55995
    },
    {
      "entropy": 6.0228760719299315,
      "epoch": 4.945249028611798,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002820277609644738,
      "loss": 5.1993,
      "mean_token_accuracy": 0.22156699746847153,
      "num_tokens": 104417513.0,
      "step": 56000
    },
    {
      "entropy": 6.072049808502197,
      "epoch": 4.945690568703638,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002819962847547948,
      "loss": 5.2828,
      "mean_token_accuracy": 0.21683287918567656,
      "num_tokens": 104426780.0,
      "step": 56005
    },
    {
      "entropy": 6.041393709182739,
      "epoch": 4.946132108795479,
      "grad_norm": 1.34375,
      "learning_rate": 0.00028196480840806255,
      "loss": 5.1855,
      "mean_token_accuracy": 0.22602000534534455,
      "num_tokens": 104435830.0,
      "step": 56010
    },
    {
      "entropy": 5.988212919235229,
      "epoch": 4.946573648887319,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00028193333192489366,
      "loss": 5.1867,
      "mean_token_accuracy": 0.23367652744054795,
      "num_tokens": 104446147.0,
      "step": 56015
    },
    {
      "entropy": 6.041865062713623,
      "epoch": 4.947015188979159,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002819018553059048,
      "loss": 5.2269,
      "mean_token_accuracy": 0.22201078683137893,
      "num_tokens": 104456210.0,
      "step": 56020
    },
    {
      "entropy": 6.090781497955322,
      "epoch": 4.947456729071,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002818703785517124,
      "loss": 5.2568,
      "mean_token_accuracy": 0.2198557436466217,
      "num_tokens": 104465375.0,
      "step": 56025
    },
    {
      "entropy": 6.045855903625489,
      "epoch": 4.94789826916284,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00028183890166293336,
      "loss": 5.1673,
      "mean_token_accuracy": 0.22812844961881637,
      "num_tokens": 104474577.0,
      "step": 56030
    },
    {
      "entropy": 6.005025339126587,
      "epoch": 4.94833980925468,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000281807424640184,
      "loss": 5.2576,
      "mean_token_accuracy": 0.2279536172747612,
      "num_tokens": 104483898.0,
      "step": 56035
    },
    {
      "entropy": 5.990642547607422,
      "epoch": 4.948781349346521,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028177594748408116,
      "loss": 5.1996,
      "mean_token_accuracy": 0.22379175573587418,
      "num_tokens": 104492788.0,
      "step": 56040
    },
    {
      "entropy": 6.036661577224732,
      "epoch": 4.949222889438361,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002817444701952413,
      "loss": 5.2438,
      "mean_token_accuracy": 0.2208358183503151,
      "num_tokens": 104501533.0,
      "step": 56045
    },
    {
      "entropy": 6.011901903152466,
      "epoch": 4.949664429530201,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028171299277428107,
      "loss": 5.201,
      "mean_token_accuracy": 0.22507418245077132,
      "num_tokens": 104511183.0,
      "step": 56050
    },
    {
      "entropy": 6.006478691101075,
      "epoch": 4.950105969622042,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002816815152218172,
      "loss": 5.2854,
      "mean_token_accuracy": 0.22193218171596527,
      "num_tokens": 104519949.0,
      "step": 56055
    },
    {
      "entropy": 5.975676536560059,
      "epoch": 4.950547509713882,
      "grad_norm": 1.40625,
      "learning_rate": 0.00028165003753846625,
      "loss": 5.2749,
      "mean_token_accuracy": 0.21858235001564025,
      "num_tokens": 104528529.0,
      "step": 56060
    },
    {
      "entropy": 6.083883428573609,
      "epoch": 4.950989049805722,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028161855972484484,
      "loss": 5.2284,
      "mean_token_accuracy": 0.22295144349336624,
      "num_tokens": 104537529.0,
      "step": 56065
    },
    {
      "entropy": 6.061558246612549,
      "epoch": 4.951430589897563,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002815870817815695,
      "loss": 5.1975,
      "mean_token_accuracy": 0.22453005760908126,
      "num_tokens": 104546137.0,
      "step": 56070
    },
    {
      "entropy": 6.057451868057251,
      "epoch": 4.951872129989403,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00028155560370925713,
      "loss": 5.2802,
      "mean_token_accuracy": 0.2185784786939621,
      "num_tokens": 104555268.0,
      "step": 56075
    },
    {
      "entropy": 6.045480108261108,
      "epoch": 4.952313670081243,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028152412550852416,
      "loss": 5.2745,
      "mean_token_accuracy": 0.2215473622083664,
      "num_tokens": 104564751.0,
      "step": 56080
    },
    {
      "entropy": 6.103043937683106,
      "epoch": 4.952755210173084,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00028149264717998717,
      "loss": 5.2905,
      "mean_token_accuracy": 0.21504993885755538,
      "num_tokens": 104573708.0,
      "step": 56085
    },
    {
      "entropy": 6.016295385360718,
      "epoch": 4.953196750264924,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028146116872426296,
      "loss": 5.1824,
      "mean_token_accuracy": 0.23006971925497055,
      "num_tokens": 104582982.0,
      "step": 56090
    },
    {
      "entropy": 6.042828893661499,
      "epoch": 4.953638290356764,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028142969014196824,
      "loss": 5.2269,
      "mean_token_accuracy": 0.22242909967899321,
      "num_tokens": 104591944.0,
      "step": 56095
    },
    {
      "entropy": 6.0100141048431395,
      "epoch": 4.954079830448605,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028139821143371947,
      "loss": 5.2333,
      "mean_token_accuracy": 0.2267719730734825,
      "num_tokens": 104601534.0,
      "step": 56100
    },
    {
      "entropy": 6.080660676956176,
      "epoch": 4.954521370540445,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002813667326001334,
      "loss": 5.2584,
      "mean_token_accuracy": 0.220837864279747,
      "num_tokens": 104610188.0,
      "step": 56105
    },
    {
      "entropy": 6.006811571121216,
      "epoch": 4.954962910632285,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002813352536418267,
      "loss": 5.2039,
      "mean_token_accuracy": 0.22129353135824203,
      "num_tokens": 104618951.0,
      "step": 56110
    },
    {
      "entropy": 6.037769937515259,
      "epoch": 4.955404450724126,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002813037745594159,
      "loss": 5.1865,
      "mean_token_accuracy": 0.23022400140762328,
      "num_tokens": 104626765.0,
      "step": 56115
    },
    {
      "entropy": 6.118412733078003,
      "epoch": 4.955845990815966,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002812722953535178,
      "loss": 5.2858,
      "mean_token_accuracy": 0.21751436293125154,
      "num_tokens": 104636453.0,
      "step": 56120
    },
    {
      "entropy": 6.104005670547485,
      "epoch": 4.956287530907806,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028124081602474895,
      "loss": 5.2662,
      "mean_token_accuracy": 0.21486400216817855,
      "num_tokens": 104645688.0,
      "step": 56125
    },
    {
      "entropy": 6.102693748474121,
      "epoch": 4.956729070999646,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002812093365737261,
      "loss": 5.2522,
      "mean_token_accuracy": 0.21631062030792236,
      "num_tokens": 104654758.0,
      "step": 56130
    },
    {
      "entropy": 6.07731351852417,
      "epoch": 4.957170611091487,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028117785700106584,
      "loss": 5.3205,
      "mean_token_accuracy": 0.21593572497367858,
      "num_tokens": 104664596.0,
      "step": 56135
    },
    {
      "entropy": 6.1210485935211185,
      "epoch": 4.957612151183327,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028114637730738496,
      "loss": 5.2215,
      "mean_token_accuracy": 0.21804869621992112,
      "num_tokens": 104674206.0,
      "step": 56140
    },
    {
      "entropy": 6.037605476379395,
      "epoch": 4.958053691275168,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002811148974933,
      "loss": 5.1512,
      "mean_token_accuracy": 0.23269506245851518,
      "num_tokens": 104683215.0,
      "step": 56145
    },
    {
      "entropy": 6.110340261459351,
      "epoch": 4.958495231367008,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002810834175594276,
      "loss": 5.2752,
      "mean_token_accuracy": 0.22269977033138275,
      "num_tokens": 104692831.0,
      "step": 56150
    },
    {
      "entropy": 6.0055732250213625,
      "epoch": 4.958936771458848,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002810519375063846,
      "loss": 5.2021,
      "mean_token_accuracy": 0.2330089122056961,
      "num_tokens": 104702311.0,
      "step": 56155
    },
    {
      "entropy": 6.068007469177246,
      "epoch": 4.959378311550688,
      "grad_norm": 1.25,
      "learning_rate": 0.00028102045733478754,
      "loss": 5.3115,
      "mean_token_accuracy": 0.21754874438047409,
      "num_tokens": 104711501.0,
      "step": 56160
    },
    {
      "entropy": 6.139487266540527,
      "epoch": 4.959819851642529,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002809889770452531,
      "loss": 5.3628,
      "mean_token_accuracy": 0.21397622376680375,
      "num_tokens": 104720640.0,
      "step": 56165
    },
    {
      "entropy": 6.061494207382202,
      "epoch": 4.960261391734369,
      "grad_norm": 1.40625,
      "learning_rate": 0.00028095749663839813,
      "loss": 5.2208,
      "mean_token_accuracy": 0.22262214869260788,
      "num_tokens": 104730400.0,
      "step": 56170
    },
    {
      "entropy": 5.97841100692749,
      "epoch": 4.9607029318262095,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002809260161148391,
      "loss": 5.1784,
      "mean_token_accuracy": 0.23912876695394517,
      "num_tokens": 104739450.0,
      "step": 56175
    },
    {
      "entropy": 6.085021209716797,
      "epoch": 4.96114447191805,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028089453547519284,
      "loss": 5.3482,
      "mean_token_accuracy": 0.2167089655995369,
      "num_tokens": 104749352.0,
      "step": 56180
    },
    {
      "entropy": 6.106247854232788,
      "epoch": 4.9615860120098905,
      "grad_norm": 1.328125,
      "learning_rate": 0.000280863054720076,
      "loss": 5.2932,
      "mean_token_accuracy": 0.21719090342521669,
      "num_tokens": 104758610.0,
      "step": 56185
    },
    {
      "entropy": 6.028858470916748,
      "epoch": 4.9620275521017305,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002808315738501052,
      "loss": 5.2035,
      "mean_token_accuracy": 0.22793878763914108,
      "num_tokens": 104768162.0,
      "step": 56190
    },
    {
      "entropy": 6.09078950881958,
      "epoch": 4.9624690921935715,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00028080009286589715,
      "loss": 5.3098,
      "mean_token_accuracy": 0.20824778228998184,
      "num_tokens": 104776805.0,
      "step": 56195
    },
    {
      "entropy": 6.029422616958618,
      "epoch": 4.9629106322854115,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00028076861176806864,
      "loss": 5.1996,
      "mean_token_accuracy": 0.22744227349758148,
      "num_tokens": 104785216.0,
      "step": 56200
    },
    {
      "entropy": 6.131026792526245,
      "epoch": 4.963352172377252,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028073713055723633,
      "loss": 5.2819,
      "mean_token_accuracy": 0.22111172527074813,
      "num_tokens": 104794176.0,
      "step": 56205
    },
    {
      "entropy": 6.0923521518707275,
      "epoch": 4.9637937124690925,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002807056492340168,
      "loss": 5.2772,
      "mean_token_accuracy": 0.2177445501089096,
      "num_tokens": 104803487.0,
      "step": 56210
    },
    {
      "entropy": 6.067073917388916,
      "epoch": 4.964235252560933,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00028067416779902686,
      "loss": 5.3253,
      "mean_token_accuracy": 0.21245105862617492,
      "num_tokens": 104813439.0,
      "step": 56215
    },
    {
      "entropy": 6.105119466781616,
      "epoch": 4.964676792652773,
      "grad_norm": 1.328125,
      "learning_rate": 0.00028064268625288324,
      "loss": 5.3326,
      "mean_token_accuracy": 0.21307367086410522,
      "num_tokens": 104823112.0,
      "step": 56220
    },
    {
      "entropy": 6.114043855667115,
      "epoch": 4.965118332744614,
      "grad_norm": 1.390625,
      "learning_rate": 0.00028061120459620266,
      "loss": 5.2095,
      "mean_token_accuracy": 0.22424016892910004,
      "num_tokens": 104832820.0,
      "step": 56225
    },
    {
      "entropy": 6.129350280761718,
      "epoch": 4.965559872836454,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00028057972282960166,
      "loss": 5.2988,
      "mean_token_accuracy": 0.22021778374910356,
      "num_tokens": 104841832.0,
      "step": 56230
    },
    {
      "entropy": 6.009437513351441,
      "epoch": 4.966001412928294,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002805482409536972,
      "loss": 5.1461,
      "mean_token_accuracy": 0.23315272033214568,
      "num_tokens": 104850917.0,
      "step": 56235
    },
    {
      "entropy": 5.954403018951416,
      "epoch": 4.966442953020135,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00028051675896910574,
      "loss": 5.1505,
      "mean_token_accuracy": 0.240226586163044,
      "num_tokens": 104860323.0,
      "step": 56240
    },
    {
      "entropy": 5.989669704437256,
      "epoch": 4.966884493111975,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00028048527687644416,
      "loss": 5.1977,
      "mean_token_accuracy": 0.22602197229862214,
      "num_tokens": 104869530.0,
      "step": 56245
    },
    {
      "entropy": 6.073773908615112,
      "epoch": 4.967326033203815,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002804537946763291,
      "loss": 5.3722,
      "mean_token_accuracy": 0.21550537198781966,
      "num_tokens": 104879582.0,
      "step": 56250
    },
    {
      "entropy": 6.058144998550415,
      "epoch": 4.967767573295655,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028042231236937733,
      "loss": 5.1967,
      "mean_token_accuracy": 0.2221479296684265,
      "num_tokens": 104888894.0,
      "step": 56255
    },
    {
      "entropy": 6.138290357589722,
      "epoch": 4.968209113387496,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002803908299562056,
      "loss": 5.313,
      "mean_token_accuracy": 0.21150096505880356,
      "num_tokens": 104899110.0,
      "step": 56260
    },
    {
      "entropy": 6.012657022476196,
      "epoch": 4.968650653479336,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00028035934743743043,
      "loss": 5.1485,
      "mean_token_accuracy": 0.23298273086547852,
      "num_tokens": 104908468.0,
      "step": 56265
    },
    {
      "entropy": 6.05804762840271,
      "epoch": 4.969092193571177,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002803278648136688,
      "loss": 5.2985,
      "mean_token_accuracy": 0.21979760229587555,
      "num_tokens": 104917579.0,
      "step": 56270
    },
    {
      "entropy": 6.032235860824585,
      "epoch": 4.969533733663017,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002802963820855373,
      "loss": 5.2809,
      "mean_token_accuracy": 0.22460612803697586,
      "num_tokens": 104927621.0,
      "step": 56275
    },
    {
      "entropy": 5.963438081741333,
      "epoch": 4.969975273754857,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002802648992536526,
      "loss": 5.1352,
      "mean_token_accuracy": 0.22513508200645446,
      "num_tokens": 104936062.0,
      "step": 56280
    },
    {
      "entropy": 5.976682519912719,
      "epoch": 4.970416813846697,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002802334163186316,
      "loss": 5.0984,
      "mean_token_accuracy": 0.23194879293441772,
      "num_tokens": 104944210.0,
      "step": 56285
    },
    {
      "entropy": 6.0351118564605715,
      "epoch": 4.970858353938538,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028020193328109097,
      "loss": 5.1044,
      "mean_token_accuracy": 0.23647112995386124,
      "num_tokens": 104953483.0,
      "step": 56290
    },
    {
      "entropy": 6.02777624130249,
      "epoch": 4.971299894030378,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002801704501416474,
      "loss": 5.2022,
      "mean_token_accuracy": 0.2262083575129509,
      "num_tokens": 104962861.0,
      "step": 56295
    },
    {
      "entropy": 6.056607675552368,
      "epoch": 4.971741434122218,
      "grad_norm": 1.34375,
      "learning_rate": 0.00028013896690091766,
      "loss": 5.2989,
      "mean_token_accuracy": 0.21264564245939255,
      "num_tokens": 104972125.0,
      "step": 56300
    },
    {
      "entropy": 6.095975399017334,
      "epoch": 4.972182974214059,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002801074835595185,
      "loss": 5.3337,
      "mean_token_accuracy": 0.2187047302722931,
      "num_tokens": 104981921.0,
      "step": 56305
    },
    {
      "entropy": 6.093904304504394,
      "epoch": 4.972624514305899,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028007600011806667,
      "loss": 5.2829,
      "mean_token_accuracy": 0.21657107323408126,
      "num_tokens": 104991807.0,
      "step": 56310
    },
    {
      "entropy": 6.090725898742676,
      "epoch": 4.973066054397739,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028004451657717884,
      "loss": 5.2561,
      "mean_token_accuracy": 0.2244846910238266,
      "num_tokens": 105000867.0,
      "step": 56315
    },
    {
      "entropy": 6.072526168823242,
      "epoch": 4.97350759448958,
      "grad_norm": 1.375,
      "learning_rate": 0.0002800130329374718,
      "loss": 5.2266,
      "mean_token_accuracy": 0.23091931343078614,
      "num_tokens": 105009877.0,
      "step": 56320
    },
    {
      "entropy": 6.031802558898926,
      "epoch": 4.97394913458142,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002799815491995623,
      "loss": 5.2091,
      "mean_token_accuracy": 0.23051048666238785,
      "num_tokens": 105018817.0,
      "step": 56325
    },
    {
      "entropy": 6.084529876708984,
      "epoch": 4.97439067467326,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000279950065364067,
      "loss": 5.2312,
      "mean_token_accuracy": 0.22807277888059616,
      "num_tokens": 105027808.0,
      "step": 56330
    },
    {
      "entropy": 6.045224332809449,
      "epoch": 4.974832214765101,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00027991858143160283,
      "loss": 5.2607,
      "mean_token_accuracy": 0.2186979040503502,
      "num_tokens": 105037327.0,
      "step": 56335
    },
    {
      "entropy": 6.147641277313232,
      "epoch": 4.975273754856941,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002798870974027865,
      "loss": 5.2568,
      "mean_token_accuracy": 0.22373672127723693,
      "num_tokens": 105045897.0,
      "step": 56340
    },
    {
      "entropy": 6.127197504043579,
      "epoch": 4.975715294948781,
      "grad_norm": 1.375,
      "learning_rate": 0.0002798556132782346,
      "loss": 5.2972,
      "mean_token_accuracy": 0.21131621152162552,
      "num_tokens": 105054471.0,
      "step": 56345
    },
    {
      "entropy": 6.0723714351654055,
      "epoch": 4.976156835040622,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000279824129058564,
      "loss": 5.2258,
      "mean_token_accuracy": 0.22368698120117186,
      "num_tokens": 105064545.0,
      "step": 56350
    },
    {
      "entropy": 5.998329782485962,
      "epoch": 4.976598375132462,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002797926447443915,
      "loss": 5.1599,
      "mean_token_accuracy": 0.2290246695280075,
      "num_tokens": 105073688.0,
      "step": 56355
    },
    {
      "entropy": 6.021358871459961,
      "epoch": 4.977039915224302,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00027976116033633384,
      "loss": 5.2043,
      "mean_token_accuracy": 0.23397961258888245,
      "num_tokens": 105082122.0,
      "step": 56360
    },
    {
      "entropy": 6.069484233856201,
      "epoch": 4.977481455316143,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00027972967583500774,
      "loss": 5.2517,
      "mean_token_accuracy": 0.22188378274440765,
      "num_tokens": 105090650.0,
      "step": 56365
    },
    {
      "entropy": 6.04574933052063,
      "epoch": 4.977922995407983,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027969819124102995,
      "loss": 5.2201,
      "mean_token_accuracy": 0.23039441257715226,
      "num_tokens": 105099622.0,
      "step": 56370
    },
    {
      "entropy": 6.114533233642578,
      "epoch": 4.978364535499823,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002796667065550173,
      "loss": 5.3062,
      "mean_token_accuracy": 0.21318623572587966,
      "num_tokens": 105109409.0,
      "step": 56375
    },
    {
      "entropy": 6.032495927810669,
      "epoch": 4.978806075591664,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002796352217775865,
      "loss": 5.1453,
      "mean_token_accuracy": 0.22875619381666185,
      "num_tokens": 105118915.0,
      "step": 56380
    },
    {
      "entropy": 6.048248434066773,
      "epoch": 4.979247615683504,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002796037369093543,
      "loss": 5.2221,
      "mean_token_accuracy": 0.22261734306812286,
      "num_tokens": 105127344.0,
      "step": 56385
    },
    {
      "entropy": 5.998430061340332,
      "epoch": 4.979689155775344,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002795722519509376,
      "loss": 5.1615,
      "mean_token_accuracy": 0.2333538368344307,
      "num_tokens": 105135672.0,
      "step": 56390
    },
    {
      "entropy": 6.15934386253357,
      "epoch": 4.980130695867185,
      "grad_norm": 1.25,
      "learning_rate": 0.00027954076690295303,
      "loss": 5.3868,
      "mean_token_accuracy": 0.20873933136463166,
      "num_tokens": 105146606.0,
      "step": 56395
    },
    {
      "entropy": 5.966761493682862,
      "epoch": 4.980572235959025,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002795092817660174,
      "loss": 5.1398,
      "mean_token_accuracy": 0.2323299303650856,
      "num_tokens": 105156674.0,
      "step": 56400
    },
    {
      "entropy": 6.1216113567352295,
      "epoch": 4.981013776050865,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002794777965407475,
      "loss": 5.3917,
      "mean_token_accuracy": 0.20828844755887985,
      "num_tokens": 105166583.0,
      "step": 56405
    },
    {
      "entropy": 6.157845497131348,
      "epoch": 4.981455316142705,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00027944631122776023,
      "loss": 5.2269,
      "mean_token_accuracy": 0.22694317251443863,
      "num_tokens": 105175456.0,
      "step": 56410
    },
    {
      "entropy": 6.1108723163604735,
      "epoch": 4.981896856234546,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00027941482582767216,
      "loss": 5.3039,
      "mean_token_accuracy": 0.2107782930135727,
      "num_tokens": 105184704.0,
      "step": 56415
    },
    {
      "entropy": 6.0376951694488525,
      "epoch": 4.982338396326386,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00027938334034110016,
      "loss": 5.1339,
      "mean_token_accuracy": 0.22927064299583436,
      "num_tokens": 105193899.0,
      "step": 56420
    },
    {
      "entropy": 6.028478193283081,
      "epoch": 4.982779936418227,
      "grad_norm": 1.359375,
      "learning_rate": 0.000279351854768661,
      "loss": 5.2513,
      "mean_token_accuracy": 0.21589391976594924,
      "num_tokens": 105202947.0,
      "step": 56425
    },
    {
      "entropy": 5.98358850479126,
      "epoch": 4.983221476510067,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002793203691109716,
      "loss": 5.196,
      "mean_token_accuracy": 0.22307593673467635,
      "num_tokens": 105211735.0,
      "step": 56430
    },
    {
      "entropy": 6.088983154296875,
      "epoch": 4.983663016601907,
      "grad_norm": 1.375,
      "learning_rate": 0.0002792888833686485,
      "loss": 5.3116,
      "mean_token_accuracy": 0.2201740860939026,
      "num_tokens": 105221003.0,
      "step": 56435
    },
    {
      "entropy": 6.0104756355285645,
      "epoch": 4.984104556693747,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002792573975423087,
      "loss": 5.1653,
      "mean_token_accuracy": 0.2317716062068939,
      "num_tokens": 105230576.0,
      "step": 56440
    },
    {
      "entropy": 6.13704023361206,
      "epoch": 4.984546096785588,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002792259116325688,
      "loss": 5.2802,
      "mean_token_accuracy": 0.21709729731082916,
      "num_tokens": 105239435.0,
      "step": 56445
    },
    {
      "entropy": 6.096299839019776,
      "epoch": 4.984987636877428,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002791944256400457,
      "loss": 5.2843,
      "mean_token_accuracy": 0.21794862002134324,
      "num_tokens": 105248672.0,
      "step": 56450
    },
    {
      "entropy": 6.05079402923584,
      "epoch": 4.9854291769692685,
      "grad_norm": 1.375,
      "learning_rate": 0.0002791629395653563,
      "loss": 5.2141,
      "mean_token_accuracy": 0.21855945736169816,
      "num_tokens": 105257073.0,
      "step": 56455
    },
    {
      "entropy": 6.052400064468384,
      "epoch": 4.985870717061109,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002791314534091172,
      "loss": 5.2692,
      "mean_token_accuracy": 0.21863967329263687,
      "num_tokens": 105266936.0,
      "step": 56460
    },
    {
      "entropy": 5.979319381713867,
      "epoch": 4.9863122571529495,
      "grad_norm": 1.375,
      "learning_rate": 0.0002790999671719452,
      "loss": 5.1463,
      "mean_token_accuracy": 0.22729051113128662,
      "num_tokens": 105275798.0,
      "step": 56465
    },
    {
      "entropy": 6.071333503723144,
      "epoch": 4.9867537972447895,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027906848085445737,
      "loss": 5.2919,
      "mean_token_accuracy": 0.22102197855710984,
      "num_tokens": 105285704.0,
      "step": 56470
    },
    {
      "entropy": 6.0678316116333,
      "epoch": 4.9871953373366305,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002790369944572702,
      "loss": 5.234,
      "mean_token_accuracy": 0.22829093486070634,
      "num_tokens": 105293870.0,
      "step": 56475
    },
    {
      "entropy": 6.089972925186157,
      "epoch": 4.9876368774284705,
      "grad_norm": 1.375,
      "learning_rate": 0.00027900550798100066,
      "loss": 5.2765,
      "mean_token_accuracy": 0.21298554837703704,
      "num_tokens": 105303339.0,
      "step": 56480
    },
    {
      "entropy": 6.084346771240234,
      "epoch": 4.988078417520311,
      "grad_norm": 1.328125,
      "learning_rate": 0.00027897402142626536,
      "loss": 5.2403,
      "mean_token_accuracy": 0.21638082414865495,
      "num_tokens": 105312232.0,
      "step": 56485
    },
    {
      "entropy": 5.967523241043091,
      "epoch": 4.9885199576121515,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002789425347936814,
      "loss": 5.1559,
      "mean_token_accuracy": 0.2337415710091591,
      "num_tokens": 105320379.0,
      "step": 56490
    },
    {
      "entropy": 6.1119099140167235,
      "epoch": 4.9889614977039916,
      "grad_norm": 1.265625,
      "learning_rate": 0.00027891104808386535,
      "loss": 5.3234,
      "mean_token_accuracy": 0.221696175634861,
      "num_tokens": 105330289.0,
      "step": 56495
    },
    {
      "entropy": 6.055273771286011,
      "epoch": 4.989403037795832,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00027887956129743415,
      "loss": 5.2453,
      "mean_token_accuracy": 0.21828257143497468,
      "num_tokens": 105338387.0,
      "step": 56500
    },
    {
      "entropy": 5.993036603927612,
      "epoch": 4.9898445778876725,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027884807443500444,
      "loss": 5.1541,
      "mean_token_accuracy": 0.23299866020679474,
      "num_tokens": 105347895.0,
      "step": 56505
    },
    {
      "entropy": 5.993123769760132,
      "epoch": 4.990286117979513,
      "grad_norm": 1.359375,
      "learning_rate": 0.00027881658749719324,
      "loss": 5.2337,
      "mean_token_accuracy": 0.21832676380872726,
      "num_tokens": 105357125.0,
      "step": 56510
    },
    {
      "entropy": 6.091261148452759,
      "epoch": 4.990727658071353,
      "grad_norm": 1.375,
      "learning_rate": 0.0002787851004846172,
      "loss": 5.2226,
      "mean_token_accuracy": 0.2240503802895546,
      "num_tokens": 105366278.0,
      "step": 56515
    },
    {
      "entropy": 6.112915945053101,
      "epoch": 4.991169198163194,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00027875361339789326,
      "loss": 5.3543,
      "mean_token_accuracy": 0.20249205976724624,
      "num_tokens": 105375970.0,
      "step": 56520
    },
    {
      "entropy": 6.0658997058868405,
      "epoch": 4.991610738255034,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002787221262376381,
      "loss": 5.2408,
      "mean_token_accuracy": 0.22836508452892304,
      "num_tokens": 105384860.0,
      "step": 56525
    },
    {
      "entropy": 6.123187160491943,
      "epoch": 4.992052278346874,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002786906390044686,
      "loss": 5.2566,
      "mean_token_accuracy": 0.22249091416597366,
      "num_tokens": 105394517.0,
      "step": 56530
    },
    {
      "entropy": 6.108393859863281,
      "epoch": 4.992493818438714,
      "grad_norm": 1.265625,
      "learning_rate": 0.00027865915169900156,
      "loss": 5.3245,
      "mean_token_accuracy": 0.21142150461673737,
      "num_tokens": 105404195.0,
      "step": 56535
    },
    {
      "entropy": 6.073452281951904,
      "epoch": 4.992935358530555,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002786276643218539,
      "loss": 5.2636,
      "mean_token_accuracy": 0.22345003485679626,
      "num_tokens": 105412550.0,
      "step": 56540
    },
    {
      "entropy": 6.053104305267334,
      "epoch": 4.993376898622395,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00027859617687364235,
      "loss": 5.2323,
      "mean_token_accuracy": 0.21846783310174941,
      "num_tokens": 105422624.0,
      "step": 56545
    },
    {
      "entropy": 6.0005372047424315,
      "epoch": 4.993818438714236,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027856468935498377,
      "loss": 5.1704,
      "mean_token_accuracy": 0.23008139580488204,
      "num_tokens": 105433234.0,
      "step": 56550
    },
    {
      "entropy": 6.028457260131836,
      "epoch": 4.994259978806076,
      "grad_norm": 1.5625,
      "learning_rate": 0.00027853320176649487,
      "loss": 5.212,
      "mean_token_accuracy": 0.22265004217624665,
      "num_tokens": 105442933.0,
      "step": 56555
    },
    {
      "entropy": 5.9414714813232425,
      "epoch": 4.994701518897916,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00027850171410879256,
      "loss": 5.1273,
      "mean_token_accuracy": 0.2230255901813507,
      "num_tokens": 105451976.0,
      "step": 56560
    },
    {
      "entropy": 6.051386070251465,
      "epoch": 4.995143058989756,
      "grad_norm": 1.359375,
      "learning_rate": 0.00027847022638249376,
      "loss": 5.2973,
      "mean_token_accuracy": 0.21482338458299638,
      "num_tokens": 105460904.0,
      "step": 56565
    },
    {
      "entropy": 6.074949836730957,
      "epoch": 4.995584599081597,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002784387385882151,
      "loss": 5.2458,
      "mean_token_accuracy": 0.2263592839241028,
      "num_tokens": 105470231.0,
      "step": 56570
    },
    {
      "entropy": 6.086312103271484,
      "epoch": 4.996026139173437,
      "grad_norm": 1.375,
      "learning_rate": 0.00027840725072657356,
      "loss": 5.2135,
      "mean_token_accuracy": 0.2290822148323059,
      "num_tokens": 105479287.0,
      "step": 56575
    },
    {
      "entropy": 6.057658958435058,
      "epoch": 4.996467679265277,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027837576279818587,
      "loss": 5.2255,
      "mean_token_accuracy": 0.2219619259238243,
      "num_tokens": 105488729.0,
      "step": 56580
    },
    {
      "entropy": 5.9667524814605715,
      "epoch": 4.996909219357118,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00027834427480366895,
      "loss": 5.1857,
      "mean_token_accuracy": 0.23266538679599763,
      "num_tokens": 105498584.0,
      "step": 56585
    },
    {
      "entropy": 6.110115909576416,
      "epoch": 4.997350759448958,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002783127867436396,
      "loss": 5.3422,
      "mean_token_accuracy": 0.21664683967828752,
      "num_tokens": 105507780.0,
      "step": 56590
    },
    {
      "entropy": 6.061344575881958,
      "epoch": 4.997792299540798,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002782812986187145,
      "loss": 5.2449,
      "mean_token_accuracy": 0.22851113528013228,
      "num_tokens": 105518275.0,
      "step": 56595
    },
    {
      "entropy": 6.006628084182739,
      "epoch": 4.998233839632639,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002782498104295108,
      "loss": 5.209,
      "mean_token_accuracy": 0.22344865649938583,
      "num_tokens": 105527548.0,
      "step": 56600
    },
    {
      "entropy": 6.00175290107727,
      "epoch": 4.998675379724479,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027821832217664506,
      "loss": 5.2509,
      "mean_token_accuracy": 0.22430136054754257,
      "num_tokens": 105538066.0,
      "step": 56605
    },
    {
      "entropy": 6.056468152999878,
      "epoch": 4.999116919816319,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002781868338607343,
      "loss": 5.2446,
      "mean_token_accuracy": 0.22693258225917817,
      "num_tokens": 105546956.0,
      "step": 56610
    },
    {
      "entropy": 6.102212905883789,
      "epoch": 4.99955845990816,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002781553454823952,
      "loss": 5.3481,
      "mean_token_accuracy": 0.21238112896680833,
      "num_tokens": 105557393.0,
      "step": 56615
    },
    {
      "entropy": 6.155296277999878,
      "epoch": 5.0,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00027812385704224476,
      "loss": 5.2889,
      "mean_token_accuracy": 0.22004973590373994,
      "num_tokens": 105565420.0,
      "step": 56620
    },
    {
      "entropy": 6.105801105499268,
      "epoch": 5.00044154009184,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002780923685408997,
      "loss": 5.2615,
      "mean_token_accuracy": 0.2195773959159851,
      "num_tokens": 105574268.0,
      "step": 56625
    },
    {
      "entropy": 6.047082996368408,
      "epoch": 5.000883080183681,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002780608799789769,
      "loss": 5.1261,
      "mean_token_accuracy": 0.23580542653799058,
      "num_tokens": 105583116.0,
      "step": 56630
    },
    {
      "entropy": 6.068490839004516,
      "epoch": 5.001324620275521,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027802939135709317,
      "loss": 5.1626,
      "mean_token_accuracy": 0.23084213584661484,
      "num_tokens": 105591763.0,
      "step": 56635
    },
    {
      "entropy": 5.992731237411499,
      "epoch": 5.001766160367361,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00027799790267586547,
      "loss": 5.2685,
      "mean_token_accuracy": 0.22272294461727143,
      "num_tokens": 105601794.0,
      "step": 56640
    },
    {
      "entropy": 6.0723566055297855,
      "epoch": 5.002207700459202,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00027796641393591044,
      "loss": 5.2264,
      "mean_token_accuracy": 0.22625560015439988,
      "num_tokens": 105611222.0,
      "step": 56645
    },
    {
      "entropy": 6.037861490249634,
      "epoch": 5.002649240551042,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00027793492513784516,
      "loss": 5.1851,
      "mean_token_accuracy": 0.22669695764780046,
      "num_tokens": 105620051.0,
      "step": 56650
    },
    {
      "entropy": 5.902747631072998,
      "epoch": 5.003090780642882,
      "grad_norm": 1.390625,
      "learning_rate": 0.00027790343628228626,
      "loss": 5.0691,
      "mean_token_accuracy": 0.24622277170419693,
      "num_tokens": 105629116.0,
      "step": 56655
    },
    {
      "entropy": 5.986675214767456,
      "epoch": 5.003532320734723,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002778719473698507,
      "loss": 5.2124,
      "mean_token_accuracy": 0.2366855874657631,
      "num_tokens": 105639210.0,
      "step": 56660
    },
    {
      "entropy": 6.118139266967773,
      "epoch": 5.003973860826563,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002778404584011554,
      "loss": 5.2384,
      "mean_token_accuracy": 0.22762082815170287,
      "num_tokens": 105647703.0,
      "step": 56665
    },
    {
      "entropy": 6.086523628234863,
      "epoch": 5.004415400918403,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002778089693768171,
      "loss": 5.2302,
      "mean_token_accuracy": 0.22769394218921662,
      "num_tokens": 105657342.0,
      "step": 56670
    },
    {
      "entropy": 6.012564373016358,
      "epoch": 5.004856941010244,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027777748029745274,
      "loss": 5.1529,
      "mean_token_accuracy": 0.2291683778166771,
      "num_tokens": 105666600.0,
      "step": 56675
    },
    {
      "entropy": 6.126947355270386,
      "epoch": 5.005298481102084,
      "grad_norm": 1.375,
      "learning_rate": 0.00027774599116367905,
      "loss": 5.3486,
      "mean_token_accuracy": 0.2154340535402298,
      "num_tokens": 105675318.0,
      "step": 56680
    },
    {
      "entropy": 6.069346618652344,
      "epoch": 5.005740021193924,
      "grad_norm": 1.296875,
      "learning_rate": 0.000277714501976113,
      "loss": 5.2482,
      "mean_token_accuracy": 0.22117656469345093,
      "num_tokens": 105684934.0,
      "step": 56685
    },
    {
      "entropy": 6.106347465515137,
      "epoch": 5.006181561285764,
      "grad_norm": 1.40625,
      "learning_rate": 0.00027768301273537133,
      "loss": 5.3369,
      "mean_token_accuracy": 0.22498101145029067,
      "num_tokens": 105694368.0,
      "step": 56690
    },
    {
      "entropy": 6.123595762252807,
      "epoch": 5.006623101377605,
      "grad_norm": 1.375,
      "learning_rate": 0.000277651523442071,
      "loss": 5.2702,
      "mean_token_accuracy": 0.21904169917106628,
      "num_tokens": 105704209.0,
      "step": 56695
    },
    {
      "entropy": 6.042940664291382,
      "epoch": 5.007064641469445,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027762003409682884,
      "loss": 5.2057,
      "mean_token_accuracy": 0.22184639126062394,
      "num_tokens": 105713073.0,
      "step": 56700
    },
    {
      "entropy": 6.116514253616333,
      "epoch": 5.007506181561285,
      "grad_norm": 1.25,
      "learning_rate": 0.0002775885447002618,
      "loss": 5.245,
      "mean_token_accuracy": 0.21958698481321334,
      "num_tokens": 105723201.0,
      "step": 56705
    },
    {
      "entropy": 6.056535291671753,
      "epoch": 5.007947721653126,
      "grad_norm": 1.359375,
      "learning_rate": 0.00027755705525298654,
      "loss": 5.2092,
      "mean_token_accuracy": 0.22744669169187545,
      "num_tokens": 105732695.0,
      "step": 56710
    },
    {
      "entropy": 6.003261613845825,
      "epoch": 5.008389261744966,
      "grad_norm": 1.484375,
      "learning_rate": 0.00027752556575562007,
      "loss": 5.1687,
      "mean_token_accuracy": 0.22885224521160125,
      "num_tokens": 105742834.0,
      "step": 56715
    },
    {
      "entropy": 6.039111852645874,
      "epoch": 5.008830801836806,
      "grad_norm": 1.328125,
      "learning_rate": 0.00027749407620877907,
      "loss": 5.2056,
      "mean_token_accuracy": 0.23333547711372377,
      "num_tokens": 105752340.0,
      "step": 56720
    },
    {
      "entropy": 6.094108295440674,
      "epoch": 5.009272341928647,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002774625866130806,
      "loss": 5.3391,
      "mean_token_accuracy": 0.21607760787010194,
      "num_tokens": 105762675.0,
      "step": 56725
    },
    {
      "entropy": 6.085579872131348,
      "epoch": 5.009713882020487,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002774310969691416,
      "loss": 5.1334,
      "mean_token_accuracy": 0.22912359088659287,
      "num_tokens": 105770611.0,
      "step": 56730
    },
    {
      "entropy": 6.034729385375977,
      "epoch": 5.0101554221123275,
      "grad_norm": 1.453125,
      "learning_rate": 0.00027739960727757865,
      "loss": 5.1411,
      "mean_token_accuracy": 0.23034260272979737,
      "num_tokens": 105780311.0,
      "step": 56735
    },
    {
      "entropy": 6.047105026245117,
      "epoch": 5.010596962204168,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002773681175390088,
      "loss": 5.2464,
      "mean_token_accuracy": 0.22533925473690034,
      "num_tokens": 105788946.0,
      "step": 56740
    },
    {
      "entropy": 6.017183876037597,
      "epoch": 5.0110385022960084,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002773366277540488,
      "loss": 5.2004,
      "mean_token_accuracy": 0.22854494154453278,
      "num_tokens": 105797594.0,
      "step": 56745
    },
    {
      "entropy": 5.923974084854126,
      "epoch": 5.0114800423878485,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027730513792331577,
      "loss": 5.104,
      "mean_token_accuracy": 0.24519113451242447,
      "num_tokens": 105806795.0,
      "step": 56750
    },
    {
      "entropy": 6.049741172790528,
      "epoch": 5.011921582479689,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00027727364804742624,
      "loss": 5.2846,
      "mean_token_accuracy": 0.21555302143096924,
      "num_tokens": 105816466.0,
      "step": 56755
    },
    {
      "entropy": 6.031032562255859,
      "epoch": 5.0123631225715295,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002772421581269973,
      "loss": 5.2541,
      "mean_token_accuracy": 0.21877897083759307,
      "num_tokens": 105825930.0,
      "step": 56760
    },
    {
      "entropy": 6.073864936828613,
      "epoch": 5.0128046626633695,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002772106681626457,
      "loss": 5.1674,
      "mean_token_accuracy": 0.23732299655675887,
      "num_tokens": 105835728.0,
      "step": 56765
    },
    {
      "entropy": 6.179848957061767,
      "epoch": 5.0132462027552105,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027717917815498844,
      "loss": 5.326,
      "mean_token_accuracy": 0.21435635089874266,
      "num_tokens": 105845607.0,
      "step": 56770
    },
    {
      "entropy": 6.041168594360352,
      "epoch": 5.0136877428470505,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027714768810464227,
      "loss": 5.2181,
      "mean_token_accuracy": 0.2196303740143776,
      "num_tokens": 105855351.0,
      "step": 56775
    },
    {
      "entropy": 5.957952833175659,
      "epoch": 5.014129282938891,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00027711619801222407,
      "loss": 5.0797,
      "mean_token_accuracy": 0.23877380788326263,
      "num_tokens": 105864889.0,
      "step": 56780
    },
    {
      "entropy": 6.064170026779175,
      "epoch": 5.0145708230307315,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002770847078783509,
      "loss": 5.2784,
      "mean_token_accuracy": 0.21880429685115815,
      "num_tokens": 105874081.0,
      "step": 56785
    },
    {
      "entropy": 5.997854471206665,
      "epoch": 5.015012363122572,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002770532177036394,
      "loss": 5.2107,
      "mean_token_accuracy": 0.22241305112838744,
      "num_tokens": 105882976.0,
      "step": 56790
    },
    {
      "entropy": 6.073395919799805,
      "epoch": 5.015453903214412,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002770217274887065,
      "loss": 5.2802,
      "mean_token_accuracy": 0.20996501296758652,
      "num_tokens": 105893367.0,
      "step": 56795
    },
    {
      "entropy": 6.157403039932251,
      "epoch": 5.015895443306253,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002769902372341692,
      "loss": 5.2739,
      "mean_token_accuracy": 0.22514159828424454,
      "num_tokens": 105903182.0,
      "step": 56800
    },
    {
      "entropy": 6.039413404464722,
      "epoch": 5.016336983398093,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002769587469406442,
      "loss": 5.1677,
      "mean_token_accuracy": 0.23617342859506607,
      "num_tokens": 105911830.0,
      "step": 56805
    },
    {
      "entropy": 6.069980955123901,
      "epoch": 5.016778523489933,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002769272566087485,
      "loss": 5.2005,
      "mean_token_accuracy": 0.2220180556178093,
      "num_tokens": 105921217.0,
      "step": 56810
    },
    {
      "entropy": 5.9880446434021,
      "epoch": 5.017220063581774,
      "grad_norm": 1.46875,
      "learning_rate": 0.00027689576623909896,
      "loss": 5.1688,
      "mean_token_accuracy": 0.23125951886177062,
      "num_tokens": 105930367.0,
      "step": 56815
    },
    {
      "entropy": 5.996271133422852,
      "epoch": 5.017661603673614,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002768642758323124,
      "loss": 5.2471,
      "mean_token_accuracy": 0.22368622124195098,
      "num_tokens": 105940321.0,
      "step": 56820
    },
    {
      "entropy": 6.099417304992675,
      "epoch": 5.018103143765454,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00027683278538900575,
      "loss": 5.2999,
      "mean_token_accuracy": 0.21121152490377426,
      "num_tokens": 105950604.0,
      "step": 56825
    },
    {
      "entropy": 6.033518886566162,
      "epoch": 5.018544683857295,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002768012949097958,
      "loss": 5.1553,
      "mean_token_accuracy": 0.22977528870105743,
      "num_tokens": 105960403.0,
      "step": 56830
    },
    {
      "entropy": 6.050705242156982,
      "epoch": 5.018986223949135,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002767698043952996,
      "loss": 5.193,
      "mean_token_accuracy": 0.22830803096294403,
      "num_tokens": 105970226.0,
      "step": 56835
    },
    {
      "entropy": 6.016075134277344,
      "epoch": 5.019427764040975,
      "grad_norm": 1.34375,
      "learning_rate": 0.000276738313846134,
      "loss": 5.1015,
      "mean_token_accuracy": 0.23584164679050446,
      "num_tokens": 105978802.0,
      "step": 56840
    },
    {
      "entropy": 6.056585025787354,
      "epoch": 5.019869304132815,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002767068232629156,
      "loss": 5.2287,
      "mean_token_accuracy": 0.22140225023031235,
      "num_tokens": 105988655.0,
      "step": 56845
    },
    {
      "entropy": 6.0929559707641605,
      "epoch": 5.020310844224656,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002766753326462617,
      "loss": 5.2631,
      "mean_token_accuracy": 0.22289710938930513,
      "num_tokens": 105998102.0,
      "step": 56850
    },
    {
      "entropy": 6.002908611297608,
      "epoch": 5.020752384316496,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027664384199678894,
      "loss": 5.1545,
      "mean_token_accuracy": 0.22854001671075821,
      "num_tokens": 106006998.0,
      "step": 56855
    },
    {
      "entropy": 6.064418601989746,
      "epoch": 5.021193924408336,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027661235131511423,
      "loss": 5.1791,
      "mean_token_accuracy": 0.22554148584604264,
      "num_tokens": 106015019.0,
      "step": 56860
    },
    {
      "entropy": 6.027379322052002,
      "epoch": 5.021635464500177,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002765808606018545,
      "loss": 5.2496,
      "mean_token_accuracy": 0.21777817159891127,
      "num_tokens": 106024887.0,
      "step": 56865
    },
    {
      "entropy": 6.0622110843658445,
      "epoch": 5.022077004592017,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002765493698576266,
      "loss": 5.2642,
      "mean_token_accuracy": 0.2315915495157242,
      "num_tokens": 106034883.0,
      "step": 56870
    },
    {
      "entropy": 6.184699058532715,
      "epoch": 5.022518544683857,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027651787908304745,
      "loss": 5.3914,
      "mean_token_accuracy": 0.20723542273044587,
      "num_tokens": 106044950.0,
      "step": 56875
    },
    {
      "entropy": 6.142701911926269,
      "epoch": 5.022960084775698,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00027648638827873387,
      "loss": 5.3459,
      "mean_token_accuracy": 0.21185731142759323,
      "num_tokens": 106054170.0,
      "step": 56880
    },
    {
      "entropy": 6.111801815032959,
      "epoch": 5.023401624867538,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002764548974453028,
      "loss": 5.3282,
      "mean_token_accuracy": 0.21078111976385117,
      "num_tokens": 106063398.0,
      "step": 56885
    },
    {
      "entropy": 5.993031740188599,
      "epoch": 5.023843164959378,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002764234065833711,
      "loss": 5.0237,
      "mean_token_accuracy": 0.24001380801200867,
      "num_tokens": 106073439.0,
      "step": 56890
    },
    {
      "entropy": 5.984681034088135,
      "epoch": 5.024284705051219,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00027639191569355567,
      "loss": 5.1237,
      "mean_token_accuracy": 0.2345047578215599,
      "num_tokens": 106082983.0,
      "step": 56895
    },
    {
      "entropy": 5.95803599357605,
      "epoch": 5.024726245143059,
      "grad_norm": 1.40625,
      "learning_rate": 0.00027636042477647345,
      "loss": 5.136,
      "mean_token_accuracy": 0.23263569474220275,
      "num_tokens": 106092384.0,
      "step": 56900
    },
    {
      "entropy": 5.981142997741699,
      "epoch": 5.025167785234899,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002763289338327413,
      "loss": 5.2928,
      "mean_token_accuracy": 0.21773302406072617,
      "num_tokens": 106102141.0,
      "step": 56905
    },
    {
      "entropy": 6.053110599517822,
      "epoch": 5.02560932532674,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000276297442862976,
      "loss": 5.2429,
      "mean_token_accuracy": 0.22592779248952866,
      "num_tokens": 106111573.0,
      "step": 56910
    },
    {
      "entropy": 6.029267597198486,
      "epoch": 5.02605086541858,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00027626595186779453,
      "loss": 5.1496,
      "mean_token_accuracy": 0.2340048670768738,
      "num_tokens": 106121081.0,
      "step": 56915
    },
    {
      "entropy": 6.071394348144532,
      "epoch": 5.02649240551042,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027623446084781383,
      "loss": 5.1918,
      "mean_token_accuracy": 0.22746396213769912,
      "num_tokens": 106130006.0,
      "step": 56920
    },
    {
      "entropy": 6.029062509536743,
      "epoch": 5.026933945602261,
      "grad_norm": 1.375,
      "learning_rate": 0.00027620296980365077,
      "loss": 5.2225,
      "mean_token_accuracy": 0.22331721633672713,
      "num_tokens": 106139768.0,
      "step": 56925
    },
    {
      "entropy": 6.026640224456787,
      "epoch": 5.027375485694101,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002761714787359222,
      "loss": 5.1795,
      "mean_token_accuracy": 0.2266572043299675,
      "num_tokens": 106149095.0,
      "step": 56930
    },
    {
      "entropy": 6.043957662582398,
      "epoch": 5.027817025785941,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000276139987645245,
      "loss": 5.1664,
      "mean_token_accuracy": 0.22969331294298173,
      "num_tokens": 106158187.0,
      "step": 56935
    },
    {
      "entropy": 5.993801927566528,
      "epoch": 5.028258565877782,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002761084965322361,
      "loss": 5.1451,
      "mean_token_accuracy": 0.23477219194173812,
      "num_tokens": 106167275.0,
      "step": 56940
    },
    {
      "entropy": 6.016649055480957,
      "epoch": 5.028700105969622,
      "grad_norm": 1.453125,
      "learning_rate": 0.00027607700539751247,
      "loss": 5.2011,
      "mean_token_accuracy": 0.23227211982011794,
      "num_tokens": 106176252.0,
      "step": 56945
    },
    {
      "entropy": 6.024848604202271,
      "epoch": 5.029141646061462,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002760455142416908,
      "loss": 5.2166,
      "mean_token_accuracy": 0.2302851215004921,
      "num_tokens": 106185262.0,
      "step": 56950
    },
    {
      "entropy": 6.0300966739654545,
      "epoch": 5.029583186153303,
      "grad_norm": 1.421875,
      "learning_rate": 0.00027601402306538813,
      "loss": 5.311,
      "mean_token_accuracy": 0.21945178210735322,
      "num_tokens": 106195059.0,
      "step": 56955
    },
    {
      "entropy": 6.081428098678589,
      "epoch": 5.030024726245143,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027598253186922133,
      "loss": 5.2146,
      "mean_token_accuracy": 0.22379807978868485,
      "num_tokens": 106204645.0,
      "step": 56960
    },
    {
      "entropy": 6.081174421310425,
      "epoch": 5.030466266336983,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027595104065380733,
      "loss": 5.1927,
      "mean_token_accuracy": 0.22293703854084015,
      "num_tokens": 106214032.0,
      "step": 56965
    },
    {
      "entropy": 6.0819450378417965,
      "epoch": 5.030907806428823,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00027591954941976295,
      "loss": 5.2422,
      "mean_token_accuracy": 0.22041977047920228,
      "num_tokens": 106222854.0,
      "step": 56970
    },
    {
      "entropy": 6.09882116317749,
      "epoch": 5.031349346520664,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002758880581677051,
      "loss": 5.2691,
      "mean_token_accuracy": 0.2184905931353569,
      "num_tokens": 106231688.0,
      "step": 56975
    },
    {
      "entropy": 6.0350183010101315,
      "epoch": 5.031790886612504,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002758565668982507,
      "loss": 5.1326,
      "mean_token_accuracy": 0.23584457188844682,
      "num_tokens": 106240175.0,
      "step": 56980
    },
    {
      "entropy": 6.072015237808228,
      "epoch": 5.032232426704344,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002758250756120167,
      "loss": 5.192,
      "mean_token_accuracy": 0.22855018973350524,
      "num_tokens": 106249286.0,
      "step": 56985
    },
    {
      "entropy": 6.128938627243042,
      "epoch": 5.032673966796185,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00027579358430961993,
      "loss": 5.3349,
      "mean_token_accuracy": 0.2130355566740036,
      "num_tokens": 106258971.0,
      "step": 56990
    },
    {
      "entropy": 6.043942260742187,
      "epoch": 5.033115506888025,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027576209299167716,
      "loss": 5.1773,
      "mean_token_accuracy": 0.22636426240205765,
      "num_tokens": 106268377.0,
      "step": 56995
    },
    {
      "entropy": 6.076539134979248,
      "epoch": 5.033557046979865,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002757306016588056,
      "loss": 5.2469,
      "mean_token_accuracy": 0.2219301238656044,
      "num_tokens": 106278101.0,
      "step": 57000
    },
    {
      "epoch": 5.033557046979865,
      "eval_entropy": 5.781136490993565,
      "eval_loss": 5.3075947761535645,
      "eval_mean_token_accuracy": 0.22784001649624225,
      "eval_num_tokens": 106278101.0,
      "eval_runtime": 20.2364,
      "eval_samples_per_second": 1745.122,
      "eval_steps_per_second": 218.171,
      "step": 57000
    },
    {
      "entropy": 6.08860411643982,
      "epoch": 5.033998587071706,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027569911031162183,
      "loss": 5.2401,
      "mean_token_accuracy": 0.21429603546857834,
      "num_tokens": 106287881.0,
      "step": 57005
    },
    {
      "entropy": 6.079791402816772,
      "epoch": 5.034440127163546,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000275667618950743,
      "loss": 5.2061,
      "mean_token_accuracy": 0.22360625118017197,
      "num_tokens": 106297990.0,
      "step": 57010
    },
    {
      "entropy": 6.049481010437011,
      "epoch": 5.034881667255386,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002756361275767858,
      "loss": 5.2877,
      "mean_token_accuracy": 0.21465180963277816,
      "num_tokens": 106307159.0,
      "step": 57015
    },
    {
      "entropy": 6.029210805892944,
      "epoch": 5.035323207347227,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002756046361903673,
      "loss": 5.2354,
      "mean_token_accuracy": 0.22508368790149688,
      "num_tokens": 106316829.0,
      "step": 57020
    },
    {
      "entropy": 5.99832501411438,
      "epoch": 5.035764747439067,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027557314479210434,
      "loss": 5.0837,
      "mean_token_accuracy": 0.23813803046941756,
      "num_tokens": 106326481.0,
      "step": 57025
    },
    {
      "entropy": 6.113915157318115,
      "epoch": 5.0362062875309075,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00027554165338261366,
      "loss": 5.262,
      "mean_token_accuracy": 0.21648513823747634,
      "num_tokens": 106335823.0,
      "step": 57030
    },
    {
      "entropy": 6.041144466400146,
      "epoch": 5.036647827622748,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002755101619625124,
      "loss": 5.1989,
      "mean_token_accuracy": 0.22939340472221376,
      "num_tokens": 106346340.0,
      "step": 57035
    },
    {
      "entropy": 6.049851036071777,
      "epoch": 5.0370893677145885,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027547867053241725,
      "loss": 5.1918,
      "mean_token_accuracy": 0.22510294616222382,
      "num_tokens": 106354947.0,
      "step": 57040
    },
    {
      "entropy": 6.026105880737305,
      "epoch": 5.0375309078064285,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00027544717909294534,
      "loss": 5.2322,
      "mean_token_accuracy": 0.22688285857439042,
      "num_tokens": 106364861.0,
      "step": 57045
    },
    {
      "entropy": 6.041514778137207,
      "epoch": 5.0379724478982695,
      "grad_norm": 1.390625,
      "learning_rate": 0.00027541568764471346,
      "loss": 5.1791,
      "mean_token_accuracy": 0.22995606660842896,
      "num_tokens": 106374348.0,
      "step": 57050
    },
    {
      "entropy": 6.066496753692627,
      "epoch": 5.0384139879901095,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002753841961883384,
      "loss": 5.285,
      "mean_token_accuracy": 0.21937502473592757,
      "num_tokens": 106384375.0,
      "step": 57055
    },
    {
      "entropy": 5.98789234161377,
      "epoch": 5.03885552808195,
      "grad_norm": 1.484375,
      "learning_rate": 0.00027535270472443726,
      "loss": 5.0995,
      "mean_token_accuracy": 0.2384343311190605,
      "num_tokens": 106393041.0,
      "step": 57060
    },
    {
      "entropy": 6.021774482727051,
      "epoch": 5.0392970681737905,
      "grad_norm": 1.25,
      "learning_rate": 0.0002753212132536268,
      "loss": 5.2068,
      "mean_token_accuracy": 0.22464186251163482,
      "num_tokens": 106402676.0,
      "step": 57065
    },
    {
      "entropy": 6.003873014450074,
      "epoch": 5.039738608265631,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00027528972177652387,
      "loss": 5.1656,
      "mean_token_accuracy": 0.22517990618944167,
      "num_tokens": 106412565.0,
      "step": 57070
    },
    {
      "entropy": 6.003065299987793,
      "epoch": 5.040180148357471,
      "grad_norm": 1.484375,
      "learning_rate": 0.00027525823029374557,
      "loss": 5.2034,
      "mean_token_accuracy": 0.22860189527273178,
      "num_tokens": 106422015.0,
      "step": 57075
    },
    {
      "entropy": 5.995194721221924,
      "epoch": 5.040621688449312,
      "grad_norm": 1.421875,
      "learning_rate": 0.00027522673880590865,
      "loss": 5.1558,
      "mean_token_accuracy": 0.22871624529361725,
      "num_tokens": 106430424.0,
      "step": 57080
    },
    {
      "entropy": 6.096951484680176,
      "epoch": 5.041063228541152,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00027519524731362997,
      "loss": 5.2873,
      "mean_token_accuracy": 0.21712989658117293,
      "num_tokens": 106438718.0,
      "step": 57085
    },
    {
      "entropy": 5.977304077148437,
      "epoch": 5.041504768632992,
      "grad_norm": 1.4375,
      "learning_rate": 0.00027516375581752666,
      "loss": 5.2021,
      "mean_token_accuracy": 0.22246607691049575,
      "num_tokens": 106447455.0,
      "step": 57090
    },
    {
      "entropy": 6.09256649017334,
      "epoch": 5.041946308724833,
      "grad_norm": 1.484375,
      "learning_rate": 0.00027513226431821535,
      "loss": 5.2699,
      "mean_token_accuracy": 0.21888320595026017,
      "num_tokens": 106456000.0,
      "step": 57095
    },
    {
      "entropy": 6.090085887908936,
      "epoch": 5.042387848816673,
      "grad_norm": 1.453125,
      "learning_rate": 0.000275100772816313,
      "loss": 5.2858,
      "mean_token_accuracy": 0.2183250293135643,
      "num_tokens": 106465077.0,
      "step": 57100
    },
    {
      "entropy": 6.072772884368897,
      "epoch": 5.042829388908513,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00027506928131243663,
      "loss": 5.2592,
      "mean_token_accuracy": 0.22021684497594834,
      "num_tokens": 106475149.0,
      "step": 57105
    },
    {
      "entropy": 6.048696899414063,
      "epoch": 5.043270929000354,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002750377898072031,
      "loss": 5.1864,
      "mean_token_accuracy": 0.22892191559076308,
      "num_tokens": 106484797.0,
      "step": 57110
    },
    {
      "entropy": 6.041065311431884,
      "epoch": 5.043712469092194,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002750062983012293,
      "loss": 5.166,
      "mean_token_accuracy": 0.2332371339201927,
      "num_tokens": 106494081.0,
      "step": 57115
    },
    {
      "entropy": 6.090987777709961,
      "epoch": 5.044154009184034,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002749748067951321,
      "loss": 5.2135,
      "mean_token_accuracy": 0.22945346385240556,
      "num_tokens": 106503116.0,
      "step": 57120
    },
    {
      "entropy": 6.067131900787354,
      "epoch": 5.044595549275874,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002749433152895284,
      "loss": 5.2941,
      "mean_token_accuracy": 0.22321651875972748,
      "num_tokens": 106512717.0,
      "step": 57125
    },
    {
      "entropy": 5.981875562667847,
      "epoch": 5.045037089367715,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002749118237850352,
      "loss": 5.2222,
      "mean_token_accuracy": 0.2250175341963768,
      "num_tokens": 106521645.0,
      "step": 57130
    },
    {
      "entropy": 6.041210126876831,
      "epoch": 5.045478629459555,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002748803322822693,
      "loss": 5.2269,
      "mean_token_accuracy": 0.22400943487882613,
      "num_tokens": 106530386.0,
      "step": 57135
    },
    {
      "entropy": 5.999428129196167,
      "epoch": 5.045920169551395,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027484884078184753,
      "loss": 5.1628,
      "mean_token_accuracy": 0.22726177424192429,
      "num_tokens": 106539490.0,
      "step": 57140
    },
    {
      "entropy": 6.111322259902954,
      "epoch": 5.046361709643236,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00027481734928438696,
      "loss": 5.2771,
      "mean_token_accuracy": 0.22146506309509278,
      "num_tokens": 106548723.0,
      "step": 57145
    },
    {
      "entropy": 6.0082799911499025,
      "epoch": 5.046803249735076,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002747858577905044,
      "loss": 5.1977,
      "mean_token_accuracy": 0.22583601921796798,
      "num_tokens": 106558335.0,
      "step": 57150
    },
    {
      "entropy": 6.070804786682129,
      "epoch": 5.047244789826916,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002747543663008169,
      "loss": 5.213,
      "mean_token_accuracy": 0.22519491463899613,
      "num_tokens": 106567413.0,
      "step": 57155
    },
    {
      "entropy": 6.058045196533203,
      "epoch": 5.047686329918757,
      "grad_norm": 1.40625,
      "learning_rate": 0.000274722874815941,
      "loss": 5.1762,
      "mean_token_accuracy": 0.22516694962978362,
      "num_tokens": 106576211.0,
      "step": 57160
    },
    {
      "entropy": 6.05512638092041,
      "epoch": 5.048127870010597,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0002746913833364939,
      "loss": 5.2505,
      "mean_token_accuracy": 0.22046177685260773,
      "num_tokens": 106585824.0,
      "step": 57165
    },
    {
      "entropy": 6.045252180099487,
      "epoch": 5.048569410102437,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027465989186309245,
      "loss": 5.2711,
      "mean_token_accuracy": 0.21763572096824646,
      "num_tokens": 106595401.0,
      "step": 57170
    },
    {
      "entropy": 6.0900157451629635,
      "epoch": 5.049010950194278,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00027462840039635346,
      "loss": 5.2394,
      "mean_token_accuracy": 0.22367216497659684,
      "num_tokens": 106605140.0,
      "step": 57175
    },
    {
      "entropy": 6.059125995635986,
      "epoch": 5.049452490286118,
      "grad_norm": 1.34375,
      "learning_rate": 0.000274596908936894,
      "loss": 5.2543,
      "mean_token_accuracy": 0.21999017745256425,
      "num_tokens": 106614505.0,
      "step": 57180
    },
    {
      "entropy": 6.084415292739868,
      "epoch": 5.049894030377958,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027456541748533075,
      "loss": 5.2345,
      "mean_token_accuracy": 0.22106605023145676,
      "num_tokens": 106622865.0,
      "step": 57185
    },
    {
      "entropy": 5.9669163703918455,
      "epoch": 5.050335570469799,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027453392604228076,
      "loss": 5.1898,
      "mean_token_accuracy": 0.22470913231372833,
      "num_tokens": 106633042.0,
      "step": 57190
    },
    {
      "entropy": 5.985907030105591,
      "epoch": 5.050777110561639,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000274502434608361,
      "loss": 5.2599,
      "mean_token_accuracy": 0.21929603070020676,
      "num_tokens": 106642042.0,
      "step": 57195
    },
    {
      "entropy": 6.06607494354248,
      "epoch": 5.051218650653479,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002744709431841882,
      "loss": 5.215,
      "mean_token_accuracy": 0.229201802611351,
      "num_tokens": 106650230.0,
      "step": 57200
    },
    {
      "entropy": 6.1293682098388675,
      "epoch": 5.05166019074532,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002744394517703793,
      "loss": 5.2414,
      "mean_token_accuracy": 0.22315734177827834,
      "num_tokens": 106658717.0,
      "step": 57205
    },
    {
      "entropy": 6.068718814849854,
      "epoch": 5.05210173083716,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00027440796036755117,
      "loss": 5.2068,
      "mean_token_accuracy": 0.22284648269414903,
      "num_tokens": 106667781.0,
      "step": 57210
    },
    {
      "entropy": 5.949995613098144,
      "epoch": 5.052543270929,
      "grad_norm": 1.375,
      "learning_rate": 0.00027437646897632083,
      "loss": 5.104,
      "mean_token_accuracy": 0.2328551098704338,
      "num_tokens": 106677089.0,
      "step": 57215
    },
    {
      "entropy": 6.062815713882446,
      "epoch": 5.052984811020841,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027434497759730507,
      "loss": 5.2785,
      "mean_token_accuracy": 0.22306747883558273,
      "num_tokens": 106685985.0,
      "step": 57220
    },
    {
      "entropy": 6.05516357421875,
      "epoch": 5.053426351112681,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002743134862311209,
      "loss": 5.2852,
      "mean_token_accuracy": 0.22506278455257417,
      "num_tokens": 106694883.0,
      "step": 57225
    },
    {
      "entropy": 5.937507772445679,
      "epoch": 5.053867891204521,
      "grad_norm": 1.265625,
      "learning_rate": 0.00027428199487838505,
      "loss": 5.1007,
      "mean_token_accuracy": 0.24541416317224501,
      "num_tokens": 106704305.0,
      "step": 57230
    },
    {
      "entropy": 6.019120740890503,
      "epoch": 5.054309431296362,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00027425050353971456,
      "loss": 5.2864,
      "mean_token_accuracy": 0.22683025300502777,
      "num_tokens": 106713494.0,
      "step": 57235
    },
    {
      "entropy": 5.99790325164795,
      "epoch": 5.054750971388202,
      "grad_norm": 1.4375,
      "learning_rate": 0.00027421901221572634,
      "loss": 5.1557,
      "mean_token_accuracy": 0.23989686518907546,
      "num_tokens": 106722621.0,
      "step": 57240
    },
    {
      "entropy": 6.042066669464111,
      "epoch": 5.055192511480042,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027418752090703713,
      "loss": 5.2022,
      "mean_token_accuracy": 0.23292670398950577,
      "num_tokens": 106731560.0,
      "step": 57245
    },
    {
      "entropy": 6.120691108703613,
      "epoch": 5.055634051571882,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002741560296142639,
      "loss": 5.2553,
      "mean_token_accuracy": 0.22731561660766603,
      "num_tokens": 106740405.0,
      "step": 57250
    },
    {
      "entropy": 6.043464422225952,
      "epoch": 5.056075591663723,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002741245383380236,
      "loss": 5.2529,
      "mean_token_accuracy": 0.22231498509645461,
      "num_tokens": 106750969.0,
      "step": 57255
    },
    {
      "entropy": 6.030443525314331,
      "epoch": 5.056517131755563,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002740930470789332,
      "loss": 5.162,
      "mean_token_accuracy": 0.22673751711845397,
      "num_tokens": 106759399.0,
      "step": 57260
    },
    {
      "entropy": 6.044418048858643,
      "epoch": 5.056958671847403,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002740615558376094,
      "loss": 5.1611,
      "mean_token_accuracy": 0.22143262773752212,
      "num_tokens": 106768597.0,
      "step": 57265
    },
    {
      "entropy": 6.019056987762451,
      "epoch": 5.057400211939244,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002740300646146692,
      "loss": 5.233,
      "mean_token_accuracy": 0.2297307163476944,
      "num_tokens": 106778239.0,
      "step": 57270
    },
    {
      "entropy": 6.031484889984131,
      "epoch": 5.057841752031084,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002739985734107295,
      "loss": 5.24,
      "mean_token_accuracy": 0.22620611637830734,
      "num_tokens": 106786893.0,
      "step": 57275
    },
    {
      "entropy": 6.097169637680054,
      "epoch": 5.058283292122924,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002739670822264072,
      "loss": 5.2045,
      "mean_token_accuracy": 0.22995937317609788,
      "num_tokens": 106795330.0,
      "step": 57280
    },
    {
      "entropy": 5.993229579925537,
      "epoch": 5.058724832214765,
      "grad_norm": 1.375,
      "learning_rate": 0.0002739355910623191,
      "loss": 5.2293,
      "mean_token_accuracy": 0.21887018531560898,
      "num_tokens": 106804126.0,
      "step": 57285
    },
    {
      "entropy": 6.019246292114258,
      "epoch": 5.059166372306605,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002739040999190822,
      "loss": 5.214,
      "mean_token_accuracy": 0.2231063202023506,
      "num_tokens": 106813180.0,
      "step": 57290
    },
    {
      "entropy": 6.073105096817017,
      "epoch": 5.059607912398445,
      "grad_norm": 1.4375,
      "learning_rate": 0.00027387260879731327,
      "loss": 5.1893,
      "mean_token_accuracy": 0.23523034155368805,
      "num_tokens": 106822428.0,
      "step": 57295
    },
    {
      "entropy": 6.077646636962891,
      "epoch": 5.060049452490286,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002738411176976294,
      "loss": 5.2283,
      "mean_token_accuracy": 0.23075138181447982,
      "num_tokens": 106831599.0,
      "step": 57300
    },
    {
      "entropy": 6.114617013931275,
      "epoch": 5.060490992582126,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002738096266206474,
      "loss": 5.221,
      "mean_token_accuracy": 0.22733762860298157,
      "num_tokens": 106841281.0,
      "step": 57305
    },
    {
      "entropy": 6.087676095962524,
      "epoch": 5.0609325326739665,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002737781355669841,
      "loss": 5.2243,
      "mean_token_accuracy": 0.22369784116744995,
      "num_tokens": 106850634.0,
      "step": 57310
    },
    {
      "entropy": 6.015677785873413,
      "epoch": 5.061374072765807,
      "grad_norm": 1.375,
      "learning_rate": 0.00027374664453725636,
      "loss": 5.191,
      "mean_token_accuracy": 0.23042788803577424,
      "num_tokens": 106860016.0,
      "step": 57315
    },
    {
      "entropy": 5.998890113830567,
      "epoch": 5.0618156128576475,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002737151535320812,
      "loss": 5.1455,
      "mean_token_accuracy": 0.2248256877064705,
      "num_tokens": 106868452.0,
      "step": 57320
    },
    {
      "entropy": 6.021964597702026,
      "epoch": 5.0622571529494875,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002736836625520755,
      "loss": 5.1623,
      "mean_token_accuracy": 0.23155375868082045,
      "num_tokens": 106877711.0,
      "step": 57325
    },
    {
      "entropy": 6.026024293899536,
      "epoch": 5.0626986930413285,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00027365217159785603,
      "loss": 5.2007,
      "mean_token_accuracy": 0.22651539742946625,
      "num_tokens": 106887181.0,
      "step": 57330
    },
    {
      "entropy": 6.094128513336182,
      "epoch": 5.0631402331331685,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002736206806700399,
      "loss": 5.2422,
      "mean_token_accuracy": 0.22013709396123887,
      "num_tokens": 106896133.0,
      "step": 57335
    },
    {
      "entropy": 5.971442365646363,
      "epoch": 5.063581773225009,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00027358918976924364,
      "loss": 5.1628,
      "mean_token_accuracy": 0.2251763641834259,
      "num_tokens": 106905470.0,
      "step": 57340
    },
    {
      "entropy": 6.061490249633789,
      "epoch": 5.0640233133168495,
      "grad_norm": 1.46875,
      "learning_rate": 0.00027355769889608446,
      "loss": 5.2076,
      "mean_token_accuracy": 0.23113714903593063,
      "num_tokens": 106914433.0,
      "step": 57345
    },
    {
      "entropy": 5.937744855880737,
      "epoch": 5.06446485340869,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00027352620805117905,
      "loss": 5.0639,
      "mean_token_accuracy": 0.2391090825200081,
      "num_tokens": 106924276.0,
      "step": 57350
    },
    {
      "entropy": 6.1152419090271,
      "epoch": 5.06490639350053,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002734947172351445,
      "loss": 5.3047,
      "mean_token_accuracy": 0.2157972276210785,
      "num_tokens": 106933910.0,
      "step": 57355
    },
    {
      "entropy": 6.073400688171387,
      "epoch": 5.065347933592371,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002734632264485975,
      "loss": 5.2528,
      "mean_token_accuracy": 0.21701523214578627,
      "num_tokens": 106942975.0,
      "step": 57360
    },
    {
      "entropy": 6.031030082702637,
      "epoch": 5.065789473684211,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00027343173569215507,
      "loss": 5.1697,
      "mean_token_accuracy": 0.23095686882734298,
      "num_tokens": 106951224.0,
      "step": 57365
    },
    {
      "entropy": 6.139675807952881,
      "epoch": 5.066231013776051,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000273400244966434,
      "loss": 5.3359,
      "mean_token_accuracy": 0.2097986102104187,
      "num_tokens": 106961398.0,
      "step": 57370
    },
    {
      "entropy": 6.082107830047607,
      "epoch": 5.066672553867892,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002733687542720513,
      "loss": 5.2132,
      "mean_token_accuracy": 0.23008606284856797,
      "num_tokens": 106970873.0,
      "step": 57375
    },
    {
      "entropy": 5.987268304824829,
      "epoch": 5.067114093959732,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002733372636096237,
      "loss": 5.1026,
      "mean_token_accuracy": 0.23452473282814026,
      "num_tokens": 106980016.0,
      "step": 57380
    },
    {
      "entropy": 6.061761426925659,
      "epoch": 5.067555634051572,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002733057729797682,
      "loss": 5.2488,
      "mean_token_accuracy": 0.22189265191555024,
      "num_tokens": 106989405.0,
      "step": 57385
    },
    {
      "entropy": 6.070483255386352,
      "epoch": 5.067997174143413,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00027327428238310166,
      "loss": 5.1891,
      "mean_token_accuracy": 0.2241707444190979,
      "num_tokens": 106998355.0,
      "step": 57390
    },
    {
      "entropy": 6.0891355037689205,
      "epoch": 5.068438714235253,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002732427918202409,
      "loss": 5.2971,
      "mean_token_accuracy": 0.21709518134593964,
      "num_tokens": 107008916.0,
      "step": 57395
    },
    {
      "entropy": 6.091312837600708,
      "epoch": 5.068880254327093,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00027321130129180296,
      "loss": 5.3019,
      "mean_token_accuracy": 0.2230485573410988,
      "num_tokens": 107018005.0,
      "step": 57400
    },
    {
      "entropy": 6.0701906204223635,
      "epoch": 5.069321794418933,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002731798107984044,
      "loss": 5.1676,
      "mean_token_accuracy": 0.22948036193847657,
      "num_tokens": 107026826.0,
      "step": 57405
    },
    {
      "entropy": 6.0929101467132565,
      "epoch": 5.069763334510774,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027314832034066244,
      "loss": 5.2551,
      "mean_token_accuracy": 0.22824761420488357,
      "num_tokens": 107035356.0,
      "step": 57410
    },
    {
      "entropy": 6.084014749526977,
      "epoch": 5.070204874602614,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00027311682991919387,
      "loss": 5.2597,
      "mean_token_accuracy": 0.22631237506866456,
      "num_tokens": 107044379.0,
      "step": 57415
    },
    {
      "entropy": 6.064893627166748,
      "epoch": 5.070646414694454,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002730853395346155,
      "loss": 5.3369,
      "mean_token_accuracy": 0.21690022349357604,
      "num_tokens": 107055168.0,
      "step": 57420
    },
    {
      "entropy": 6.054228210449219,
      "epoch": 5.071087954786295,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027305384918754417,
      "loss": 5.2942,
      "mean_token_accuracy": 0.2203395038843155,
      "num_tokens": 107064557.0,
      "step": 57425
    },
    {
      "entropy": 6.111990976333618,
      "epoch": 5.071529494878135,
      "grad_norm": 1.515625,
      "learning_rate": 0.00027302235887859686,
      "loss": 5.2386,
      "mean_token_accuracy": 0.22222268134355544,
      "num_tokens": 107073095.0,
      "step": 57430
    },
    {
      "entropy": 6.127697563171386,
      "epoch": 5.071971034969975,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027299086860839043,
      "loss": 5.2509,
      "mean_token_accuracy": 0.2186145156621933,
      "num_tokens": 107082837.0,
      "step": 57435
    },
    {
      "entropy": 6.0206379890441895,
      "epoch": 5.072412575061816,
      "grad_norm": 1.375,
      "learning_rate": 0.00027295937837754184,
      "loss": 5.1784,
      "mean_token_accuracy": 0.23329028487205505,
      "num_tokens": 107091194.0,
      "step": 57440
    },
    {
      "entropy": 6.05669379234314,
      "epoch": 5.072854115153656,
      "grad_norm": 1.6875,
      "learning_rate": 0.0002729278881866677,
      "loss": 5.2142,
      "mean_token_accuracy": 0.23525483906269073,
      "num_tokens": 107100169.0,
      "step": 57445
    },
    {
      "entropy": 6.015271997451782,
      "epoch": 5.073295655245496,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002728963980363851,
      "loss": 5.2367,
      "mean_token_accuracy": 0.22153684347867966,
      "num_tokens": 107109406.0,
      "step": 57450
    },
    {
      "entropy": 6.122568511962891,
      "epoch": 5.073737195337337,
      "grad_norm": 1.390625,
      "learning_rate": 0.00027286490792731097,
      "loss": 5.2874,
      "mean_token_accuracy": 0.2150283470749855,
      "num_tokens": 107118386.0,
      "step": 57455
    },
    {
      "entropy": 6.066212797164917,
      "epoch": 5.074178735429177,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000272833417860062,
      "loss": 5.2446,
      "mean_token_accuracy": 0.2273940086364746,
      "num_tokens": 107127866.0,
      "step": 57460
    },
    {
      "entropy": 6.129306507110596,
      "epoch": 5.074620275521017,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00027280192783525517,
      "loss": 5.258,
      "mean_token_accuracy": 0.2208525210618973,
      "num_tokens": 107136026.0,
      "step": 57465
    },
    {
      "entropy": 6.033658123016357,
      "epoch": 5.075061815612858,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002727704378535072,
      "loss": 5.2134,
      "mean_token_accuracy": 0.22219108343124389,
      "num_tokens": 107145237.0,
      "step": 57470
    },
    {
      "entropy": 6.074203968048096,
      "epoch": 5.075503355704698,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027273894791543524,
      "loss": 5.2663,
      "mean_token_accuracy": 0.22507865875959396,
      "num_tokens": 107154789.0,
      "step": 57475
    },
    {
      "entropy": 6.135494995117187,
      "epoch": 5.075944895796538,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027270745802165594,
      "loss": 5.3073,
      "mean_token_accuracy": 0.21515426635742188,
      "num_tokens": 107163294.0,
      "step": 57480
    },
    {
      "entropy": 6.116348552703857,
      "epoch": 5.076386435888379,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027267596817278625,
      "loss": 5.2736,
      "mean_token_accuracy": 0.2178521141409874,
      "num_tokens": 107172315.0,
      "step": 57485
    },
    {
      "entropy": 6.039710664749146,
      "epoch": 5.076827975980219,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027264447836944296,
      "loss": 5.1876,
      "mean_token_accuracy": 0.22666075825691223,
      "num_tokens": 107182144.0,
      "step": 57490
    },
    {
      "entropy": 5.974835634231567,
      "epoch": 5.077269516072059,
      "grad_norm": 1.359375,
      "learning_rate": 0.00027261298861224305,
      "loss": 5.1276,
      "mean_token_accuracy": 0.23372730016708373,
      "num_tokens": 107191255.0,
      "step": 57495
    },
    {
      "entropy": 6.075478458404541,
      "epoch": 5.0777110561639,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002725814989018033,
      "loss": 5.2623,
      "mean_token_accuracy": 0.22115224003791809,
      "num_tokens": 107199950.0,
      "step": 57500
    },
    {
      "entropy": 6.028783369064331,
      "epoch": 5.07815259625574,
      "grad_norm": 1.25,
      "learning_rate": 0.0002725500092387407,
      "loss": 5.2439,
      "mean_token_accuracy": 0.2199856847524643,
      "num_tokens": 107210262.0,
      "step": 57505
    },
    {
      "entropy": 6.059226846694946,
      "epoch": 5.07859413634758,
      "grad_norm": 1.375,
      "learning_rate": 0.000272518519623672,
      "loss": 5.214,
      "mean_token_accuracy": 0.22559371888637542,
      "num_tokens": 107219343.0,
      "step": 57510
    },
    {
      "entropy": 6.120079755783081,
      "epoch": 5.079035676439421,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027248703005721396,
      "loss": 5.3167,
      "mean_token_accuracy": 0.21302568018436432,
      "num_tokens": 107228332.0,
      "step": 57515
    },
    {
      "entropy": 5.988225221633911,
      "epoch": 5.079477216531261,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002724555405399837,
      "loss": 5.098,
      "mean_token_accuracy": 0.24264044016599656,
      "num_tokens": 107237844.0,
      "step": 57520
    },
    {
      "entropy": 5.959729337692261,
      "epoch": 5.079918756623101,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000272424051072598,
      "loss": 5.2243,
      "mean_token_accuracy": 0.2279964342713356,
      "num_tokens": 107247398.0,
      "step": 57525
    },
    {
      "entropy": 5.9529776096344,
      "epoch": 5.080360296714941,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002723925616556736,
      "loss": 5.069,
      "mean_token_accuracy": 0.23132589757442473,
      "num_tokens": 107256937.0,
      "step": 57530
    },
    {
      "entropy": 6.081384134292603,
      "epoch": 5.080801836806782,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002723610722898274,
      "loss": 5.2693,
      "mean_token_accuracy": 0.2209930405020714,
      "num_tokens": 107265667.0,
      "step": 57535
    },
    {
      "entropy": 6.078723669052124,
      "epoch": 5.081243376898622,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00027232958297567633,
      "loss": 5.289,
      "mean_token_accuracy": 0.2240409553050995,
      "num_tokens": 107275099.0,
      "step": 57540
    },
    {
      "entropy": 5.985892248153687,
      "epoch": 5.081684916990462,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00027229809371383723,
      "loss": 5.2209,
      "mean_token_accuracy": 0.22797100991010666,
      "num_tokens": 107284674.0,
      "step": 57545
    },
    {
      "entropy": 6.0490740776062015,
      "epoch": 5.082126457082303,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002722666045049269,
      "loss": 5.225,
      "mean_token_accuracy": 0.22393065989017485,
      "num_tokens": 107294954.0,
      "step": 57550
    },
    {
      "entropy": 6.016070699691772,
      "epoch": 5.082567997174143,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002722351153495622,
      "loss": 5.1479,
      "mean_token_accuracy": 0.22244638353586196,
      "num_tokens": 107303975.0,
      "step": 57555
    },
    {
      "entropy": 6.0460076332092285,
      "epoch": 5.083009537265983,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002722036262483601,
      "loss": 5.188,
      "mean_token_accuracy": 0.22634099423885345,
      "num_tokens": 107313198.0,
      "step": 57560
    },
    {
      "entropy": 5.989384889602661,
      "epoch": 5.083451077357824,
      "grad_norm": 1.359375,
      "learning_rate": 0.00027217213720193736,
      "loss": 5.2097,
      "mean_token_accuracy": 0.22869783788919448,
      "num_tokens": 107322970.0,
      "step": 57565
    },
    {
      "entropy": 5.9572492122650145,
      "epoch": 5.083892617449664,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002721406482109108,
      "loss": 5.1205,
      "mean_token_accuracy": 0.22628855407238008,
      "num_tokens": 107331211.0,
      "step": 57570
    },
    {
      "entropy": 6.096937656402588,
      "epoch": 5.084334157541504,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002721091592758974,
      "loss": 5.3257,
      "mean_token_accuracy": 0.21379570960998534,
      "num_tokens": 107340283.0,
      "step": 57575
    },
    {
      "entropy": 6.140096235275268,
      "epoch": 5.084775697633345,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002720776703975138,
      "loss": 5.2613,
      "mean_token_accuracy": 0.2187095269560814,
      "num_tokens": 107349614.0,
      "step": 57580
    },
    {
      "entropy": 6.065349817276001,
      "epoch": 5.085217237725185,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00027204618157637705,
      "loss": 5.1331,
      "mean_token_accuracy": 0.22999337017536164,
      "num_tokens": 107358006.0,
      "step": 57585
    },
    {
      "entropy": 6.017925119400024,
      "epoch": 5.0856587778170255,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002720146928131039,
      "loss": 5.2569,
      "mean_token_accuracy": 0.22037787288427352,
      "num_tokens": 107367990.0,
      "step": 57590
    },
    {
      "entropy": 6.060958814620972,
      "epoch": 5.086100317908866,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027198320410831125,
      "loss": 5.2893,
      "mean_token_accuracy": 0.22009919136762618,
      "num_tokens": 107377076.0,
      "step": 57595
    },
    {
      "entropy": 6.122621059417725,
      "epoch": 5.0865418580007065,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002719517154626159,
      "loss": 5.1933,
      "mean_token_accuracy": 0.22695891559123993,
      "num_tokens": 107386169.0,
      "step": 57600
    },
    {
      "entropy": 6.002755498886108,
      "epoch": 5.0869833980925465,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002719202268766347,
      "loss": 5.1267,
      "mean_token_accuracy": 0.22924428880214692,
      "num_tokens": 107395659.0,
      "step": 57605
    },
    {
      "entropy": 6.078211879730224,
      "epoch": 5.0874249381843875,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002718887383509845,
      "loss": 5.2938,
      "mean_token_accuracy": 0.2143993154168129,
      "num_tokens": 107405546.0,
      "step": 57610
    },
    {
      "entropy": 6.057095718383789,
      "epoch": 5.0878664782762275,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002718572498862822,
      "loss": 5.164,
      "mean_token_accuracy": 0.23430154025554656,
      "num_tokens": 107415162.0,
      "step": 57615
    },
    {
      "entropy": 6.072339820861816,
      "epoch": 5.088308018368068,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00027182576148314445,
      "loss": 5.19,
      "mean_token_accuracy": 0.22816912680864335,
      "num_tokens": 107425383.0,
      "step": 57620
    },
    {
      "entropy": 6.099550485610962,
      "epoch": 5.0887495584599085,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00027179427314218834,
      "loss": 5.279,
      "mean_token_accuracy": 0.21706322729587554,
      "num_tokens": 107434438.0,
      "step": 57625
    },
    {
      "entropy": 6.033328723907471,
      "epoch": 5.089191098551749,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002717627848640306,
      "loss": 5.1629,
      "mean_token_accuracy": 0.22458149045705794,
      "num_tokens": 107443122.0,
      "step": 57630
    },
    {
      "entropy": 5.9889702796936035,
      "epoch": 5.089632638643589,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002717312966492881,
      "loss": 5.2098,
      "mean_token_accuracy": 0.22317602783441542,
      "num_tokens": 107452751.0,
      "step": 57635
    },
    {
      "entropy": 6.066564559936523,
      "epoch": 5.0900741787354296,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00027169980849857763,
      "loss": 5.3069,
      "mean_token_accuracy": 0.21557349860668182,
      "num_tokens": 107462744.0,
      "step": 57640
    },
    {
      "entropy": 6.092546892166138,
      "epoch": 5.09051571882727,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002716683204125159,
      "loss": 5.2737,
      "mean_token_accuracy": 0.22608109563589096,
      "num_tokens": 107472056.0,
      "step": 57645
    },
    {
      "entropy": 6.094057989120484,
      "epoch": 5.09095725891911,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00027163683239171996,
      "loss": 5.2078,
      "mean_token_accuracy": 0.2275050774216652,
      "num_tokens": 107481218.0,
      "step": 57650
    },
    {
      "entropy": 6.073882722854615,
      "epoch": 5.091398799010951,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002716053444368066,
      "loss": 5.2129,
      "mean_token_accuracy": 0.22458608001470565,
      "num_tokens": 107490817.0,
      "step": 57655
    },
    {
      "entropy": 6.0074629306793215,
      "epoch": 5.091840339102791,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00027157385654839254,
      "loss": 5.2433,
      "mean_token_accuracy": 0.21537226736545562,
      "num_tokens": 107499500.0,
      "step": 57660
    },
    {
      "entropy": 5.989596891403198,
      "epoch": 5.092281879194631,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002715423687270947,
      "loss": 5.2026,
      "mean_token_accuracy": 0.2230004832148552,
      "num_tokens": 107510243.0,
      "step": 57665
    },
    {
      "entropy": 6.020297384262085,
      "epoch": 5.092723419286472,
      "grad_norm": 1.40625,
      "learning_rate": 0.00027151088097352995,
      "loss": 5.1341,
      "mean_token_accuracy": 0.23862782269716262,
      "num_tokens": 107518912.0,
      "step": 57670
    },
    {
      "entropy": 6.066211223602295,
      "epoch": 5.093164959378312,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000271479393288315,
      "loss": 5.2568,
      "mean_token_accuracy": 0.22926558554172516,
      "num_tokens": 107528492.0,
      "step": 57675
    },
    {
      "entropy": 6.1009539604187015,
      "epoch": 5.093606499470152,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00027144790567206677,
      "loss": 5.2768,
      "mean_token_accuracy": 0.2163035050034523,
      "num_tokens": 107538202.0,
      "step": 57680
    },
    {
      "entropy": 6.066462564468384,
      "epoch": 5.094048039561992,
      "grad_norm": 1.453125,
      "learning_rate": 0.00027141641812540205,
      "loss": 5.1823,
      "mean_token_accuracy": 0.2252688631415367,
      "num_tokens": 107546461.0,
      "step": 57685
    },
    {
      "entropy": 5.96748104095459,
      "epoch": 5.094489579653833,
      "grad_norm": 1.421875,
      "learning_rate": 0.00027138493064893753,
      "loss": 5.1617,
      "mean_token_accuracy": 0.23489903211593627,
      "num_tokens": 107554986.0,
      "step": 57690
    },
    {
      "entropy": 6.072757005691528,
      "epoch": 5.094931119745673,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002713534432432904,
      "loss": 5.2161,
      "mean_token_accuracy": 0.23007556200027465,
      "num_tokens": 107564124.0,
      "step": 57695
    },
    {
      "entropy": 6.070644092559815,
      "epoch": 5.095372659837513,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002713219559090771,
      "loss": 5.1718,
      "mean_token_accuracy": 0.2255056843161583,
      "num_tokens": 107572676.0,
      "step": 57700
    },
    {
      "entropy": 6.084677600860596,
      "epoch": 5.095814199929354,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002712904686469146,
      "loss": 5.1934,
      "mean_token_accuracy": 0.2271731451153755,
      "num_tokens": 107581396.0,
      "step": 57705
    },
    {
      "entropy": 6.021796798706054,
      "epoch": 5.096255740021194,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002712589814574197,
      "loss": 5.1689,
      "mean_token_accuracy": 0.23124755471944808,
      "num_tokens": 107590332.0,
      "step": 57710
    },
    {
      "entropy": 6.0692458152771,
      "epoch": 5.096697280113034,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027122749434120927,
      "loss": 5.2694,
      "mean_token_accuracy": 0.2171960785984993,
      "num_tokens": 107599922.0,
      "step": 57715
    },
    {
      "entropy": 6.049526023864746,
      "epoch": 5.097138820204875,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027119600729890006,
      "loss": 5.2113,
      "mean_token_accuracy": 0.2282872125506401,
      "num_tokens": 107608781.0,
      "step": 57720
    },
    {
      "entropy": 6.008285570144653,
      "epoch": 5.097580360296715,
      "grad_norm": 1.359375,
      "learning_rate": 0.00027116452033110883,
      "loss": 5.1835,
      "mean_token_accuracy": 0.2261463537812233,
      "num_tokens": 107617701.0,
      "step": 57725
    },
    {
      "entropy": 5.970264196395874,
      "epoch": 5.098021900388555,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027113303343845254,
      "loss": 5.1734,
      "mean_token_accuracy": 0.22493974417448043,
      "num_tokens": 107627066.0,
      "step": 57730
    },
    {
      "entropy": 6.0820784091949465,
      "epoch": 5.098463440480396,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002711015466215479,
      "loss": 5.2253,
      "mean_token_accuracy": 0.22200553864240646,
      "num_tokens": 107636897.0,
      "step": 57735
    },
    {
      "entropy": 6.1113018035888675,
      "epoch": 5.098904980572236,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027107005988101176,
      "loss": 5.3043,
      "mean_token_accuracy": 0.21669484972953795,
      "num_tokens": 107645460.0,
      "step": 57740
    },
    {
      "entropy": 6.066132354736328,
      "epoch": 5.099346520664076,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027103857321746094,
      "loss": 5.2359,
      "mean_token_accuracy": 0.22391301691532134,
      "num_tokens": 107655175.0,
      "step": 57745
    },
    {
      "entropy": 6.001733875274658,
      "epoch": 5.099788060755917,
      "grad_norm": 1.359375,
      "learning_rate": 0.00027100708663151227,
      "loss": 5.0748,
      "mean_token_accuracy": 0.23482443541288375,
      "num_tokens": 107663477.0,
      "step": 57750
    },
    {
      "entropy": 6.042233419418335,
      "epoch": 5.100229600847757,
      "grad_norm": 1.4375,
      "learning_rate": 0.00027097560012378233,
      "loss": 5.1821,
      "mean_token_accuracy": 0.22045822888612748,
      "num_tokens": 107672583.0,
      "step": 57755
    },
    {
      "entropy": 5.970422840118408,
      "epoch": 5.100671140939597,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002709441136948882,
      "loss": 5.219,
      "mean_token_accuracy": 0.22564164400100709,
      "num_tokens": 107682575.0,
      "step": 57760
    },
    {
      "entropy": 5.996292352676392,
      "epoch": 5.101112681031438,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002709126273454466,
      "loss": 5.2343,
      "mean_token_accuracy": 0.2247144728899002,
      "num_tokens": 107691599.0,
      "step": 57765
    },
    {
      "entropy": 6.034738683700562,
      "epoch": 5.101554221123278,
      "grad_norm": 1.359375,
      "learning_rate": 0.00027088114107607427,
      "loss": 5.209,
      "mean_token_accuracy": 0.2367088869214058,
      "num_tokens": 107701479.0,
      "step": 57770
    },
    {
      "entropy": 5.899065351486206,
      "epoch": 5.101995761215118,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027084965488738795,
      "loss": 5.0494,
      "mean_token_accuracy": 0.24003540128469467,
      "num_tokens": 107710384.0,
      "step": 57775
    },
    {
      "entropy": 6.018415069580078,
      "epoch": 5.102437301306959,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00027081816878000466,
      "loss": 5.1359,
      "mean_token_accuracy": 0.2301793321967125,
      "num_tokens": 107719697.0,
      "step": 57780
    },
    {
      "entropy": 6.020245838165283,
      "epoch": 5.102878841398799,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002707866827545411,
      "loss": 5.1187,
      "mean_token_accuracy": 0.2310629442334175,
      "num_tokens": 107728223.0,
      "step": 57785
    },
    {
      "entropy": 6.079818964004517,
      "epoch": 5.103320381490639,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027075519681161385,
      "loss": 5.2437,
      "mean_token_accuracy": 0.21473153978586196,
      "num_tokens": 107737770.0,
      "step": 57790
    },
    {
      "entropy": 6.058513593673706,
      "epoch": 5.10376192158248,
      "grad_norm": 1.453125,
      "learning_rate": 0.00027072371095184,
      "loss": 5.2572,
      "mean_token_accuracy": 0.21885233521461486,
      "num_tokens": 107746401.0,
      "step": 57795
    },
    {
      "entropy": 5.953838920593261,
      "epoch": 5.10420346167432,
      "grad_norm": 1.4375,
      "learning_rate": 0.00027069222517583626,
      "loss": 5.0934,
      "mean_token_accuracy": 0.23781318664550782,
      "num_tokens": 107755285.0,
      "step": 57800
    },
    {
      "entropy": 5.994574165344238,
      "epoch": 5.10464500176616,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002706607394842193,
      "loss": 5.2221,
      "mean_token_accuracy": 0.22332217395305634,
      "num_tokens": 107765178.0,
      "step": 57805
    },
    {
      "entropy": 6.087968635559082,
      "epoch": 5.105086541858,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000270629253877606,
      "loss": 5.2844,
      "mean_token_accuracy": 0.22504662871360778,
      "num_tokens": 107774495.0,
      "step": 57810
    },
    {
      "entropy": 6.086825561523438,
      "epoch": 5.105528081949841,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002705977683566132,
      "loss": 5.2508,
      "mean_token_accuracy": 0.21799319833517075,
      "num_tokens": 107783710.0,
      "step": 57815
    },
    {
      "entropy": 6.018869829177857,
      "epoch": 5.105969622041681,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027056628292185746,
      "loss": 5.1328,
      "mean_token_accuracy": 0.23352475613355636,
      "num_tokens": 107792153.0,
      "step": 57820
    },
    {
      "entropy": 6.1091328144073485,
      "epoch": 5.106411162133521,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002705347975739559,
      "loss": 5.2859,
      "mean_token_accuracy": 0.22088534384965897,
      "num_tokens": 107801503.0,
      "step": 57825
    },
    {
      "entropy": 5.923882865905762,
      "epoch": 5.106852702225362,
      "grad_norm": 1.359375,
      "learning_rate": 0.000270503312313525,
      "loss": 5.1402,
      "mean_token_accuracy": 0.2345053419470787,
      "num_tokens": 107810086.0,
      "step": 57830
    },
    {
      "entropy": 6.005446863174439,
      "epoch": 5.107294242317202,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002704718271411817,
      "loss": 5.1845,
      "mean_token_accuracy": 0.22291427254676818,
      "num_tokens": 107819943.0,
      "step": 57835
    },
    {
      "entropy": 5.963482189178467,
      "epoch": 5.107735782409042,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027044034205754266,
      "loss": 5.1144,
      "mean_token_accuracy": 0.23480138927698135,
      "num_tokens": 107829073.0,
      "step": 57840
    },
    {
      "entropy": 6.05488805770874,
      "epoch": 5.108177322500883,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027040885706322484,
      "loss": 5.2366,
      "mean_token_accuracy": 0.22336840033531188,
      "num_tokens": 107839523.0,
      "step": 57845
    },
    {
      "entropy": 6.08737096786499,
      "epoch": 5.108618862592723,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027037737215884484,
      "loss": 5.286,
      "mean_token_accuracy": 0.2206320807337761,
      "num_tokens": 107848565.0,
      "step": 57850
    },
    {
      "entropy": 5.972881841659546,
      "epoch": 5.109060402684563,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00027034588734501947,
      "loss": 5.1753,
      "mean_token_accuracy": 0.22504884153604507,
      "num_tokens": 107857440.0,
      "step": 57855
    },
    {
      "entropy": 6.022005653381347,
      "epoch": 5.109501942776404,
      "grad_norm": 1.421875,
      "learning_rate": 0.00027031440262236553,
      "loss": 5.2016,
      "mean_token_accuracy": 0.22852162420749664,
      "num_tokens": 107866164.0,
      "step": 57860
    },
    {
      "entropy": 6.077906131744385,
      "epoch": 5.109943482868244,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027028291799149976,
      "loss": 5.2844,
      "mean_token_accuracy": 0.221753354370594,
      "num_tokens": 107875359.0,
      "step": 57865
    },
    {
      "entropy": 6.0889770030975345,
      "epoch": 5.1103850229600845,
      "grad_norm": 1.3125,
      "learning_rate": 0.000270251433453039,
      "loss": 5.2808,
      "mean_token_accuracy": 0.22011569887399673,
      "num_tokens": 107884606.0,
      "step": 57870
    },
    {
      "entropy": 6.091895723342896,
      "epoch": 5.110826563051925,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002702199490075999,
      "loss": 5.238,
      "mean_token_accuracy": 0.2155725121498108,
      "num_tokens": 107894133.0,
      "step": 57875
    },
    {
      "entropy": 6.0826832294464115,
      "epoch": 5.1112681031437655,
      "grad_norm": 1.4375,
      "learning_rate": 0.00027018846465579933,
      "loss": 5.2598,
      "mean_token_accuracy": 0.22293177843093873,
      "num_tokens": 107903134.0,
      "step": 57880
    },
    {
      "entropy": 5.908328723907471,
      "epoch": 5.1117096432356055,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002701569803982539,
      "loss": 5.0119,
      "mean_token_accuracy": 0.2472485527396202,
      "num_tokens": 107912183.0,
      "step": 57885
    },
    {
      "entropy": 5.982635164260865,
      "epoch": 5.1121511833274464,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002701254962355806,
      "loss": 5.2259,
      "mean_token_accuracy": 0.22255186289548873,
      "num_tokens": 107921239.0,
      "step": 57890
    },
    {
      "entropy": 5.9395280361175535,
      "epoch": 5.1125927234192865,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027009401216839603,
      "loss": 5.1651,
      "mean_token_accuracy": 0.23688606172800064,
      "num_tokens": 107930468.0,
      "step": 57895
    },
    {
      "entropy": 6.035775136947632,
      "epoch": 5.1130342635111266,
      "grad_norm": 1.390625,
      "learning_rate": 0.00027006252819731694,
      "loss": 5.2088,
      "mean_token_accuracy": 0.2267150491476059,
      "num_tokens": 107939475.0,
      "step": 57900
    },
    {
      "entropy": 5.9869225025177,
      "epoch": 5.1134758036029675,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027003104432296013,
      "loss": 5.2133,
      "mean_token_accuracy": 0.23150868862867355,
      "num_tokens": 107949095.0,
      "step": 57905
    },
    {
      "entropy": 5.962857055664062,
      "epoch": 5.1139173436948075,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00026999956054594233,
      "loss": 5.0767,
      "mean_token_accuracy": 0.23945170044898986,
      "num_tokens": 107957738.0,
      "step": 57910
    },
    {
      "entropy": 6.093236064910888,
      "epoch": 5.114358883786648,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002699680768668803,
      "loss": 5.235,
      "mean_token_accuracy": 0.21775096505880356,
      "num_tokens": 107967454.0,
      "step": 57915
    },
    {
      "entropy": 6.063222074508667,
      "epoch": 5.1148004238784885,
      "grad_norm": 1.40625,
      "learning_rate": 0.00026993659328639074,
      "loss": 5.1871,
      "mean_token_accuracy": 0.2241704061627388,
      "num_tokens": 107976075.0,
      "step": 57920
    },
    {
      "entropy": 6.030957365036011,
      "epoch": 5.115241963970329,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026990510980509056,
      "loss": 5.1741,
      "mean_token_accuracy": 0.22832833975553513,
      "num_tokens": 107985593.0,
      "step": 57925
    },
    {
      "entropy": 6.0492377281188965,
      "epoch": 5.115683504062169,
      "grad_norm": 1.5,
      "learning_rate": 0.0002698736264235962,
      "loss": 5.2389,
      "mean_token_accuracy": 0.22737445682287216,
      "num_tokens": 107995236.0,
      "step": 57930
    },
    {
      "entropy": 6.011835527420044,
      "epoch": 5.11612504415401,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002698421431425247,
      "loss": 5.1959,
      "mean_token_accuracy": 0.2301377311348915,
      "num_tokens": 108005984.0,
      "step": 57935
    },
    {
      "entropy": 6.06799201965332,
      "epoch": 5.11656658424585,
      "grad_norm": 1.25,
      "learning_rate": 0.00026981065996249263,
      "loss": 5.239,
      "mean_token_accuracy": 0.21719661951065064,
      "num_tokens": 108015476.0,
      "step": 57940
    },
    {
      "entropy": 6.1197902202606205,
      "epoch": 5.11700812433769,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002697791768841169,
      "loss": 5.3118,
      "mean_token_accuracy": 0.20940975099802017,
      "num_tokens": 108024999.0,
      "step": 57945
    },
    {
      "entropy": 6.122086238861084,
      "epoch": 5.117449664429531,
      "grad_norm": 1.40625,
      "learning_rate": 0.000269747693908014,
      "loss": 5.2422,
      "mean_token_accuracy": 0.2230202317237854,
      "num_tokens": 108033984.0,
      "step": 57950
    },
    {
      "entropy": 6.165675020217895,
      "epoch": 5.117891204521371,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00026971621103480083,
      "loss": 5.3183,
      "mean_token_accuracy": 0.2102182537317276,
      "num_tokens": 108043307.0,
      "step": 57955
    },
    {
      "entropy": 6.042755889892578,
      "epoch": 5.118332744613211,
      "grad_norm": 1.375,
      "learning_rate": 0.000269684728265094,
      "loss": 5.201,
      "mean_token_accuracy": 0.23006541579961776,
      "num_tokens": 108052644.0,
      "step": 57960
    },
    {
      "entropy": 5.984759902954101,
      "epoch": 5.118774284705051,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00026965324559951047,
      "loss": 5.1916,
      "mean_token_accuracy": 0.2337382525205612,
      "num_tokens": 108061551.0,
      "step": 57965
    },
    {
      "entropy": 6.004194688796997,
      "epoch": 5.119215824796892,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002696217630386667,
      "loss": 5.2322,
      "mean_token_accuracy": 0.22641228437423705,
      "num_tokens": 108069613.0,
      "step": 57970
    },
    {
      "entropy": 6.082559585571289,
      "epoch": 5.119657364888732,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002695902805831796,
      "loss": 5.2425,
      "mean_token_accuracy": 0.22010931223630906,
      "num_tokens": 108079954.0,
      "step": 57975
    },
    {
      "entropy": 6.126623678207397,
      "epoch": 5.120098904980572,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002695587982336658,
      "loss": 5.2563,
      "mean_token_accuracy": 0.22083953768014908,
      "num_tokens": 108088785.0,
      "step": 57980
    },
    {
      "entropy": 6.0112508773803714,
      "epoch": 5.120540445072413,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000269527315990742,
      "loss": 5.1955,
      "mean_token_accuracy": 0.23082282543182372,
      "num_tokens": 108097759.0,
      "step": 57985
    },
    {
      "entropy": 6.050992774963379,
      "epoch": 5.120981985164253,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002694958338550251,
      "loss": 5.2768,
      "mean_token_accuracy": 0.22060371786355973,
      "num_tokens": 108107268.0,
      "step": 57990
    },
    {
      "entropy": 6.036056280136108,
      "epoch": 5.121423525256093,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002694643518271315,
      "loss": 5.28,
      "mean_token_accuracy": 0.2212240681052208,
      "num_tokens": 108116734.0,
      "step": 57995
    },
    {
      "entropy": 6.047697734832764,
      "epoch": 5.121865065347934,
      "grad_norm": 1.375,
      "learning_rate": 0.00026943286990767823,
      "loss": 5.2228,
      "mean_token_accuracy": 0.22801891416311265,
      "num_tokens": 108126480.0,
      "step": 58000
    },
    {
      "entropy": 6.15278811454773,
      "epoch": 5.122306605439774,
      "grad_norm": 1.4375,
      "learning_rate": 0.00026940138809728186,
      "loss": 5.2819,
      "mean_token_accuracy": 0.22205127626657487,
      "num_tokens": 108135533.0,
      "step": 58005
    },
    {
      "entropy": 6.0539963722229,
      "epoch": 5.122748145531614,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002693699063965592,
      "loss": 5.2425,
      "mean_token_accuracy": 0.21949058175086975,
      "num_tokens": 108144437.0,
      "step": 58010
    },
    {
      "entropy": 6.110585117340088,
      "epoch": 5.123189685623455,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002693384248061267,
      "loss": 5.3059,
      "mean_token_accuracy": 0.21868440508842468,
      "num_tokens": 108153542.0,
      "step": 58015
    },
    {
      "entropy": 6.1057204246521,
      "epoch": 5.123631225715295,
      "grad_norm": 1.4375,
      "learning_rate": 0.00026930694332660127,
      "loss": 5.2475,
      "mean_token_accuracy": 0.22371246367692948,
      "num_tokens": 108162488.0,
      "step": 58020
    },
    {
      "entropy": 6.100756072998047,
      "epoch": 5.124072765807135,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002692754619585997,
      "loss": 5.1928,
      "mean_token_accuracy": 0.2293983981013298,
      "num_tokens": 108171594.0,
      "step": 58025
    },
    {
      "entropy": 6.06680383682251,
      "epoch": 5.124514305898976,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002692439807027385,
      "loss": 5.3086,
      "mean_token_accuracy": 0.2121575340628624,
      "num_tokens": 108181525.0,
      "step": 58030
    },
    {
      "entropy": 6.041836404800415,
      "epoch": 5.124955845990816,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00026921249955963434,
      "loss": 5.2252,
      "mean_token_accuracy": 0.22860667407512664,
      "num_tokens": 108191058.0,
      "step": 58035
    },
    {
      "entropy": 6.053577518463134,
      "epoch": 5.125397386082656,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026918101852990413,
      "loss": 5.2588,
      "mean_token_accuracy": 0.2209423378109932,
      "num_tokens": 108200244.0,
      "step": 58040
    },
    {
      "entropy": 5.944131469726562,
      "epoch": 5.125838926174497,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002691495376141645,
      "loss": 5.1118,
      "mean_token_accuracy": 0.23557327389717103,
      "num_tokens": 108209600.0,
      "step": 58045
    },
    {
      "entropy": 6.105034828186035,
      "epoch": 5.126280466266337,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00026911805681303203,
      "loss": 5.3177,
      "mean_token_accuracy": 0.21893067806959152,
      "num_tokens": 108219225.0,
      "step": 58050
    },
    {
      "entropy": 6.182063388824463,
      "epoch": 5.126722006358177,
      "grad_norm": 1.5,
      "learning_rate": 0.0002690865761271235,
      "loss": 5.301,
      "mean_token_accuracy": 0.21952827274799347,
      "num_tokens": 108227782.0,
      "step": 58055
    },
    {
      "entropy": 6.052910089492798,
      "epoch": 5.127163546450018,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002690550955570555,
      "loss": 5.2231,
      "mean_token_accuracy": 0.22304531186819077,
      "num_tokens": 108237226.0,
      "step": 58060
    },
    {
      "entropy": 5.98926682472229,
      "epoch": 5.127605086541858,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002690236151034449,
      "loss": 5.1943,
      "mean_token_accuracy": 0.22598605006933212,
      "num_tokens": 108246105.0,
      "step": 58065
    },
    {
      "entropy": 6.037521839141846,
      "epoch": 5.128046626633698,
      "grad_norm": 1.375,
      "learning_rate": 0.0002689921347669083,
      "loss": 5.2248,
      "mean_token_accuracy": 0.22180511504411698,
      "num_tokens": 108255890.0,
      "step": 58070
    },
    {
      "entropy": 5.966853761672974,
      "epoch": 5.128488166725539,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002689606545480624,
      "loss": 5.0575,
      "mean_token_accuracy": 0.2404019922018051,
      "num_tokens": 108265009.0,
      "step": 58075
    },
    {
      "entropy": 6.052152872085571,
      "epoch": 5.128929706817379,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002689291744475237,
      "loss": 5.2098,
      "mean_token_accuracy": 0.22870263755321502,
      "num_tokens": 108274547.0,
      "step": 58080
    },
    {
      "entropy": 5.9988258361816404,
      "epoch": 5.129371246909219,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002688976944659091,
      "loss": 5.1882,
      "mean_token_accuracy": 0.23079932630062103,
      "num_tokens": 108283942.0,
      "step": 58085
    },
    {
      "entropy": 6.0111010551452635,
      "epoch": 5.129812787001059,
      "grad_norm": 1.5,
      "learning_rate": 0.00026886621460383523,
      "loss": 5.2308,
      "mean_token_accuracy": 0.22126525342464448,
      "num_tokens": 108293514.0,
      "step": 58090
    },
    {
      "entropy": 6.078436374664307,
      "epoch": 5.1302543270929,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002688347348619187,
      "loss": 5.2866,
      "mean_token_accuracy": 0.22236368060112,
      "num_tokens": 108302692.0,
      "step": 58095
    },
    {
      "entropy": 6.004593276977539,
      "epoch": 5.13069586718474,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002688032552407761,
      "loss": 5.1659,
      "mean_token_accuracy": 0.2265709787607193,
      "num_tokens": 108311659.0,
      "step": 58100
    },
    {
      "entropy": 6.046520566940307,
      "epoch": 5.13113740727658,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002687717757410243,
      "loss": 5.1988,
      "mean_token_accuracy": 0.2229372575879097,
      "num_tokens": 108321248.0,
      "step": 58105
    },
    {
      "entropy": 6.038823652267456,
      "epoch": 5.131578947368421,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002687402963632799,
      "loss": 5.2788,
      "mean_token_accuracy": 0.22341022938489913,
      "num_tokens": 108330610.0,
      "step": 58110
    },
    {
      "entropy": 6.083142566680908,
      "epoch": 5.132020487460261,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00026870881710815944,
      "loss": 5.2001,
      "mean_token_accuracy": 0.22988732010126114,
      "num_tokens": 108338891.0,
      "step": 58115
    },
    {
      "entropy": 6.049809837341309,
      "epoch": 5.132462027552101,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026867733797627976,
      "loss": 5.2436,
      "mean_token_accuracy": 0.22149487286806108,
      "num_tokens": 108348120.0,
      "step": 58120
    },
    {
      "entropy": 6.066640567779541,
      "epoch": 5.132903567643942,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026864585896825734,
      "loss": 5.3209,
      "mean_token_accuracy": 0.21298032104969025,
      "num_tokens": 108357859.0,
      "step": 58125
    },
    {
      "entropy": 6.07613697052002,
      "epoch": 5.133345107735782,
      "grad_norm": 1.515625,
      "learning_rate": 0.000268614380084709,
      "loss": 5.3095,
      "mean_token_accuracy": 0.21713941246271135,
      "num_tokens": 108366472.0,
      "step": 58130
    },
    {
      "entropy": 6.1392096996307375,
      "epoch": 5.133786647827622,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026858290132625133,
      "loss": 5.345,
      "mean_token_accuracy": 0.21183254569768906,
      "num_tokens": 108375865.0,
      "step": 58135
    },
    {
      "entropy": 6.085149240493775,
      "epoch": 5.134228187919463,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026855142269350103,
      "loss": 5.2246,
      "mean_token_accuracy": 0.2232057124376297,
      "num_tokens": 108386387.0,
      "step": 58140
    },
    {
      "entropy": 6.076584959030152,
      "epoch": 5.134669728011303,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00026851994418707455,
      "loss": 5.2328,
      "mean_token_accuracy": 0.226354418694973,
      "num_tokens": 108396915.0,
      "step": 58145
    },
    {
      "entropy": 6.054689073562622,
      "epoch": 5.1351112681031434,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002684884658075888,
      "loss": 5.2274,
      "mean_token_accuracy": 0.22709985077381134,
      "num_tokens": 108405721.0,
      "step": 58150
    },
    {
      "entropy": 5.977194786071777,
      "epoch": 5.135552808194984,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002684569875556602,
      "loss": 5.2286,
      "mean_token_accuracy": 0.21931505352258682,
      "num_tokens": 108414588.0,
      "step": 58155
    },
    {
      "entropy": 5.997914409637451,
      "epoch": 5.135994348286824,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026842550943190566,
      "loss": 5.1899,
      "mean_token_accuracy": 0.23108907043933868,
      "num_tokens": 108424121.0,
      "step": 58160
    },
    {
      "entropy": 6.093493986129761,
      "epoch": 5.1364358883786645,
      "grad_norm": 1.421875,
      "learning_rate": 0.00026839403143694146,
      "loss": 5.2594,
      "mean_token_accuracy": 0.2193574383854866,
      "num_tokens": 108433921.0,
      "step": 58165
    },
    {
      "entropy": 6.02805347442627,
      "epoch": 5.136877428470505,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00026836255357138444,
      "loss": 5.2282,
      "mean_token_accuracy": 0.22123686969280243,
      "num_tokens": 108443957.0,
      "step": 58170
    },
    {
      "entropy": 6.018652248382568,
      "epoch": 5.1373189685623455,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002683310758358513,
      "loss": 5.1967,
      "mean_token_accuracy": 0.22458928376436232,
      "num_tokens": 108452460.0,
      "step": 58175
    },
    {
      "entropy": 6.073056554794311,
      "epoch": 5.1377605086541855,
      "grad_norm": 1.421875,
      "learning_rate": 0.00026829959823095864,
      "loss": 5.2467,
      "mean_token_accuracy": 0.22413255274295807,
      "num_tokens": 108461490.0,
      "step": 58180
    },
    {
      "entropy": 6.0446981430053714,
      "epoch": 5.1382020487460265,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000268268120757323,
      "loss": 5.1463,
      "mean_token_accuracy": 0.22650267332792282,
      "num_tokens": 108470619.0,
      "step": 58185
    },
    {
      "entropy": 6.079906415939331,
      "epoch": 5.1386435888378665,
      "grad_norm": 1.375,
      "learning_rate": 0.00026823664341556094,
      "loss": 5.2893,
      "mean_token_accuracy": 0.21154992729425431,
      "num_tokens": 108479657.0,
      "step": 58190
    },
    {
      "entropy": 6.012883710861206,
      "epoch": 5.139085128929707,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002682051662062893,
      "loss": 5.2242,
      "mean_token_accuracy": 0.22448742091655732,
      "num_tokens": 108489037.0,
      "step": 58195
    },
    {
      "entropy": 6.059903144836426,
      "epoch": 5.1395266690215475,
      "grad_norm": 1.375,
      "learning_rate": 0.0002681736891301245,
      "loss": 5.3017,
      "mean_token_accuracy": 0.21988268196582794,
      "num_tokens": 108497616.0,
      "step": 58200
    },
    {
      "entropy": 6.041230344772339,
      "epoch": 5.139968209113388,
      "grad_norm": 1.421875,
      "learning_rate": 0.00026814221218768335,
      "loss": 5.2686,
      "mean_token_accuracy": 0.2250336542725563,
      "num_tokens": 108506061.0,
      "step": 58205
    },
    {
      "entropy": 6.076915264129639,
      "epoch": 5.140409749205228,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026811073537958227,
      "loss": 5.2921,
      "mean_token_accuracy": 0.21910804659128189,
      "num_tokens": 108516460.0,
      "step": 58210
    },
    {
      "entropy": 6.023511266708374,
      "epoch": 5.140851289297069,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000268079258706438,
      "loss": 5.1773,
      "mean_token_accuracy": 0.23041654229164124,
      "num_tokens": 108526589.0,
      "step": 58215
    },
    {
      "entropy": 6.04552640914917,
      "epoch": 5.141292829388909,
      "grad_norm": 1.375,
      "learning_rate": 0.0002680477821688672,
      "loss": 5.2571,
      "mean_token_accuracy": 0.2230142682790756,
      "num_tokens": 108536023.0,
      "step": 58220
    },
    {
      "entropy": 5.9643806457519535,
      "epoch": 5.141734369480749,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00026801630576748635,
      "loss": 5.1431,
      "mean_token_accuracy": 0.23615486025810242,
      "num_tokens": 108545782.0,
      "step": 58225
    },
    {
      "entropy": 5.991047620773315,
      "epoch": 5.14217590957259,
      "grad_norm": 1.46875,
      "learning_rate": 0.000267984829502912,
      "loss": 5.1598,
      "mean_token_accuracy": 0.23027846664190293,
      "num_tokens": 108554013.0,
      "step": 58230
    },
    {
      "entropy": 6.018269634246826,
      "epoch": 5.14261744966443,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002679533533757609,
      "loss": 5.2074,
      "mean_token_accuracy": 0.22292208522558213,
      "num_tokens": 108563576.0,
      "step": 58235
    },
    {
      "entropy": 5.948701095581055,
      "epoch": 5.14305898975627,
      "grad_norm": 1.5,
      "learning_rate": 0.0002679218773866497,
      "loss": 5.1778,
      "mean_token_accuracy": 0.23219663202762603,
      "num_tokens": 108573317.0,
      "step": 58240
    },
    {
      "entropy": 6.034775352478027,
      "epoch": 5.14350052984811,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002678904015361948,
      "loss": 5.1492,
      "mean_token_accuracy": 0.22364792078733445,
      "num_tokens": 108582609.0,
      "step": 58245
    },
    {
      "entropy": 6.001080894470215,
      "epoch": 5.143942069939951,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002678589258250129,
      "loss": 5.1738,
      "mean_token_accuracy": 0.22519802302122116,
      "num_tokens": 108592131.0,
      "step": 58250
    },
    {
      "entropy": 6.070180177688599,
      "epoch": 5.144383610031791,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026782745025372053,
      "loss": 5.2234,
      "mean_token_accuracy": 0.22362836748361586,
      "num_tokens": 108601580.0,
      "step": 58255
    },
    {
      "entropy": 6.049284744262695,
      "epoch": 5.144825150123631,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026779597482293443,
      "loss": 5.1932,
      "mean_token_accuracy": 0.22538652420043945,
      "num_tokens": 108610672.0,
      "step": 58260
    },
    {
      "entropy": 5.998813438415527,
      "epoch": 5.145266690215472,
      "grad_norm": 1.484375,
      "learning_rate": 0.00026776449953327106,
      "loss": 5.1798,
      "mean_token_accuracy": 0.22993281483650208,
      "num_tokens": 108619226.0,
      "step": 58265
    },
    {
      "entropy": 6.048073053359985,
      "epoch": 5.145708230307312,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002677330243853471,
      "loss": 5.2409,
      "mean_token_accuracy": 0.22202516794204713,
      "num_tokens": 108627776.0,
      "step": 58270
    },
    {
      "entropy": 6.067367601394653,
      "epoch": 5.146149770399152,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002677015493797789,
      "loss": 5.203,
      "mean_token_accuracy": 0.2310940220952034,
      "num_tokens": 108637086.0,
      "step": 58275
    },
    {
      "entropy": 6.0189026355743405,
      "epoch": 5.146591310490993,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00026767007451718324,
      "loss": 5.1559,
      "mean_token_accuracy": 0.23043530583381652,
      "num_tokens": 108645922.0,
      "step": 58280
    },
    {
      "entropy": 6.078170585632324,
      "epoch": 5.147032850582833,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026763859979817673,
      "loss": 5.2779,
      "mean_token_accuracy": 0.2149080753326416,
      "num_tokens": 108655144.0,
      "step": 58285
    },
    {
      "entropy": 6.053775024414063,
      "epoch": 5.147474390674673,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026760712522337583,
      "loss": 5.2397,
      "mean_token_accuracy": 0.22339043021202087,
      "num_tokens": 108664500.0,
      "step": 58290
    },
    {
      "entropy": 6.01724853515625,
      "epoch": 5.147915930766514,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00026757565079339713,
      "loss": 5.2133,
      "mean_token_accuracy": 0.22780964970588685,
      "num_tokens": 108674457.0,
      "step": 58295
    },
    {
      "entropy": 6.041658163070679,
      "epoch": 5.148357470858354,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00026754417650885716,
      "loss": 5.1754,
      "mean_token_accuracy": 0.23195933997631074,
      "num_tokens": 108683543.0,
      "step": 58300
    },
    {
      "entropy": 6.093363666534424,
      "epoch": 5.148799010950194,
      "grad_norm": 1.421875,
      "learning_rate": 0.00026751270237037255,
      "loss": 5.2277,
      "mean_token_accuracy": 0.2256547823548317,
      "num_tokens": 108692305.0,
      "step": 58305
    },
    {
      "entropy": 6.095085954666137,
      "epoch": 5.149240551042035,
      "grad_norm": 1.375,
      "learning_rate": 0.0002674812283785599,
      "loss": 5.3535,
      "mean_token_accuracy": 0.2207064837217331,
      "num_tokens": 108703046.0,
      "step": 58310
    },
    {
      "entropy": 5.9852033138275145,
      "epoch": 5.149682091133875,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002674497545340357,
      "loss": 5.1395,
      "mean_token_accuracy": 0.23279359340667724,
      "num_tokens": 108712426.0,
      "step": 58315
    },
    {
      "entropy": 5.928853321075439,
      "epoch": 5.150123631225715,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002674182808374164,
      "loss": 5.1508,
      "mean_token_accuracy": 0.2311267375946045,
      "num_tokens": 108721021.0,
      "step": 58320
    },
    {
      "entropy": 6.029537010192871,
      "epoch": 5.150565171317556,
      "grad_norm": 1.5,
      "learning_rate": 0.0002673868072893189,
      "loss": 5.1814,
      "mean_token_accuracy": 0.2234216570854187,
      "num_tokens": 108730557.0,
      "step": 58325
    },
    {
      "entropy": 6.102909851074219,
      "epoch": 5.151006711409396,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002673553338903593,
      "loss": 5.2757,
      "mean_token_accuracy": 0.2231506735086441,
      "num_tokens": 108740946.0,
      "step": 58330
    },
    {
      "entropy": 5.94796085357666,
      "epoch": 5.151448251501236,
      "grad_norm": 1.375,
      "learning_rate": 0.0002673238606411545,
      "loss": 5.0873,
      "mean_token_accuracy": 0.23854904919862746,
      "num_tokens": 108750215.0,
      "step": 58335
    },
    {
      "entropy": 6.100114202499389,
      "epoch": 5.151889791593077,
      "grad_norm": 1.4375,
      "learning_rate": 0.00026729238754232075,
      "loss": 5.2184,
      "mean_token_accuracy": 0.22840608209371566,
      "num_tokens": 108759768.0,
      "step": 58340
    },
    {
      "entropy": 6.092281532287598,
      "epoch": 5.152331331684917,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002672609145944749,
      "loss": 5.1883,
      "mean_token_accuracy": 0.21641179770231248,
      "num_tokens": 108769300.0,
      "step": 58345
    },
    {
      "entropy": 6.094103193283081,
      "epoch": 5.152772871776757,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026722944179823326,
      "loss": 5.305,
      "mean_token_accuracy": 0.21631038337945938,
      "num_tokens": 108779471.0,
      "step": 58350
    },
    {
      "entropy": 6.012993192672729,
      "epoch": 5.153214411868598,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026719796915421245,
      "loss": 5.1954,
      "mean_token_accuracy": 0.2296552687883377,
      "num_tokens": 108789581.0,
      "step": 58355
    },
    {
      "entropy": 6.064988803863526,
      "epoch": 5.153655951960438,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000267166496663029,
      "loss": 5.1979,
      "mean_token_accuracy": 0.2228770971298218,
      "num_tokens": 108798495.0,
      "step": 58360
    },
    {
      "entropy": 6.079225349426269,
      "epoch": 5.154097492052278,
      "grad_norm": 1.390625,
      "learning_rate": 0.00026713502432529937,
      "loss": 5.2621,
      "mean_token_accuracy": 0.21572437584400178,
      "num_tokens": 108807799.0,
      "step": 58365
    },
    {
      "entropy": 6.005072593688965,
      "epoch": 5.154539032144118,
      "grad_norm": 1.515625,
      "learning_rate": 0.00026710355214164017,
      "loss": 5.1384,
      "mean_token_accuracy": 0.23010132610797882,
      "num_tokens": 108816779.0,
      "step": 58370
    },
    {
      "entropy": 6.040118312835693,
      "epoch": 5.154980572235959,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026707208011266794,
      "loss": 5.1854,
      "mean_token_accuracy": 0.22916667759418488,
      "num_tokens": 108826068.0,
      "step": 58375
    },
    {
      "entropy": 6.062417364120483,
      "epoch": 5.155422112327799,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002670406082389991,
      "loss": 5.2108,
      "mean_token_accuracy": 0.22404149025678635,
      "num_tokens": 108835413.0,
      "step": 58380
    },
    {
      "entropy": 6.017290163040161,
      "epoch": 5.155863652419639,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002670091365212502,
      "loss": 5.1881,
      "mean_token_accuracy": 0.22699471861124038,
      "num_tokens": 108845233.0,
      "step": 58385
    },
    {
      "entropy": 6.0314476013183596,
      "epoch": 5.15630519251148,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002669776649600378,
      "loss": 5.1817,
      "mean_token_accuracy": 0.2263755276799202,
      "num_tokens": 108855742.0,
      "step": 58390
    },
    {
      "entropy": 6.016028928756714,
      "epoch": 5.15674673260332,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026694619355597843,
      "loss": 5.2342,
      "mean_token_accuracy": 0.2208236813545227,
      "num_tokens": 108865364.0,
      "step": 58395
    },
    {
      "entropy": 6.061778926849366,
      "epoch": 5.15718827269516,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002669147223096885,
      "loss": 5.2687,
      "mean_token_accuracy": 0.21272174417972564,
      "num_tokens": 108875380.0,
      "step": 58400
    },
    {
      "entropy": 5.989071941375732,
      "epoch": 5.157629812787001,
      "grad_norm": 1.5,
      "learning_rate": 0.00026688325122178457,
      "loss": 5.1349,
      "mean_token_accuracy": 0.23163420855998992,
      "num_tokens": 108883919.0,
      "step": 58405
    },
    {
      "entropy": 6.0131457328796385,
      "epoch": 5.158071352878841,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002668517802928831,
      "loss": 5.2038,
      "mean_token_accuracy": 0.2274141177535057,
      "num_tokens": 108894349.0,
      "step": 58410
    },
    {
      "entropy": 6.0556220531463625,
      "epoch": 5.158512892970681,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002668203095236006,
      "loss": 5.2643,
      "mean_token_accuracy": 0.22209099680185318,
      "num_tokens": 108903249.0,
      "step": 58415
    },
    {
      "entropy": 6.09345965385437,
      "epoch": 5.158954433062522,
      "grad_norm": 1.578125,
      "learning_rate": 0.00026678883891455366,
      "loss": 5.2776,
      "mean_token_accuracy": 0.21419717371463776,
      "num_tokens": 108911212.0,
      "step": 58420
    },
    {
      "entropy": 6.0210484027862545,
      "epoch": 5.159395973154362,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00026675736846635867,
      "loss": 5.2248,
      "mean_token_accuracy": 0.23092602789402009,
      "num_tokens": 108920252.0,
      "step": 58425
    },
    {
      "entropy": 6.090162658691407,
      "epoch": 5.159837513246202,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002667258981796321,
      "loss": 5.2454,
      "mean_token_accuracy": 0.21999965459108353,
      "num_tokens": 108929569.0,
      "step": 58430
    },
    {
      "entropy": 6.0632177829742435,
      "epoch": 5.160279053338043,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026669442805499053,
      "loss": 5.1811,
      "mean_token_accuracy": 0.2260158032178879,
      "num_tokens": 108938497.0,
      "step": 58435
    },
    {
      "entropy": 6.008658218383789,
      "epoch": 5.160720593429883,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002666629580930504,
      "loss": 5.2322,
      "mean_token_accuracy": 0.22981365472078324,
      "num_tokens": 108948069.0,
      "step": 58440
    },
    {
      "entropy": 6.072429466247558,
      "epoch": 5.1611621335217235,
      "grad_norm": 1.375,
      "learning_rate": 0.0002666314882944282,
      "loss": 5.2641,
      "mean_token_accuracy": 0.2210446536540985,
      "num_tokens": 108957285.0,
      "step": 58445
    },
    {
      "entropy": 6.014968633651733,
      "epoch": 5.161603673613564,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002666000186597403,
      "loss": 5.182,
      "mean_token_accuracy": 0.22941473573446275,
      "num_tokens": 108966180.0,
      "step": 58450
    },
    {
      "entropy": 6.047918224334717,
      "epoch": 5.1620452137054045,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00026656854918960326,
      "loss": 5.1579,
      "mean_token_accuracy": 0.23465794175863267,
      "num_tokens": 108974643.0,
      "step": 58455
    },
    {
      "entropy": 6.082137107849121,
      "epoch": 5.1624867537972445,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026653707988463364,
      "loss": 5.3099,
      "mean_token_accuracy": 0.2191835895180702,
      "num_tokens": 108984899.0,
      "step": 58460
    },
    {
      "entropy": 6.024218273162842,
      "epoch": 5.1629282938890855,
      "grad_norm": 1.4375,
      "learning_rate": 0.00026650561074544776,
      "loss": 5.1654,
      "mean_token_accuracy": 0.22962578684091567,
      "num_tokens": 108993392.0,
      "step": 58465
    },
    {
      "entropy": 6.000665140151978,
      "epoch": 5.1633698339809255,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002664741417726622,
      "loss": 5.263,
      "mean_token_accuracy": 0.23020392060279846,
      "num_tokens": 109003399.0,
      "step": 58470
    },
    {
      "entropy": 6.049236488342285,
      "epoch": 5.163811374072766,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002664426729668932,
      "loss": 5.1931,
      "mean_token_accuracy": 0.22697603404521943,
      "num_tokens": 109012766.0,
      "step": 58475
    },
    {
      "entropy": 6.072697257995605,
      "epoch": 5.1642529141646065,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002664112043287575,
      "loss": 5.2361,
      "mean_token_accuracy": 0.2233826607465744,
      "num_tokens": 109022454.0,
      "step": 58480
    },
    {
      "entropy": 6.064543056488037,
      "epoch": 5.164694454256447,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026637973585887136,
      "loss": 5.1944,
      "mean_token_accuracy": 0.226271615922451,
      "num_tokens": 109031607.0,
      "step": 58485
    },
    {
      "entropy": 6.1480590343475345,
      "epoch": 5.165135994348287,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002663482675578513,
      "loss": 5.2902,
      "mean_token_accuracy": 0.2210153117775917,
      "num_tokens": 109040599.0,
      "step": 58490
    },
    {
      "entropy": 6.062751340866089,
      "epoch": 5.165577534440128,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002663167994263138,
      "loss": 5.2187,
      "mean_token_accuracy": 0.2267448365688324,
      "num_tokens": 109050101.0,
      "step": 58495
    },
    {
      "entropy": 6.089010000228882,
      "epoch": 5.166019074531968,
      "grad_norm": 1.421875,
      "learning_rate": 0.00026628533146487514,
      "loss": 5.2543,
      "mean_token_accuracy": 0.2352498933672905,
      "num_tokens": 109059625.0,
      "step": 58500
    },
    {
      "entropy": 6.08569712638855,
      "epoch": 5.166460614623808,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000266253863674152,
      "loss": 5.2542,
      "mean_token_accuracy": 0.2161970853805542,
      "num_tokens": 109069093.0,
      "step": 58505
    },
    {
      "entropy": 5.999886846542358,
      "epoch": 5.166902154715649,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002662223960547607,
      "loss": 5.2089,
      "mean_token_accuracy": 0.22819710820913314,
      "num_tokens": 109078389.0,
      "step": 58510
    },
    {
      "entropy": 6.023866891860962,
      "epoch": 5.167343694807489,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00026619092860731753,
      "loss": 5.1963,
      "mean_token_accuracy": 0.22373077422380447,
      "num_tokens": 109087642.0,
      "step": 58515
    },
    {
      "entropy": 6.084073734283447,
      "epoch": 5.167785234899329,
      "grad_norm": 1.390625,
      "learning_rate": 0.00026615946133243923,
      "loss": 5.2653,
      "mean_token_accuracy": 0.22363041788339616,
      "num_tokens": 109096285.0,
      "step": 58520
    },
    {
      "entropy": 6.025940227508545,
      "epoch": 5.168226774991169,
      "grad_norm": 1.28125,
      "learning_rate": 0.000266127994230742,
      "loss": 5.1803,
      "mean_token_accuracy": 0.2313343793153763,
      "num_tokens": 109106513.0,
      "step": 58525
    },
    {
      "entropy": 6.075152444839477,
      "epoch": 5.16866831508301,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00026609652730284235,
      "loss": 5.2375,
      "mean_token_accuracy": 0.22422468960285186,
      "num_tokens": 109116830.0,
      "step": 58530
    },
    {
      "entropy": 6.019080018997192,
      "epoch": 5.16910985517485,
      "grad_norm": 1.53125,
      "learning_rate": 0.00026606506054935666,
      "loss": 5.2158,
      "mean_token_accuracy": 0.2215456858277321,
      "num_tokens": 109124952.0,
      "step": 58535
    },
    {
      "entropy": 6.051059055328369,
      "epoch": 5.16955139526669,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002660335939709012,
      "loss": 5.2059,
      "mean_token_accuracy": 0.22335304915905,
      "num_tokens": 109133784.0,
      "step": 58540
    },
    {
      "entropy": 6.063338041305542,
      "epoch": 5.169992935358531,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002660021275680927,
      "loss": 5.2459,
      "mean_token_accuracy": 0.22314477115869522,
      "num_tokens": 109142681.0,
      "step": 58545
    },
    {
      "entropy": 5.96950135231018,
      "epoch": 5.170434475450371,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026597066134154737,
      "loss": 5.2009,
      "mean_token_accuracy": 0.23326812088489532,
      "num_tokens": 109153170.0,
      "step": 58550
    },
    {
      "entropy": 6.01894211769104,
      "epoch": 5.170876015542211,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002659391952918816,
      "loss": 5.1064,
      "mean_token_accuracy": 0.23499225825071335,
      "num_tokens": 109161207.0,
      "step": 58555
    },
    {
      "entropy": 6.062561702728272,
      "epoch": 5.171317555634052,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002659077294197119,
      "loss": 5.2296,
      "mean_token_accuracy": 0.22227906286716462,
      "num_tokens": 109170132.0,
      "step": 58560
    },
    {
      "entropy": 5.986817359924316,
      "epoch": 5.171759095725892,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002658762637256545,
      "loss": 5.1702,
      "mean_token_accuracy": 0.2361446276307106,
      "num_tokens": 109180151.0,
      "step": 58565
    },
    {
      "entropy": 6.108633184432984,
      "epoch": 5.172200635817732,
      "grad_norm": 1.421875,
      "learning_rate": 0.000265844798210326,
      "loss": 5.2841,
      "mean_token_accuracy": 0.21651095449924468,
      "num_tokens": 109189775.0,
      "step": 58570
    },
    {
      "entropy": 6.142046070098877,
      "epoch": 5.172642175909573,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002658133328743427,
      "loss": 5.3444,
      "mean_token_accuracy": 0.20777871012687682,
      "num_tokens": 109200320.0,
      "step": 58575
    },
    {
      "entropy": 5.994301414489746,
      "epoch": 5.173083716001413,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002657818677183209,
      "loss": 5.1085,
      "mean_token_accuracy": 0.23503157049417495,
      "num_tokens": 109210264.0,
      "step": 58580
    },
    {
      "entropy": 6.019202995300293,
      "epoch": 5.173525256093253,
      "grad_norm": 1.375,
      "learning_rate": 0.0002657504027428772,
      "loss": 5.2226,
      "mean_token_accuracy": 0.22181497514247894,
      "num_tokens": 109219839.0,
      "step": 58585
    },
    {
      "entropy": 6.02592191696167,
      "epoch": 5.173966796185094,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002657189379486278,
      "loss": 5.2002,
      "mean_token_accuracy": 0.23268163651227952,
      "num_tokens": 109230357.0,
      "step": 58590
    },
    {
      "entropy": 6.0058714866638185,
      "epoch": 5.174408336276934,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026568747333618917,
      "loss": 5.0956,
      "mean_token_accuracy": 0.2408140257000923,
      "num_tokens": 109239403.0,
      "step": 58595
    },
    {
      "entropy": 6.062946796417236,
      "epoch": 5.174849876368774,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026565600890617757,
      "loss": 5.1933,
      "mean_token_accuracy": 0.22221199423074722,
      "num_tokens": 109248927.0,
      "step": 58600
    },
    {
      "entropy": 6.019367313385009,
      "epoch": 5.175291416460615,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002656245446592094,
      "loss": 5.2496,
      "mean_token_accuracy": 0.22782927006483078,
      "num_tokens": 109258895.0,
      "step": 58605
    },
    {
      "entropy": 6.112438535690307,
      "epoch": 5.175732956552455,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026559308059590116,
      "loss": 5.2913,
      "mean_token_accuracy": 0.21466986685991288,
      "num_tokens": 109268024.0,
      "step": 58610
    },
    {
      "entropy": 6.053680896759033,
      "epoch": 5.176174496644295,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002655616167168691,
      "loss": 5.2043,
      "mean_token_accuracy": 0.23077867180109024,
      "num_tokens": 109277098.0,
      "step": 58615
    },
    {
      "entropy": 6.007803583145142,
      "epoch": 5.176616036736136,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002655301530227295,
      "loss": 5.1654,
      "mean_token_accuracy": 0.23037771880626678,
      "num_tokens": 109286257.0,
      "step": 58620
    },
    {
      "entropy": 6.103433275222779,
      "epoch": 5.177057576827976,
      "grad_norm": 1.421875,
      "learning_rate": 0.00026549868951409895,
      "loss": 5.3248,
      "mean_token_accuracy": 0.21540609896183013,
      "num_tokens": 109295196.0,
      "step": 58625
    },
    {
      "entropy": 6.059934759140015,
      "epoch": 5.177499116919816,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002654672261915936,
      "loss": 5.295,
      "mean_token_accuracy": 0.2192981317639351,
      "num_tokens": 109304149.0,
      "step": 58630
    },
    {
      "entropy": 5.984541749954223,
      "epoch": 5.177940657011657,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002654357630558299,
      "loss": 5.1347,
      "mean_token_accuracy": 0.2291536659002304,
      "num_tokens": 109312783.0,
      "step": 58635
    },
    {
      "entropy": 6.067545938491821,
      "epoch": 5.178382197103497,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026540430010742416,
      "loss": 5.2613,
      "mean_token_accuracy": 0.2194505363702774,
      "num_tokens": 109322106.0,
      "step": 58640
    },
    {
      "entropy": 6.078786611557007,
      "epoch": 5.178823737195337,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026537283734699273,
      "loss": 5.1773,
      "mean_token_accuracy": 0.22459367513656617,
      "num_tokens": 109331368.0,
      "step": 58645
    },
    {
      "entropy": 5.9674194812774655,
      "epoch": 5.179265277287177,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002653413747751519,
      "loss": 5.2154,
      "mean_token_accuracy": 0.22809458374977112,
      "num_tokens": 109340251.0,
      "step": 58650
    },
    {
      "entropy": 5.970209217071533,
      "epoch": 5.179706817379018,
      "grad_norm": 1.421875,
      "learning_rate": 0.00026530991239251807,
      "loss": 5.1835,
      "mean_token_accuracy": 0.2255791276693344,
      "num_tokens": 109349004.0,
      "step": 58655
    },
    {
      "entropy": 6.066261672973633,
      "epoch": 5.180148357470858,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002652784501997075,
      "loss": 5.2152,
      "mean_token_accuracy": 0.2184986338019371,
      "num_tokens": 109359203.0,
      "step": 58660
    },
    {
      "entropy": 6.046020364761352,
      "epoch": 5.180589897562698,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002652469881973367,
      "loss": 5.2077,
      "mean_token_accuracy": 0.2227126970887184,
      "num_tokens": 109368500.0,
      "step": 58665
    },
    {
      "entropy": 5.977790498733521,
      "epoch": 5.181031437654539,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002652155263860217,
      "loss": 5.0924,
      "mean_token_accuracy": 0.2424722984433174,
      "num_tokens": 109377323.0,
      "step": 58670
    },
    {
      "entropy": 6.028044939041138,
      "epoch": 5.181472977746379,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000265184064766379,
      "loss": 5.2461,
      "mean_token_accuracy": 0.22121296375989913,
      "num_tokens": 109385991.0,
      "step": 58675
    },
    {
      "entropy": 6.074767065048218,
      "epoch": 5.181914517838219,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002651526033390249,
      "loss": 5.2973,
      "mean_token_accuracy": 0.22061156779527663,
      "num_tokens": 109396290.0,
      "step": 58680
    },
    {
      "entropy": 6.140890121459961,
      "epoch": 5.18235605793006,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00026512114210457577,
      "loss": 5.3371,
      "mean_token_accuracy": 0.21112188398838044,
      "num_tokens": 109405717.0,
      "step": 58685
    },
    {
      "entropy": 6.074089527130127,
      "epoch": 5.1827975980219,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002650896810636478,
      "loss": 5.2797,
      "mean_token_accuracy": 0.22157852500677108,
      "num_tokens": 109414318.0,
      "step": 58690
    },
    {
      "entropy": 6.017424392700195,
      "epoch": 5.18323913811374,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002650582202168573,
      "loss": 5.1538,
      "mean_token_accuracy": 0.23296236693859101,
      "num_tokens": 109423767.0,
      "step": 58695
    },
    {
      "entropy": 5.976594161987305,
      "epoch": 5.183680678205581,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002650267595648207,
      "loss": 5.2296,
      "mean_token_accuracy": 0.23000459671020507,
      "num_tokens": 109432579.0,
      "step": 58700
    },
    {
      "entropy": 6.089522409439087,
      "epoch": 5.184122218297421,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026499529910815417,
      "loss": 5.3264,
      "mean_token_accuracy": 0.21111815720796584,
      "num_tokens": 109442906.0,
      "step": 58705
    },
    {
      "entropy": 6.108569145202637,
      "epoch": 5.184563758389261,
      "grad_norm": 1.375,
      "learning_rate": 0.00026496383884747407,
      "loss": 5.1819,
      "mean_token_accuracy": 0.2265142619609833,
      "num_tokens": 109452008.0,
      "step": 58710
    },
    {
      "entropy": 6.146604728698731,
      "epoch": 5.185005298481102,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002649323787833965,
      "loss": 5.2882,
      "mean_token_accuracy": 0.218049618601799,
      "num_tokens": 109461327.0,
      "step": 58715
    },
    {
      "entropy": 5.963629198074341,
      "epoch": 5.185446838572942,
      "grad_norm": 1.453125,
      "learning_rate": 0.00026490091891653806,
      "loss": 5.2066,
      "mean_token_accuracy": 0.22918720841407775,
      "num_tokens": 109470850.0,
      "step": 58720
    },
    {
      "entropy": 5.990911388397217,
      "epoch": 5.1858883786647825,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002648694592475149,
      "loss": 5.1638,
      "mean_token_accuracy": 0.2216727316379547,
      "num_tokens": 109479735.0,
      "step": 58725
    },
    {
      "entropy": 6.0538171291351315,
      "epoch": 5.186329918756623,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002648379997769432,
      "loss": 5.3054,
      "mean_token_accuracy": 0.22265393733978273,
      "num_tokens": 109489626.0,
      "step": 58730
    },
    {
      "entropy": 6.047466659545899,
      "epoch": 5.1867714588484635,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00026480654050543926,
      "loss": 5.0971,
      "mean_token_accuracy": 0.2366907224059105,
      "num_tokens": 109498709.0,
      "step": 58735
    },
    {
      "entropy": 6.001376056671143,
      "epoch": 5.1872129989403035,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002647750814336195,
      "loss": 5.1424,
      "mean_token_accuracy": 0.22511292845010758,
      "num_tokens": 109508040.0,
      "step": 58740
    },
    {
      "entropy": 6.066401100158691,
      "epoch": 5.1876545390321445,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00026474362256209997,
      "loss": 5.2236,
      "mean_token_accuracy": 0.22563302218914033,
      "num_tokens": 109517664.0,
      "step": 58745
    },
    {
      "entropy": 5.924116325378418,
      "epoch": 5.1880960791239845,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00026471216389149717,
      "loss": 5.0816,
      "mean_token_accuracy": 0.2416619911789894,
      "num_tokens": 109526320.0,
      "step": 58750
    },
    {
      "entropy": 6.033281660079956,
      "epoch": 5.188537619215825,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002646807054224271,
      "loss": 5.2353,
      "mean_token_accuracy": 0.22127437740564346,
      "num_tokens": 109536196.0,
      "step": 58755
    },
    {
      "entropy": 6.0598907470703125,
      "epoch": 5.1889791593076655,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002646492471555062,
      "loss": 5.2148,
      "mean_token_accuracy": 0.229184390604496,
      "num_tokens": 109545588.0,
      "step": 58760
    },
    {
      "entropy": 6.082006788253784,
      "epoch": 5.189420699399506,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002646177890913507,
      "loss": 5.309,
      "mean_token_accuracy": 0.21730370074510574,
      "num_tokens": 109554559.0,
      "step": 58765
    },
    {
      "entropy": 6.035778665542603,
      "epoch": 5.189862239491346,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002645863312305768,
      "loss": 5.1862,
      "mean_token_accuracy": 0.23652290254831315,
      "num_tokens": 109564195.0,
      "step": 58770
    },
    {
      "entropy": 6.008813428878784,
      "epoch": 5.190303779583187,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002645548735738007,
      "loss": 5.2197,
      "mean_token_accuracy": 0.22778207212686538,
      "num_tokens": 109573321.0,
      "step": 58775
    },
    {
      "entropy": 6.105978345870971,
      "epoch": 5.190745319675027,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002645234161216387,
      "loss": 5.2515,
      "mean_token_accuracy": 0.227383491396904,
      "num_tokens": 109582684.0,
      "step": 58780
    },
    {
      "entropy": 6.101035261154175,
      "epoch": 5.191186859766867,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000264491958874707,
      "loss": 5.2803,
      "mean_token_accuracy": 0.21628399193286896,
      "num_tokens": 109593176.0,
      "step": 58785
    },
    {
      "entropy": 6.1347592830657955,
      "epoch": 5.191628399858708,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002644605018336219,
      "loss": 5.2767,
      "mean_token_accuracy": 0.22098368704319,
      "num_tokens": 109602201.0,
      "step": 58790
    },
    {
      "entropy": 6.079919815063477,
      "epoch": 5.192069939950548,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026442904499899953,
      "loss": 5.2663,
      "mean_token_accuracy": 0.21515865623950958,
      "num_tokens": 109612496.0,
      "step": 58795
    },
    {
      "entropy": 6.087953996658325,
      "epoch": 5.192511480042388,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00026439758837145614,
      "loss": 5.2271,
      "mean_token_accuracy": 0.22555198073387145,
      "num_tokens": 109622282.0,
      "step": 58800
    },
    {
      "entropy": 6.09975037574768,
      "epoch": 5.192953020134228,
      "grad_norm": 1.4375,
      "learning_rate": 0.00026436613195160804,
      "loss": 5.2948,
      "mean_token_accuracy": 0.21699911803007127,
      "num_tokens": 109631751.0,
      "step": 58805
    },
    {
      "entropy": 6.03333854675293,
      "epoch": 5.193394560226069,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002643346757400713,
      "loss": 5.2285,
      "mean_token_accuracy": 0.22164877206087114,
      "num_tokens": 109641068.0,
      "step": 58810
    },
    {
      "entropy": 5.977601385116577,
      "epoch": 5.193836100317909,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026430321973746224,
      "loss": 5.2031,
      "mean_token_accuracy": 0.23137539327144624,
      "num_tokens": 109652041.0,
      "step": 58815
    },
    {
      "entropy": 6.0775144577026365,
      "epoch": 5.194277640409749,
      "grad_norm": 1.375,
      "learning_rate": 0.000264271763944397,
      "loss": 5.2915,
      "mean_token_accuracy": 0.2216573491692543,
      "num_tokens": 109662702.0,
      "step": 58820
    },
    {
      "entropy": 6.11164813041687,
      "epoch": 5.19471918050159,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002642403083614918,
      "loss": 5.2056,
      "mean_token_accuracy": 0.21897186040878297,
      "num_tokens": 109671074.0,
      "step": 58825
    },
    {
      "entropy": 6.050094413757324,
      "epoch": 5.19516072059343,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002642088529893629,
      "loss": 5.2664,
      "mean_token_accuracy": 0.21699047237634658,
      "num_tokens": 109680963.0,
      "step": 58830
    },
    {
      "entropy": 6.0111805438995365,
      "epoch": 5.19560226068527,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026417739782862637,
      "loss": 5.1903,
      "mean_token_accuracy": 0.22845541089773178,
      "num_tokens": 109689415.0,
      "step": 58835
    },
    {
      "entropy": 6.005054712295532,
      "epoch": 5.196043800777111,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002641459428798985,
      "loss": 5.2011,
      "mean_token_accuracy": 0.2304435357451439,
      "num_tokens": 109697720.0,
      "step": 58840
    },
    {
      "entropy": 6.08939037322998,
      "epoch": 5.196485340868951,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002641144881437954,
      "loss": 5.269,
      "mean_token_accuracy": 0.22272133231163024,
      "num_tokens": 109706938.0,
      "step": 58845
    },
    {
      "entropy": 6.016390323638916,
      "epoch": 5.196926880960791,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002640830336209332,
      "loss": 5.1559,
      "mean_token_accuracy": 0.23038191348314285,
      "num_tokens": 109715885.0,
      "step": 58850
    },
    {
      "entropy": 5.999240827560425,
      "epoch": 5.197368421052632,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002640515793119283,
      "loss": 5.149,
      "mean_token_accuracy": 0.2287047564983368,
      "num_tokens": 109725623.0,
      "step": 58855
    },
    {
      "entropy": 5.998471212387085,
      "epoch": 5.197809961144472,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002640201252173968,
      "loss": 5.2013,
      "mean_token_accuracy": 0.23082427233457564,
      "num_tokens": 109736685.0,
      "step": 58860
    },
    {
      "entropy": 5.9917566776275635,
      "epoch": 5.198251501236312,
      "grad_norm": 1.390625,
      "learning_rate": 0.00026398867133795457,
      "loss": 5.1774,
      "mean_token_accuracy": 0.22801507413387298,
      "num_tokens": 109745498.0,
      "step": 58865
    },
    {
      "entropy": 6.069150781631469,
      "epoch": 5.198693041328153,
      "grad_norm": 1.546875,
      "learning_rate": 0.00026395721767421816,
      "loss": 5.1676,
      "mean_token_accuracy": 0.23381781429052353,
      "num_tokens": 109753980.0,
      "step": 58870
    },
    {
      "entropy": 5.983275079727173,
      "epoch": 5.199134581419993,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002639257642268035,
      "loss": 5.2441,
      "mean_token_accuracy": 0.22432216703891755,
      "num_tokens": 109763536.0,
      "step": 58875
    },
    {
      "entropy": 5.982507753372192,
      "epoch": 5.199576121511833,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026389431099632687,
      "loss": 5.1357,
      "mean_token_accuracy": 0.22102082371711732,
      "num_tokens": 109771951.0,
      "step": 58880
    },
    {
      "entropy": 6.027839517593383,
      "epoch": 5.200017661603674,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026386285798340445,
      "loss": 5.2771,
      "mean_token_accuracy": 0.2175005167722702,
      "num_tokens": 109781696.0,
      "step": 58885
    },
    {
      "entropy": 6.139021921157837,
      "epoch": 5.200459201695514,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002638314051886521,
      "loss": 5.3125,
      "mean_token_accuracy": 0.20596616566181183,
      "num_tokens": 109791700.0,
      "step": 58890
    },
    {
      "entropy": 6.087416172027588,
      "epoch": 5.200900741787354,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002637999526126864,
      "loss": 5.2691,
      "mean_token_accuracy": 0.2195592626929283,
      "num_tokens": 109800919.0,
      "step": 58895
    },
    {
      "entropy": 6.085106277465821,
      "epoch": 5.201342281879195,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026376850025612315,
      "loss": 5.2486,
      "mean_token_accuracy": 0.22446715235710143,
      "num_tokens": 109810799.0,
      "step": 58900
    },
    {
      "entropy": 6.031948041915894,
      "epoch": 5.201783821971035,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002637370481195786,
      "loss": 5.201,
      "mean_token_accuracy": 0.23228852748870848,
      "num_tokens": 109819859.0,
      "step": 58905
    },
    {
      "entropy": 6.007193470001221,
      "epoch": 5.202225362062875,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002637055962036688,
      "loss": 5.3167,
      "mean_token_accuracy": 0.21494730859994887,
      "num_tokens": 109829707.0,
      "step": 58910
    },
    {
      "entropy": 5.99403657913208,
      "epoch": 5.202666902154716,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026367414450901005,
      "loss": 5.1919,
      "mean_token_accuracy": 0.2300949811935425,
      "num_tokens": 109838444.0,
      "step": 58915
    },
    {
      "entropy": 6.048970556259155,
      "epoch": 5.203108442246556,
      "grad_norm": 1.390625,
      "learning_rate": 0.00026364269303621833,
      "loss": 5.2886,
      "mean_token_accuracy": 0.21874046474695205,
      "num_tokens": 109847575.0,
      "step": 58920
    },
    {
      "entropy": 6.070389080047607,
      "epoch": 5.203549982338396,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00026361124178590984,
      "loss": 5.31,
      "mean_token_accuracy": 0.21602606624364853,
      "num_tokens": 109856377.0,
      "step": 58925
    },
    {
      "entropy": 6.033780908584594,
      "epoch": 5.203991522430236,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00026357979075870055,
      "loss": 5.156,
      "mean_token_accuracy": 0.23023621439933778,
      "num_tokens": 109865004.0,
      "step": 58930
    },
    {
      "entropy": 6.111438512802124,
      "epoch": 5.204433062522077,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00026354833995520673,
      "loss": 5.2568,
      "mean_token_accuracy": 0.2180517539381981,
      "num_tokens": 109873892.0,
      "step": 58935
    },
    {
      "entropy": 6.044076824188233,
      "epoch": 5.204874602613917,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00026351688937604437,
      "loss": 5.2164,
      "mean_token_accuracy": 0.2250634402036667,
      "num_tokens": 109883413.0,
      "step": 58940
    },
    {
      "entropy": 5.976048994064331,
      "epoch": 5.205316142705757,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002634854390218297,
      "loss": 5.1934,
      "mean_token_accuracy": 0.2291681408882141,
      "num_tokens": 109893062.0,
      "step": 58945
    },
    {
      "entropy": 6.053479480743408,
      "epoch": 5.205757682797598,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002634539888931787,
      "loss": 5.2403,
      "mean_token_accuracy": 0.22112203240394593,
      "num_tokens": 109901782.0,
      "step": 58950
    },
    {
      "entropy": 6.0974406719207765,
      "epoch": 5.206199222889438,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002634225389907074,
      "loss": 5.3036,
      "mean_token_accuracy": 0.2206927001476288,
      "num_tokens": 109909981.0,
      "step": 58955
    },
    {
      "entropy": 6.101591396331787,
      "epoch": 5.206640762981278,
      "grad_norm": 1.375,
      "learning_rate": 0.000263391089315032,
      "loss": 5.2161,
      "mean_token_accuracy": 0.2185104340314865,
      "num_tokens": 109918625.0,
      "step": 58960
    },
    {
      "entropy": 6.018256998062133,
      "epoch": 5.207082303073119,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002633596398667687,
      "loss": 5.1826,
      "mean_token_accuracy": 0.2298142671585083,
      "num_tokens": 109927224.0,
      "step": 58965
    },
    {
      "entropy": 6.099199962615967,
      "epoch": 5.207523843164959,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00026332819064653335,
      "loss": 5.2545,
      "mean_token_accuracy": 0.22377737015485763,
      "num_tokens": 109936554.0,
      "step": 58970
    },
    {
      "entropy": 6.078859901428222,
      "epoch": 5.207965383256799,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026329674165494204,
      "loss": 5.2714,
      "mean_token_accuracy": 0.21445172727108003,
      "num_tokens": 109945925.0,
      "step": 58975
    },
    {
      "entropy": 5.975079679489136,
      "epoch": 5.20840692334864,
      "grad_norm": 1.453125,
      "learning_rate": 0.000263265292892611,
      "loss": 5.1466,
      "mean_token_accuracy": 0.23040404617786409,
      "num_tokens": 109955144.0,
      "step": 58980
    },
    {
      "entropy": 6.048996257781982,
      "epoch": 5.20884846344048,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002632338443601562,
      "loss": 5.1914,
      "mean_token_accuracy": 0.21971047818660736,
      "num_tokens": 109964146.0,
      "step": 58985
    },
    {
      "entropy": 5.9671470642089846,
      "epoch": 5.20929000353232,
      "grad_norm": 1.46875,
      "learning_rate": 0.00026320239605819365,
      "loss": 5.1288,
      "mean_token_accuracy": 0.2342725545167923,
      "num_tokens": 109973092.0,
      "step": 58990
    },
    {
      "entropy": 5.9897441387176515,
      "epoch": 5.209731543624161,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002631709479873394,
      "loss": 5.1481,
      "mean_token_accuracy": 0.23373684287071228,
      "num_tokens": 109981844.0,
      "step": 58995
    },
    {
      "entropy": 5.931584787368775,
      "epoch": 5.210173083716001,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026313950014820965,
      "loss": 5.1365,
      "mean_token_accuracy": 0.2365485265851021,
      "num_tokens": 109991475.0,
      "step": 59000
    },
    {
      "entropy": 6.120680522918701,
      "epoch": 5.2106146238078415,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002631080525414203,
      "loss": 5.3373,
      "mean_token_accuracy": 0.2185873955488205,
      "num_tokens": 110000580.0,
      "step": 59005
    },
    {
      "entropy": 6.025818538665772,
      "epoch": 5.211056163899682,
      "grad_norm": 1.375,
      "learning_rate": 0.0002630766051675875,
      "loss": 5.1339,
      "mean_token_accuracy": 0.22955281436443328,
      "num_tokens": 110009492.0,
      "step": 59010
    },
    {
      "entropy": 6.089053106307984,
      "epoch": 5.2114977039915225,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026304515802732717,
      "loss": 5.2545,
      "mean_token_accuracy": 0.22186901718378066,
      "num_tokens": 110019172.0,
      "step": 59015
    },
    {
      "entropy": 6.012477779388428,
      "epoch": 5.2119392440833625,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002630137111212553,
      "loss": 5.1778,
      "mean_token_accuracy": 0.22579750120639802,
      "num_tokens": 110028104.0,
      "step": 59020
    },
    {
      "entropy": 6.01051607131958,
      "epoch": 5.2123807841752035,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026298226444998815,
      "loss": 5.1785,
      "mean_token_accuracy": 0.23453581631183623,
      "num_tokens": 110037513.0,
      "step": 59025
    },
    {
      "entropy": 5.998272275924682,
      "epoch": 5.2128223242670435,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002629508180141416,
      "loss": 5.2406,
      "mean_token_accuracy": 0.22600959986448288,
      "num_tokens": 110046993.0,
      "step": 59030
    },
    {
      "entropy": 6.0856410503387455,
      "epoch": 5.213263864358884,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002629193718143316,
      "loss": 5.3057,
      "mean_token_accuracy": 0.2112346649169922,
      "num_tokens": 110057364.0,
      "step": 59035
    },
    {
      "entropy": 6.0858238697052,
      "epoch": 5.2137054044507245,
      "grad_norm": 1.375,
      "learning_rate": 0.0002628879258511742,
      "loss": 5.2112,
      "mean_token_accuracy": 0.22878143638372422,
      "num_tokens": 110067070.0,
      "step": 59040
    },
    {
      "entropy": 6.011166191101074,
      "epoch": 5.2141469445425646,
      "grad_norm": 1.53125,
      "learning_rate": 0.00026285648012528556,
      "loss": 5.0806,
      "mean_token_accuracy": 0.24414756745100022,
      "num_tokens": 110076897.0,
      "step": 59045
    },
    {
      "entropy": 6.115145254135132,
      "epoch": 5.214588484634405,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002628250346372816,
      "loss": 5.3449,
      "mean_token_accuracy": 0.21540362238883973,
      "num_tokens": 110086334.0,
      "step": 59050
    },
    {
      "entropy": 6.054930400848389,
      "epoch": 5.2150300247262455,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002627935893877782,
      "loss": 5.2538,
      "mean_token_accuracy": 0.2267913728952408,
      "num_tokens": 110096006.0,
      "step": 59055
    },
    {
      "entropy": 6.142521142959595,
      "epoch": 5.215471564818086,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002627621443773914,
      "loss": 5.2852,
      "mean_token_accuracy": 0.2186047539114952,
      "num_tokens": 110104750.0,
      "step": 59060
    },
    {
      "entropy": 6.145096349716186,
      "epoch": 5.215913104909926,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002627306996067372,
      "loss": 5.326,
      "mean_token_accuracy": 0.2146193727850914,
      "num_tokens": 110113460.0,
      "step": 59065
    },
    {
      "entropy": 6.068349361419678,
      "epoch": 5.216354645001767,
      "grad_norm": 1.515625,
      "learning_rate": 0.00026269925507643174,
      "loss": 5.2008,
      "mean_token_accuracy": 0.22896523177623748,
      "num_tokens": 110121969.0,
      "step": 59070
    },
    {
      "entropy": 5.988681030273438,
      "epoch": 5.216796185093607,
      "grad_norm": 1.375,
      "learning_rate": 0.00026266781078709085,
      "loss": 5.179,
      "mean_token_accuracy": 0.22919145077466965,
      "num_tokens": 110131264.0,
      "step": 59075
    },
    {
      "entropy": 5.9720337867736815,
      "epoch": 5.217237725185447,
      "grad_norm": 1.578125,
      "learning_rate": 0.00026263636673933054,
      "loss": 5.0963,
      "mean_token_accuracy": 0.24129415303468704,
      "num_tokens": 110139746.0,
      "step": 59080
    },
    {
      "entropy": 6.023876142501831,
      "epoch": 5.217679265277287,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002626049229337667,
      "loss": 5.1371,
      "mean_token_accuracy": 0.23261846452951432,
      "num_tokens": 110148864.0,
      "step": 59085
    },
    {
      "entropy": 6.011609220504761,
      "epoch": 5.218120805369128,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002625734793710154,
      "loss": 5.2066,
      "mean_token_accuracy": 0.22845807522535325,
      "num_tokens": 110158793.0,
      "step": 59090
    },
    {
      "entropy": 6.068686771392822,
      "epoch": 5.218562345460968,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002625420360516927,
      "loss": 5.2591,
      "mean_token_accuracy": 0.22313473075628282,
      "num_tokens": 110168017.0,
      "step": 59095
    },
    {
      "entropy": 6.1215935230255125,
      "epoch": 5.219003885552808,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002625105929764143,
      "loss": 5.3162,
      "mean_token_accuracy": 0.211555777490139,
      "num_tokens": 110177884.0,
      "step": 59100
    },
    {
      "entropy": 6.026191473007202,
      "epoch": 5.219445425644649,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002624791501457963,
      "loss": 5.2185,
      "mean_token_accuracy": 0.22420741617679596,
      "num_tokens": 110186660.0,
      "step": 59105
    },
    {
      "entropy": 6.00239634513855,
      "epoch": 5.219886965736489,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026244770756045463,
      "loss": 5.1312,
      "mean_token_accuracy": 0.23957661241292955,
      "num_tokens": 110196458.0,
      "step": 59110
    },
    {
      "entropy": 6.065647220611572,
      "epoch": 5.220328505828329,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002624162652210053,
      "loss": 5.2198,
      "mean_token_accuracy": 0.23398562967777253,
      "num_tokens": 110205440.0,
      "step": 59115
    },
    {
      "entropy": 6.046412515640259,
      "epoch": 5.22077004592017,
      "grad_norm": 1.390625,
      "learning_rate": 0.00026238482312806403,
      "loss": 5.2954,
      "mean_token_accuracy": 0.21611884236335754,
      "num_tokens": 110214180.0,
      "step": 59120
    },
    {
      "entropy": 6.026818895339966,
      "epoch": 5.22121158601201,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000262353381282247,
      "loss": 5.2226,
      "mean_token_accuracy": 0.2260102689266205,
      "num_tokens": 110225080.0,
      "step": 59125
    },
    {
      "entropy": 6.009346628189087,
      "epoch": 5.22165312610385,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00026232193968417003,
      "loss": 5.1468,
      "mean_token_accuracy": 0.2320275828242302,
      "num_tokens": 110233812.0,
      "step": 59130
    },
    {
      "entropy": 6.074969673156739,
      "epoch": 5.222094666195691,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000262290498334449,
      "loss": 5.1633,
      "mean_token_accuracy": 0.23498228788375855,
      "num_tokens": 110242487.0,
      "step": 59135
    },
    {
      "entropy": 6.062996959686279,
      "epoch": 5.222536206287531,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002622590572336999,
      "loss": 5.2089,
      "mean_token_accuracy": 0.23124281018972398,
      "num_tokens": 110251093.0,
      "step": 59140
    },
    {
      "entropy": 6.059306859970093,
      "epoch": 5.222977746379371,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026222761638253873,
      "loss": 5.2764,
      "mean_token_accuracy": 0.2225502237677574,
      "num_tokens": 110260259.0,
      "step": 59145
    },
    {
      "entropy": 6.007097005844116,
      "epoch": 5.223419286471212,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002621961757815811,
      "loss": 5.1456,
      "mean_token_accuracy": 0.23255994766950608,
      "num_tokens": 110269585.0,
      "step": 59150
    },
    {
      "entropy": 6.029016733169556,
      "epoch": 5.223860826563052,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002621647354314432,
      "loss": 5.2043,
      "mean_token_accuracy": 0.2197168990969658,
      "num_tokens": 110279776.0,
      "step": 59155
    },
    {
      "entropy": 6.038393974304199,
      "epoch": 5.224302366654892,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002621332953327409,
      "loss": 5.2439,
      "mean_token_accuracy": 0.22783226519823074,
      "num_tokens": 110289931.0,
      "step": 59160
    },
    {
      "entropy": 6.040189170837403,
      "epoch": 5.224743906746733,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00026210185548609,
      "loss": 5.2754,
      "mean_token_accuracy": 0.21632067412137984,
      "num_tokens": 110299463.0,
      "step": 59165
    },
    {
      "entropy": 6.0198441505432125,
      "epoch": 5.225185446838573,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002620704158921063,
      "loss": 5.1888,
      "mean_token_accuracy": 0.22727908343076705,
      "num_tokens": 110308181.0,
      "step": 59170
    },
    {
      "entropy": 6.073774766921997,
      "epoch": 5.225626986930413,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002620389765514059,
      "loss": 5.1996,
      "mean_token_accuracy": 0.2236030161380768,
      "num_tokens": 110317586.0,
      "step": 59175
    },
    {
      "entropy": 6.140529632568359,
      "epoch": 5.226068527022254,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002620075374646046,
      "loss": 5.3424,
      "mean_token_accuracy": 0.21253828555345536,
      "num_tokens": 110327110.0,
      "step": 59180
    },
    {
      "entropy": 6.119946718215942,
      "epoch": 5.226510067114094,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002619760986323182,
      "loss": 5.2657,
      "mean_token_accuracy": 0.2126380905508995,
      "num_tokens": 110336362.0,
      "step": 59185
    },
    {
      "entropy": 6.075910377502441,
      "epoch": 5.226951607205934,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026194466005516266,
      "loss": 5.2824,
      "mean_token_accuracy": 0.22056616097688675,
      "num_tokens": 110345835.0,
      "step": 59190
    },
    {
      "entropy": 6.076473760604858,
      "epoch": 5.227393147297775,
      "grad_norm": 1.5,
      "learning_rate": 0.0002619132217337537,
      "loss": 5.2018,
      "mean_token_accuracy": 0.22005849927663804,
      "num_tokens": 110354389.0,
      "step": 59195
    },
    {
      "entropy": 6.035341501235962,
      "epoch": 5.227834687389615,
      "grad_norm": 1.65625,
      "learning_rate": 0.00026188178366870734,
      "loss": 5.1693,
      "mean_token_accuracy": 0.22839803397655487,
      "num_tokens": 110363104.0,
      "step": 59200
    },
    {
      "entropy": 6.020644235610962,
      "epoch": 5.228276227481455,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026185034586063944,
      "loss": 5.1861,
      "mean_token_accuracy": 0.2372784897685051,
      "num_tokens": 110372952.0,
      "step": 59205
    },
    {
      "entropy": 5.990231847763061,
      "epoch": 5.228717767573295,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002618189083101658,
      "loss": 5.1941,
      "mean_token_accuracy": 0.22684166133403777,
      "num_tokens": 110382105.0,
      "step": 59210
    },
    {
      "entropy": 6.062443637847901,
      "epoch": 5.229159307665136,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026178747101790207,
      "loss": 5.2161,
      "mean_token_accuracy": 0.2271784648299217,
      "num_tokens": 110391812.0,
      "step": 59215
    },
    {
      "entropy": 6.058978939056397,
      "epoch": 5.229600847756976,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026175603398446447,
      "loss": 5.2023,
      "mean_token_accuracy": 0.2261226400732994,
      "num_tokens": 110400483.0,
      "step": 59220
    },
    {
      "entropy": 6.047107744216919,
      "epoch": 5.230042387848816,
      "grad_norm": 1.421875,
      "learning_rate": 0.00026172459721046854,
      "loss": 5.2277,
      "mean_token_accuracy": 0.22310669869184493,
      "num_tokens": 110408947.0,
      "step": 59225
    },
    {
      "entropy": 6.044586801528931,
      "epoch": 5.230483927940657,
      "grad_norm": 1.453125,
      "learning_rate": 0.00026169316069653026,
      "loss": 5.2267,
      "mean_token_accuracy": 0.2263523519039154,
      "num_tokens": 110418643.0,
      "step": 59230
    },
    {
      "entropy": 5.947828054428101,
      "epoch": 5.230925468032497,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00026166172444326525,
      "loss": 5.185,
      "mean_token_accuracy": 0.23015898764133452,
      "num_tokens": 110428495.0,
      "step": 59235
    },
    {
      "entropy": 6.070070791244507,
      "epoch": 5.231367008124337,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002616302884512896,
      "loss": 5.2489,
      "mean_token_accuracy": 0.22100357115268707,
      "num_tokens": 110437027.0,
      "step": 59240
    },
    {
      "entropy": 6.057007217407227,
      "epoch": 5.231808548216178,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026159885272121903,
      "loss": 5.2081,
      "mean_token_accuracy": 0.2195697084069252,
      "num_tokens": 110446448.0,
      "step": 59245
    },
    {
      "entropy": 6.028631782531738,
      "epoch": 5.232250088308018,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002615674172536692,
      "loss": 5.2017,
      "mean_token_accuracy": 0.22901894450187682,
      "num_tokens": 110455915.0,
      "step": 59250
    },
    {
      "entropy": 6.025280857086182,
      "epoch": 5.232691628399858,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002615359820492561,
      "loss": 5.1888,
      "mean_token_accuracy": 0.2317181795835495,
      "num_tokens": 110465925.0,
      "step": 59255
    },
    {
      "entropy": 6.0821709632873535,
      "epoch": 5.233133168491699,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002615045471085954,
      "loss": 5.1886,
      "mean_token_accuracy": 0.22259167432785035,
      "num_tokens": 110474454.0,
      "step": 59260
    },
    {
      "entropy": 6.077139806747437,
      "epoch": 5.233574708583539,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026147311243230296,
      "loss": 5.269,
      "mean_token_accuracy": 0.22632376104593277,
      "num_tokens": 110482973.0,
      "step": 59265
    },
    {
      "entropy": 5.9820959091186525,
      "epoch": 5.234016248675379,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002614416780209946,
      "loss": 5.1618,
      "mean_token_accuracy": 0.22865193337202072,
      "num_tokens": 110492246.0,
      "step": 59270
    },
    {
      "entropy": 5.998512935638428,
      "epoch": 5.23445778876722,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002614102438752861,
      "loss": 5.1517,
      "mean_token_accuracy": 0.2254369705915451,
      "num_tokens": 110500680.0,
      "step": 59275
    },
    {
      "entropy": 6.188348054885864,
      "epoch": 5.23489932885906,
      "grad_norm": 1.515625,
      "learning_rate": 0.000261378809995793,
      "loss": 5.3462,
      "mean_token_accuracy": 0.21262193024158477,
      "num_tokens": 110510183.0,
      "step": 59280
    },
    {
      "entropy": 6.095116853713989,
      "epoch": 5.2353408689509005,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002613473763831314,
      "loss": 5.16,
      "mean_token_accuracy": 0.23158311247825622,
      "num_tokens": 110519510.0,
      "step": 59285
    },
    {
      "entropy": 6.051781368255615,
      "epoch": 5.235782409042741,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00026131594303791694,
      "loss": 5.2239,
      "mean_token_accuracy": 0.22270375043153762,
      "num_tokens": 110527898.0,
      "step": 59290
    },
    {
      "entropy": 6.068691062927246,
      "epoch": 5.2362239491345814,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026128450996076543,
      "loss": 5.2566,
      "mean_token_accuracy": 0.22899987250566484,
      "num_tokens": 110538271.0,
      "step": 59295
    },
    {
      "entropy": 6.049211931228638,
      "epoch": 5.2366654892264215,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002612530771522924,
      "loss": 5.2537,
      "mean_token_accuracy": 0.22055220007896423,
      "num_tokens": 110547545.0,
      "step": 59300
    },
    {
      "entropy": 6.079239177703857,
      "epoch": 5.237107029318262,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002612216446131139,
      "loss": 5.198,
      "mean_token_accuracy": 0.22853266596794128,
      "num_tokens": 110558022.0,
      "step": 59305
    },
    {
      "entropy": 6.119705629348755,
      "epoch": 5.2375485694101025,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002611902123438455,
      "loss": 5.2588,
      "mean_token_accuracy": 0.22270744740962983,
      "num_tokens": 110567197.0,
      "step": 59310
    },
    {
      "entropy": 5.990456056594849,
      "epoch": 5.2379901095019425,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000261158780345103,
      "loss": 5.1135,
      "mean_token_accuracy": 0.2401810258626938,
      "num_tokens": 110576474.0,
      "step": 59315
    },
    {
      "entropy": 6.0463203430175785,
      "epoch": 5.2384316495937835,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00026112734861750216,
      "loss": 5.1856,
      "mean_token_accuracy": 0.2339848354458809,
      "num_tokens": 110584923.0,
      "step": 59320
    },
    {
      "entropy": 5.943973731994629,
      "epoch": 5.2388731896856235,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002610959171616586,
      "loss": 5.1501,
      "mean_token_accuracy": 0.23563266694545745,
      "num_tokens": 110594570.0,
      "step": 59325
    },
    {
      "entropy": 6.054224157333374,
      "epoch": 5.239314729777464,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002610644859781881,
      "loss": 5.2143,
      "mean_token_accuracy": 0.22400575429201125,
      "num_tokens": 110604016.0,
      "step": 59330
    },
    {
      "entropy": 6.025666236877441,
      "epoch": 5.2397562698693045,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002610330550677064,
      "loss": 5.1524,
      "mean_token_accuracy": 0.2275429770350456,
      "num_tokens": 110612930.0,
      "step": 59335
    },
    {
      "entropy": 6.106694841384888,
      "epoch": 5.240197809961145,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002610016244308293,
      "loss": 5.2461,
      "mean_token_accuracy": 0.2282422214746475,
      "num_tokens": 110621739.0,
      "step": 59340
    },
    {
      "entropy": 5.992775583267212,
      "epoch": 5.240639350052985,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002609701940681723,
      "loss": 5.1817,
      "mean_token_accuracy": 0.22819234281778336,
      "num_tokens": 110632082.0,
      "step": 59345
    },
    {
      "entropy": 6.144306945800781,
      "epoch": 5.241080890144826,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002609387639803513,
      "loss": 5.333,
      "mean_token_accuracy": 0.207715405523777,
      "num_tokens": 110641353.0,
      "step": 59350
    },
    {
      "entropy": 6.107177734375,
      "epoch": 5.241522430236666,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002609073341679819,
      "loss": 5.2473,
      "mean_token_accuracy": 0.22682132422924042,
      "num_tokens": 110650389.0,
      "step": 59355
    },
    {
      "entropy": 6.110825395584106,
      "epoch": 5.241963970328506,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002608759046316798,
      "loss": 5.3006,
      "mean_token_accuracy": 0.21752808392047882,
      "num_tokens": 110659279.0,
      "step": 59360
    },
    {
      "entropy": 6.050821924209595,
      "epoch": 5.242405510420346,
      "grad_norm": 1.421875,
      "learning_rate": 0.00026084447537206076,
      "loss": 5.1547,
      "mean_token_accuracy": 0.2337836131453514,
      "num_tokens": 110668415.0,
      "step": 59365
    },
    {
      "entropy": 6.0557512760162355,
      "epoch": 5.242847050512187,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002608130463897403,
      "loss": 5.1935,
      "mean_token_accuracy": 0.22913022339344025,
      "num_tokens": 110677884.0,
      "step": 59370
    },
    {
      "entropy": 6.0843418598175045,
      "epoch": 5.243288590604027,
      "grad_norm": 1.375,
      "learning_rate": 0.0002607816176853342,
      "loss": 5.3495,
      "mean_token_accuracy": 0.2164295494556427,
      "num_tokens": 110687461.0,
      "step": 59375
    },
    {
      "entropy": 6.047536897659302,
      "epoch": 5.243730130695867,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026075018925945817,
      "loss": 5.2103,
      "mean_token_accuracy": 0.22530848681926727,
      "num_tokens": 110696777.0,
      "step": 59380
    },
    {
      "entropy": 6.0865747928619385,
      "epoch": 5.244171670787708,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002607187611127278,
      "loss": 5.2477,
      "mean_token_accuracy": 0.22283905595541,
      "num_tokens": 110706016.0,
      "step": 59385
    },
    {
      "entropy": 6.0487922668457035,
      "epoch": 5.244613210879548,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026068733324575874,
      "loss": 5.2272,
      "mean_token_accuracy": 0.22083297818899156,
      "num_tokens": 110715131.0,
      "step": 59390
    },
    {
      "entropy": 6.075422620773315,
      "epoch": 5.245054750971388,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00026065590565916674,
      "loss": 5.2628,
      "mean_token_accuracy": 0.22323612570762635,
      "num_tokens": 110724436.0,
      "step": 59395
    },
    {
      "entropy": 6.029212141036988,
      "epoch": 5.245496291063229,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002606244783535674,
      "loss": 5.2158,
      "mean_token_accuracy": 0.2250267133116722,
      "num_tokens": 110734341.0,
      "step": 59400
    },
    {
      "entropy": 6.021995639801025,
      "epoch": 5.245937831155069,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002605930513295764,
      "loss": 5.2227,
      "mean_token_accuracy": 0.23042683750391008,
      "num_tokens": 110743738.0,
      "step": 59405
    },
    {
      "entropy": 5.967891454696655,
      "epoch": 5.246379371246909,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002605616245878092,
      "loss": 5.1283,
      "mean_token_accuracy": 0.226123908162117,
      "num_tokens": 110753197.0,
      "step": 59410
    },
    {
      "entropy": 5.969487857818604,
      "epoch": 5.24682091133875,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002605301981288816,
      "loss": 5.1171,
      "mean_token_accuracy": 0.23592667281627655,
      "num_tokens": 110762766.0,
      "step": 59415
    },
    {
      "entropy": 5.991319179534912,
      "epoch": 5.24726245143059,
      "grad_norm": 1.53125,
      "learning_rate": 0.00026049877195340925,
      "loss": 5.1241,
      "mean_token_accuracy": 0.23192215859889984,
      "num_tokens": 110771840.0,
      "step": 59420
    },
    {
      "entropy": 6.009267282485962,
      "epoch": 5.24770399152243,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026046734606200773,
      "loss": 5.2197,
      "mean_token_accuracy": 0.22875605523586273,
      "num_tokens": 110781671.0,
      "step": 59425
    },
    {
      "entropy": 6.062819147109986,
      "epoch": 5.248145531614271,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026043592045529255,
      "loss": 5.2135,
      "mean_token_accuracy": 0.22771085053682327,
      "num_tokens": 110791287.0,
      "step": 59430
    },
    {
      "entropy": 6.043314743041992,
      "epoch": 5.248587071706111,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026040449513387937,
      "loss": 5.1818,
      "mean_token_accuracy": 0.22658920288085938,
      "num_tokens": 110801622.0,
      "step": 59435
    },
    {
      "entropy": 6.1096189498901365,
      "epoch": 5.249028611797951,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00026037307009838396,
      "loss": 5.2336,
      "mean_token_accuracy": 0.22342415153980255,
      "num_tokens": 110810156.0,
      "step": 59440
    },
    {
      "entropy": 6.02425799369812,
      "epoch": 5.249470151889792,
      "grad_norm": 1.484375,
      "learning_rate": 0.00026034164534942167,
      "loss": 5.1879,
      "mean_token_accuracy": 0.22258393466472626,
      "num_tokens": 110819680.0,
      "step": 59445
    },
    {
      "entropy": 6.033130788803101,
      "epoch": 5.249911691981632,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002603102208876083,
      "loss": 5.1319,
      "mean_token_accuracy": 0.2313705414533615,
      "num_tokens": 110828395.0,
      "step": 59450
    },
    {
      "entropy": 6.005770683288574,
      "epoch": 5.250353232073472,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026027879671355924,
      "loss": 5.1827,
      "mean_token_accuracy": 0.2266995996236801,
      "num_tokens": 110837505.0,
      "step": 59455
    },
    {
      "entropy": 6.00320782661438,
      "epoch": 5.250794772165313,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026024737282789027,
      "loss": 5.193,
      "mean_token_accuracy": 0.22458989322185516,
      "num_tokens": 110847627.0,
      "step": 59460
    },
    {
      "entropy": 6.093921852111817,
      "epoch": 5.251236312257153,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002602159492312169,
      "loss": 5.3157,
      "mean_token_accuracy": 0.21674961894750594,
      "num_tokens": 110857433.0,
      "step": 59465
    },
    {
      "entropy": 6.012055683135986,
      "epoch": 5.251677852348993,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026018452592415465,
      "loss": 5.1666,
      "mean_token_accuracy": 0.2328389510512352,
      "num_tokens": 110866493.0,
      "step": 59470
    },
    {
      "entropy": 6.1116614818573,
      "epoch": 5.252119392440834,
      "grad_norm": 1.375,
      "learning_rate": 0.000260153102907319,
      "loss": 5.3092,
      "mean_token_accuracy": 0.22734017670154572,
      "num_tokens": 110875220.0,
      "step": 59475
    },
    {
      "entropy": 6.0661811351776125,
      "epoch": 5.252560932532674,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026012168018132574,
      "loss": 5.281,
      "mean_token_accuracy": 0.22240131944417954,
      "num_tokens": 110885223.0,
      "step": 59480
    },
    {
      "entropy": 5.994140720367431,
      "epoch": 5.253002472624514,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002600902577467903,
      "loss": 5.1317,
      "mean_token_accuracy": 0.22552909702062607,
      "num_tokens": 110894223.0,
      "step": 59485
    },
    {
      "entropy": 5.98375391960144,
      "epoch": 5.253444012716354,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002600588356043283,
      "loss": 5.2015,
      "mean_token_accuracy": 0.22596197575330734,
      "num_tokens": 110903593.0,
      "step": 59490
    },
    {
      "entropy": 6.103166961669922,
      "epoch": 5.253885552808195,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026002741375455505,
      "loss": 5.2918,
      "mean_token_accuracy": 0.22599630802869797,
      "num_tokens": 110913917.0,
      "step": 59495
    },
    {
      "entropy": 6.059743499755859,
      "epoch": 5.254327092900035,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00025999599219808634,
      "loss": 5.2687,
      "mean_token_accuracy": 0.22586900889873504,
      "num_tokens": 110924753.0,
      "step": 59500
    },
    {
      "entropy": 5.9524720191955565,
      "epoch": 5.254768632991876,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002599645709355376,
      "loss": 5.1296,
      "mean_token_accuracy": 0.23185429871082305,
      "num_tokens": 110934723.0,
      "step": 59505
    },
    {
      "entropy": 5.995367574691772,
      "epoch": 5.255210173083716,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002599331499675243,
      "loss": 5.1888,
      "mean_token_accuracy": 0.2277670234441757,
      "num_tokens": 110943540.0,
      "step": 59510
    },
    {
      "entropy": 6.08364520072937,
      "epoch": 5.255651713175556,
      "grad_norm": 1.375,
      "learning_rate": 0.0002599017292946621,
      "loss": 5.2938,
      "mean_token_accuracy": 0.2196209728717804,
      "num_tokens": 110953194.0,
      "step": 59515
    },
    {
      "entropy": 6.121569728851318,
      "epoch": 5.256093253267396,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002598703089175664,
      "loss": 5.2775,
      "mean_token_accuracy": 0.22558927088975905,
      "num_tokens": 110962102.0,
      "step": 59520
    },
    {
      "entropy": 6.118563556671143,
      "epoch": 5.256534793359237,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002598388888368527,
      "loss": 5.2734,
      "mean_token_accuracy": 0.2183632254600525,
      "num_tokens": 110970884.0,
      "step": 59525
    },
    {
      "entropy": 6.022595167160034,
      "epoch": 5.256976333451077,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002598074690531367,
      "loss": 5.2049,
      "mean_token_accuracy": 0.2237844467163086,
      "num_tokens": 110979701.0,
      "step": 59530
    },
    {
      "entropy": 6.036863994598389,
      "epoch": 5.257417873542917,
      "grad_norm": 1.453125,
      "learning_rate": 0.00025977604956703367,
      "loss": 5.2257,
      "mean_token_accuracy": 0.224261049926281,
      "num_tokens": 110988119.0,
      "step": 59535
    },
    {
      "entropy": 6.096504878997803,
      "epoch": 5.257859413634758,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002597446303791591,
      "loss": 5.2235,
      "mean_token_accuracy": 0.22412670254707337,
      "num_tokens": 110997908.0,
      "step": 59540
    },
    {
      "entropy": 5.999405336380005,
      "epoch": 5.258300953726598,
      "grad_norm": 1.375,
      "learning_rate": 0.0002597132114901286,
      "loss": 5.1777,
      "mean_token_accuracy": 0.22988596856594085,
      "num_tokens": 111007520.0,
      "step": 59545
    },
    {
      "entropy": 6.067533111572265,
      "epoch": 5.258742493818438,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002596817929005576,
      "loss": 5.2143,
      "mean_token_accuracy": 0.22329814583063126,
      "num_tokens": 111016664.0,
      "step": 59550
    },
    {
      "entropy": 6.087304210662841,
      "epoch": 5.259184033910279,
      "grad_norm": 1.421875,
      "learning_rate": 0.00025965037461106157,
      "loss": 5.2443,
      "mean_token_accuracy": 0.21763961613178254,
      "num_tokens": 111026111.0,
      "step": 59555
    },
    {
      "entropy": 6.05186710357666,
      "epoch": 5.259625574002119,
      "grad_norm": 1.515625,
      "learning_rate": 0.00025961895662225594,
      "loss": 5.2151,
      "mean_token_accuracy": 0.2210829272866249,
      "num_tokens": 111035398.0,
      "step": 59560
    },
    {
      "entropy": 5.93880295753479,
      "epoch": 5.260067114093959,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002595875389347562,
      "loss": 5.1112,
      "mean_token_accuracy": 0.24198360443115235,
      "num_tokens": 111043803.0,
      "step": 59565
    },
    {
      "entropy": 6.061573266983032,
      "epoch": 5.2605086541858,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002595561215491779,
      "loss": 5.2624,
      "mean_token_accuracy": 0.22440509200096131,
      "num_tokens": 111053360.0,
      "step": 59570
    },
    {
      "entropy": 6.0593750953674315,
      "epoch": 5.26095019427764,
      "grad_norm": 1.453125,
      "learning_rate": 0.00025952470446613627,
      "loss": 5.2146,
      "mean_token_accuracy": 0.22671715170145035,
      "num_tokens": 111061410.0,
      "step": 59575
    },
    {
      "entropy": 6.0421899318695065,
      "epoch": 5.2613917343694805,
      "grad_norm": 1.375,
      "learning_rate": 0.0002594932876862469,
      "loss": 5.2387,
      "mean_token_accuracy": 0.22925255745649337,
      "num_tokens": 111070389.0,
      "step": 59580
    },
    {
      "entropy": 6.0751550674438475,
      "epoch": 5.261833274461321,
      "grad_norm": 1.484375,
      "learning_rate": 0.00025946187121012515,
      "loss": 5.2292,
      "mean_token_accuracy": 0.22081447392702103,
      "num_tokens": 111079243.0,
      "step": 59585
    },
    {
      "entropy": 6.020857095718384,
      "epoch": 5.2622748145531615,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002594304550383866,
      "loss": 5.1444,
      "mean_token_accuracy": 0.22743354141712188,
      "num_tokens": 111087419.0,
      "step": 59590
    },
    {
      "entropy": 6.067612361907959,
      "epoch": 5.2627163546450015,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002593990391716466,
      "loss": 5.3005,
      "mean_token_accuracy": 0.21669476628303527,
      "num_tokens": 111096802.0,
      "step": 59595
    },
    {
      "entropy": 6.004194116592407,
      "epoch": 5.2631578947368425,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002593676236105205,
      "loss": 5.194,
      "mean_token_accuracy": 0.23173576295375825,
      "num_tokens": 111105978.0,
      "step": 59600
    },
    {
      "entropy": 5.9261890888214115,
      "epoch": 5.2635994348286825,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002593362083556238,
      "loss": 5.1486,
      "mean_token_accuracy": 0.23147414326667787,
      "num_tokens": 111115945.0,
      "step": 59605
    },
    {
      "entropy": 6.045053625106812,
      "epoch": 5.264040974920523,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002593047934075718,
      "loss": 5.2215,
      "mean_token_accuracy": 0.22285182178020477,
      "num_tokens": 111125555.0,
      "step": 59610
    },
    {
      "entropy": 6.027329015731811,
      "epoch": 5.2644825150123635,
      "grad_norm": 1.25,
      "learning_rate": 0.00025927337876698004,
      "loss": 5.267,
      "mean_token_accuracy": 0.22038078606128692,
      "num_tokens": 111136133.0,
      "step": 59615
    },
    {
      "entropy": 6.088502025604248,
      "epoch": 5.264924055104204,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002592419644344639,
      "loss": 5.1779,
      "mean_token_accuracy": 0.22397976219654084,
      "num_tokens": 111144647.0,
      "step": 59620
    },
    {
      "entropy": 6.060219240188599,
      "epoch": 5.265365595196044,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025921055041063857,
      "loss": 5.202,
      "mean_token_accuracy": 0.22834513634443282,
      "num_tokens": 111153791.0,
      "step": 59625
    },
    {
      "entropy": 6.149212312698364,
      "epoch": 5.265807135287885,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025917913669611965,
      "loss": 5.2691,
      "mean_token_accuracy": 0.21873089522123337,
      "num_tokens": 111163416.0,
      "step": 59630
    },
    {
      "entropy": 6.036468887329102,
      "epoch": 5.266248675379725,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002591477232915225,
      "loss": 5.1935,
      "mean_token_accuracy": 0.22906496971845627,
      "num_tokens": 111172468.0,
      "step": 59635
    },
    {
      "entropy": 6.02874116897583,
      "epoch": 5.266690215471565,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00025911631019746227,
      "loss": 5.2333,
      "mean_token_accuracy": 0.21838191747665406,
      "num_tokens": 111182260.0,
      "step": 59640
    },
    {
      "entropy": 6.024715995788574,
      "epoch": 5.267131755563405,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025908489741455464,
      "loss": 5.2361,
      "mean_token_accuracy": 0.21894582659006118,
      "num_tokens": 111191955.0,
      "step": 59645
    },
    {
      "entropy": 6.04401707649231,
      "epoch": 5.267573295655246,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00025905348494341466,
      "loss": 5.2114,
      "mean_token_accuracy": 0.2282543271780014,
      "num_tokens": 111200743.0,
      "step": 59650
    },
    {
      "entropy": 6.079119873046875,
      "epoch": 5.268014835747086,
      "grad_norm": 1.296875,
      "learning_rate": 0.000259022072784658,
      "loss": 5.2828,
      "mean_token_accuracy": 0.21480244547128677,
      "num_tokens": 111210975.0,
      "step": 59655
    },
    {
      "entropy": 6.095824241638184,
      "epoch": 5.268456375838926,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002589906609388998,
      "loss": 5.2534,
      "mean_token_accuracy": 0.2185666009783745,
      "num_tokens": 111220455.0,
      "step": 59660
    },
    {
      "entropy": 6.037643575668335,
      "epoch": 5.268897915930767,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002589592494067555,
      "loss": 5.1902,
      "mean_token_accuracy": 0.22799817174673082,
      "num_tokens": 111229446.0,
      "step": 59665
    },
    {
      "entropy": 6.10402545928955,
      "epoch": 5.269339456022607,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025892783818884043,
      "loss": 5.2503,
      "mean_token_accuracy": 0.2216830775141716,
      "num_tokens": 111238599.0,
      "step": 59670
    },
    {
      "entropy": 6.0108331680297855,
      "epoch": 5.269780996114447,
      "grad_norm": 1.390625,
      "learning_rate": 0.00025889642728576964,
      "loss": 5.1932,
      "mean_token_accuracy": 0.2218190237879753,
      "num_tokens": 111247609.0,
      "step": 59675
    },
    {
      "entropy": 6.129221630096436,
      "epoch": 5.270222536206288,
      "grad_norm": 1.375,
      "learning_rate": 0.00025886501669815886,
      "loss": 5.3742,
      "mean_token_accuracy": 0.2088504523038864,
      "num_tokens": 111258985.0,
      "step": 59680
    },
    {
      "entropy": 6.017303228378296,
      "epoch": 5.270664076298128,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002588336064266232,
      "loss": 5.2032,
      "mean_token_accuracy": 0.2243396297097206,
      "num_tokens": 111268437.0,
      "step": 59685
    },
    {
      "entropy": 6.05005955696106,
      "epoch": 5.271105616389968,
      "grad_norm": 1.34375,
      "learning_rate": 0.00025880219647177795,
      "loss": 5.1751,
      "mean_token_accuracy": 0.22758544683456422,
      "num_tokens": 111277256.0,
      "step": 59690
    },
    {
      "entropy": 6.1188623905181885,
      "epoch": 5.271547156481809,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025877078683423846,
      "loss": 5.3803,
      "mean_token_accuracy": 0.21462995409965516,
      "num_tokens": 111286787.0,
      "step": 59695
    },
    {
      "entropy": 6.10723557472229,
      "epoch": 5.271988696573649,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00025873937751462006,
      "loss": 5.2778,
      "mean_token_accuracy": 0.2202487975358963,
      "num_tokens": 111296340.0,
      "step": 59700
    },
    {
      "entropy": 6.063771200180054,
      "epoch": 5.272430236665489,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000258707968513538,
      "loss": 5.2355,
      "mean_token_accuracy": 0.218394236266613,
      "num_tokens": 111305501.0,
      "step": 59705
    },
    {
      "entropy": 6.0305102348327635,
      "epoch": 5.27287177675733,
      "grad_norm": 1.515625,
      "learning_rate": 0.00025867655983160756,
      "loss": 5.1732,
      "mean_token_accuracy": 0.2281005173921585,
      "num_tokens": 111314834.0,
      "step": 59710
    },
    {
      "entropy": 6.003125190734863,
      "epoch": 5.27331331684917,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000258645151469444,
      "loss": 5.213,
      "mean_token_accuracy": 0.22283125519752503,
      "num_tokens": 111324465.0,
      "step": 59715
    },
    {
      "entropy": 6.0710982322692875,
      "epoch": 5.27375485694101,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002586137434276626,
      "loss": 5.3325,
      "mean_token_accuracy": 0.21677003502845765,
      "num_tokens": 111334657.0,
      "step": 59720
    },
    {
      "entropy": 6.067590761184692,
      "epoch": 5.274196397032851,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002585823357068788,
      "loss": 5.2051,
      "mean_token_accuracy": 0.22607736885547638,
      "num_tokens": 111343598.0,
      "step": 59725
    },
    {
      "entropy": 5.932564163208008,
      "epoch": 5.274637937124691,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002585509283077076,
      "loss": 5.0816,
      "mean_token_accuracy": 0.24177492111921312,
      "num_tokens": 111352922.0,
      "step": 59730
    },
    {
      "entropy": 6.096740055084228,
      "epoch": 5.275079477216531,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002585195212307644,
      "loss": 5.2832,
      "mean_token_accuracy": 0.22197161465883256,
      "num_tokens": 111363148.0,
      "step": 59735
    },
    {
      "entropy": 6.086122751235962,
      "epoch": 5.275521017308372,
      "grad_norm": 1.484375,
      "learning_rate": 0.00025848811447666434,
      "loss": 5.2368,
      "mean_token_accuracy": 0.2162653237581253,
      "num_tokens": 111371550.0,
      "step": 59740
    },
    {
      "entropy": 6.054261302947998,
      "epoch": 5.275962557400212,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002584567080460228,
      "loss": 5.2696,
      "mean_token_accuracy": 0.22414608299732208,
      "num_tokens": 111381317.0,
      "step": 59745
    },
    {
      "entropy": 6.063138675689697,
      "epoch": 5.276404097492052,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002584253019394549,
      "loss": 5.1953,
      "mean_token_accuracy": 0.2228836879134178,
      "num_tokens": 111390624.0,
      "step": 59750
    },
    {
      "entropy": 6.060435152053833,
      "epoch": 5.276845637583893,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00025839389615757587,
      "loss": 5.2202,
      "mean_token_accuracy": 0.22762832939624786,
      "num_tokens": 111398912.0,
      "step": 59755
    },
    {
      "entropy": 6.074517107009887,
      "epoch": 5.277287177675733,
      "grad_norm": 1.234375,
      "learning_rate": 0.000258362490701001,
      "loss": 5.2343,
      "mean_token_accuracy": 0.21386268883943557,
      "num_tokens": 111408190.0,
      "step": 59760
    },
    {
      "entropy": 5.984902477264404,
      "epoch": 5.277728717767573,
      "grad_norm": 1.421875,
      "learning_rate": 0.00025833108557034545,
      "loss": 5.1316,
      "mean_token_accuracy": 0.23119004368782042,
      "num_tokens": 111417700.0,
      "step": 59765
    },
    {
      "entropy": 6.0362548828125,
      "epoch": 5.278170257859413,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002582996807662245,
      "loss": 5.2122,
      "mean_token_accuracy": 0.22337282747030257,
      "num_tokens": 111426899.0,
      "step": 59770
    },
    {
      "entropy": 6.043412780761718,
      "epoch": 5.278611797951254,
      "grad_norm": 1.546875,
      "learning_rate": 0.00025826827628925333,
      "loss": 5.2759,
      "mean_token_accuracy": 0.21592916399240494,
      "num_tokens": 111435292.0,
      "step": 59775
    },
    {
      "entropy": 6.051729345321656,
      "epoch": 5.279053338043094,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000258236872140047,
      "loss": 5.1839,
      "mean_token_accuracy": 0.2286922737956047,
      "num_tokens": 111444493.0,
      "step": 59780
    },
    {
      "entropy": 6.049395370483398,
      "epoch": 5.279494878134935,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002582054683192209,
      "loss": 5.2399,
      "mean_token_accuracy": 0.2280182048678398,
      "num_tokens": 111453244.0,
      "step": 59785
    },
    {
      "entropy": 6.099921798706054,
      "epoch": 5.279936418226775,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00025817406482739016,
      "loss": 5.2616,
      "mean_token_accuracy": 0.21521342992782594,
      "num_tokens": 111462569.0,
      "step": 59790
    },
    {
      "entropy": 5.99062967300415,
      "epoch": 5.280377958318615,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025814266166516986,
      "loss": 5.2213,
      "mean_token_accuracy": 0.21960498690605162,
      "num_tokens": 111472729.0,
      "step": 59795
    },
    {
      "entropy": 6.011597728729248,
      "epoch": 5.280819498410455,
      "grad_norm": 1.21875,
      "learning_rate": 0.00025811125883317527,
      "loss": 5.1634,
      "mean_token_accuracy": 0.2285621702671051,
      "num_tokens": 111481857.0,
      "step": 59800
    },
    {
      "entropy": 6.0225257396698,
      "epoch": 5.281261038502296,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002580798563320214,
      "loss": 5.2143,
      "mean_token_accuracy": 0.2276562348008156,
      "num_tokens": 111490182.0,
      "step": 59805
    },
    {
      "entropy": 5.976329135894775,
      "epoch": 5.281702578594136,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025804845416232367,
      "loss": 5.1807,
      "mean_token_accuracy": 0.2309132322669029,
      "num_tokens": 111500575.0,
      "step": 59810
    },
    {
      "entropy": 6.0559868812561035,
      "epoch": 5.282144118685976,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000258017052324697,
      "loss": 5.2913,
      "mean_token_accuracy": 0.22396857142448426,
      "num_tokens": 111511107.0,
      "step": 59815
    },
    {
      "entropy": 6.006028509140014,
      "epoch": 5.282585658777817,
      "grad_norm": 1.5625,
      "learning_rate": 0.00025798565081975667,
      "loss": 5.1952,
      "mean_token_accuracy": 0.23588674664497375,
      "num_tokens": 111519969.0,
      "step": 59820
    },
    {
      "entropy": 6.054263687133789,
      "epoch": 5.283027198869657,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002579542496481177,
      "loss": 5.2593,
      "mean_token_accuracy": 0.2219328463077545,
      "num_tokens": 111529213.0,
      "step": 59825
    },
    {
      "entropy": 6.057608366012573,
      "epoch": 5.283468738961497,
      "grad_norm": 1.375,
      "learning_rate": 0.00025792284881039524,
      "loss": 5.2112,
      "mean_token_accuracy": 0.22692868262529373,
      "num_tokens": 111538183.0,
      "step": 59830
    },
    {
      "entropy": 6.0307722091674805,
      "epoch": 5.283910279053338,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002578914483072046,
      "loss": 5.1908,
      "mean_token_accuracy": 0.22493842840194703,
      "num_tokens": 111547928.0,
      "step": 59835
    },
    {
      "entropy": 6.082479858398438,
      "epoch": 5.284351819145178,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002578600481391607,
      "loss": 5.3016,
      "mean_token_accuracy": 0.22012601047754288,
      "num_tokens": 111556842.0,
      "step": 59840
    },
    {
      "entropy": 6.073219680786133,
      "epoch": 5.284793359237018,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00025782864830687876,
      "loss": 5.2152,
      "mean_token_accuracy": 0.22211705297231674,
      "num_tokens": 111566091.0,
      "step": 59845
    },
    {
      "entropy": 6.019358825683594,
      "epoch": 5.285234899328859,
      "grad_norm": 1.46875,
      "learning_rate": 0.00025779724881097365,
      "loss": 5.168,
      "mean_token_accuracy": 0.22895945608615875,
      "num_tokens": 111574608.0,
      "step": 59850
    },
    {
      "entropy": 6.000295639038086,
      "epoch": 5.285676439420699,
      "grad_norm": 1.46875,
      "learning_rate": 0.00025776584965206086,
      "loss": 5.2229,
      "mean_token_accuracy": 0.22063653469085692,
      "num_tokens": 111584217.0,
      "step": 59855
    },
    {
      "entropy": 6.081241178512573,
      "epoch": 5.2861179795125395,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002577344508307552,
      "loss": 5.2444,
      "mean_token_accuracy": 0.22555772513151168,
      "num_tokens": 111592935.0,
      "step": 59860
    },
    {
      "entropy": 6.074546813964844,
      "epoch": 5.28655951960438,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025770305234767185,
      "loss": 5.2479,
      "mean_token_accuracy": 0.22272471636533736,
      "num_tokens": 111602231.0,
      "step": 59865
    },
    {
      "entropy": 6.026153326034546,
      "epoch": 5.2870010596962205,
      "grad_norm": 1.375,
      "learning_rate": 0.00025767165420342576,
      "loss": 5.2002,
      "mean_token_accuracy": 0.23002442121505737,
      "num_tokens": 111611894.0,
      "step": 59870
    },
    {
      "entropy": 6.036501550674439,
      "epoch": 5.2874425997880605,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002576402563986322,
      "loss": 5.1494,
      "mean_token_accuracy": 0.2323340117931366,
      "num_tokens": 111621542.0,
      "step": 59875
    },
    {
      "entropy": 5.990242862701416,
      "epoch": 5.2878841398799015,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002576088589339062,
      "loss": 5.2292,
      "mean_token_accuracy": 0.2217501223087311,
      "num_tokens": 111631403.0,
      "step": 59880
    },
    {
      "entropy": 6.029236364364624,
      "epoch": 5.2883256799717415,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002575774618098626,
      "loss": 5.2443,
      "mean_token_accuracy": 0.22460977882146835,
      "num_tokens": 111640454.0,
      "step": 59885
    },
    {
      "entropy": 6.028171634674072,
      "epoch": 5.288767220063582,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002575460650271167,
      "loss": 5.27,
      "mean_token_accuracy": 0.22438787966966628,
      "num_tokens": 111650233.0,
      "step": 59890
    },
    {
      "entropy": 6.045134592056274,
      "epoch": 5.2892087601554225,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002575146685862835,
      "loss": 5.2112,
      "mean_token_accuracy": 0.22549945265054702,
      "num_tokens": 111659635.0,
      "step": 59895
    },
    {
      "entropy": 6.086396551132202,
      "epoch": 5.289650300247263,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002574832724879779,
      "loss": 5.2749,
      "mean_token_accuracy": 0.21462691128253936,
      "num_tokens": 111669010.0,
      "step": 59900
    },
    {
      "entropy": 6.140405082702637,
      "epoch": 5.290091840339103,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00025745187673281516,
      "loss": 5.2761,
      "mean_token_accuracy": 0.22220934927463531,
      "num_tokens": 111678602.0,
      "step": 59905
    },
    {
      "entropy": 6.081107950210571,
      "epoch": 5.290533380430944,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002574204813214101,
      "loss": 5.2777,
      "mean_token_accuracy": 0.21500156223773956,
      "num_tokens": 111688534.0,
      "step": 59910
    },
    {
      "entropy": 6.075704908370971,
      "epoch": 5.290974920522784,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00025738908625437773,
      "loss": 5.2753,
      "mean_token_accuracy": 0.2171427220106125,
      "num_tokens": 111698350.0,
      "step": 59915
    },
    {
      "entropy": 6.106607055664062,
      "epoch": 5.291416460614624,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002573576915323333,
      "loss": 5.2837,
      "mean_token_accuracy": 0.22420414835214614,
      "num_tokens": 111707174.0,
      "step": 59920
    },
    {
      "entropy": 6.016745519638062,
      "epoch": 5.291858000706464,
      "grad_norm": 1.484375,
      "learning_rate": 0.00025732629715589163,
      "loss": 5.0632,
      "mean_token_accuracy": 0.24054811745882035,
      "num_tokens": 111715231.0,
      "step": 59925
    },
    {
      "entropy": 6.021278381347656,
      "epoch": 5.292299540798305,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002572949031256677,
      "loss": 5.1778,
      "mean_token_accuracy": 0.2246862009167671,
      "num_tokens": 111724143.0,
      "step": 59930
    },
    {
      "entropy": 5.96147050857544,
      "epoch": 5.292741080890145,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002572635094422765,
      "loss": 5.1576,
      "mean_token_accuracy": 0.23317955285310746,
      "num_tokens": 111733248.0,
      "step": 59935
    },
    {
      "entropy": 6.084982919692993,
      "epoch": 5.293182620981985,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002572321161063332,
      "loss": 5.2282,
      "mean_token_accuracy": 0.22291140258312225,
      "num_tokens": 111742657.0,
      "step": 59940
    },
    {
      "entropy": 6.07096037864685,
      "epoch": 5.293624161073826,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00025720072311845256,
      "loss": 5.2416,
      "mean_token_accuracy": 0.22295560389757157,
      "num_tokens": 111751693.0,
      "step": 59945
    },
    {
      "entropy": 6.044959878921508,
      "epoch": 5.294065701165666,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002571693304792496,
      "loss": 5.272,
      "mean_token_accuracy": 0.2238932430744171,
      "num_tokens": 111760196.0,
      "step": 59950
    },
    {
      "entropy": 6.0602428913116455,
      "epoch": 5.294507241257506,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002571379381893393,
      "loss": 5.2892,
      "mean_token_accuracy": 0.22125948518514632,
      "num_tokens": 111769191.0,
      "step": 59955
    },
    {
      "entropy": 6.1885809898376465,
      "epoch": 5.294948781349347,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00025710654624933667,
      "loss": 5.2984,
      "mean_token_accuracy": 0.21663240492343902,
      "num_tokens": 111779265.0,
      "step": 59960
    },
    {
      "entropy": 6.062135314941406,
      "epoch": 5.295390321441187,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002570751546598567,
      "loss": 5.2567,
      "mean_token_accuracy": 0.22138111293315887,
      "num_tokens": 111788030.0,
      "step": 59965
    },
    {
      "entropy": 6.078354597091675,
      "epoch": 5.295831861533027,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002570437634215142,
      "loss": 5.2778,
      "mean_token_accuracy": 0.2154680758714676,
      "num_tokens": 111796893.0,
      "step": 59970
    },
    {
      "entropy": 6.049232769012451,
      "epoch": 5.296273401624868,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002570123725349242,
      "loss": 5.2267,
      "mean_token_accuracy": 0.22843310385942459,
      "num_tokens": 111807549.0,
      "step": 59975
    },
    {
      "entropy": 6.019101572036743,
      "epoch": 5.296714941716708,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002569809820007015,
      "loss": 5.1966,
      "mean_token_accuracy": 0.22940220534801484,
      "num_tokens": 111816712.0,
      "step": 59980
    },
    {
      "entropy": 6.067951679229736,
      "epoch": 5.297156481808548,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025694959181946114,
      "loss": 5.2025,
      "mean_token_accuracy": 0.2260554924607277,
      "num_tokens": 111827029.0,
      "step": 59985
    },
    {
      "entropy": 6.114413547515869,
      "epoch": 5.297598021900389,
      "grad_norm": 1.46875,
      "learning_rate": 0.00025691820199181804,
      "loss": 5.2917,
      "mean_token_accuracy": 0.2152247577905655,
      "num_tokens": 111837334.0,
      "step": 59990
    },
    {
      "entropy": 6.062362289428711,
      "epoch": 5.298039561992229,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002568868125183871,
      "loss": 5.2647,
      "mean_token_accuracy": 0.2236577108502388,
      "num_tokens": 111847137.0,
      "step": 59995
    },
    {
      "entropy": 6.066639518737793,
      "epoch": 5.298481102084069,
      "grad_norm": 1.453125,
      "learning_rate": 0.00025685542339978307,
      "loss": 5.272,
      "mean_token_accuracy": 0.2192596361041069,
      "num_tokens": 111856422.0,
      "step": 60000
    },
    {
      "epoch": 5.298481102084069,
      "eval_entropy": 5.7345574150539065,
      "eval_loss": 5.302523612976074,
      "eval_mean_token_accuracy": 0.22849614446389985,
      "eval_num_tokens": 111856422.0,
      "eval_runtime": 20.1318,
      "eval_samples_per_second": 1754.186,
      "eval_steps_per_second": 219.304,
      "step": 60000
    },
    {
      "entropy": 6.0365479469299315,
      "epoch": 5.29892264217591,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002568240346366211,
      "loss": 5.2537,
      "mean_token_accuracy": 0.222467902302742,
      "num_tokens": 111866122.0,
      "step": 60005
    },
    {
      "entropy": 6.029616212844848,
      "epoch": 5.29936418226775,
      "grad_norm": 1.421875,
      "learning_rate": 0.00025679264622951596,
      "loss": 5.1715,
      "mean_token_accuracy": 0.23323997259140014,
      "num_tokens": 111874708.0,
      "step": 60010
    },
    {
      "entropy": 6.00296368598938,
      "epoch": 5.29980572235959,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002567612581790825,
      "loss": 5.1545,
      "mean_token_accuracy": 0.22819686383008958,
      "num_tokens": 111884673.0,
      "step": 60015
    },
    {
      "entropy": 6.021663045883178,
      "epoch": 5.300247262451431,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002567298704859356,
      "loss": 5.3039,
      "mean_token_accuracy": 0.21899803578853608,
      "num_tokens": 111893623.0,
      "step": 60020
    },
    {
      "entropy": 6.046892213821411,
      "epoch": 5.300688802543271,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002566984831506901,
      "loss": 5.2774,
      "mean_token_accuracy": 0.22501010447740555,
      "num_tokens": 111902536.0,
      "step": 60025
    },
    {
      "entropy": 6.135685729980469,
      "epoch": 5.301130342635111,
      "grad_norm": 1.34375,
      "learning_rate": 0.000256667096173961,
      "loss": 5.2904,
      "mean_token_accuracy": 0.2222877562046051,
      "num_tokens": 111911915.0,
      "step": 60030
    },
    {
      "entropy": 6.1322774410247805,
      "epoch": 5.301571882726952,
      "grad_norm": 1.4375,
      "learning_rate": 0.00025663570955636306,
      "loss": 5.3443,
      "mean_token_accuracy": 0.21167702674865724,
      "num_tokens": 111920303.0,
      "step": 60035
    },
    {
      "entropy": 6.16318941116333,
      "epoch": 5.302013422818792,
      "grad_norm": 1.453125,
      "learning_rate": 0.00025660432329851113,
      "loss": 5.4163,
      "mean_token_accuracy": 0.20531315058469773,
      "num_tokens": 111930196.0,
      "step": 60040
    },
    {
      "entropy": 6.080794811248779,
      "epoch": 5.302454962910632,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002565729374010199,
      "loss": 5.2494,
      "mean_token_accuracy": 0.22003195136785508,
      "num_tokens": 111940314.0,
      "step": 60045
    },
    {
      "entropy": 6.125058126449585,
      "epoch": 5.302896503002472,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002565415518645045,
      "loss": 5.3153,
      "mean_token_accuracy": 0.2157591015100479,
      "num_tokens": 111949300.0,
      "step": 60050
    },
    {
      "entropy": 6.069022464752197,
      "epoch": 5.303338043094313,
      "grad_norm": 1.421875,
      "learning_rate": 0.00025651016668957956,
      "loss": 5.205,
      "mean_token_accuracy": 0.2299453765153885,
      "num_tokens": 111958096.0,
      "step": 60055
    },
    {
      "entropy": 5.994094324111939,
      "epoch": 5.303779583186153,
      "grad_norm": 1.4375,
      "learning_rate": 0.00025647878187685995,
      "loss": 5.1806,
      "mean_token_accuracy": 0.22929944545030595,
      "num_tokens": 111967116.0,
      "step": 60060
    },
    {
      "entropy": 6.068012809753418,
      "epoch": 5.304221123277994,
      "grad_norm": 1.375,
      "learning_rate": 0.00025644739742696046,
      "loss": 5.2467,
      "mean_token_accuracy": 0.22122009545564653,
      "num_tokens": 111975717.0,
      "step": 60065
    },
    {
      "entropy": 6.020433568954468,
      "epoch": 5.304662663369834,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025641601334049584,
      "loss": 5.1782,
      "mean_token_accuracy": 0.2280845284461975,
      "num_tokens": 111985475.0,
      "step": 60070
    },
    {
      "entropy": 6.003778600692749,
      "epoch": 5.305104203461674,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025638462961808105,
      "loss": 5.1693,
      "mean_token_accuracy": 0.24132649302482606,
      "num_tokens": 111995739.0,
      "step": 60075
    },
    {
      "entropy": 6.036219787597656,
      "epoch": 5.305545743553514,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002563532462603307,
      "loss": 5.2802,
      "mean_token_accuracy": 0.22142065763473512,
      "num_tokens": 112004129.0,
      "step": 60080
    },
    {
      "entropy": 6.030439567565918,
      "epoch": 5.305987283645355,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002563218632678597,
      "loss": 5.1693,
      "mean_token_accuracy": 0.22899944186210633,
      "num_tokens": 112012757.0,
      "step": 60085
    },
    {
      "entropy": 6.014446210861206,
      "epoch": 5.306428823737195,
      "grad_norm": 1.5625,
      "learning_rate": 0.00025629048064128274,
      "loss": 5.1559,
      "mean_token_accuracy": 0.2334291011095047,
      "num_tokens": 112021512.0,
      "step": 60090
    },
    {
      "entropy": 6.055612277984619,
      "epoch": 5.306870363829035,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00025625909838121463,
      "loss": 5.1715,
      "mean_token_accuracy": 0.2356986001133919,
      "num_tokens": 112030258.0,
      "step": 60095
    },
    {
      "entropy": 5.970679092407226,
      "epoch": 5.307311903920876,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00025622771648827023,
      "loss": 5.2197,
      "mean_token_accuracy": 0.23255113810300826,
      "num_tokens": 112039503.0,
      "step": 60100
    },
    {
      "entropy": 6.030249834060669,
      "epoch": 5.307753444012716,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002561963349630641,
      "loss": 5.2305,
      "mean_token_accuracy": 0.23047108054161072,
      "num_tokens": 112048812.0,
      "step": 60105
    },
    {
      "entropy": 5.9323265075683596,
      "epoch": 5.308194984104556,
      "grad_norm": 1.421875,
      "learning_rate": 0.00025616495380621096,
      "loss": 5.1682,
      "mean_token_accuracy": 0.234182471036911,
      "num_tokens": 112058670.0,
      "step": 60110
    },
    {
      "entropy": 6.004081058502197,
      "epoch": 5.308636524196397,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002561335730183258,
      "loss": 5.19,
      "mean_token_accuracy": 0.22875068783760072,
      "num_tokens": 112067518.0,
      "step": 60115
    },
    {
      "entropy": 6.086654901504517,
      "epoch": 5.309078064288237,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002561021926000232,
      "loss": 5.3782,
      "mean_token_accuracy": 0.21284040361642836,
      "num_tokens": 112076343.0,
      "step": 60120
    },
    {
      "entropy": 6.076177263259888,
      "epoch": 5.309519604380077,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025607081255191795,
      "loss": 5.1811,
      "mean_token_accuracy": 0.22468704283237456,
      "num_tokens": 112087213.0,
      "step": 60125
    },
    {
      "entropy": 6.074078321456909,
      "epoch": 5.309961144471918,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00025603943287462455,
      "loss": 5.1919,
      "mean_token_accuracy": 0.23051467090845107,
      "num_tokens": 112095710.0,
      "step": 60130
    },
    {
      "entropy": 6.044302177429199,
      "epoch": 5.310402684563758,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000256008053568758,
      "loss": 5.2797,
      "mean_token_accuracy": 0.21798965185880662,
      "num_tokens": 112105443.0,
      "step": 60135
    },
    {
      "entropy": 6.038471221923828,
      "epoch": 5.3108442246555985,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002559766746349329,
      "loss": 5.3251,
      "mean_token_accuracy": 0.2139790713787079,
      "num_tokens": 112115881.0,
      "step": 60140
    },
    {
      "entropy": 5.992483854293823,
      "epoch": 5.311285764747439,
      "grad_norm": 1.28125,
      "learning_rate": 0.00025594529607376386,
      "loss": 5.155,
      "mean_token_accuracy": 0.2329930230975151,
      "num_tokens": 112125349.0,
      "step": 60145
    },
    {
      "entropy": 6.0590917587280275,
      "epoch": 5.3117273048392795,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002559139178858657,
      "loss": 5.2088,
      "mean_token_accuracy": 0.22118598222732544,
      "num_tokens": 112134357.0,
      "step": 60150
    },
    {
      "entropy": 6.077356481552124,
      "epoch": 5.3121688449311195,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00025588254007185294,
      "loss": 5.2387,
      "mean_token_accuracy": 0.2224220246076584,
      "num_tokens": 112143815.0,
      "step": 60155
    },
    {
      "entropy": 6.061232328414917,
      "epoch": 5.3126103850229605,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00025585116263234035,
      "loss": 5.178,
      "mean_token_accuracy": 0.22632226943969727,
      "num_tokens": 112152335.0,
      "step": 60160
    },
    {
      "entropy": 6.028223705291748,
      "epoch": 5.3130519251148005,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00025581978556794274,
      "loss": 5.2669,
      "mean_token_accuracy": 0.2177998259663582,
      "num_tokens": 112161463.0,
      "step": 60165
    },
    {
      "entropy": 6.026059293746949,
      "epoch": 5.313493465206641,
      "grad_norm": 1.25,
      "learning_rate": 0.0002557884088792745,
      "loss": 5.266,
      "mean_token_accuracy": 0.21581969559192657,
      "num_tokens": 112172654.0,
      "step": 60170
    },
    {
      "entropy": 6.092182159423828,
      "epoch": 5.3139350052984815,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002557570325669504,
      "loss": 5.2193,
      "mean_token_accuracy": 0.22419993877410888,
      "num_tokens": 112182670.0,
      "step": 60175
    },
    {
      "entropy": 6.083679437637329,
      "epoch": 5.314376545390322,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002557256566315851,
      "loss": 5.242,
      "mean_token_accuracy": 0.22482632100582123,
      "num_tokens": 112192814.0,
      "step": 60180
    },
    {
      "entropy": 5.965578508377075,
      "epoch": 5.314818085482162,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002556942810737933,
      "loss": 5.1475,
      "mean_token_accuracy": 0.23788630366325378,
      "num_tokens": 112202760.0,
      "step": 60185
    },
    {
      "entropy": 6.064625072479248,
      "epoch": 5.3152596255740026,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002556629058941895,
      "loss": 5.2337,
      "mean_token_accuracy": 0.2232731357216835,
      "num_tokens": 112212793.0,
      "step": 60190
    },
    {
      "entropy": 6.135086631774902,
      "epoch": 5.315701165665843,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025563153109338827,
      "loss": 5.2627,
      "mean_token_accuracy": 0.21989407539367675,
      "num_tokens": 112223007.0,
      "step": 60195
    },
    {
      "entropy": 6.077649927139282,
      "epoch": 5.316142705757683,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002556001566720045,
      "loss": 5.2219,
      "mean_token_accuracy": 0.22421513199806214,
      "num_tokens": 112231968.0,
      "step": 60200
    },
    {
      "entropy": 6.0680280208587645,
      "epoch": 5.316584245849523,
      "grad_norm": 1.4375,
      "learning_rate": 0.00025556878263065247,
      "loss": 5.2537,
      "mean_token_accuracy": 0.2269832208752632,
      "num_tokens": 112241755.0,
      "step": 60205
    },
    {
      "entropy": 6.032734394073486,
      "epoch": 5.317025785941364,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002555374089699471,
      "loss": 5.2511,
      "mean_token_accuracy": 0.223533733189106,
      "num_tokens": 112250284.0,
      "step": 60210
    },
    {
      "entropy": 6.031100797653198,
      "epoch": 5.317467326033204,
      "grad_norm": 1.40625,
      "learning_rate": 0.00025550603569050276,
      "loss": 5.291,
      "mean_token_accuracy": 0.22002929151058198,
      "num_tokens": 112259841.0,
      "step": 60215
    },
    {
      "entropy": 6.01732234954834,
      "epoch": 5.317908866125044,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00025547466279293404,
      "loss": 5.1534,
      "mean_token_accuracy": 0.23148889243602752,
      "num_tokens": 112268493.0,
      "step": 60220
    },
    {
      "entropy": 6.165205717086792,
      "epoch": 5.318350406216885,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025544329027785557,
      "loss": 5.3486,
      "mean_token_accuracy": 0.21257535368204117,
      "num_tokens": 112277782.0,
      "step": 60225
    },
    {
      "entropy": 6.130924272537231,
      "epoch": 5.318791946308725,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000255411918145882,
      "loss": 5.3175,
      "mean_token_accuracy": 0.2196976900100708,
      "num_tokens": 112287018.0,
      "step": 60230
    },
    {
      "entropy": 6.018941879272461,
      "epoch": 5.319233486400565,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002553805463976278,
      "loss": 5.1471,
      "mean_token_accuracy": 0.23431486636400223,
      "num_tokens": 112296409.0,
      "step": 60235
    },
    {
      "entropy": 6.047210168838501,
      "epoch": 5.319675026492406,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00025534917503370746,
      "loss": 5.2125,
      "mean_token_accuracy": 0.2201052188873291,
      "num_tokens": 112305064.0,
      "step": 60240
    },
    {
      "entropy": 6.004211044311523,
      "epoch": 5.320116566584246,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025531780405473563,
      "loss": 5.2055,
      "mean_token_accuracy": 0.22155147343873977,
      "num_tokens": 112313412.0,
      "step": 60245
    },
    {
      "entropy": 6.071744298934936,
      "epoch": 5.320558106676086,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002552864334613269,
      "loss": 5.2551,
      "mean_token_accuracy": 0.22162658125162124,
      "num_tokens": 112322992.0,
      "step": 60250
    },
    {
      "entropy": 6.078303432464599,
      "epoch": 5.320999646767927,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002552550632540957,
      "loss": 5.3594,
      "mean_token_accuracy": 0.21130720973014833,
      "num_tokens": 112332601.0,
      "step": 60255
    },
    {
      "entropy": 6.073573589324951,
      "epoch": 5.321441186859767,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002552236934336565,
      "loss": 5.2448,
      "mean_token_accuracy": 0.22533912360668182,
      "num_tokens": 112341254.0,
      "step": 60260
    },
    {
      "entropy": 6.01397557258606,
      "epoch": 5.321882726951607,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00025519232400062397,
      "loss": 5.1408,
      "mean_token_accuracy": 0.22804878950119017,
      "num_tokens": 112351077.0,
      "step": 60265
    },
    {
      "entropy": 6.075030374526977,
      "epoch": 5.322324267043448,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002551609549556125,
      "loss": 5.2264,
      "mean_token_accuracy": 0.22830682694911958,
      "num_tokens": 112361153.0,
      "step": 60270
    },
    {
      "entropy": 6.031240081787109,
      "epoch": 5.322765807135288,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002551295862992367,
      "loss": 5.1934,
      "mean_token_accuracy": 0.22595476508140563,
      "num_tokens": 112370644.0,
      "step": 60275
    },
    {
      "entropy": 6.007279539108277,
      "epoch": 5.323207347227128,
      "grad_norm": 1.421875,
      "learning_rate": 0.000255098218032111,
      "loss": 5.1748,
      "mean_token_accuracy": 0.21999400854110718,
      "num_tokens": 112380612.0,
      "step": 60280
    },
    {
      "entropy": 5.957664632797242,
      "epoch": 5.323648887318969,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002550668501548498,
      "loss": 5.1819,
      "mean_token_accuracy": 0.22122908085584642,
      "num_tokens": 112390047.0,
      "step": 60285
    },
    {
      "entropy": 5.9876738548278805,
      "epoch": 5.324090427410809,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025503548266806776,
      "loss": 5.2077,
      "mean_token_accuracy": 0.22928987741470336,
      "num_tokens": 112399352.0,
      "step": 60290
    },
    {
      "entropy": 5.956270360946656,
      "epoch": 5.324531967502649,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002550041155723793,
      "loss": 5.1269,
      "mean_token_accuracy": 0.2333345130085945,
      "num_tokens": 112408618.0,
      "step": 60295
    },
    {
      "entropy": 6.092015743255615,
      "epoch": 5.32497350759449,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002549727488683988,
      "loss": 5.2559,
      "mean_token_accuracy": 0.21960047632455826,
      "num_tokens": 112417876.0,
      "step": 60300
    },
    {
      "entropy": 6.080139064788819,
      "epoch": 5.32541504768633,
      "grad_norm": 1.421875,
      "learning_rate": 0.00025494138255674064,
      "loss": 5.2175,
      "mean_token_accuracy": 0.22911669462919235,
      "num_tokens": 112426645.0,
      "step": 60305
    },
    {
      "entropy": 5.991890621185303,
      "epoch": 5.32585658777817,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00025491001663801953,
      "loss": 5.1662,
      "mean_token_accuracy": 0.22891554087400437,
      "num_tokens": 112435840.0,
      "step": 60310
    },
    {
      "entropy": 6.006267023086548,
      "epoch": 5.326298127870011,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002548786511128497,
      "loss": 5.2645,
      "mean_token_accuracy": 0.22067474573850632,
      "num_tokens": 112444621.0,
      "step": 60315
    },
    {
      "entropy": 6.0355761528015135,
      "epoch": 5.326739667961851,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00025484728598184573,
      "loss": 5.2089,
      "mean_token_accuracy": 0.2313089221715927,
      "num_tokens": 112453777.0,
      "step": 60320
    },
    {
      "entropy": 6.058231925964355,
      "epoch": 5.327181208053691,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002548159212456219,
      "loss": 5.2587,
      "mean_token_accuracy": 0.2259373039007187,
      "num_tokens": 112463953.0,
      "step": 60325
    },
    {
      "entropy": 6.025503206253052,
      "epoch": 5.327622748145531,
      "grad_norm": 1.578125,
      "learning_rate": 0.00025478455690479266,
      "loss": 5.1706,
      "mean_token_accuracy": 0.23060597777366637,
      "num_tokens": 112473191.0,
      "step": 60330
    },
    {
      "entropy": 6.0794837951660154,
      "epoch": 5.328064288237372,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002547531929599725,
      "loss": 5.3706,
      "mean_token_accuracy": 0.22088454365730287,
      "num_tokens": 112482964.0,
      "step": 60335
    },
    {
      "entropy": 6.019461441040039,
      "epoch": 5.328505828329212,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002547218294117758,
      "loss": 5.2039,
      "mean_token_accuracy": 0.2261048138141632,
      "num_tokens": 112493051.0,
      "step": 60340
    },
    {
      "entropy": 5.984277820587158,
      "epoch": 5.328947368421053,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002546904662608169,
      "loss": 5.1041,
      "mean_token_accuracy": 0.241952084004879,
      "num_tokens": 112502625.0,
      "step": 60345
    },
    {
      "entropy": 6.075881576538086,
      "epoch": 5.329388908512893,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00025465910350771023,
      "loss": 5.2736,
      "mean_token_accuracy": 0.22588704973459245,
      "num_tokens": 112511302.0,
      "step": 60350
    },
    {
      "entropy": 5.873527097702026,
      "epoch": 5.329830448604733,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002546277411530701,
      "loss": 5.0696,
      "mean_token_accuracy": 0.24287248849868776,
      "num_tokens": 112520161.0,
      "step": 60355
    },
    {
      "entropy": 6.0367570400238035,
      "epoch": 5.330271988696573,
      "grad_norm": 1.265625,
      "learning_rate": 0.00025459637919751106,
      "loss": 5.2642,
      "mean_token_accuracy": 0.2209914356470108,
      "num_tokens": 112530301.0,
      "step": 60360
    },
    {
      "entropy": 6.087601804733277,
      "epoch": 5.330713528788414,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002545650176416473,
      "loss": 5.2941,
      "mean_token_accuracy": 0.21537033319473267,
      "num_tokens": 112539858.0,
      "step": 60365
    },
    {
      "entropy": 6.1484307765960695,
      "epoch": 5.331155068880254,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025453365648609306,
      "loss": 5.227,
      "mean_token_accuracy": 0.22114695161581038,
      "num_tokens": 112549468.0,
      "step": 60370
    },
    {
      "entropy": 6.096394777297974,
      "epoch": 5.331596608972094,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000254502295731463,
      "loss": 5.2468,
      "mean_token_accuracy": 0.22536102384328843,
      "num_tokens": 112558782.0,
      "step": 60375
    },
    {
      "entropy": 6.062069940567016,
      "epoch": 5.332038149063935,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002544709353783713,
      "loss": 5.2244,
      "mean_token_accuracy": 0.22292190045118332,
      "num_tokens": 112568030.0,
      "step": 60380
    },
    {
      "entropy": 6.038879489898681,
      "epoch": 5.332479689155775,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025443957542743234,
      "loss": 5.2686,
      "mean_token_accuracy": 0.21444010287523269,
      "num_tokens": 112577466.0,
      "step": 60385
    },
    {
      "entropy": 5.974626302719116,
      "epoch": 5.332921229247615,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025440821587926034,
      "loss": 5.1886,
      "mean_token_accuracy": 0.2344556525349617,
      "num_tokens": 112586976.0,
      "step": 60390
    },
    {
      "entropy": 6.072126007080078,
      "epoch": 5.333362769339456,
      "grad_norm": 1.4375,
      "learning_rate": 0.00025437685673446967,
      "loss": 5.2104,
      "mean_token_accuracy": 0.22327377200126647,
      "num_tokens": 112595688.0,
      "step": 60395
    },
    {
      "entropy": 6.113717699050904,
      "epoch": 5.333804309431296,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002543454979936746,
      "loss": 5.2664,
      "mean_token_accuracy": 0.2236168637871742,
      "num_tokens": 112604299.0,
      "step": 60400
    },
    {
      "entropy": 6.098137664794922,
      "epoch": 5.334245849523136,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002543141396574896,
      "loss": 5.3105,
      "mean_token_accuracy": 0.2160210371017456,
      "num_tokens": 112615060.0,
      "step": 60405
    },
    {
      "entropy": 6.091606473922729,
      "epoch": 5.334687389614977,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002542827817265288,
      "loss": 5.2126,
      "mean_token_accuracy": 0.22371581494808196,
      "num_tokens": 112624705.0,
      "step": 60410
    },
    {
      "entropy": 6.113559913635254,
      "epoch": 5.335128929706817,
      "grad_norm": 1.421875,
      "learning_rate": 0.00025425142420140647,
      "loss": 5.2285,
      "mean_token_accuracy": 0.2239419236779213,
      "num_tokens": 112634046.0,
      "step": 60415
    },
    {
      "entropy": 6.011914205551148,
      "epoch": 5.3355704697986575,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000254220067082737,
      "loss": 5.1922,
      "mean_token_accuracy": 0.23110329061746598,
      "num_tokens": 112644129.0,
      "step": 60420
    },
    {
      "entropy": 6.066092252731323,
      "epoch": 5.336012009890498,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00025418871037113456,
      "loss": 5.2315,
      "mean_token_accuracy": 0.22451080232858658,
      "num_tokens": 112653562.0,
      "step": 60425
    },
    {
      "entropy": 5.923864650726318,
      "epoch": 5.3364535499823385,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002541573540672135,
      "loss": 5.1298,
      "mean_token_accuracy": 0.24115063846111298,
      "num_tokens": 112662400.0,
      "step": 60430
    },
    {
      "entropy": 6.020350790023803,
      "epoch": 5.3368950900741785,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002541259981715879,
      "loss": 5.1554,
      "mean_token_accuracy": 0.23052266091108323,
      "num_tokens": 112671622.0,
      "step": 60435
    },
    {
      "entropy": 6.022075653076172,
      "epoch": 5.3373366301660194,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00025409464268487224,
      "loss": 5.2076,
      "mean_token_accuracy": 0.22797441631555557,
      "num_tokens": 112680725.0,
      "step": 60440
    },
    {
      "entropy": 6.095391035079956,
      "epoch": 5.3377781702578595,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002540632876076807,
      "loss": 5.3437,
      "mean_token_accuracy": 0.2129685491323471,
      "num_tokens": 112689725.0,
      "step": 60445
    },
    {
      "entropy": 5.96876745223999,
      "epoch": 5.3382197103496996,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002540319329406273,
      "loss": 5.1069,
      "mean_token_accuracy": 0.2420957401394844,
      "num_tokens": 112698949.0,
      "step": 60450
    },
    {
      "entropy": 6.088597774505615,
      "epoch": 5.3386612504415405,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025400057868432647,
      "loss": 5.3294,
      "mean_token_accuracy": 0.21690808236598969,
      "num_tokens": 112709202.0,
      "step": 60455
    },
    {
      "entropy": 6.047358894348145,
      "epoch": 5.3391027905333805,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002539692248393923,
      "loss": 5.1994,
      "mean_token_accuracy": 0.221555095911026,
      "num_tokens": 112718364.0,
      "step": 60460
    },
    {
      "entropy": 5.99072527885437,
      "epoch": 5.339544330625221,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002539378714064391,
      "loss": 5.1692,
      "mean_token_accuracy": 0.22939686775207518,
      "num_tokens": 112727427.0,
      "step": 60465
    },
    {
      "entropy": 6.034748983383179,
      "epoch": 5.3399858707170615,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002539065183860811,
      "loss": 5.226,
      "mean_token_accuracy": 0.2208361178636551,
      "num_tokens": 112736257.0,
      "step": 60470
    },
    {
      "entropy": 6.060145616531372,
      "epoch": 5.340427410808902,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00025387516577893233,
      "loss": 5.267,
      "mean_token_accuracy": 0.22061607092618943,
      "num_tokens": 112746658.0,
      "step": 60475
    },
    {
      "entropy": 6.113250827789306,
      "epoch": 5.340868950900742,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000253843813585607,
      "loss": 5.2489,
      "mean_token_accuracy": 0.22283517867326735,
      "num_tokens": 112757171.0,
      "step": 60480
    },
    {
      "entropy": 6.090609121322632,
      "epoch": 5.341310490992582,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002538124618067194,
      "loss": 5.174,
      "mean_token_accuracy": 0.2298719674348831,
      "num_tokens": 112766351.0,
      "step": 60485
    },
    {
      "entropy": 6.03028130531311,
      "epoch": 5.341752031084423,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00025378111044288364,
      "loss": 5.1611,
      "mean_token_accuracy": 0.23351406008005143,
      "num_tokens": 112776253.0,
      "step": 60490
    },
    {
      "entropy": 6.03301591873169,
      "epoch": 5.342193571176263,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00025374975949471386,
      "loss": 5.1858,
      "mean_token_accuracy": 0.2245427429676056,
      "num_tokens": 112784986.0,
      "step": 60495
    },
    {
      "entropy": 6.022006559371948,
      "epoch": 5.342635111268103,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002537184089628241,
      "loss": 5.1964,
      "mean_token_accuracy": 0.22736707478761672,
      "num_tokens": 112792973.0,
      "step": 60500
    },
    {
      "entropy": 5.9884978294372555,
      "epoch": 5.343076651359944,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002536870588478287,
      "loss": 5.1594,
      "mean_token_accuracy": 0.2307494804263115,
      "num_tokens": 112801804.0,
      "step": 60505
    },
    {
      "entropy": 5.969391393661499,
      "epoch": 5.343518191451784,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002536557091503417,
      "loss": 5.1028,
      "mean_token_accuracy": 0.2338630422949791,
      "num_tokens": 112810487.0,
      "step": 60510
    },
    {
      "entropy": 6.043009662628174,
      "epoch": 5.343959731543624,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002536243598709772,
      "loss": 5.1623,
      "mean_token_accuracy": 0.23184710294008254,
      "num_tokens": 112819117.0,
      "step": 60515
    },
    {
      "entropy": 6.020269966125488,
      "epoch": 5.344401271635465,
      "grad_norm": 1.375,
      "learning_rate": 0.0002535930110103493,
      "loss": 5.1926,
      "mean_token_accuracy": 0.2270648419857025,
      "num_tokens": 112828013.0,
      "step": 60520
    },
    {
      "entropy": 6.052876901626587,
      "epoch": 5.344842811727305,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00025356166256907216,
      "loss": 5.17,
      "mean_token_accuracy": 0.23300502300262452,
      "num_tokens": 112836833.0,
      "step": 60525
    },
    {
      "entropy": 6.009183406829834,
      "epoch": 5.345284351819145,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002535303145477598,
      "loss": 5.2981,
      "mean_token_accuracy": 0.22231378853321077,
      "num_tokens": 112846543.0,
      "step": 60530
    },
    {
      "entropy": 5.97885479927063,
      "epoch": 5.345725891910986,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002534989669470265,
      "loss": 5.1257,
      "mean_token_accuracy": 0.23139157444238662,
      "num_tokens": 112854848.0,
      "step": 60535
    },
    {
      "entropy": 6.029416561126709,
      "epoch": 5.346167432002826,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002534676197674861,
      "loss": 5.2218,
      "mean_token_accuracy": 0.22208908796310425,
      "num_tokens": 112862965.0,
      "step": 60540
    },
    {
      "entropy": 5.986669635772705,
      "epoch": 5.346608972094666,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00025343627300975283,
      "loss": 5.0941,
      "mean_token_accuracy": 0.231774266064167,
      "num_tokens": 112872770.0,
      "step": 60545
    },
    {
      "entropy": 6.11428017616272,
      "epoch": 5.347050512186507,
      "grad_norm": 1.25,
      "learning_rate": 0.00025340492667444073,
      "loss": 5.2272,
      "mean_token_accuracy": 0.222216834127903,
      "num_tokens": 112882647.0,
      "step": 60550
    },
    {
      "entropy": 6.101854610443115,
      "epoch": 5.347492052278347,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00025337358076216377,
      "loss": 5.2819,
      "mean_token_accuracy": 0.21719713360071183,
      "num_tokens": 112892733.0,
      "step": 60555
    },
    {
      "entropy": 6.103162622451782,
      "epoch": 5.347933592370187,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002533422352735361,
      "loss": 5.3279,
      "mean_token_accuracy": 0.21013472378253936,
      "num_tokens": 112901707.0,
      "step": 60560
    },
    {
      "entropy": 6.035892295837402,
      "epoch": 5.348375132462028,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002533108902091718,
      "loss": 5.2993,
      "mean_token_accuracy": 0.21974569261074067,
      "num_tokens": 112911791.0,
      "step": 60565
    },
    {
      "entropy": 6.038190937042236,
      "epoch": 5.348816672553868,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00025327954556968467,
      "loss": 5.281,
      "mean_token_accuracy": 0.22160248458385468,
      "num_tokens": 112920854.0,
      "step": 60570
    },
    {
      "entropy": 6.049878931045532,
      "epoch": 5.349258212645708,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000253248201355689,
      "loss": 5.2347,
      "mean_token_accuracy": 0.21919489353895188,
      "num_tokens": 112930884.0,
      "step": 60575
    },
    {
      "entropy": 6.076404094696045,
      "epoch": 5.349699752737549,
      "grad_norm": 1.4375,
      "learning_rate": 0.00025321685756779863,
      "loss": 5.2579,
      "mean_token_accuracy": 0.2242264375090599,
      "num_tokens": 112940693.0,
      "step": 60580
    },
    {
      "entropy": 6.043005847930909,
      "epoch": 5.350141292829389,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00025318551420662765,
      "loss": 5.2691,
      "mean_token_accuracy": 0.2267079845070839,
      "num_tokens": 112949980.0,
      "step": 60585
    },
    {
      "entropy": 6.1070136547088625,
      "epoch": 5.350582832921229,
      "grad_norm": 1.265625,
      "learning_rate": 0.00025315417127279007,
      "loss": 5.3175,
      "mean_token_accuracy": 0.221352282166481,
      "num_tokens": 112960484.0,
      "step": 60590
    },
    {
      "entropy": 5.976442527770996,
      "epoch": 5.35102437301307,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002531228287668998,
      "loss": 5.026,
      "mean_token_accuracy": 0.24636048078536987,
      "num_tokens": 112969663.0,
      "step": 60595
    },
    {
      "entropy": 5.963383674621582,
      "epoch": 5.35146591310491,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00025309148668957096,
      "loss": 5.137,
      "mean_token_accuracy": 0.2374045416712761,
      "num_tokens": 112978559.0,
      "step": 60600
    },
    {
      "entropy": 5.94689621925354,
      "epoch": 5.35190745319675,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002530601450414174,
      "loss": 5.137,
      "mean_token_accuracy": 0.22589555233716965,
      "num_tokens": 112988153.0,
      "step": 60605
    },
    {
      "entropy": 6.03077073097229,
      "epoch": 5.35234899328859,
      "grad_norm": 1.28125,
      "learning_rate": 0.00025302880382305296,
      "loss": 5.2021,
      "mean_token_accuracy": 0.2305827796459198,
      "num_tokens": 112998543.0,
      "step": 60610
    },
    {
      "entropy": 6.081029224395752,
      "epoch": 5.352790533380431,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00025299746303509196,
      "loss": 5.176,
      "mean_token_accuracy": 0.22912287563085557,
      "num_tokens": 113006525.0,
      "step": 60615
    },
    {
      "entropy": 6.043404531478882,
      "epoch": 5.353232073472271,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002529661226781481,
      "loss": 5.2531,
      "mean_token_accuracy": 0.2211180567741394,
      "num_tokens": 113015558.0,
      "step": 60620
    },
    {
      "entropy": 5.98813157081604,
      "epoch": 5.353673613564112,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002529347827528354,
      "loss": 5.1433,
      "mean_token_accuracy": 0.23252544701099395,
      "num_tokens": 113025313.0,
      "step": 60625
    },
    {
      "entropy": 6.124172210693359,
      "epoch": 5.354115153655952,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002529034432597677,
      "loss": 5.253,
      "mean_token_accuracy": 0.2206597864627838,
      "num_tokens": 113034243.0,
      "step": 60630
    },
    {
      "entropy": 6.0313033103942875,
      "epoch": 5.354556693747792,
      "grad_norm": 1.328125,
      "learning_rate": 0.00025287210419955893,
      "loss": 5.2394,
      "mean_token_accuracy": 0.22442435324192048,
      "num_tokens": 113044026.0,
      "step": 60635
    },
    {
      "entropy": 6.0416727542877195,
      "epoch": 5.354998233839632,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002528407655728231,
      "loss": 5.2701,
      "mean_token_accuracy": 0.22278929501771927,
      "num_tokens": 113052974.0,
      "step": 60640
    },
    {
      "entropy": 6.06078872680664,
      "epoch": 5.355439773931473,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002528094273801741,
      "loss": 5.2175,
      "mean_token_accuracy": 0.2249728947877884,
      "num_tokens": 113062824.0,
      "step": 60645
    },
    {
      "entropy": 6.060472202301026,
      "epoch": 5.355881314023313,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002527780896222258,
      "loss": 5.2704,
      "mean_token_accuracy": 0.2224893182516098,
      "num_tokens": 113071926.0,
      "step": 60650
    },
    {
      "entropy": 6.045025730133057,
      "epoch": 5.356322854115153,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000252746752299592,
      "loss": 5.296,
      "mean_token_accuracy": 0.21797877699136733,
      "num_tokens": 113081203.0,
      "step": 60655
    },
    {
      "entropy": 6.007857418060302,
      "epoch": 5.356764394206994,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00025271541541288674,
      "loss": 5.1871,
      "mean_token_accuracy": 0.2263427421450615,
      "num_tokens": 113090832.0,
      "step": 60660
    },
    {
      "entropy": 6.117470979690552,
      "epoch": 5.357205934298834,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002526840789627239,
      "loss": 5.2094,
      "mean_token_accuracy": 0.22414618581533433,
      "num_tokens": 113100224.0,
      "step": 60665
    },
    {
      "entropy": 6.082813501358032,
      "epoch": 5.357647474390674,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002526527429497171,
      "loss": 5.2685,
      "mean_token_accuracy": 0.21953411847352983,
      "num_tokens": 113110127.0,
      "step": 60670
    },
    {
      "entropy": 6.095746088027954,
      "epoch": 5.358089014482515,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00025262140737448036,
      "loss": 5.2995,
      "mean_token_accuracy": 0.2225226566195488,
      "num_tokens": 113120096.0,
      "step": 60675
    },
    {
      "entropy": 6.039718580245972,
      "epoch": 5.358530554574355,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00025259007223762763,
      "loss": 5.1838,
      "mean_token_accuracy": 0.22701687216758729,
      "num_tokens": 113128641.0,
      "step": 60680
    },
    {
      "entropy": 6.094959497451782,
      "epoch": 5.358972094666195,
      "grad_norm": 1.453125,
      "learning_rate": 0.00025255873753977264,
      "loss": 5.3336,
      "mean_token_accuracy": 0.21049435138702394,
      "num_tokens": 113136938.0,
      "step": 60685
    },
    {
      "entropy": 6.055631256103515,
      "epoch": 5.359413634758036,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00025252740328152916,
      "loss": 5.2218,
      "mean_token_accuracy": 0.22239466160535812,
      "num_tokens": 113146440.0,
      "step": 60690
    },
    {
      "entropy": 6.132568788528443,
      "epoch": 5.359855174849876,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025249606946351117,
      "loss": 5.2603,
      "mean_token_accuracy": 0.22172476053237916,
      "num_tokens": 113155910.0,
      "step": 60695
    },
    {
      "entropy": 6.1214131832122805,
      "epoch": 5.3602967149417164,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002524647360863322,
      "loss": 5.2358,
      "mean_token_accuracy": 0.2248570516705513,
      "num_tokens": 113164676.0,
      "step": 60700
    },
    {
      "entropy": 6.0614067077636715,
      "epoch": 5.360738255033557,
      "grad_norm": 1.375,
      "learning_rate": 0.00025243340315060636,
      "loss": 5.2126,
      "mean_token_accuracy": 0.22364506721496583,
      "num_tokens": 113173867.0,
      "step": 60705
    },
    {
      "entropy": 6.012321424484253,
      "epoch": 5.361179795125397,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025240207065694724,
      "loss": 5.2263,
      "mean_token_accuracy": 0.22106801718473434,
      "num_tokens": 113182221.0,
      "step": 60710
    },
    {
      "entropy": 6.037744855880737,
      "epoch": 5.3616213352172375,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00025237073860596884,
      "loss": 5.2271,
      "mean_token_accuracy": 0.22292625308036804,
      "num_tokens": 113191558.0,
      "step": 60715
    },
    {
      "entropy": 6.0166667938232425,
      "epoch": 5.362062875309078,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002523394069982846,
      "loss": 5.2404,
      "mean_token_accuracy": 0.23205175399780273,
      "num_tokens": 113200426.0,
      "step": 60720
    },
    {
      "entropy": 6.110444021224976,
      "epoch": 5.3625044154009185,
      "grad_norm": 1.5,
      "learning_rate": 0.0002523080758345086,
      "loss": 5.203,
      "mean_token_accuracy": 0.22087440341711045,
      "num_tokens": 113209504.0,
      "step": 60725
    },
    {
      "entropy": 6.053078889846802,
      "epoch": 5.3629459554927585,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002522767451152545,
      "loss": 5.2534,
      "mean_token_accuracy": 0.21976675987243652,
      "num_tokens": 113218877.0,
      "step": 60730
    },
    {
      "entropy": 6.083275604248047,
      "epoch": 5.3633874955845995,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002522454148411361,
      "loss": 5.376,
      "mean_token_accuracy": 0.21151868849992753,
      "num_tokens": 113229079.0,
      "step": 60735
    },
    {
      "entropy": 6.08402647972107,
      "epoch": 5.3638290356764395,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025221408501276696,
      "loss": 5.3177,
      "mean_token_accuracy": 0.21968963146209716,
      "num_tokens": 113238622.0,
      "step": 60740
    },
    {
      "entropy": 6.116844129562378,
      "epoch": 5.36427057576828,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025218275563076097,
      "loss": 5.2296,
      "mean_token_accuracy": 0.21649052649736406,
      "num_tokens": 113248566.0,
      "step": 60745
    },
    {
      "entropy": 6.100872468948364,
      "epoch": 5.3647121158601205,
      "grad_norm": 1.390625,
      "learning_rate": 0.00025215142669573185,
      "loss": 5.2282,
      "mean_token_accuracy": 0.23371145278215408,
      "num_tokens": 113258367.0,
      "step": 60750
    },
    {
      "entropy": 6.031077146530151,
      "epoch": 5.365153655951961,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002521200982082933,
      "loss": 5.2185,
      "mean_token_accuracy": 0.21950019747018815,
      "num_tokens": 113267654.0,
      "step": 60755
    },
    {
      "entropy": 6.076569128036499,
      "epoch": 5.365595196043801,
      "grad_norm": 1.46875,
      "learning_rate": 0.00025208877016905904,
      "loss": 5.2592,
      "mean_token_accuracy": 0.2217062532901764,
      "num_tokens": 113275997.0,
      "step": 60760
    },
    {
      "entropy": 6.038567066192627,
      "epoch": 5.366036736135641,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002520574425786426,
      "loss": 5.2089,
      "mean_token_accuracy": 0.22781105786561967,
      "num_tokens": 113285494.0,
      "step": 60765
    },
    {
      "entropy": 6.082100248336792,
      "epoch": 5.366478276227482,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025202611543765794,
      "loss": 5.2394,
      "mean_token_accuracy": 0.23139816373586655,
      "num_tokens": 113295818.0,
      "step": 60770
    },
    {
      "entropy": 6.025233316421509,
      "epoch": 5.366919816319322,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00025199478874671856,
      "loss": 5.2709,
      "mean_token_accuracy": 0.22725419253110885,
      "num_tokens": 113305327.0,
      "step": 60775
    },
    {
      "entropy": 6.058161115646362,
      "epoch": 5.367361356411162,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002519634625064382,
      "loss": 5.1494,
      "mean_token_accuracy": 0.23073663413524628,
      "num_tokens": 113314578.0,
      "step": 60780
    },
    {
      "entropy": 6.020532417297363,
      "epoch": 5.367802896503003,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025193213671743047,
      "loss": 5.1572,
      "mean_token_accuracy": 0.23062872141599655,
      "num_tokens": 113323700.0,
      "step": 60785
    },
    {
      "entropy": 6.017740678787232,
      "epoch": 5.368244436594843,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025190081138030913,
      "loss": 5.1547,
      "mean_token_accuracy": 0.22816307246685028,
      "num_tokens": 113333051.0,
      "step": 60790
    },
    {
      "entropy": 6.077167940139771,
      "epoch": 5.368685976686683,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002518694864956877,
      "loss": 5.2924,
      "mean_token_accuracy": 0.21839062869548798,
      "num_tokens": 113342299.0,
      "step": 60795
    },
    {
      "entropy": 6.079062509536743,
      "epoch": 5.369127516778524,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025183816206417994,
      "loss": 5.2437,
      "mean_token_accuracy": 0.228866109251976,
      "num_tokens": 113351858.0,
      "step": 60800
    },
    {
      "entropy": 6.047565221786499,
      "epoch": 5.369569056870364,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00025180683808639936,
      "loss": 5.2349,
      "mean_token_accuracy": 0.22546037286520004,
      "num_tokens": 113360509.0,
      "step": 60805
    },
    {
      "entropy": 6.107026386260986,
      "epoch": 5.370010596962204,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002517755145629596,
      "loss": 5.2791,
      "mean_token_accuracy": 0.22445527762174605,
      "num_tokens": 113370790.0,
      "step": 60810
    },
    {
      "entropy": 6.055635547637939,
      "epoch": 5.370452137054045,
      "grad_norm": 1.4375,
      "learning_rate": 0.00025174419149447434,
      "loss": 5.2582,
      "mean_token_accuracy": 0.22846523076295852,
      "num_tokens": 113379712.0,
      "step": 60815
    },
    {
      "entropy": 5.942818117141724,
      "epoch": 5.370893677145885,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002517128688815572,
      "loss": 5.066,
      "mean_token_accuracy": 0.2394164800643921,
      "num_tokens": 113388156.0,
      "step": 60820
    },
    {
      "entropy": 6.016356515884399,
      "epoch": 5.371335217237725,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002516815467248217,
      "loss": 5.1931,
      "mean_token_accuracy": 0.22139729112386702,
      "num_tokens": 113398817.0,
      "step": 60825
    },
    {
      "entropy": 6.094228458404541,
      "epoch": 5.371776757329566,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00025165022502488134,
      "loss": 5.2824,
      "mean_token_accuracy": 0.2213185116648674,
      "num_tokens": 113408862.0,
      "step": 60830
    },
    {
      "entropy": 6.043175554275512,
      "epoch": 5.372218297421406,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00025161890378234985,
      "loss": 5.1475,
      "mean_token_accuracy": 0.24117099791765212,
      "num_tokens": 113416632.0,
      "step": 60835
    },
    {
      "entropy": 6.067741632461548,
      "epoch": 5.372659837513246,
      "grad_norm": 1.453125,
      "learning_rate": 0.00025158758299784076,
      "loss": 5.2393,
      "mean_token_accuracy": 0.2247203692793846,
      "num_tokens": 113425050.0,
      "step": 60840
    },
    {
      "entropy": 5.905238246917724,
      "epoch": 5.373101377605087,
      "grad_norm": 1.4375,
      "learning_rate": 0.00025155626267196757,
      "loss": 5.0789,
      "mean_token_accuracy": 0.2402191072702408,
      "num_tokens": 113433467.0,
      "step": 60845
    },
    {
      "entropy": 6.02354998588562,
      "epoch": 5.373542917696927,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025152494280534387,
      "loss": 5.2625,
      "mean_token_accuracy": 0.2214444324374199,
      "num_tokens": 113443606.0,
      "step": 60850
    },
    {
      "entropy": 5.9352034568786625,
      "epoch": 5.373984457788767,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002514936233985832,
      "loss": 5.0649,
      "mean_token_accuracy": 0.23993799835443497,
      "num_tokens": 113452845.0,
      "step": 60855
    },
    {
      "entropy": 6.0492835521698,
      "epoch": 5.374425997880608,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002514623044522991,
      "loss": 5.2054,
      "mean_token_accuracy": 0.22666447311639787,
      "num_tokens": 113461092.0,
      "step": 60860
    },
    {
      "entropy": 6.0348598003387455,
      "epoch": 5.374867537972448,
      "grad_norm": 1.421875,
      "learning_rate": 0.000251430985967105,
      "loss": 5.2204,
      "mean_token_accuracy": 0.22266903817653655,
      "num_tokens": 113469971.0,
      "step": 60865
    },
    {
      "entropy": 6.061641502380371,
      "epoch": 5.375309078064288,
      "grad_norm": 1.5,
      "learning_rate": 0.00025139966794361456,
      "loss": 5.2046,
      "mean_token_accuracy": 0.22689270675182344,
      "num_tokens": 113478597.0,
      "step": 60870
    },
    {
      "entropy": 6.066545534133911,
      "epoch": 5.375750618156129,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025136835038244113,
      "loss": 5.2288,
      "mean_token_accuracy": 0.2192302644252777,
      "num_tokens": 113488709.0,
      "step": 60875
    },
    {
      "entropy": 6.097975254058838,
      "epoch": 5.376192158247969,
      "grad_norm": 1.453125,
      "learning_rate": 0.00025133703328419834,
      "loss": 5.289,
      "mean_token_accuracy": 0.22271567583084106,
      "num_tokens": 113498560.0,
      "step": 60880
    },
    {
      "entropy": 6.043340635299683,
      "epoch": 5.376633698339809,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00025130571664949957,
      "loss": 5.2446,
      "mean_token_accuracy": 0.22285159975290297,
      "num_tokens": 113507741.0,
      "step": 60885
    },
    {
      "entropy": 6.059183025360108,
      "epoch": 5.377075238431649,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002512744004789584,
      "loss": 5.2696,
      "mean_token_accuracy": 0.2143764466047287,
      "num_tokens": 113517407.0,
      "step": 60890
    },
    {
      "entropy": 6.046245050430298,
      "epoch": 5.37751677852349,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002512430847731881,
      "loss": 5.279,
      "mean_token_accuracy": 0.2186632826924324,
      "num_tokens": 113527163.0,
      "step": 60895
    },
    {
      "entropy": 6.035612487792969,
      "epoch": 5.37795831861533,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002512117695328023,
      "loss": 5.2091,
      "mean_token_accuracy": 0.22814281284809113,
      "num_tokens": 113536980.0,
      "step": 60900
    },
    {
      "entropy": 6.027620649337768,
      "epoch": 5.378399858707171,
      "grad_norm": 1.53125,
      "learning_rate": 0.00025118045475841446,
      "loss": 5.1754,
      "mean_token_accuracy": 0.23404147624969482,
      "num_tokens": 113546123.0,
      "step": 60905
    },
    {
      "entropy": 6.123595666885376,
      "epoch": 5.378841398799011,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000251149140450638,
      "loss": 5.2978,
      "mean_token_accuracy": 0.21455518752336503,
      "num_tokens": 113555978.0,
      "step": 60910
    },
    {
      "entropy": 6.076695585250855,
      "epoch": 5.379282938890851,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002511178266100861,
      "loss": 5.1847,
      "mean_token_accuracy": 0.22494272291660308,
      "num_tokens": 113565259.0,
      "step": 60915
    },
    {
      "entropy": 6.095316648483276,
      "epoch": 5.379724478982691,
      "grad_norm": 1.46875,
      "learning_rate": 0.00025108651323737266,
      "loss": 5.2422,
      "mean_token_accuracy": 0.22662412673234938,
      "num_tokens": 113575053.0,
      "step": 60920
    },
    {
      "entropy": 6.066705560684204,
      "epoch": 5.380166019074532,
      "grad_norm": 1.375,
      "learning_rate": 0.00025105520033311075,
      "loss": 5.2843,
      "mean_token_accuracy": 0.22195328027009964,
      "num_tokens": 113584524.0,
      "step": 60925
    },
    {
      "entropy": 6.005106687545776,
      "epoch": 5.380607559166372,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002510238878979138,
      "loss": 5.1963,
      "mean_token_accuracy": 0.22293533086776735,
      "num_tokens": 113593964.0,
      "step": 60930
    },
    {
      "entropy": 6.031959533691406,
      "epoch": 5.381049099258212,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002509925759323953,
      "loss": 5.242,
      "mean_token_accuracy": 0.22517872005701065,
      "num_tokens": 113603932.0,
      "step": 60935
    },
    {
      "entropy": 6.0532325267791744,
      "epoch": 5.381490639350053,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002509612644371685,
      "loss": 5.1338,
      "mean_token_accuracy": 0.23277105242013932,
      "num_tokens": 113612730.0,
      "step": 60940
    },
    {
      "entropy": 6.070697069168091,
      "epoch": 5.381932179441893,
      "grad_norm": 1.4375,
      "learning_rate": 0.00025092995341284693,
      "loss": 5.2188,
      "mean_token_accuracy": 0.21743446588516235,
      "num_tokens": 113620974.0,
      "step": 60945
    },
    {
      "entropy": 5.868153810501099,
      "epoch": 5.382373719533733,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002508986428600439,
      "loss": 5.1323,
      "mean_token_accuracy": 0.24624827057123183,
      "num_tokens": 113630398.0,
      "step": 60950
    },
    {
      "entropy": 6.027488803863525,
      "epoch": 5.382815259625574,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002508673327793727,
      "loss": 5.2376,
      "mean_token_accuracy": 0.227632075548172,
      "num_tokens": 113641168.0,
      "step": 60955
    },
    {
      "entropy": 6.060127305984497,
      "epoch": 5.383256799717414,
      "grad_norm": 1.421875,
      "learning_rate": 0.00025083602317144664,
      "loss": 5.1883,
      "mean_token_accuracy": 0.23258500546216965,
      "num_tokens": 113650288.0,
      "step": 60960
    },
    {
      "entropy": 6.074800825119018,
      "epoch": 5.383698339809254,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002508047140368793,
      "loss": 5.242,
      "mean_token_accuracy": 0.22469713538885117,
      "num_tokens": 113660345.0,
      "step": 60965
    },
    {
      "entropy": 6.039338254928589,
      "epoch": 5.384139879901095,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00025077340537628374,
      "loss": 5.1739,
      "mean_token_accuracy": 0.2278781682252884,
      "num_tokens": 113669533.0,
      "step": 60970
    },
    {
      "entropy": 6.122875499725342,
      "epoch": 5.384581419992935,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002507420971902735,
      "loss": 5.2557,
      "mean_token_accuracy": 0.2184277892112732,
      "num_tokens": 113678770.0,
      "step": 60975
    },
    {
      "entropy": 6.0989316463470455,
      "epoch": 5.385022960084775,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00025071078947946164,
      "loss": 5.2702,
      "mean_token_accuracy": 0.22260887622833253,
      "num_tokens": 113688187.0,
      "step": 60980
    },
    {
      "entropy": 6.054841184616089,
      "epoch": 5.385464500176616,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00025067948224446167,
      "loss": 5.2404,
      "mean_token_accuracy": 0.22236336767673492,
      "num_tokens": 113697030.0,
      "step": 60985
    },
    {
      "entropy": 6.016594076156617,
      "epoch": 5.385906040268456,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00025064817548588685,
      "loss": 5.1923,
      "mean_token_accuracy": 0.23107357174158097,
      "num_tokens": 113705769.0,
      "step": 60990
    },
    {
      "entropy": 6.08399772644043,
      "epoch": 5.3863475803602965,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002506168692043505,
      "loss": 5.2149,
      "mean_token_accuracy": 0.22976410388946533,
      "num_tokens": 113714660.0,
      "step": 60995
    },
    {
      "entropy": 5.994329929351807,
      "epoch": 5.386789120452137,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00025058556340046566,
      "loss": 5.1723,
      "mean_token_accuracy": 0.2272121086716652,
      "num_tokens": 113724537.0,
      "step": 61000
    },
    {
      "entropy": 5.959210920333862,
      "epoch": 5.3872306605439775,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002505542580748458,
      "loss": 5.1277,
      "mean_token_accuracy": 0.23474522680044174,
      "num_tokens": 113733390.0,
      "step": 61005
    },
    {
      "entropy": 5.99512209892273,
      "epoch": 5.3876722006358175,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002505229532281041,
      "loss": 5.181,
      "mean_token_accuracy": 0.2277516558766365,
      "num_tokens": 113742650.0,
      "step": 61010
    },
    {
      "entropy": 6.03376259803772,
      "epoch": 5.3881137407276585,
      "grad_norm": 1.421875,
      "learning_rate": 0.00025049164886085387,
      "loss": 5.2797,
      "mean_token_accuracy": 0.22163023203611373,
      "num_tokens": 113751646.0,
      "step": 61015
    },
    {
      "entropy": 6.03362627029419,
      "epoch": 5.3885552808194985,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002504603449737082,
      "loss": 5.1602,
      "mean_token_accuracy": 0.2321920871734619,
      "num_tokens": 113760396.0,
      "step": 61020
    },
    {
      "entropy": 6.142240571975708,
      "epoch": 5.388996820911339,
      "grad_norm": 1.375,
      "learning_rate": 0.0002504290415672805,
      "loss": 5.309,
      "mean_token_accuracy": 0.21427830308675766,
      "num_tokens": 113770015.0,
      "step": 61025
    },
    {
      "entropy": 6.050463771820068,
      "epoch": 5.3894383610031795,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002503977386421839,
      "loss": 5.2808,
      "mean_token_accuracy": 0.21361314058303832,
      "num_tokens": 113780204.0,
      "step": 61030
    },
    {
      "entropy": 5.967915773391724,
      "epoch": 5.38987990109502,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002503664361990316,
      "loss": 5.1626,
      "mean_token_accuracy": 0.234660342335701,
      "num_tokens": 113789437.0,
      "step": 61035
    },
    {
      "entropy": 6.0666858673095705,
      "epoch": 5.39032144118686,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00025033513423843685,
      "loss": 5.2728,
      "mean_token_accuracy": 0.22035941034555434,
      "num_tokens": 113798624.0,
      "step": 61040
    },
    {
      "entropy": 6.005956172943115,
      "epoch": 5.3907629812787,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025030383276101277,
      "loss": 5.1761,
      "mean_token_accuracy": 0.22329329550266266,
      "num_tokens": 113807128.0,
      "step": 61045
    },
    {
      "entropy": 6.061503076553345,
      "epoch": 5.391204521370541,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002502725317673725,
      "loss": 5.1795,
      "mean_token_accuracy": 0.2304622307419777,
      "num_tokens": 113815183.0,
      "step": 61050
    },
    {
      "entropy": 6.06678729057312,
      "epoch": 5.391646061462381,
      "grad_norm": 1.390625,
      "learning_rate": 0.00025024123125812924,
      "loss": 5.2565,
      "mean_token_accuracy": 0.22017364501953124,
      "num_tokens": 113824972.0,
      "step": 61055
    },
    {
      "entropy": 6.064811897277832,
      "epoch": 5.392087601554221,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002502099312338963,
      "loss": 5.2005,
      "mean_token_accuracy": 0.2295891210436821,
      "num_tokens": 113833587.0,
      "step": 61060
    },
    {
      "entropy": 5.975347805023193,
      "epoch": 5.392529141646062,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002501786316952867,
      "loss": 5.1486,
      "mean_token_accuracy": 0.23780195116996766,
      "num_tokens": 113842375.0,
      "step": 61065
    },
    {
      "entropy": 6.028278589248657,
      "epoch": 5.392970681737902,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002501473326429135,
      "loss": 5.2038,
      "mean_token_accuracy": 0.22557576298713683,
      "num_tokens": 113851889.0,
      "step": 61070
    },
    {
      "entropy": 6.082150840759278,
      "epoch": 5.393412221829742,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00025011603407738993,
      "loss": 5.2626,
      "mean_token_accuracy": 0.21641744673252106,
      "num_tokens": 113861065.0,
      "step": 61075
    },
    {
      "entropy": 6.042938995361328,
      "epoch": 5.393853761921583,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00025008473599932916,
      "loss": 5.1479,
      "mean_token_accuracy": 0.22776872664690018,
      "num_tokens": 113870164.0,
      "step": 61080
    },
    {
      "entropy": 5.992004442214966,
      "epoch": 5.394295302013423,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002500534384093442,
      "loss": 5.1711,
      "mean_token_accuracy": 0.2272490456700325,
      "num_tokens": 113879715.0,
      "step": 61085
    },
    {
      "entropy": 5.990101718902588,
      "epoch": 5.394736842105263,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002500221413080482,
      "loss": 5.0908,
      "mean_token_accuracy": 0.23407143354415894,
      "num_tokens": 113888139.0,
      "step": 61090
    },
    {
      "entropy": 5.9985963821411135,
      "epoch": 5.395178382197104,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00024999084469605427,
      "loss": 5.2493,
      "mean_token_accuracy": 0.22340281158685685,
      "num_tokens": 113897748.0,
      "step": 61095
    },
    {
      "entropy": 5.992952680587768,
      "epoch": 5.395619922288944,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002499595485739754,
      "loss": 5.2327,
      "mean_token_accuracy": 0.22736496180295945,
      "num_tokens": 113906785.0,
      "step": 61100
    },
    {
      "entropy": 6.079975509643555,
      "epoch": 5.396061462380784,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00024992825294242483,
      "loss": 5.2402,
      "mean_token_accuracy": 0.22204923778772354,
      "num_tokens": 113916367.0,
      "step": 61105
    },
    {
      "entropy": 5.971091842651367,
      "epoch": 5.396503002472625,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002498969578020155,
      "loss": 5.0996,
      "mean_token_accuracy": 0.23154597878456115,
      "num_tokens": 113925058.0,
      "step": 61110
    },
    {
      "entropy": 6.086994647979736,
      "epoch": 5.396944542564465,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002498656631533604,
      "loss": 5.2652,
      "mean_token_accuracy": 0.22575172334909438,
      "num_tokens": 113933641.0,
      "step": 61115
    },
    {
      "entropy": 6.0245680809021,
      "epoch": 5.397386082656305,
      "grad_norm": 1.453125,
      "learning_rate": 0.00024983436899707275,
      "loss": 5.206,
      "mean_token_accuracy": 0.2204352453351021,
      "num_tokens": 113943681.0,
      "step": 61120
    },
    {
      "entropy": 6.057355690002441,
      "epoch": 5.397827622748146,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00024980307533376553,
      "loss": 5.1807,
      "mean_token_accuracy": 0.2260803386569023,
      "num_tokens": 113952975.0,
      "step": 61125
    },
    {
      "entropy": 5.959685659408569,
      "epoch": 5.398269162839986,
      "grad_norm": 1.5,
      "learning_rate": 0.0002497717821640517,
      "loss": 5.1314,
      "mean_token_accuracy": 0.2413389652967453,
      "num_tokens": 113962064.0,
      "step": 61130
    },
    {
      "entropy": 6.046669578552246,
      "epoch": 5.398710702931826,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00024974048948854424,
      "loss": 5.3169,
      "mean_token_accuracy": 0.215122090280056,
      "num_tokens": 113970904.0,
      "step": 61135
    },
    {
      "entropy": 6.085514783859253,
      "epoch": 5.399152243023667,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002497091973078563,
      "loss": 5.2658,
      "mean_token_accuracy": 0.21589116454124452,
      "num_tokens": 113979771.0,
      "step": 61140
    },
    {
      "entropy": 6.049578762054443,
      "epoch": 5.399593783115507,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002496779056226008,
      "loss": 5.2744,
      "mean_token_accuracy": 0.221530182659626,
      "num_tokens": 113989747.0,
      "step": 61145
    },
    {
      "entropy": 6.0580321788787845,
      "epoch": 5.400035323207347,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002496466144333907,
      "loss": 5.1895,
      "mean_token_accuracy": 0.23021225035190582,
      "num_tokens": 113998742.0,
      "step": 61150
    },
    {
      "entropy": 6.133547353744507,
      "epoch": 5.400476863299188,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024961532374083895,
      "loss": 5.3031,
      "mean_token_accuracy": 0.21403149515390396,
      "num_tokens": 114008758.0,
      "step": 61155
    },
    {
      "entropy": 6.138278532028198,
      "epoch": 5.400918403391028,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002495840335455587,
      "loss": 5.2643,
      "mean_token_accuracy": 0.22500773072242736,
      "num_tokens": 114017658.0,
      "step": 61160
    },
    {
      "entropy": 6.060484552383423,
      "epoch": 5.401359943482868,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002495527438481627,
      "loss": 5.2512,
      "mean_token_accuracy": 0.22961549162864686,
      "num_tokens": 114027715.0,
      "step": 61165
    },
    {
      "entropy": 6.049768543243408,
      "epoch": 5.401801483574708,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024952145464926404,
      "loss": 5.2064,
      "mean_token_accuracy": 0.22737342715263367,
      "num_tokens": 114037113.0,
      "step": 61170
    },
    {
      "entropy": 6.07215051651001,
      "epoch": 5.402243023666549,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024949016594947557,
      "loss": 5.2515,
      "mean_token_accuracy": 0.22548391073942184,
      "num_tokens": 114046907.0,
      "step": 61175
    },
    {
      "entropy": 6.057642650604248,
      "epoch": 5.402684563758389,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002494588777494101,
      "loss": 5.2433,
      "mean_token_accuracy": 0.23212765008211136,
      "num_tokens": 114056673.0,
      "step": 61180
    },
    {
      "entropy": 6.056851387023926,
      "epoch": 5.40312610385023,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002494275900496808,
      "loss": 5.2666,
      "mean_token_accuracy": 0.2169455274939537,
      "num_tokens": 114066880.0,
      "step": 61185
    },
    {
      "entropy": 6.090019798278808,
      "epoch": 5.40356764394207,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002493963028509005,
      "loss": 5.2473,
      "mean_token_accuracy": 0.2206387996673584,
      "num_tokens": 114077161.0,
      "step": 61190
    },
    {
      "entropy": 6.071571588516235,
      "epoch": 5.40400918403391,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024936501615368205,
      "loss": 5.2413,
      "mean_token_accuracy": 0.2282167300581932,
      "num_tokens": 114086356.0,
      "step": 61195
    },
    {
      "entropy": 6.031814765930176,
      "epoch": 5.40445072412575,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002493337299586383,
      "loss": 5.1352,
      "mean_token_accuracy": 0.23052586168050765,
      "num_tokens": 114095745.0,
      "step": 61200
    },
    {
      "entropy": 5.976234006881714,
      "epoch": 5.404892264217591,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002493024442663822,
      "loss": 5.2098,
      "mean_token_accuracy": 0.2345723405480385,
      "num_tokens": 114104858.0,
      "step": 61205
    },
    {
      "entropy": 6.036595869064331,
      "epoch": 5.405333804309431,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024927115907752667,
      "loss": 5.1914,
      "mean_token_accuracy": 0.2283271297812462,
      "num_tokens": 114115997.0,
      "step": 61210
    },
    {
      "entropy": 6.1234198093414305,
      "epoch": 5.405775344401271,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002492398743926844,
      "loss": 5.2535,
      "mean_token_accuracy": 0.21597384214401244,
      "num_tokens": 114125225.0,
      "step": 61215
    },
    {
      "entropy": 6.060633993148803,
      "epoch": 5.406216884493112,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024920859021246837,
      "loss": 5.1937,
      "mean_token_accuracy": 0.21904040575027467,
      "num_tokens": 114135831.0,
      "step": 61220
    },
    {
      "entropy": 6.054720211029053,
      "epoch": 5.406658424584952,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024917730653749147,
      "loss": 5.2227,
      "mean_token_accuracy": 0.22428526133298873,
      "num_tokens": 114145026.0,
      "step": 61225
    },
    {
      "entropy": 6.01779260635376,
      "epoch": 5.407099964676792,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002491460233683663,
      "loss": 5.1906,
      "mean_token_accuracy": 0.22922585904598236,
      "num_tokens": 114153371.0,
      "step": 61230
    },
    {
      "entropy": 5.999640226364136,
      "epoch": 5.407541504768633,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000249114740705706,
      "loss": 5.1464,
      "mean_token_accuracy": 0.2311359867453575,
      "num_tokens": 114162399.0,
      "step": 61235
    },
    {
      "entropy": 6.055850028991699,
      "epoch": 5.407983044860473,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024908345855012315,
      "loss": 5.2739,
      "mean_token_accuracy": 0.22438858896493913,
      "num_tokens": 114172333.0,
      "step": 61240
    },
    {
      "entropy": 6.055154418945312,
      "epoch": 5.408424584952313,
      "grad_norm": 1.5,
      "learning_rate": 0.00024905217690223054,
      "loss": 5.2567,
      "mean_token_accuracy": 0.21823464184999466,
      "num_tokens": 114181330.0,
      "step": 61245
    },
    {
      "entropy": 5.96801791191101,
      "epoch": 5.408866125044154,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024902089576264106,
      "loss": 5.0657,
      "mean_token_accuracy": 0.23676405102014542,
      "num_tokens": 114190444.0,
      "step": 61250
    },
    {
      "entropy": 6.120291090011596,
      "epoch": 5.409307665135994,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024898961513196747,
      "loss": 5.2369,
      "mean_token_accuracy": 0.22262672632932662,
      "num_tokens": 114199817.0,
      "step": 61255
    },
    {
      "entropy": 6.036653661727906,
      "epoch": 5.409749205227834,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002489583350108226,
      "loss": 5.182,
      "mean_token_accuracy": 0.23288676887750626,
      "num_tokens": 114209446.0,
      "step": 61260
    },
    {
      "entropy": 6.057039737701416,
      "epoch": 5.410190745319675,
      "grad_norm": 1.421875,
      "learning_rate": 0.000248927055399819,
      "loss": 5.3391,
      "mean_token_accuracy": 0.21664735674858093,
      "num_tokens": 114217940.0,
      "step": 61265
    },
    {
      "entropy": 5.971650171279907,
      "epoch": 5.410632285411515,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002488957762995696,
      "loss": 5.1977,
      "mean_token_accuracy": 0.22750129103660582,
      "num_tokens": 114227857.0,
      "step": 61270
    },
    {
      "entropy": 6.063978052139282,
      "epoch": 5.4110738255033555,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002488644977106871,
      "loss": 5.2864,
      "mean_token_accuracy": 0.22483311146497725,
      "num_tokens": 114236656.0,
      "step": 61275
    },
    {
      "entropy": 6.072184038162232,
      "epoch": 5.411515365595196,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002488332196337843,
      "loss": 5.2689,
      "mean_token_accuracy": 0.21962662637233735,
      "num_tokens": 114246372.0,
      "step": 61280
    },
    {
      "entropy": 6.103810453414917,
      "epoch": 5.4119569056870365,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002488019420694738,
      "loss": 5.2188,
      "mean_token_accuracy": 0.2260892480611801,
      "num_tokens": 114254741.0,
      "step": 61285
    },
    {
      "entropy": 6.112741136550904,
      "epoch": 5.4123984457788765,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024877066501836827,
      "loss": 5.2394,
      "mean_token_accuracy": 0.22609854340553284,
      "num_tokens": 114263760.0,
      "step": 61290
    },
    {
      "entropy": 6.089585542678833,
      "epoch": 5.4128399858707175,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002487393884810806,
      "loss": 5.2375,
      "mean_token_accuracy": 0.2258844032883644,
      "num_tokens": 114273524.0,
      "step": 61295
    },
    {
      "entropy": 6.00205626487732,
      "epoch": 5.4132815259625575,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00024870811245822333,
      "loss": 5.1871,
      "mean_token_accuracy": 0.22380489110946655,
      "num_tokens": 114282530.0,
      "step": 61300
    },
    {
      "entropy": 6.003279113769532,
      "epoch": 5.413723066054398,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024867683695040923,
      "loss": 5.2186,
      "mean_token_accuracy": 0.22596501410007477,
      "num_tokens": 114291629.0,
      "step": 61305
    },
    {
      "entropy": 6.01701774597168,
      "epoch": 5.4141646061462385,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00024864556195825085,
      "loss": 5.146,
      "mean_token_accuracy": 0.23264367282390594,
      "num_tokens": 114300109.0,
      "step": 61310
    },
    {
      "entropy": 5.986947822570801,
      "epoch": 5.414606146238079,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024861428748236097,
      "loss": 5.1538,
      "mean_token_accuracy": 0.22877017706632613,
      "num_tokens": 114308991.0,
      "step": 61315
    },
    {
      "entropy": 5.967794132232666,
      "epoch": 5.415047686329919,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002485830135233522,
      "loss": 5.2048,
      "mean_token_accuracy": 0.2261102095246315,
      "num_tokens": 114318791.0,
      "step": 61320
    },
    {
      "entropy": 5.989448976516724,
      "epoch": 5.415489226421759,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002485517400818372,
      "loss": 5.176,
      "mean_token_accuracy": 0.22546836137771606,
      "num_tokens": 114327738.0,
      "step": 61325
    },
    {
      "entropy": 5.866352701187134,
      "epoch": 5.4159307665136,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00024852046715842844,
      "loss": 4.9661,
      "mean_token_accuracy": 0.25347079932689665,
      "num_tokens": 114336706.0,
      "step": 61330
    },
    {
      "entropy": 5.957741641998291,
      "epoch": 5.41637230660544,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00024848919475373876,
      "loss": 5.1351,
      "mean_token_accuracy": 0.23543955534696578,
      "num_tokens": 114345961.0,
      "step": 61335
    },
    {
      "entropy": 6.041811418533325,
      "epoch": 5.41681384669728,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002484579228683807,
      "loss": 5.301,
      "mean_token_accuracy": 0.2180149018764496,
      "num_tokens": 114356048.0,
      "step": 61340
    },
    {
      "entropy": 5.982124328613281,
      "epoch": 5.417255386789121,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002484266515029668,
      "loss": 5.257,
      "mean_token_accuracy": 0.22373687326908112,
      "num_tokens": 114364832.0,
      "step": 61345
    },
    {
      "entropy": 6.029757404327393,
      "epoch": 5.417696926880961,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002483953806581097,
      "loss": 5.1318,
      "mean_token_accuracy": 0.23791854679584504,
      "num_tokens": 114374264.0,
      "step": 61350
    },
    {
      "entropy": 6.0377260684967045,
      "epoch": 5.418138466972801,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002483641103344219,
      "loss": 5.2118,
      "mean_token_accuracy": 0.22447644174098969,
      "num_tokens": 114383370.0,
      "step": 61355
    },
    {
      "entropy": 5.961506032943726,
      "epoch": 5.418580007064642,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000248332840532516,
      "loss": 5.2249,
      "mean_token_accuracy": 0.2320706933736801,
      "num_tokens": 114393160.0,
      "step": 61360
    },
    {
      "entropy": 6.099381303787231,
      "epoch": 5.419021547156482,
      "grad_norm": 1.5,
      "learning_rate": 0.0002483015712530048,
      "loss": 5.3023,
      "mean_token_accuracy": 0.220154070854187,
      "num_tokens": 114402273.0,
      "step": 61365
    },
    {
      "entropy": 5.959041500091553,
      "epoch": 5.419463087248322,
      "grad_norm": 1.453125,
      "learning_rate": 0.00024827030249650043,
      "loss": 5.1103,
      "mean_token_accuracy": 0.23608842939138414,
      "num_tokens": 114410942.0,
      "step": 61370
    },
    {
      "entropy": 6.09998083114624,
      "epoch": 5.419904627340163,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00024823903426361574,
      "loss": 5.3354,
      "mean_token_accuracy": 0.21285803914070128,
      "num_tokens": 114420595.0,
      "step": 61375
    },
    {
      "entropy": 6.024990463256836,
      "epoch": 5.420346167432003,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002482077665549631,
      "loss": 5.2138,
      "mean_token_accuracy": 0.2299349620938301,
      "num_tokens": 114429918.0,
      "step": 61380
    },
    {
      "entropy": 6.07735857963562,
      "epoch": 5.420787707523843,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002481764993711551,
      "loss": 5.2319,
      "mean_token_accuracy": 0.21788738220930098,
      "num_tokens": 114439894.0,
      "step": 61385
    },
    {
      "entropy": 6.04951548576355,
      "epoch": 5.421229247615684,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002481452327128043,
      "loss": 5.2076,
      "mean_token_accuracy": 0.226434588432312,
      "num_tokens": 114449264.0,
      "step": 61390
    },
    {
      "entropy": 6.09732027053833,
      "epoch": 5.421670787707524,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024811396658052287,
      "loss": 5.27,
      "mean_token_accuracy": 0.21918446719646453,
      "num_tokens": 114458928.0,
      "step": 61395
    },
    {
      "entropy": 6.010128307342529,
      "epoch": 5.422112327799364,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002480827009749237,
      "loss": 5.197,
      "mean_token_accuracy": 0.22789528965950012,
      "num_tokens": 114467735.0,
      "step": 61400
    },
    {
      "entropy": 6.092724084854126,
      "epoch": 5.422553867891205,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002480514358966192,
      "loss": 5.2395,
      "mean_token_accuracy": 0.22538168132305145,
      "num_tokens": 114477130.0,
      "step": 61405
    },
    {
      "entropy": 6.000969409942627,
      "epoch": 5.422995407983045,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002480201713462216,
      "loss": 5.1501,
      "mean_token_accuracy": 0.2321245476603508,
      "num_tokens": 114486342.0,
      "step": 61410
    },
    {
      "entropy": 6.013962697982788,
      "epoch": 5.423436948074885,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024798890732434357,
      "loss": 5.2502,
      "mean_token_accuracy": 0.2194196566939354,
      "num_tokens": 114496022.0,
      "step": 61415
    },
    {
      "entropy": 6.006428384780884,
      "epoch": 5.423878488166726,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024795764383159745,
      "loss": 5.2712,
      "mean_token_accuracy": 0.21564616560935973,
      "num_tokens": 114505818.0,
      "step": 61420
    },
    {
      "entropy": 6.030517435073852,
      "epoch": 5.424320028258566,
      "grad_norm": 1.46875,
      "learning_rate": 0.00024792638086859576,
      "loss": 5.1629,
      "mean_token_accuracy": 0.2284172758460045,
      "num_tokens": 114514398.0,
      "step": 61425
    },
    {
      "entropy": 6.018305253982544,
      "epoch": 5.424761568350406,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024789511843595086,
      "loss": 5.1954,
      "mean_token_accuracy": 0.22864645719528198,
      "num_tokens": 114524324.0,
      "step": 61430
    },
    {
      "entropy": 5.942876768112183,
      "epoch": 5.425203108442247,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002478638565342751,
      "loss": 5.1456,
      "mean_token_accuracy": 0.23953498303890228,
      "num_tokens": 114533664.0,
      "step": 61435
    },
    {
      "entropy": 6.069459772109985,
      "epoch": 5.425644648534087,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000247832595164181,
      "loss": 5.2343,
      "mean_token_accuracy": 0.22387616336345673,
      "num_tokens": 114543590.0,
      "step": 61440
    },
    {
      "entropy": 6.111136627197266,
      "epoch": 5.426086188625927,
      "grad_norm": 1.53125,
      "learning_rate": 0.00024780133432628093,
      "loss": 5.3845,
      "mean_token_accuracy": 0.21219249665737153,
      "num_tokens": 114553914.0,
      "step": 61445
    },
    {
      "entropy": 6.134344244003296,
      "epoch": 5.426527728717767,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002477700740211873,
      "loss": 5.2558,
      "mean_token_accuracy": 0.22031670808792114,
      "num_tokens": 114563889.0,
      "step": 61450
    },
    {
      "entropy": 6.139947986602783,
      "epoch": 5.426969268809608,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002477388142495124,
      "loss": 5.2618,
      "mean_token_accuracy": 0.21820999234914779,
      "num_tokens": 114573129.0,
      "step": 61455
    },
    {
      "entropy": 6.063434600830078,
      "epoch": 5.427410808901448,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002477075550118685,
      "loss": 5.1571,
      "mean_token_accuracy": 0.2303298145532608,
      "num_tokens": 114582956.0,
      "step": 61460
    },
    {
      "entropy": 6.038194561004639,
      "epoch": 5.427852348993289,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00024767629630886823,
      "loss": 5.2077,
      "mean_token_accuracy": 0.22388033866882323,
      "num_tokens": 114592439.0,
      "step": 61465
    },
    {
      "entropy": 5.94417815208435,
      "epoch": 5.428293889085129,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00024764503814112373,
      "loss": 5.1962,
      "mean_token_accuracy": 0.22647866755723953,
      "num_tokens": 114601025.0,
      "step": 61470
    },
    {
      "entropy": 5.9231466293334964,
      "epoch": 5.428735429176969,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00024761378050924734,
      "loss": 5.1108,
      "mean_token_accuracy": 0.24086050689220428,
      "num_tokens": 114609669.0,
      "step": 61475
    },
    {
      "entropy": 6.078621864318848,
      "epoch": 5.429176969268809,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002475825234138515,
      "loss": 5.2935,
      "mean_token_accuracy": 0.21837198734283447,
      "num_tokens": 114619644.0,
      "step": 61480
    },
    {
      "entropy": 6.038014698028564,
      "epoch": 5.42961850936065,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002475512668555482,
      "loss": 5.1777,
      "mean_token_accuracy": 0.23812734633684157,
      "num_tokens": 114629117.0,
      "step": 61485
    },
    {
      "entropy": 6.073904132843017,
      "epoch": 5.43006004945249,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002475200108349502,
      "loss": 5.1311,
      "mean_token_accuracy": 0.2330441430211067,
      "num_tokens": 114638648.0,
      "step": 61490
    },
    {
      "entropy": 6.01399793624878,
      "epoch": 5.43050158954433,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024748875535266954,
      "loss": 5.2087,
      "mean_token_accuracy": 0.23356544375419616,
      "num_tokens": 114648313.0,
      "step": 61495
    },
    {
      "entropy": 6.065545320510864,
      "epoch": 5.430943129636171,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002474575004093185,
      "loss": 5.2365,
      "mean_token_accuracy": 0.22587068676948546,
      "num_tokens": 114658130.0,
      "step": 61500
    },
    {
      "entropy": 6.109068870544434,
      "epoch": 5.431384669728011,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002474262460055093,
      "loss": 5.2766,
      "mean_token_accuracy": 0.21932830214500426,
      "num_tokens": 114667152.0,
      "step": 61505
    },
    {
      "entropy": 6.045371580123901,
      "epoch": 5.431826209819851,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002473949921418543,
      "loss": 5.2379,
      "mean_token_accuracy": 0.21873826384544373,
      "num_tokens": 114676104.0,
      "step": 61510
    },
    {
      "entropy": 6.074635171890259,
      "epoch": 5.432267749911692,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024736373881896576,
      "loss": 5.1956,
      "mean_token_accuracy": 0.22624377757310868,
      "num_tokens": 114685472.0,
      "step": 61515
    },
    {
      "entropy": 6.107772016525269,
      "epoch": 5.432709290003532,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024733248603745584,
      "loss": 5.2703,
      "mean_token_accuracy": 0.21934105306863785,
      "num_tokens": 114694703.0,
      "step": 61520
    },
    {
      "entropy": 6.014826059341431,
      "epoch": 5.433150830095372,
      "grad_norm": 1.375,
      "learning_rate": 0.0002473012337979367,
      "loss": 5.2208,
      "mean_token_accuracy": 0.2375827446579933,
      "num_tokens": 114704934.0,
      "step": 61525
    },
    {
      "entropy": 6.03592209815979,
      "epoch": 5.433592370187213,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002472699821010206,
      "loss": 5.2753,
      "mean_token_accuracy": 0.2286689817905426,
      "num_tokens": 114714572.0,
      "step": 61530
    },
    {
      "entropy": 6.019181632995606,
      "epoch": 5.434033910279053,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024723873094731997,
      "loss": 5.2,
      "mean_token_accuracy": 0.22777336090803146,
      "num_tokens": 114724051.0,
      "step": 61535
    },
    {
      "entropy": 6.041088199615478,
      "epoch": 5.434475450370893,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002472074803374467,
      "loss": 5.2521,
      "mean_token_accuracy": 0.21690381467342376,
      "num_tokens": 114733119.0,
      "step": 61540
    },
    {
      "entropy": 5.992226266860962,
      "epoch": 5.434916990462734,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002471762302720131,
      "loss": 5.1807,
      "mean_token_accuracy": 0.2295840248465538,
      "num_tokens": 114742260.0,
      "step": 61545
    },
    {
      "entropy": 6.146010446548462,
      "epoch": 5.435358530554574,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002471449807516312,
      "loss": 5.3618,
      "mean_token_accuracy": 0.21226837784051894,
      "num_tokens": 114751579.0,
      "step": 61550
    },
    {
      "entropy": 6.087926435470581,
      "epoch": 5.4358000706464145,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00024711373177691346,
      "loss": 5.2492,
      "mean_token_accuracy": 0.21904585659503936,
      "num_tokens": 114760676.0,
      "step": 61555
    },
    {
      "entropy": 6.019922351837158,
      "epoch": 5.436241610738255,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00024708248334847185,
      "loss": 5.1814,
      "mean_token_accuracy": 0.22998992800712587,
      "num_tokens": 114770827.0,
      "step": 61560
    },
    {
      "entropy": 6.060814237594604,
      "epoch": 5.4366831508300955,
      "grad_norm": 1.59375,
      "learning_rate": 0.00024705123546691845,
      "loss": 5.1837,
      "mean_token_accuracy": 0.2275385782122612,
      "num_tokens": 114780991.0,
      "step": 61565
    },
    {
      "entropy": 6.0988929748535154,
      "epoch": 5.4371246909219355,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002470199881328654,
      "loss": 5.2656,
      "mean_token_accuracy": 0.22471841126680375,
      "num_tokens": 114790370.0,
      "step": 61570
    },
    {
      "entropy": 6.09516921043396,
      "epoch": 5.4375662310137765,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024698874134692494,
      "loss": 5.2813,
      "mean_token_accuracy": 0.22227708548307418,
      "num_tokens": 114799472.0,
      "step": 61575
    },
    {
      "entropy": 6.06783618927002,
      "epoch": 5.4380077711056165,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002469574951097091,
      "loss": 5.2654,
      "mean_token_accuracy": 0.2209443375468254,
      "num_tokens": 114808723.0,
      "step": 61580
    },
    {
      "entropy": 6.154822492599488,
      "epoch": 5.438449311197457,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024692624942183,
      "loss": 5.3217,
      "mean_token_accuracy": 0.21495793163776397,
      "num_tokens": 114818132.0,
      "step": 61585
    },
    {
      "entropy": 6.095607089996338,
      "epoch": 5.4388908512892975,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024689500428389964,
      "loss": 5.2281,
      "mean_token_accuracy": 0.22937208712100982,
      "num_tokens": 114827314.0,
      "step": 61590
    },
    {
      "entropy": 6.0049803256988525,
      "epoch": 5.4393323913811376,
      "grad_norm": 1.46875,
      "learning_rate": 0.00024686375969653024,
      "loss": 5.19,
      "mean_token_accuracy": 0.22799888253211975,
      "num_tokens": 114836239.0,
      "step": 61595
    },
    {
      "entropy": 6.015345716476441,
      "epoch": 5.439773931472978,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024683251566033377,
      "loss": 5.2303,
      "mean_token_accuracy": 0.23173866271972657,
      "num_tokens": 114845404.0,
      "step": 61600
    },
    {
      "entropy": 6.041878747940063,
      "epoch": 5.440215471564818,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024680127217592226,
      "loss": 5.2244,
      "mean_token_accuracy": 0.22293916791677476,
      "num_tokens": 114854758.0,
      "step": 61605
    },
    {
      "entropy": 6.037006425857544,
      "epoch": 5.440657011656659,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024677002924390786,
      "loss": 5.2682,
      "mean_token_accuracy": 0.2261832371354103,
      "num_tokens": 114864379.0,
      "step": 61610
    },
    {
      "entropy": 6.009396123886108,
      "epoch": 5.441098551748499,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024673878686490237,
      "loss": 5.2175,
      "mean_token_accuracy": 0.23320864737033845,
      "num_tokens": 114873966.0,
      "step": 61615
    },
    {
      "entropy": 6.047862911224366,
      "epoch": 5.441540091840339,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002467075450395181,
      "loss": 5.2554,
      "mean_token_accuracy": 0.2203705742955208,
      "num_tokens": 114884688.0,
      "step": 61620
    },
    {
      "entropy": 6.007942199707031,
      "epoch": 5.44198163193218,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000246676303768367,
      "loss": 5.2375,
      "mean_token_accuracy": 0.22617800980806352,
      "num_tokens": 114893802.0,
      "step": 61625
    },
    {
      "entropy": 6.072159719467163,
      "epoch": 5.44242317202402,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002466450630520609,
      "loss": 5.2425,
      "mean_token_accuracy": 0.22401743084192277,
      "num_tokens": 114903515.0,
      "step": 61630
    },
    {
      "entropy": 6.0996991157531735,
      "epoch": 5.44286471211586,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002466138228912118,
      "loss": 5.3899,
      "mean_token_accuracy": 0.22033312022686005,
      "num_tokens": 114913951.0,
      "step": 61635
    },
    {
      "entropy": 6.102444648742676,
      "epoch": 5.443306252207701,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002465825832864319,
      "loss": 5.2406,
      "mean_token_accuracy": 0.22598912119865416,
      "num_tokens": 114922831.0,
      "step": 61640
    },
    {
      "entropy": 6.005637979507446,
      "epoch": 5.443747792299541,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000246551344238333,
      "loss": 5.117,
      "mean_token_accuracy": 0.22854546457529068,
      "num_tokens": 114931787.0,
      "step": 61645
    },
    {
      "entropy": 6.078301906585693,
      "epoch": 5.444189332391381,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002465201057475271,
      "loss": 5.2815,
      "mean_token_accuracy": 0.2204708382487297,
      "num_tokens": 114941620.0,
      "step": 61650
    },
    {
      "entropy": 6.067985200881958,
      "epoch": 5.444630872483222,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002464888678146262,
      "loss": 5.2979,
      "mean_token_accuracy": 0.22416184693574906,
      "num_tokens": 114951647.0,
      "step": 61655
    },
    {
      "entropy": 6.062508630752563,
      "epoch": 5.445072412575062,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024645763044024206,
      "loss": 5.1852,
      "mean_token_accuracy": 0.22441217452287673,
      "num_tokens": 114960557.0,
      "step": 61660
    },
    {
      "entropy": 5.98695969581604,
      "epoch": 5.445513952666902,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002464263936249867,
      "loss": 5.1571,
      "mean_token_accuracy": 0.23012548834085464,
      "num_tokens": 114970474.0,
      "step": 61665
    },
    {
      "entropy": 5.988839721679687,
      "epoch": 5.445955492758743,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024639515736947206,
      "loss": 5.2009,
      "mean_token_accuracy": 0.22131024897098542,
      "num_tokens": 114979699.0,
      "step": 61670
    },
    {
      "entropy": 6.053602504730224,
      "epoch": 5.446397032850583,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002463639216743101,
      "loss": 5.2081,
      "mean_token_accuracy": 0.22835762947797775,
      "num_tokens": 114988932.0,
      "step": 61675
    },
    {
      "entropy": 6.011814928054809,
      "epoch": 5.446838572942423,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002463326865401125,
      "loss": 5.1347,
      "mean_token_accuracy": 0.23265593349933625,
      "num_tokens": 114997936.0,
      "step": 61680
    },
    {
      "entropy": 6.069163990020752,
      "epoch": 5.447280113034264,
      "grad_norm": 1.375,
      "learning_rate": 0.0002463014519674913,
      "loss": 5.3205,
      "mean_token_accuracy": 0.2141149401664734,
      "num_tokens": 115007669.0,
      "step": 61685
    },
    {
      "entropy": 6.091994333267212,
      "epoch": 5.447721653126104,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002462702179570584,
      "loss": 5.2225,
      "mean_token_accuracy": 0.22658989876508712,
      "num_tokens": 115016192.0,
      "step": 61690
    },
    {
      "entropy": 6.03215274810791,
      "epoch": 5.448163193217944,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00024623898450942556,
      "loss": 5.264,
      "mean_token_accuracy": 0.22729405015707016,
      "num_tokens": 115026863.0,
      "step": 61695
    },
    {
      "entropy": 6.046065092086792,
      "epoch": 5.448604733309785,
      "grad_norm": 1.625,
      "learning_rate": 0.0002462077516252046,
      "loss": 5.2659,
      "mean_token_accuracy": 0.21944993138313293,
      "num_tokens": 115035780.0,
      "step": 61700
    },
    {
      "entropy": 6.014527225494385,
      "epoch": 5.449046273401625,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00024617651930500744,
      "loss": 5.2317,
      "mean_token_accuracy": 0.22374074459075927,
      "num_tokens": 115044478.0,
      "step": 61705
    },
    {
      "entropy": 6.024570369720459,
      "epoch": 5.449487813493465,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002461452875494459,
      "loss": 5.1552,
      "mean_token_accuracy": 0.23295242190361024,
      "num_tokens": 115053936.0,
      "step": 61710
    },
    {
      "entropy": 6.004432106018067,
      "epoch": 5.449929353585306,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002461140563591318,
      "loss": 5.2393,
      "mean_token_accuracy": 0.22396209985017776,
      "num_tokens": 115063155.0,
      "step": 61715
    },
    {
      "entropy": 6.078683519363404,
      "epoch": 5.450370893677146,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002460828257346768,
      "loss": 5.2464,
      "mean_token_accuracy": 0.21966800093650818,
      "num_tokens": 115072405.0,
      "step": 61720
    },
    {
      "entropy": 6.043862152099609,
      "epoch": 5.450812433768986,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002460515956766929,
      "loss": 5.2307,
      "mean_token_accuracy": 0.22791988253593445,
      "num_tokens": 115081828.0,
      "step": 61725
    },
    {
      "entropy": 6.006490087509155,
      "epoch": 5.451253973860826,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002460203661857917,
      "loss": 5.2295,
      "mean_token_accuracy": 0.2240167036652565,
      "num_tokens": 115091255.0,
      "step": 61730
    },
    {
      "entropy": 6.071682357788086,
      "epoch": 5.451695513952667,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024598913726258515,
      "loss": 5.1895,
      "mean_token_accuracy": 0.22928118854761123,
      "num_tokens": 115099677.0,
      "step": 61735
    },
    {
      "entropy": 6.026135873794556,
      "epoch": 5.452137054044507,
      "grad_norm": 1.484375,
      "learning_rate": 0.00024595790890768484,
      "loss": 5.1622,
      "mean_token_accuracy": 0.2314505934715271,
      "num_tokens": 115108299.0,
      "step": 61740
    },
    {
      "entropy": 6.050188779830933,
      "epoch": 5.452578594136348,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002459266811217025,
      "loss": 5.3131,
      "mean_token_accuracy": 0.22343717366456986,
      "num_tokens": 115117301.0,
      "step": 61745
    },
    {
      "entropy": 6.014244508743286,
      "epoch": 5.453020134228188,
      "grad_norm": 1.546875,
      "learning_rate": 0.00024589545390525,
      "loss": 5.2563,
      "mean_token_accuracy": 0.22221506386995316,
      "num_tokens": 115126643.0,
      "step": 61750
    },
    {
      "entropy": 6.117557048797607,
      "epoch": 5.453461674320028,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002458642272589391,
      "loss": 5.3327,
      "mean_token_accuracy": 0.20836355835199355,
      "num_tokens": 115136650.0,
      "step": 61755
    },
    {
      "entropy": 6.159934616088867,
      "epoch": 5.453903214411868,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024583300118338134,
      "loss": 5.2547,
      "mean_token_accuracy": 0.23007752746343613,
      "num_tokens": 115145861.0,
      "step": 61760
    },
    {
      "entropy": 6.013709926605225,
      "epoch": 5.454344754503709,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00024580177567918855,
      "loss": 5.1761,
      "mean_token_accuracy": 0.22635306864976884,
      "num_tokens": 115156058.0,
      "step": 61765
    },
    {
      "entropy": 5.951263284683227,
      "epoch": 5.454786294595549,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024577055074697225,
      "loss": 5.076,
      "mean_token_accuracy": 0.24498169422149657,
      "num_tokens": 115165183.0,
      "step": 61770
    },
    {
      "entropy": 6.075048780441284,
      "epoch": 5.455227834687389,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002457393263873443,
      "loss": 5.2632,
      "mean_token_accuracy": 0.21915347576141359,
      "num_tokens": 115175767.0,
      "step": 61775
    },
    {
      "entropy": 6.099623441696167,
      "epoch": 5.45566937477923,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002457081026009163,
      "loss": 5.2036,
      "mean_token_accuracy": 0.2260573074221611,
      "num_tokens": 115185480.0,
      "step": 61780
    },
    {
      "entropy": 6.07134518623352,
      "epoch": 5.45611091487107,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024567687938829993,
      "loss": 5.3052,
      "mean_token_accuracy": 0.22489404082298278,
      "num_tokens": 115195348.0,
      "step": 61785
    },
    {
      "entropy": 6.078722953796387,
      "epoch": 5.45655245496291,
      "grad_norm": 1.375,
      "learning_rate": 0.00024564565675010673,
      "loss": 5.2909,
      "mean_token_accuracy": 0.2193239837884903,
      "num_tokens": 115204517.0,
      "step": 61790
    },
    {
      "entropy": 6.096750640869141,
      "epoch": 5.456993995054751,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002456144346869485,
      "loss": 5.1732,
      "mean_token_accuracy": 0.22814128994941713,
      "num_tokens": 115213883.0,
      "step": 61795
    },
    {
      "entropy": 6.133403158187866,
      "epoch": 5.457435535146591,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002455832131994368,
      "loss": 5.2523,
      "mean_token_accuracy": 0.21825806498527528,
      "num_tokens": 115222826.0,
      "step": 61800
    },
    {
      "entropy": 6.091535425186157,
      "epoch": 5.457877075238431,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002455519922881832,
      "loss": 5.2168,
      "mean_token_accuracy": 0.21609001010656356,
      "num_tokens": 115231949.0,
      "step": 61805
    },
    {
      "entropy": 6.0225378513336185,
      "epoch": 5.458318615330272,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002455207719537992,
      "loss": 5.2168,
      "mean_token_accuracy": 0.22485591173171998,
      "num_tokens": 115240617.0,
      "step": 61810
    },
    {
      "entropy": 5.982728815078735,
      "epoch": 5.458760155422112,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002454895521968967,
      "loss": 5.1268,
      "mean_token_accuracy": 0.24217121601104735,
      "num_tokens": 115249702.0,
      "step": 61815
    },
    {
      "entropy": 6.055751514434815,
      "epoch": 5.459201695513952,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00024545833301808706,
      "loss": 5.2406,
      "mean_token_accuracy": 0.22538674473762513,
      "num_tokens": 115259706.0,
      "step": 61820
    },
    {
      "entropy": 6.063770055770874,
      "epoch": 5.459643235605793,
      "grad_norm": 1.453125,
      "learning_rate": 0.00024542711441798186,
      "loss": 5.3096,
      "mean_token_accuracy": 0.21931319683790207,
      "num_tokens": 115269594.0,
      "step": 61825
    },
    {
      "entropy": 5.956035089492798,
      "epoch": 5.460084775697633,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024539589639719267,
      "loss": 5.0501,
      "mean_token_accuracy": 0.2490328937768936,
      "num_tokens": 115279318.0,
      "step": 61830
    },
    {
      "entropy": 6.07448410987854,
      "epoch": 5.4605263157894735,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024536467895633097,
      "loss": 5.3007,
      "mean_token_accuracy": 0.21693418622016908,
      "num_tokens": 115289431.0,
      "step": 61835
    },
    {
      "entropy": 6.066574621200561,
      "epoch": 5.460967855881314,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002453334620960084,
      "loss": 5.2073,
      "mean_token_accuracy": 0.2269717574119568,
      "num_tokens": 115298570.0,
      "step": 61840
    },
    {
      "entropy": 6.089998579025268,
      "epoch": 5.4614093959731544,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002453022458168365,
      "loss": 5.2649,
      "mean_token_accuracy": 0.21549892723560332,
      "num_tokens": 115307804.0,
      "step": 61845
    },
    {
      "entropy": 6.070803833007813,
      "epoch": 5.4618509360649945,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024527103011942657,
      "loss": 5.2549,
      "mean_token_accuracy": 0.2179350346326828,
      "num_tokens": 115317670.0,
      "step": 61850
    },
    {
      "entropy": 5.997021055221557,
      "epoch": 5.462292476156835,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002452398150043903,
      "loss": 5.1122,
      "mean_token_accuracy": 0.2317805290222168,
      "num_tokens": 115326653.0,
      "step": 61855
    },
    {
      "entropy": 6.077196264266968,
      "epoch": 5.4627340162486755,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00024520860047233913,
      "loss": 5.2255,
      "mean_token_accuracy": 0.2207217574119568,
      "num_tokens": 115336221.0,
      "step": 61860
    },
    {
      "entropy": 6.077456951141357,
      "epoch": 5.4631755563405155,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002451773865238846,
      "loss": 5.2684,
      "mean_token_accuracy": 0.21595600992441177,
      "num_tokens": 115345297.0,
      "step": 61865
    },
    {
      "entropy": 6.129289340972901,
      "epoch": 5.4636170964323565,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002451461731596381,
      "loss": 5.3137,
      "mean_token_accuracy": 0.21464509814977645,
      "num_tokens": 115354327.0,
      "step": 61870
    },
    {
      "entropy": 6.053257465362549,
      "epoch": 5.4640586365241965,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024511496038021096,
      "loss": 5.1923,
      "mean_token_accuracy": 0.2206509530544281,
      "num_tokens": 115362961.0,
      "step": 61875
    },
    {
      "entropy": 6.050921058654785,
      "epoch": 5.464500176616037,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024508374818621486,
      "loss": 5.1916,
      "mean_token_accuracy": 0.22824226021766664,
      "num_tokens": 115373716.0,
      "step": 61880
    },
    {
      "entropy": 6.1358170986175535,
      "epoch": 5.464941716707877,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024505253657826115,
      "loss": 5.3147,
      "mean_token_accuracy": 0.22046783566474915,
      "num_tokens": 115383158.0,
      "step": 61885
    },
    {
      "entropy": 6.032471704483032,
      "epoch": 5.465383256799718,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002450213255569612,
      "loss": 5.1922,
      "mean_token_accuracy": 0.22773023545742035,
      "num_tokens": 115392080.0,
      "step": 61890
    },
    {
      "entropy": 6.033509922027588,
      "epoch": 5.465824796891558,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002449901151229264,
      "loss": 5.299,
      "mean_token_accuracy": 0.21313064694404601,
      "num_tokens": 115401361.0,
      "step": 61895
    },
    {
      "entropy": 6.057472896575928,
      "epoch": 5.466266336983398,
      "grad_norm": 1.625,
      "learning_rate": 0.00024495890527676813,
      "loss": 5.2303,
      "mean_token_accuracy": 0.21797715723514557,
      "num_tokens": 115410158.0,
      "step": 61900
    },
    {
      "entropy": 6.087020206451416,
      "epoch": 5.466707877075239,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002449276960190979,
      "loss": 5.315,
      "mean_token_accuracy": 0.21403346955776215,
      "num_tokens": 115420246.0,
      "step": 61905
    },
    {
      "entropy": 6.074652576446534,
      "epoch": 5.467149417167079,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000244896487350527,
      "loss": 5.2242,
      "mean_token_accuracy": 0.22012778520584106,
      "num_tokens": 115429702.0,
      "step": 61910
    },
    {
      "entropy": 6.083765506744385,
      "epoch": 5.467590957258919,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00024486527927166675,
      "loss": 5.1931,
      "mean_token_accuracy": 0.22461662888526918,
      "num_tokens": 115439316.0,
      "step": 61915
    },
    {
      "entropy": 6.079676866531372,
      "epoch": 5.46803249735076,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002448340717831285,
      "loss": 5.2704,
      "mean_token_accuracy": 0.21643882244825363,
      "num_tokens": 115448516.0,
      "step": 61920
    },
    {
      "entropy": 6.010709238052368,
      "epoch": 5.4684740374426,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002448028648855237,
      "loss": 5.2059,
      "mean_token_accuracy": 0.2284141331911087,
      "num_tokens": 115457786.0,
      "step": 61925
    },
    {
      "entropy": 5.961758756637574,
      "epoch": 5.46891557753444,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002447716585794636,
      "loss": 5.2045,
      "mean_token_accuracy": 0.2290767550468445,
      "num_tokens": 115467181.0,
      "step": 61930
    },
    {
      "entropy": 6.069272470474243,
      "epoch": 5.469357117626281,
      "grad_norm": 1.5,
      "learning_rate": 0.0002447404528655595,
      "loss": 5.2022,
      "mean_token_accuracy": 0.22804715782403945,
      "num_tokens": 115475320.0,
      "step": 61935
    },
    {
      "entropy": 5.994935369491577,
      "epoch": 5.469798657718121,
      "grad_norm": 1.53125,
      "learning_rate": 0.00024470924774442265,
      "loss": 5.153,
      "mean_token_accuracy": 0.23173559308052064,
      "num_tokens": 115484419.0,
      "step": 61940
    },
    {
      "entropy": 5.991590166091919,
      "epoch": 5.470240197809961,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002446780432166645,
      "loss": 5.1643,
      "mean_token_accuracy": 0.229911969602108,
      "num_tokens": 115492982.0,
      "step": 61945
    },
    {
      "entropy": 6.0394891738891605,
      "epoch": 5.470681737901802,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002446468392828962,
      "loss": 5.2346,
      "mean_token_accuracy": 0.22343563437461852,
      "num_tokens": 115502517.0,
      "step": 61950
    },
    {
      "entropy": 6.11904296875,
      "epoch": 5.471123277993642,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002446156359437291,
      "loss": 5.2922,
      "mean_token_accuracy": 0.21444427967071533,
      "num_tokens": 115511512.0,
      "step": 61955
    },
    {
      "entropy": 5.997283697128296,
      "epoch": 5.471564818085482,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024458443319977445,
      "loss": 5.1935,
      "mean_token_accuracy": 0.23227821588516234,
      "num_tokens": 115522494.0,
      "step": 61960
    },
    {
      "entropy": 6.023033189773559,
      "epoch": 5.472006358177323,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002445532310516433,
      "loss": 5.2189,
      "mean_token_accuracy": 0.22645225375890732,
      "num_tokens": 115532270.0,
      "step": 61965
    },
    {
      "entropy": 6.078926944732666,
      "epoch": 5.472447898269163,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002445220294999472,
      "loss": 5.1735,
      "mean_token_accuracy": 0.23176121562719346,
      "num_tokens": 115543057.0,
      "step": 61970
    },
    {
      "entropy": 6.1007922172546385,
      "epoch": 5.472889438361003,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024449082854529714,
      "loss": 5.2447,
      "mean_token_accuracy": 0.22663369476795198,
      "num_tokens": 115551551.0,
      "step": 61975
    },
    {
      "entropy": 6.067565631866455,
      "epoch": 5.473330978452844,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024445962818830437,
      "loss": 5.2506,
      "mean_token_accuracy": 0.22989346086978912,
      "num_tokens": 115560911.0,
      "step": 61980
    },
    {
      "entropy": 5.993834829330444,
      "epoch": 5.473772518544684,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002444284284295801,
      "loss": 5.1637,
      "mean_token_accuracy": 0.23605379313230515,
      "num_tokens": 115569735.0,
      "step": 61985
    },
    {
      "entropy": 6.057575130462647,
      "epoch": 5.474214058636524,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002443972292697355,
      "loss": 5.2587,
      "mean_token_accuracy": 0.22591546624898912,
      "num_tokens": 115578384.0,
      "step": 61990
    },
    {
      "entropy": 5.986876535415649,
      "epoch": 5.474655598728365,
      "grad_norm": 1.375,
      "learning_rate": 0.0002443660307093819,
      "loss": 5.2048,
      "mean_token_accuracy": 0.2256135329604149,
      "num_tokens": 115588353.0,
      "step": 61995
    },
    {
      "entropy": 6.0893189907073975,
      "epoch": 5.475097138820205,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002443348327491303,
      "loss": 5.2468,
      "mean_token_accuracy": 0.23365252017974852,
      "num_tokens": 115597744.0,
      "step": 62000
    },
    {
      "entropy": 5.989856958389282,
      "epoch": 5.475538678912045,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002443036353895919,
      "loss": 5.1597,
      "mean_token_accuracy": 0.23682999461889268,
      "num_tokens": 115607012.0,
      "step": 62005
    },
    {
      "entropy": 6.003364944458008,
      "epoch": 5.475980219003885,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002442724386313777,
      "loss": 5.2066,
      "mean_token_accuracy": 0.22164494544267654,
      "num_tokens": 115616341.0,
      "step": 62010
    },
    {
      "entropy": 6.00749831199646,
      "epoch": 5.476421759095726,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024424124247509905,
      "loss": 5.2657,
      "mean_token_accuracy": 0.22038172781467438,
      "num_tokens": 115626612.0,
      "step": 62015
    },
    {
      "entropy": 6.089195775985718,
      "epoch": 5.476863299187566,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000244210046921367,
      "loss": 5.2919,
      "mean_token_accuracy": 0.2163772001862526,
      "num_tokens": 115635240.0,
      "step": 62020
    },
    {
      "entropy": 6.0362877368927,
      "epoch": 5.477304839279407,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002441788519707926,
      "loss": 5.2268,
      "mean_token_accuracy": 0.22232712656259537,
      "num_tokens": 115644900.0,
      "step": 62025
    },
    {
      "entropy": 6.027635669708252,
      "epoch": 5.477746379371247,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002441476576239869,
      "loss": 5.1982,
      "mean_token_accuracy": 0.22615486532449722,
      "num_tokens": 115654886.0,
      "step": 62030
    },
    {
      "entropy": 6.038482427597046,
      "epoch": 5.478187919463087,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00024411646388156104,
      "loss": 5.1722,
      "mean_token_accuracy": 0.22689111083745955,
      "num_tokens": 115664230.0,
      "step": 62035
    },
    {
      "entropy": 6.063598728179931,
      "epoch": 5.478629459554927,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00024408527074412618,
      "loss": 5.2356,
      "mean_token_accuracy": 0.21690384149551392,
      "num_tokens": 115673181.0,
      "step": 62040
    },
    {
      "entropy": 6.016487455368042,
      "epoch": 5.479070999646768,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002440540782122932,
      "loss": 5.1361,
      "mean_token_accuracy": 0.23114736527204513,
      "num_tokens": 115682484.0,
      "step": 62045
    },
    {
      "entropy": 6.014018535614014,
      "epoch": 5.479512539738608,
      "grad_norm": 1.484375,
      "learning_rate": 0.00024402288628667325,
      "loss": 5.1947,
      "mean_token_accuracy": 0.22379042208194733,
      "num_tokens": 115691536.0,
      "step": 62050
    },
    {
      "entropy": 6.04052767753601,
      "epoch": 5.479954079830448,
      "grad_norm": 1.46875,
      "learning_rate": 0.00024399169496787732,
      "loss": 5.2344,
      "mean_token_accuracy": 0.22235572189092637,
      "num_tokens": 115701211.0,
      "step": 62055
    },
    {
      "entropy": 6.066923570632935,
      "epoch": 5.480395619922289,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002439605042565165,
      "loss": 5.1903,
      "mean_token_accuracy": 0.22814004868268967,
      "num_tokens": 115709890.0,
      "step": 62060
    },
    {
      "entropy": 6.084840297698975,
      "epoch": 5.480837160014129,
      "grad_norm": 1.453125,
      "learning_rate": 0.00024392931415320173,
      "loss": 5.2954,
      "mean_token_accuracy": 0.21888307780027388,
      "num_tokens": 115719306.0,
      "step": 62065
    },
    {
      "entropy": 6.043415260314942,
      "epoch": 5.481278700105969,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00024389812465854405,
      "loss": 5.1803,
      "mean_token_accuracy": 0.22142709344625472,
      "num_tokens": 115728307.0,
      "step": 62070
    },
    {
      "entropy": 6.069724988937378,
      "epoch": 5.48172024019781,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024386693577315433,
      "loss": 5.1728,
      "mean_token_accuracy": 0.23193855136632918,
      "num_tokens": 115737748.0,
      "step": 62075
    },
    {
      "entropy": 6.026749134063721,
      "epoch": 5.48216178028965,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002438357474976437,
      "loss": 5.1455,
      "mean_token_accuracy": 0.23831393122673034,
      "num_tokens": 115746863.0,
      "step": 62080
    },
    {
      "entropy": 6.075042390823365,
      "epoch": 5.48260332038149,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024380455983262307,
      "loss": 5.2926,
      "mean_token_accuracy": 0.22318018972873688,
      "num_tokens": 115755905.0,
      "step": 62085
    },
    {
      "entropy": 5.9983703136444095,
      "epoch": 5.483044860473331,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024377337277870339,
      "loss": 5.2271,
      "mean_token_accuracy": 0.22391680330038072,
      "num_tokens": 115764637.0,
      "step": 62090
    },
    {
      "entropy": 6.053933477401733,
      "epoch": 5.483486400565171,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00024374218633649547,
      "loss": 5.1979,
      "mean_token_accuracy": 0.22060836404561995,
      "num_tokens": 115774125.0,
      "step": 62095
    },
    {
      "entropy": 6.015043115615844,
      "epoch": 5.483927940657011,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00024371100050661043,
      "loss": 5.1618,
      "mean_token_accuracy": 0.23368758857250213,
      "num_tokens": 115783282.0,
      "step": 62100
    },
    {
      "entropy": 6.057418298721314,
      "epoch": 5.484369480748852,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024367981528965904,
      "loss": 5.2481,
      "mean_token_accuracy": 0.2239992216229439,
      "num_tokens": 115792854.0,
      "step": 62105
    },
    {
      "entropy": 6.06398639678955,
      "epoch": 5.484811020840692,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002436486306862523,
      "loss": 5.2579,
      "mean_token_accuracy": 0.22466578483581542,
      "num_tokens": 115801771.0,
      "step": 62110
    },
    {
      "entropy": 6.027759647369384,
      "epoch": 5.485252560932532,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024361744669700104,
      "loss": 5.246,
      "mean_token_accuracy": 0.2251509353518486,
      "num_tokens": 115811822.0,
      "step": 62115
    },
    {
      "entropy": 6.009767961502075,
      "epoch": 5.485694101024373,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002435862633225161,
      "loss": 5.1587,
      "mean_token_accuracy": 0.23276099264621736,
      "num_tokens": 115821283.0,
      "step": 62120
    },
    {
      "entropy": 6.059946966171265,
      "epoch": 5.486135641116213,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024355508056340846,
      "loss": 5.1694,
      "mean_token_accuracy": 0.2242455691099167,
      "num_tokens": 115830356.0,
      "step": 62125
    },
    {
      "entropy": 6.077281808853149,
      "epoch": 5.4865771812080535,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024352389842028895,
      "loss": 5.1917,
      "mean_token_accuracy": 0.2288808971643448,
      "num_tokens": 115839274.0,
      "step": 62130
    },
    {
      "entropy": 5.99035611152649,
      "epoch": 5.487018721299894,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024349271689376833,
      "loss": 5.0979,
      "mean_token_accuracy": 0.23366983830928803,
      "num_tokens": 115847932.0,
      "step": 62135
    },
    {
      "entropy": 5.997437477111816,
      "epoch": 5.4874602613917345,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024346153598445736,
      "loss": 5.1672,
      "mean_token_accuracy": 0.23202474117279054,
      "num_tokens": 115857220.0,
      "step": 62140
    },
    {
      "entropy": 6.018846607208252,
      "epoch": 5.4879018014835745,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024343035569296712,
      "loss": 5.2174,
      "mean_token_accuracy": 0.22857537120580673,
      "num_tokens": 115866847.0,
      "step": 62145
    },
    {
      "entropy": 6.058346891403199,
      "epoch": 5.4883433415754155,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002433991760199082,
      "loss": 5.2219,
      "mean_token_accuracy": 0.23154670000076294,
      "num_tokens": 115876365.0,
      "step": 62150
    },
    {
      "entropy": 6.061969804763794,
      "epoch": 5.4887848816672555,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024336799696589146,
      "loss": 5.1856,
      "mean_token_accuracy": 0.22999742478132248,
      "num_tokens": 115884758.0,
      "step": 62155
    },
    {
      "entropy": 6.024302673339844,
      "epoch": 5.489226421759096,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024333681853152755,
      "loss": 5.2493,
      "mean_token_accuracy": 0.22022691369056702,
      "num_tokens": 115894066.0,
      "step": 62160
    },
    {
      "entropy": 5.992680501937866,
      "epoch": 5.489667961850936,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002433056407174275,
      "loss": 5.1765,
      "mean_token_accuracy": 0.22934453785419465,
      "num_tokens": 115903752.0,
      "step": 62165
    },
    {
      "entropy": 6.119137334823608,
      "epoch": 5.490109501942777,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024327446352420185,
      "loss": 5.2797,
      "mean_token_accuracy": 0.2212293490767479,
      "num_tokens": 115913328.0,
      "step": 62170
    },
    {
      "entropy": 6.087964391708374,
      "epoch": 5.490551042034617,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002432432869524615,
      "loss": 5.2201,
      "mean_token_accuracy": 0.2300078898668289,
      "num_tokens": 115922815.0,
      "step": 62175
    },
    {
      "entropy": 6.0593348979949955,
      "epoch": 5.490992582126457,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000243212111002817,
      "loss": 5.2687,
      "mean_token_accuracy": 0.21864124685525893,
      "num_tokens": 115933335.0,
      "step": 62180
    },
    {
      "entropy": 6.026180839538574,
      "epoch": 5.491434122218298,
      "grad_norm": 1.46875,
      "learning_rate": 0.00024318093567587918,
      "loss": 5.2484,
      "mean_token_accuracy": 0.2281866803765297,
      "num_tokens": 115942858.0,
      "step": 62185
    },
    {
      "entropy": 5.998095321655273,
      "epoch": 5.491875662310138,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024314976097225877,
      "loss": 5.296,
      "mean_token_accuracy": 0.22805932015180588,
      "num_tokens": 115953259.0,
      "step": 62190
    },
    {
      "entropy": 6.045404100418091,
      "epoch": 5.492317202401978,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024311858689256649,
      "loss": 5.2686,
      "mean_token_accuracy": 0.22717025876045227,
      "num_tokens": 115962160.0,
      "step": 62195
    },
    {
      "entropy": 6.069490385055542,
      "epoch": 5.492758742493819,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024308741343741293,
      "loss": 5.2141,
      "mean_token_accuracy": 0.22683506608009338,
      "num_tokens": 115971539.0,
      "step": 62200
    },
    {
      "entropy": 6.17658576965332,
      "epoch": 5.493200282585659,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024305624060740866,
      "loss": 5.3023,
      "mean_token_accuracy": 0.2191744014620781,
      "num_tokens": 115980986.0,
      "step": 62205
    },
    {
      "entropy": 6.063413858413696,
      "epoch": 5.493641822677499,
      "grad_norm": 1.453125,
      "learning_rate": 0.00024302506840316463,
      "loss": 5.1958,
      "mean_token_accuracy": 0.2232072174549103,
      "num_tokens": 115989887.0,
      "step": 62210
    },
    {
      "entropy": 6.140634632110595,
      "epoch": 5.49408336276934,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024299389682529122,
      "loss": 5.3278,
      "mean_token_accuracy": 0.2110798552632332,
      "num_tokens": 116000639.0,
      "step": 62215
    },
    {
      "entropy": 6.092266845703125,
      "epoch": 5.49452490286118,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024296272587439927,
      "loss": 5.3221,
      "mean_token_accuracy": 0.22088171243667604,
      "num_tokens": 116010401.0,
      "step": 62220
    },
    {
      "entropy": 6.001838159561157,
      "epoch": 5.49496644295302,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024293155555109912,
      "loss": 5.1906,
      "mean_token_accuracy": 0.23140318542718888,
      "num_tokens": 116020095.0,
      "step": 62225
    },
    {
      "entropy": 6.1178374767303465,
      "epoch": 5.495407983044861,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002429003858560017,
      "loss": 5.3649,
      "mean_token_accuracy": 0.2109069675207138,
      "num_tokens": 116030047.0,
      "step": 62230
    },
    {
      "entropy": 6.033942127227784,
      "epoch": 5.495849523136701,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024286921678971747,
      "loss": 5.1814,
      "mean_token_accuracy": 0.22907767742872237,
      "num_tokens": 116039191.0,
      "step": 62235
    },
    {
      "entropy": 6.0886506080627445,
      "epoch": 5.496291063228541,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000242838048352857,
      "loss": 5.2543,
      "mean_token_accuracy": 0.2174023672938347,
      "num_tokens": 116048402.0,
      "step": 62240
    },
    {
      "entropy": 6.039603185653687,
      "epoch": 5.496732603320382,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00024280688054603084,
      "loss": 5.2067,
      "mean_token_accuracy": 0.225368832051754,
      "num_tokens": 116056678.0,
      "step": 62245
    },
    {
      "entropy": 6.016619491577148,
      "epoch": 5.497174143412222,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024277571336984955,
      "loss": 5.1993,
      "mean_token_accuracy": 0.2313400164246559,
      "num_tokens": 116065654.0,
      "step": 62250
    },
    {
      "entropy": 6.102425050735474,
      "epoch": 5.497615683504062,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024274454682492375,
      "loss": 5.292,
      "mean_token_accuracy": 0.21568685173988342,
      "num_tokens": 116075080.0,
      "step": 62255
    },
    {
      "entropy": 6.045001554489136,
      "epoch": 5.498057223595903,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024271338091186396,
      "loss": 5.271,
      "mean_token_accuracy": 0.2209614932537079,
      "num_tokens": 116083518.0,
      "step": 62260
    },
    {
      "entropy": 6.026791572570801,
      "epoch": 5.498498763687743,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024268221563128069,
      "loss": 5.2304,
      "mean_token_accuracy": 0.22872748523950576,
      "num_tokens": 116092863.0,
      "step": 62265
    },
    {
      "entropy": 6.111660385131836,
      "epoch": 5.498940303779583,
      "grad_norm": 1.609375,
      "learning_rate": 0.0002426510509837843,
      "loss": 5.2614,
      "mean_token_accuracy": 0.22451234757900237,
      "num_tokens": 116100880.0,
      "step": 62270
    },
    {
      "entropy": 6.063886404037476,
      "epoch": 5.499381843871424,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024261988696998554,
      "loss": 5.2278,
      "mean_token_accuracy": 0.22635110318660737,
      "num_tokens": 116110360.0,
      "step": 62275
    },
    {
      "entropy": 6.0925047397613525,
      "epoch": 5.499823383963264,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024258872359049478,
      "loss": 5.283,
      "mean_token_accuracy": 0.2196325033903122,
      "num_tokens": 116119528.0,
      "step": 62280
    },
    {
      "entropy": 5.996968030929565,
      "epoch": 5.500264924055104,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002425575608459225,
      "loss": 5.1526,
      "mean_token_accuracy": 0.2249797984957695,
      "num_tokens": 116129322.0,
      "step": 62285
    },
    {
      "entropy": 6.053278255462646,
      "epoch": 5.500706464146944,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024252639873687903,
      "loss": 5.1891,
      "mean_token_accuracy": 0.2241920456290245,
      "num_tokens": 116139091.0,
      "step": 62290
    },
    {
      "entropy": 6.038626480102539,
      "epoch": 5.501148004238785,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002424952372639751,
      "loss": 5.1706,
      "mean_token_accuracy": 0.2322185829281807,
      "num_tokens": 116147893.0,
      "step": 62295
    },
    {
      "entropy": 6.070420455932617,
      "epoch": 5.501589544330625,
      "grad_norm": 1.4375,
      "learning_rate": 0.00024246407642782087,
      "loss": 5.2809,
      "mean_token_accuracy": 0.21160894632339478,
      "num_tokens": 116157068.0,
      "step": 62300
    },
    {
      "entropy": 6.030198383331299,
      "epoch": 5.502031084422466,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024243291622902697,
      "loss": 5.1443,
      "mean_token_accuracy": 0.2327238067984581,
      "num_tokens": 116166192.0,
      "step": 62305
    },
    {
      "entropy": 6.0860999584197994,
      "epoch": 5.502472624514306,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00024240175666820368,
      "loss": 5.2709,
      "mean_token_accuracy": 0.22437336891889573,
      "num_tokens": 116175376.0,
      "step": 62310
    },
    {
      "entropy": 6.034239864349365,
      "epoch": 5.502914164606146,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00024237059774596137,
      "loss": 5.2225,
      "mean_token_accuracy": 0.221588596701622,
      "num_tokens": 116185472.0,
      "step": 62315
    },
    {
      "entropy": 6.071870517730713,
      "epoch": 5.503355704697986,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024233943946291053,
      "loss": 5.1824,
      "mean_token_accuracy": 0.2237004294991493,
      "num_tokens": 116194811.0,
      "step": 62320
    },
    {
      "entropy": 6.109658193588257,
      "epoch": 5.503797244789827,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024230828181966157,
      "loss": 5.2445,
      "mean_token_accuracy": 0.21939802318811416,
      "num_tokens": 116204062.0,
      "step": 62325
    },
    {
      "entropy": 6.105697965621948,
      "epoch": 5.504238784881667,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024227712481682473,
      "loss": 5.2819,
      "mean_token_accuracy": 0.21881347447633742,
      "num_tokens": 116213626.0,
      "step": 62330
    },
    {
      "entropy": 6.029559755325318,
      "epoch": 5.504680324973507,
      "grad_norm": 1.4375,
      "learning_rate": 0.00024224596845501034,
      "loss": 5.271,
      "mean_token_accuracy": 0.21989672482013703,
      "num_tokens": 116222815.0,
      "step": 62335
    },
    {
      "entropy": 6.099753999710083,
      "epoch": 5.505121865065348,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024221481273482888,
      "loss": 5.3078,
      "mean_token_accuracy": 0.21935114562511443,
      "num_tokens": 116232633.0,
      "step": 62340
    },
    {
      "entropy": 6.0804449081420895,
      "epoch": 5.505563405157188,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002421836576568906,
      "loss": 5.2089,
      "mean_token_accuracy": 0.22915463745594025,
      "num_tokens": 116241250.0,
      "step": 62345
    },
    {
      "entropy": 6.059580612182617,
      "epoch": 5.506004945249028,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024215250322180581,
      "loss": 5.2284,
      "mean_token_accuracy": 0.22293934375047683,
      "num_tokens": 116251014.0,
      "step": 62350
    },
    {
      "entropy": 6.097855043411255,
      "epoch": 5.506446485340869,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024212134943018472,
      "loss": 5.2917,
      "mean_token_accuracy": 0.21383731663227082,
      "num_tokens": 116259541.0,
      "step": 62355
    },
    {
      "entropy": 5.910884380340576,
      "epoch": 5.506888025432709,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00024209019628263772,
      "loss": 5.1705,
      "mean_token_accuracy": 0.2385655388236046,
      "num_tokens": 116268138.0,
      "step": 62360
    },
    {
      "entropy": 5.944452905654908,
      "epoch": 5.507329565524549,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024205904377977511,
      "loss": 5.2262,
      "mean_token_accuracy": 0.22971484065055847,
      "num_tokens": 116277808.0,
      "step": 62365
    },
    {
      "entropy": 6.070479822158814,
      "epoch": 5.50777110561639,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002420278919222071,
      "loss": 5.2174,
      "mean_token_accuracy": 0.22051677107810974,
      "num_tokens": 116286605.0,
      "step": 62370
    },
    {
      "entropy": 6.056458282470703,
      "epoch": 5.50821264570823,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002419967407105439,
      "loss": 5.1494,
      "mean_token_accuracy": 0.23069103360176085,
      "num_tokens": 116295685.0,
      "step": 62375
    },
    {
      "entropy": 6.053239536285401,
      "epoch": 5.50865418580007,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002419655901453957,
      "loss": 5.2446,
      "mean_token_accuracy": 0.21575284451246263,
      "num_tokens": 116305134.0,
      "step": 62380
    },
    {
      "entropy": 6.025068569183349,
      "epoch": 5.509095725891911,
      "grad_norm": 1.484375,
      "learning_rate": 0.00024193444022737288,
      "loss": 5.1769,
      "mean_token_accuracy": 0.23222784101963043,
      "num_tokens": 116313907.0,
      "step": 62385
    },
    {
      "entropy": 6.057039022445679,
      "epoch": 5.509537265983751,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00024190329095708562,
      "loss": 5.3057,
      "mean_token_accuracy": 0.21624220460653304,
      "num_tokens": 116323375.0,
      "step": 62390
    },
    {
      "entropy": 6.087482166290283,
      "epoch": 5.509978806075591,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024187214233514404,
      "loss": 5.2429,
      "mean_token_accuracy": 0.21670276075601577,
      "num_tokens": 116332591.0,
      "step": 62395
    },
    {
      "entropy": 6.060337066650391,
      "epoch": 5.510420346167432,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002418409943621583,
      "loss": 5.2108,
      "mean_token_accuracy": 0.22630608528852464,
      "num_tokens": 116341246.0,
      "step": 62400
    },
    {
      "entropy": 6.026460027694702,
      "epoch": 5.510861886259272,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024180984703873866,
      "loss": 5.22,
      "mean_token_accuracy": 0.2205394133925438,
      "num_tokens": 116350571.0,
      "step": 62405
    },
    {
      "entropy": 6.105708217620849,
      "epoch": 5.5113034263511125,
      "grad_norm": 1.5,
      "learning_rate": 0.00024177870036549526,
      "loss": 5.351,
      "mean_token_accuracy": 0.21739721298217773,
      "num_tokens": 116359331.0,
      "step": 62410
    },
    {
      "entropy": 6.125597858428955,
      "epoch": 5.5117449664429525,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002417475543430382,
      "loss": 5.2953,
      "mean_token_accuracy": 0.22049131095409394,
      "num_tokens": 116368866.0,
      "step": 62415
    },
    {
      "entropy": 6.077592563629151,
      "epoch": 5.5121865065347935,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00024171640897197754,
      "loss": 5.2484,
      "mean_token_accuracy": 0.22010924518108368,
      "num_tokens": 116378319.0,
      "step": 62420
    },
    {
      "entropy": 6.0937563419342045,
      "epoch": 5.5126280466266335,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024168526425292363,
      "loss": 5.2357,
      "mean_token_accuracy": 0.22272469103336334,
      "num_tokens": 116387488.0,
      "step": 62425
    },
    {
      "entropy": 6.050289630889893,
      "epoch": 5.5130695867184745,
      "grad_norm": 1.765625,
      "learning_rate": 0.00024165412018648641,
      "loss": 5.2248,
      "mean_token_accuracy": 0.2192055955529213,
      "num_tokens": 116396416.0,
      "step": 62430
    },
    {
      "entropy": 6.058220529556275,
      "epoch": 5.5135111268103145,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024162297677327604,
      "loss": 5.2015,
      "mean_token_accuracy": 0.2218688502907753,
      "num_tokens": 116406324.0,
      "step": 62435
    },
    {
      "entropy": 6.084795618057251,
      "epoch": 5.513952666902155,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024159183401390255,
      "loss": 5.2391,
      "mean_token_accuracy": 0.21519234776496887,
      "num_tokens": 116416071.0,
      "step": 62440
    },
    {
      "entropy": 6.001453256607055,
      "epoch": 5.514394206993995,
      "grad_norm": 1.46875,
      "learning_rate": 0.000241560691908976,
      "loss": 5.1468,
      "mean_token_accuracy": 0.2278449222445488,
      "num_tokens": 116425283.0,
      "step": 62445
    },
    {
      "entropy": 6.016279983520508,
      "epoch": 5.514835747085836,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024152955045910653,
      "loss": 5.2112,
      "mean_token_accuracy": 0.23094690442085267,
      "num_tokens": 116434209.0,
      "step": 62450
    },
    {
      "entropy": 6.017763090133667,
      "epoch": 5.515277287177676,
      "grad_norm": 1.515625,
      "learning_rate": 0.00024149840966490412,
      "loss": 5.2458,
      "mean_token_accuracy": 0.22085121124982834,
      "num_tokens": 116442674.0,
      "step": 62455
    },
    {
      "entropy": 6.171484088897705,
      "epoch": 5.515718827269517,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00024146726952697884,
      "loss": 5.2953,
      "mean_token_accuracy": 0.22005636841058732,
      "num_tokens": 116451406.0,
      "step": 62460
    },
    {
      "entropy": 6.069138097763061,
      "epoch": 5.516160367361357,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002414361300459405,
      "loss": 5.1927,
      "mean_token_accuracy": 0.23112663477659226,
      "num_tokens": 116460651.0,
      "step": 62465
    },
    {
      "entropy": 5.991749048233032,
      "epoch": 5.516601907453197,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024140499122239946,
      "loss": 5.1197,
      "mean_token_accuracy": 0.2239631786942482,
      "num_tokens": 116468856.0,
      "step": 62470
    },
    {
      "entropy": 6.012980985641479,
      "epoch": 5.517043447545037,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00024137385305696553,
      "loss": 5.2482,
      "mean_token_accuracy": 0.22281948179006578,
      "num_tokens": 116478452.0,
      "step": 62475
    },
    {
      "entropy": 6.104845571517944,
      "epoch": 5.517484987636878,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002413427155502486,
      "loss": 5.2631,
      "mean_token_accuracy": 0.21733763366937636,
      "num_tokens": 116487912.0,
      "step": 62480
    },
    {
      "entropy": 6.0699324131011965,
      "epoch": 5.517926527728718,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024131157870285885,
      "loss": 5.1575,
      "mean_token_accuracy": 0.22895239293575287,
      "num_tokens": 116496876.0,
      "step": 62485
    },
    {
      "entropy": 6.022778844833374,
      "epoch": 5.518368067820558,
      "grad_norm": 1.375,
      "learning_rate": 0.00024128044251540604,
      "loss": 5.2184,
      "mean_token_accuracy": 0.21867665350437165,
      "num_tokens": 116506464.0,
      "step": 62490
    },
    {
      "entropy": 5.992070436477661,
      "epoch": 5.518809607912399,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00024124930698850018,
      "loss": 5.2339,
      "mean_token_accuracy": 0.21880560368299484,
      "num_tokens": 116516057.0,
      "step": 62495
    },
    {
      "entropy": 6.0393805503845215,
      "epoch": 5.519251148004239,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002412181721227512,
      "loss": 5.2059,
      "mean_token_accuracy": 0.22528675347566604,
      "num_tokens": 116524882.0,
      "step": 62500
    },
    {
      "entropy": 6.089426136016845,
      "epoch": 5.519692688096079,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002411870379187691,
      "loss": 5.2285,
      "mean_token_accuracy": 0.2278471916913986,
      "num_tokens": 116534018.0,
      "step": 62505
    },
    {
      "entropy": 6.087402200698852,
      "epoch": 5.52013422818792,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00024115590437716355,
      "loss": 5.3054,
      "mean_token_accuracy": 0.2139840915799141,
      "num_tokens": 116542772.0,
      "step": 62510
    },
    {
      "entropy": 6.066777849197388,
      "epoch": 5.52057576827976,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00024112477149854466,
      "loss": 5.2577,
      "mean_token_accuracy": 0.22026961892843247,
      "num_tokens": 116551827.0,
      "step": 62515
    },
    {
      "entropy": 6.143025159835815,
      "epoch": 5.5210173083716,
      "grad_norm": 1.375,
      "learning_rate": 0.00024109363928352228,
      "loss": 5.3407,
      "mean_token_accuracy": 0.22305194288492203,
      "num_tokens": 116562028.0,
      "step": 62520
    },
    {
      "entropy": 6.097217559814453,
      "epoch": 5.521458848463441,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024106250773270622,
      "loss": 5.2181,
      "mean_token_accuracy": 0.22127556353807448,
      "num_tokens": 116570711.0,
      "step": 62525
    },
    {
      "entropy": 6.030672883987426,
      "epoch": 5.521900388555281,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024103137684670624,
      "loss": 5.1753,
      "mean_token_accuracy": 0.23060980588197708,
      "num_tokens": 116580343.0,
      "step": 62530
    },
    {
      "entropy": 6.074433183670044,
      "epoch": 5.522341928647121,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024100024662613228,
      "loss": 5.2037,
      "mean_token_accuracy": 0.22433370649814605,
      "num_tokens": 116590100.0,
      "step": 62535
    },
    {
      "entropy": 6.032867479324341,
      "epoch": 5.522783468738962,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024096911707159426,
      "loss": 5.1769,
      "mean_token_accuracy": 0.23369849920272828,
      "num_tokens": 116600368.0,
      "step": 62540
    },
    {
      "entropy": 6.1149028778076175,
      "epoch": 5.523225008830802,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024093798818370176,
      "loss": 5.2611,
      "mean_token_accuracy": 0.2262421265244484,
      "num_tokens": 116610401.0,
      "step": 62545
    },
    {
      "entropy": 6.1352368831634525,
      "epoch": 5.523666548922642,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024090685996306483,
      "loss": 5.301,
      "mean_token_accuracy": 0.21980213969945908,
      "num_tokens": 116619601.0,
      "step": 62550
    },
    {
      "entropy": 6.045227432250977,
      "epoch": 5.524108089014483,
      "grad_norm": 1.203125,
      "learning_rate": 0.000240875732410293,
      "loss": 5.285,
      "mean_token_accuracy": 0.22341656535863877,
      "num_tokens": 116629460.0,
      "step": 62555
    },
    {
      "entropy": 6.0532063961029055,
      "epoch": 5.524549629106323,
      "grad_norm": 1.640625,
      "learning_rate": 0.0002408446055259962,
      "loss": 5.1429,
      "mean_token_accuracy": 0.22647874057292938,
      "num_tokens": 116637139.0,
      "step": 62560
    },
    {
      "entropy": 6.0971191883087155,
      "epoch": 5.524991169198163,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002408134793107842,
      "loss": 5.3163,
      "mean_token_accuracy": 0.2169114053249359,
      "num_tokens": 116646566.0,
      "step": 62565
    },
    {
      "entropy": 6.016827249526978,
      "epoch": 5.525432709290003,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00024078235376526663,
      "loss": 5.2156,
      "mean_token_accuracy": 0.22587411850690842,
      "num_tokens": 116656360.0,
      "step": 62570
    },
    {
      "entropy": 6.0550758838653564,
      "epoch": 5.525874249381844,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002407512288900533,
      "loss": 5.2203,
      "mean_token_accuracy": 0.22613471895456314,
      "num_tokens": 116666256.0,
      "step": 62575
    },
    {
      "entropy": 6.085453367233276,
      "epoch": 5.526315789473684,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024072010468575395,
      "loss": 5.2796,
      "mean_token_accuracy": 0.22053480446338652,
      "num_tokens": 116675779.0,
      "step": 62580
    },
    {
      "entropy": 6.031354856491089,
      "epoch": 5.526757329565525,
      "grad_norm": 1.375,
      "learning_rate": 0.00024068898115297828,
      "loss": 5.163,
      "mean_token_accuracy": 0.2364737570285797,
      "num_tokens": 116684976.0,
      "step": 62585
    },
    {
      "entropy": 6.113326978683472,
      "epoch": 5.527198869657365,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024065785829233594,
      "loss": 5.2972,
      "mean_token_accuracy": 0.21997741907835006,
      "num_tokens": 116694147.0,
      "step": 62590
    },
    {
      "entropy": 5.99093451499939,
      "epoch": 5.527640409749205,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00024062673610443646,
      "loss": 5.1746,
      "mean_token_accuracy": 0.2274283304810524,
      "num_tokens": 116703088.0,
      "step": 62595
    },
    {
      "entropy": 6.082138872146606,
      "epoch": 5.528081949841045,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00024059561458988988,
      "loss": 5.2535,
      "mean_token_accuracy": 0.22236556112766265,
      "num_tokens": 116711794.0,
      "step": 62600
    },
    {
      "entropy": 6.039030694961548,
      "epoch": 5.528523489932886,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024056449374930562,
      "loss": 5.2046,
      "mean_token_accuracy": 0.22855071276426314,
      "num_tokens": 116721845.0,
      "step": 62605
    },
    {
      "entropy": 6.058062028884888,
      "epoch": 5.528965030024726,
      "grad_norm": 1.4375,
      "learning_rate": 0.00024053337358329326,
      "loss": 5.216,
      "mean_token_accuracy": 0.2218458652496338,
      "num_tokens": 116731625.0,
      "step": 62610
    },
    {
      "entropy": 6.100890207290649,
      "epoch": 5.529406570116566,
      "grad_norm": 1.515625,
      "learning_rate": 0.00024050225409246256,
      "loss": 5.193,
      "mean_token_accuracy": 0.23057294487953187,
      "num_tokens": 116740230.0,
      "step": 62615
    },
    {
      "entropy": 6.030095481872559,
      "epoch": 5.529848110208407,
      "grad_norm": 1.484375,
      "learning_rate": 0.00024047113527742305,
      "loss": 5.155,
      "mean_token_accuracy": 0.2269255131483078,
      "num_tokens": 116749057.0,
      "step": 62620
    },
    {
      "entropy": 6.017797183990479,
      "epoch": 5.530289650300247,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002404400171387844,
      "loss": 5.2745,
      "mean_token_accuracy": 0.22145338207483292,
      "num_tokens": 116758402.0,
      "step": 62625
    },
    {
      "entropy": 6.031425046920776,
      "epoch": 5.530731190392087,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024040889967715613,
      "loss": 5.2224,
      "mean_token_accuracy": 0.21702628284692765,
      "num_tokens": 116768667.0,
      "step": 62630
    },
    {
      "entropy": 6.036007356643677,
      "epoch": 5.531172730483928,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024037778289314789,
      "loss": 5.1657,
      "mean_token_accuracy": 0.23071809262037277,
      "num_tokens": 116778220.0,
      "step": 62635
    },
    {
      "entropy": 6.062730216979981,
      "epoch": 5.531614270575768,
      "grad_norm": 1.546875,
      "learning_rate": 0.00024034666678736906,
      "loss": 5.1708,
      "mean_token_accuracy": 0.2296672597527504,
      "num_tokens": 116787186.0,
      "step": 62640
    },
    {
      "entropy": 6.036794900894165,
      "epoch": 5.532055810667608,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024031555136042943,
      "loss": 5.2184,
      "mean_token_accuracy": 0.22316948026418687,
      "num_tokens": 116797204.0,
      "step": 62645
    },
    {
      "entropy": 6.045986175537109,
      "epoch": 5.532497350759449,
      "grad_norm": 1.46875,
      "learning_rate": 0.00024028443661293846,
      "loss": 5.1644,
      "mean_token_accuracy": 0.22580112814903258,
      "num_tokens": 116806160.0,
      "step": 62650
    },
    {
      "entropy": 5.977301597595215,
      "epoch": 5.532938890851289,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024025332254550558,
      "loss": 5.1126,
      "mean_token_accuracy": 0.24137209802865983,
      "num_tokens": 116816029.0,
      "step": 62655
    },
    {
      "entropy": 6.092135667800903,
      "epoch": 5.533380430943129,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024022220915874027,
      "loss": 5.3179,
      "mean_token_accuracy": 0.2168172314763069,
      "num_tokens": 116825519.0,
      "step": 62660
    },
    {
      "entropy": 6.059766817092895,
      "epoch": 5.53382197103497,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002401910964532522,
      "loss": 5.1897,
      "mean_token_accuracy": 0.22940254360437393,
      "num_tokens": 116834803.0,
      "step": 62665
    },
    {
      "entropy": 6.078421497344971,
      "epoch": 5.53426351112681,
      "grad_norm": 1.4375,
      "learning_rate": 0.00024015998442965082,
      "loss": 5.2889,
      "mean_token_accuracy": 0.21773108392953872,
      "num_tokens": 116844747.0,
      "step": 62670
    },
    {
      "entropy": 6.089973402023316,
      "epoch": 5.53470505121865,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00024012887308854535,
      "loss": 5.2219,
      "mean_token_accuracy": 0.21747233867645263,
      "num_tokens": 116853795.0,
      "step": 62675
    },
    {
      "entropy": 6.072410678863525,
      "epoch": 5.535146591310491,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002400977624305456,
      "loss": 5.2878,
      "mean_token_accuracy": 0.22073231786489486,
      "num_tokens": 116862459.0,
      "step": 62680
    },
    {
      "entropy": 6.049811840057373,
      "epoch": 5.535588131402331,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00024006665245626075,
      "loss": 5.2056,
      "mean_token_accuracy": 0.22118400484323503,
      "num_tokens": 116872780.0,
      "step": 62685
    },
    {
      "entropy": 5.987762451171875,
      "epoch": 5.5360296714941715,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024003554316630028,
      "loss": 5.1885,
      "mean_token_accuracy": 0.22823083102703096,
      "num_tokens": 116882979.0,
      "step": 62690
    },
    {
      "entropy": 6.124750995635987,
      "epoch": 5.5364712115860115,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002400044345612737,
      "loss": 5.3012,
      "mean_token_accuracy": 0.21565325558185577,
      "num_tokens": 116892074.0,
      "step": 62695
    },
    {
      "entropy": 6.089807224273682,
      "epoch": 5.5369127516778525,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00023997332664179034,
      "loss": 5.1804,
      "mean_token_accuracy": 0.22911562770605087,
      "num_tokens": 116901156.0,
      "step": 62700
    },
    {
      "entropy": 6.079555034637451,
      "epoch": 5.5373542917696925,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023994221940845951,
      "loss": 5.2193,
      "mean_token_accuracy": 0.2254370391368866,
      "num_tokens": 116911651.0,
      "step": 62705
    },
    {
      "entropy": 6.077653932571411,
      "epoch": 5.5377958318615335,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023991111286189072,
      "loss": 5.304,
      "mean_token_accuracy": 0.2207689628005028,
      "num_tokens": 116921646.0,
      "step": 62710
    },
    {
      "entropy": 6.02825436592102,
      "epoch": 5.5382373719533735,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002398800070026933,
      "loss": 5.1828,
      "mean_token_accuracy": 0.22705733329057692,
      "num_tokens": 116930398.0,
      "step": 62715
    },
    {
      "entropy": 6.028886413574218,
      "epoch": 5.538678912045214,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023984890183147657,
      "loss": 5.1889,
      "mean_token_accuracy": 0.2281762331724167,
      "num_tokens": 116939902.0,
      "step": 62720
    },
    {
      "entropy": 6.051761722564697,
      "epoch": 5.539120452137054,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023981779734884984,
      "loss": 5.2349,
      "mean_token_accuracy": 0.22814268618822098,
      "num_tokens": 116949756.0,
      "step": 62725
    },
    {
      "entropy": 6.047085285186768,
      "epoch": 5.539561992228895,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023978669355542237,
      "loss": 5.1291,
      "mean_token_accuracy": 0.23668964356184005,
      "num_tokens": 116958950.0,
      "step": 62730
    },
    {
      "entropy": 6.0537700176239015,
      "epoch": 5.540003532320735,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002397555904518036,
      "loss": 5.2177,
      "mean_token_accuracy": 0.2195592075586319,
      "num_tokens": 116968318.0,
      "step": 62735
    },
    {
      "entropy": 6.071507740020752,
      "epoch": 5.5404450724125756,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002397244880386028,
      "loss": 5.2994,
      "mean_token_accuracy": 0.21885300576686859,
      "num_tokens": 116978195.0,
      "step": 62740
    },
    {
      "entropy": 6.044378423690796,
      "epoch": 5.540886612504416,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023969338631642923,
      "loss": 5.2436,
      "mean_token_accuracy": 0.22041541188955308,
      "num_tokens": 116988345.0,
      "step": 62745
    },
    {
      "entropy": 6.114583110809326,
      "epoch": 5.541328152596256,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023966228528589213,
      "loss": 5.3217,
      "mean_token_accuracy": 0.21173076331615448,
      "num_tokens": 116998679.0,
      "step": 62750
    },
    {
      "entropy": 6.101701545715332,
      "epoch": 5.541769692688096,
      "grad_norm": 1.546875,
      "learning_rate": 0.00023963118494760072,
      "loss": 5.1935,
      "mean_token_accuracy": 0.23335042148828505,
      "num_tokens": 117007504.0,
      "step": 62755
    },
    {
      "entropy": 6.103125333786011,
      "epoch": 5.542211232779937,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002396000853021643,
      "loss": 5.2372,
      "mean_token_accuracy": 0.22407971173524857,
      "num_tokens": 117016526.0,
      "step": 62760
    },
    {
      "entropy": 6.084678840637207,
      "epoch": 5.542652772871777,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023956898635019215,
      "loss": 5.1644,
      "mean_token_accuracy": 0.2261310636997223,
      "num_tokens": 117024886.0,
      "step": 62765
    },
    {
      "entropy": 6.0280708312988285,
      "epoch": 5.543094312963617,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00023953788809229326,
      "loss": 5.1497,
      "mean_token_accuracy": 0.2221749544143677,
      "num_tokens": 117033509.0,
      "step": 62770
    },
    {
      "entropy": 5.922865962982177,
      "epoch": 5.543535853055458,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002395067905290771,
      "loss": 5.0421,
      "mean_token_accuracy": 0.2476954147219658,
      "num_tokens": 117042866.0,
      "step": 62775
    },
    {
      "entropy": 6.029966974258423,
      "epoch": 5.543977393147298,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002394756936611527,
      "loss": 5.1727,
      "mean_token_accuracy": 0.2284366548061371,
      "num_tokens": 117051539.0,
      "step": 62780
    },
    {
      "entropy": 6.080309867858887,
      "epoch": 5.544418933239138,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023944459748912928,
      "loss": 5.2243,
      "mean_token_accuracy": 0.22167014479637145,
      "num_tokens": 117061235.0,
      "step": 62785
    },
    {
      "entropy": 5.986869525909424,
      "epoch": 5.544860473330979,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023941350201361596,
      "loss": 5.1764,
      "mean_token_accuracy": 0.23041078597307205,
      "num_tokens": 117070743.0,
      "step": 62790
    },
    {
      "entropy": 6.085215950012207,
      "epoch": 5.545302013422819,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023938240723522176,
      "loss": 5.274,
      "mean_token_accuracy": 0.21843780428171158,
      "num_tokens": 117079885.0,
      "step": 62795
    },
    {
      "entropy": 6.100526809692383,
      "epoch": 5.545743553514659,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002393513131545561,
      "loss": 5.2562,
      "mean_token_accuracy": 0.22414913177490234,
      "num_tokens": 117090281.0,
      "step": 62800
    },
    {
      "entropy": 6.007891130447388,
      "epoch": 5.5461850936065,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023932021977222785,
      "loss": 5.1418,
      "mean_token_accuracy": 0.22690737545490264,
      "num_tokens": 117099667.0,
      "step": 62805
    },
    {
      "entropy": 6.076144456863403,
      "epoch": 5.54662663369834,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023928912708884632,
      "loss": 5.2758,
      "mean_token_accuracy": 0.21463903188705444,
      "num_tokens": 117108758.0,
      "step": 62810
    },
    {
      "entropy": 5.998135948181153,
      "epoch": 5.54706817379018,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023925803510502032,
      "loss": 5.1517,
      "mean_token_accuracy": 0.22582096457481385,
      "num_tokens": 117117264.0,
      "step": 62815
    },
    {
      "entropy": 6.037234878540039,
      "epoch": 5.547509713882021,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023922694382135916,
      "loss": 5.1986,
      "mean_token_accuracy": 0.2236717864871025,
      "num_tokens": 117126010.0,
      "step": 62820
    },
    {
      "entropy": 6.029758548736572,
      "epoch": 5.547951253973861,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023919585323847192,
      "loss": 5.2171,
      "mean_token_accuracy": 0.22710440158843995,
      "num_tokens": 117135911.0,
      "step": 62825
    },
    {
      "entropy": 6.0315264701843265,
      "epoch": 5.548392794065701,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023916476335696752,
      "loss": 5.1589,
      "mean_token_accuracy": 0.22981197237968445,
      "num_tokens": 117145551.0,
      "step": 62830
    },
    {
      "entropy": 6.03686408996582,
      "epoch": 5.548834334157542,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00023913367417745485,
      "loss": 5.2056,
      "mean_token_accuracy": 0.2298249661922455,
      "num_tokens": 117155298.0,
      "step": 62835
    },
    {
      "entropy": 6.021312999725342,
      "epoch": 5.549275874249382,
      "grad_norm": 1.375,
      "learning_rate": 0.00023910258570054328,
      "loss": 5.2003,
      "mean_token_accuracy": 0.2265899658203125,
      "num_tokens": 117164277.0,
      "step": 62840
    },
    {
      "entropy": 6.0606121063232425,
      "epoch": 5.549717414341222,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023907149792684162,
      "loss": 5.2224,
      "mean_token_accuracy": 0.23214369863271714,
      "num_tokens": 117173072.0,
      "step": 62845
    },
    {
      "entropy": 6.080872535705566,
      "epoch": 5.550158954433062,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023904041085695888,
      "loss": 5.2257,
      "mean_token_accuracy": 0.22521917819976806,
      "num_tokens": 117182413.0,
      "step": 62850
    },
    {
      "entropy": 6.010768890380859,
      "epoch": 5.550600494524903,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023900932449150402,
      "loss": 5.1427,
      "mean_token_accuracy": 0.2370156764984131,
      "num_tokens": 117191479.0,
      "step": 62855
    },
    {
      "entropy": 6.062417316436767,
      "epoch": 5.551042034616743,
      "grad_norm": 1.390625,
      "learning_rate": 0.000238978238831086,
      "loss": 5.2438,
      "mean_token_accuracy": 0.22632353156805038,
      "num_tokens": 117200407.0,
      "step": 62860
    },
    {
      "entropy": 6.020398712158203,
      "epoch": 5.551483574708584,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023894715387631383,
      "loss": 5.2479,
      "mean_token_accuracy": 0.22392435669898986,
      "num_tokens": 117208807.0,
      "step": 62865
    },
    {
      "entropy": 5.965640497207642,
      "epoch": 5.551925114800424,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023891606962779638,
      "loss": 5.0831,
      "mean_token_accuracy": 0.24054908454418183,
      "num_tokens": 117217057.0,
      "step": 62870
    },
    {
      "entropy": 6.06219744682312,
      "epoch": 5.552366654892264,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023888498608614262,
      "loss": 5.2648,
      "mean_token_accuracy": 0.21967781484127044,
      "num_tokens": 117225830.0,
      "step": 62875
    },
    {
      "entropy": 6.018804407119751,
      "epoch": 5.552808194984104,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023885390325196137,
      "loss": 5.1967,
      "mean_token_accuracy": 0.22941365092992783,
      "num_tokens": 117234614.0,
      "step": 62880
    },
    {
      "entropy": 6.074917840957641,
      "epoch": 5.553249735075945,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023882282112586167,
      "loss": 5.2268,
      "mean_token_accuracy": 0.21970055252313614,
      "num_tokens": 117245216.0,
      "step": 62885
    },
    {
      "entropy": 6.025595808029175,
      "epoch": 5.553691275167785,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023879173970845242,
      "loss": 5.1739,
      "mean_token_accuracy": 0.22771623432636262,
      "num_tokens": 117254414.0,
      "step": 62890
    },
    {
      "entropy": 6.068880701065064,
      "epoch": 5.554132815259625,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023876065900034233,
      "loss": 5.2216,
      "mean_token_accuracy": 0.22303633391857147,
      "num_tokens": 117263664.0,
      "step": 62895
    },
    {
      "entropy": 6.045490789413452,
      "epoch": 5.554574355351466,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002387295790021402,
      "loss": 5.2381,
      "mean_token_accuracy": 0.22445165663957595,
      "num_tokens": 117273191.0,
      "step": 62900
    },
    {
      "entropy": 5.982860660552978,
      "epoch": 5.555015895443306,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023869849971445512,
      "loss": 5.1042,
      "mean_token_accuracy": 0.2405182644724846,
      "num_tokens": 117282924.0,
      "step": 62905
    },
    {
      "entropy": 6.0322614192962645,
      "epoch": 5.555457435535146,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00023866742113789574,
      "loss": 5.2142,
      "mean_token_accuracy": 0.22541653513908386,
      "num_tokens": 117292464.0,
      "step": 62910
    },
    {
      "entropy": 6.045757579803467,
      "epoch": 5.555898975626987,
      "grad_norm": 1.296875,
      "learning_rate": 0.000238636343273071,
      "loss": 5.2282,
      "mean_token_accuracy": 0.22727230936288834,
      "num_tokens": 117301798.0,
      "step": 62915
    },
    {
      "entropy": 6.044881772994995,
      "epoch": 5.556340515718827,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023860526612058948,
      "loss": 5.2343,
      "mean_token_accuracy": 0.22266731411218643,
      "num_tokens": 117311121.0,
      "step": 62920
    },
    {
      "entropy": 6.023078584671021,
      "epoch": 5.556782055810667,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023857418968106016,
      "loss": 5.1532,
      "mean_token_accuracy": 0.24099140465259553,
      "num_tokens": 117321468.0,
      "step": 62925
    },
    {
      "entropy": 6.050618696212768,
      "epoch": 5.557223595902508,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023854311395509175,
      "loss": 5.2562,
      "mean_token_accuracy": 0.2237761065363884,
      "num_tokens": 117330938.0,
      "step": 62930
    },
    {
      "entropy": 6.0561206340789795,
      "epoch": 5.557665135994348,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023851203894329298,
      "loss": 5.2426,
      "mean_token_accuracy": 0.2225588768720627,
      "num_tokens": 117340278.0,
      "step": 62935
    },
    {
      "entropy": 6.068589782714843,
      "epoch": 5.558106676086188,
      "grad_norm": 1.53125,
      "learning_rate": 0.00023848096464627267,
      "loss": 5.2246,
      "mean_token_accuracy": 0.22208599299192427,
      "num_tokens": 117348896.0,
      "step": 62940
    },
    {
      "entropy": 6.036426591873169,
      "epoch": 5.558548216178029,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00023844989106463937,
      "loss": 5.1596,
      "mean_token_accuracy": 0.2352440521121025,
      "num_tokens": 117357409.0,
      "step": 62945
    },
    {
      "entropy": 6.0688323974609375,
      "epoch": 5.558989756269869,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023841881819900202,
      "loss": 5.2362,
      "mean_token_accuracy": 0.22196311205625535,
      "num_tokens": 117366863.0,
      "step": 62950
    },
    {
      "entropy": 6.137952184677124,
      "epoch": 5.559431296361709,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002383877460499692,
      "loss": 5.3456,
      "mean_token_accuracy": 0.20584270656108855,
      "num_tokens": 117378812.0,
      "step": 62955
    },
    {
      "entropy": 6.136503124237061,
      "epoch": 5.55987283645355,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002383566746181497,
      "loss": 5.2558,
      "mean_token_accuracy": 0.2205076977610588,
      "num_tokens": 117388983.0,
      "step": 62960
    },
    {
      "entropy": 6.068946886062622,
      "epoch": 5.56031437654539,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00023832560390415203,
      "loss": 5.2396,
      "mean_token_accuracy": 0.21994371116161346,
      "num_tokens": 117397631.0,
      "step": 62965
    },
    {
      "entropy": 5.945316028594971,
      "epoch": 5.5607559166372305,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023829453390858486,
      "loss": 5.1242,
      "mean_token_accuracy": 0.23328791409730912,
      "num_tokens": 117406734.0,
      "step": 62970
    },
    {
      "entropy": 5.9617565155029295,
      "epoch": 5.5611974567290705,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023826346463205696,
      "loss": 5.1005,
      "mean_token_accuracy": 0.2360785722732544,
      "num_tokens": 117415930.0,
      "step": 62975
    },
    {
      "entropy": 6.029478979110718,
      "epoch": 5.5616389968209115,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002382323960751769,
      "loss": 5.2973,
      "mean_token_accuracy": 0.22758392840623856,
      "num_tokens": 117424864.0,
      "step": 62980
    },
    {
      "entropy": 5.995655059814453,
      "epoch": 5.5620805369127515,
      "grad_norm": 1.59375,
      "learning_rate": 0.00023820132823855323,
      "loss": 5.1313,
      "mean_token_accuracy": 0.22958495914936067,
      "num_tokens": 117433236.0,
      "step": 62985
    },
    {
      "entropy": 6.01356725692749,
      "epoch": 5.5625220770045924,
      "grad_norm": 1.25,
      "learning_rate": 0.00023817026112279462,
      "loss": 5.2125,
      "mean_token_accuracy": 0.22891737818717955,
      "num_tokens": 117442499.0,
      "step": 62990
    },
    {
      "entropy": 6.024672031402588,
      "epoch": 5.5629636170964325,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002381391947285097,
      "loss": 5.2605,
      "mean_token_accuracy": 0.2185487389564514,
      "num_tokens": 117451187.0,
      "step": 62995
    },
    {
      "entropy": 5.943870258331299,
      "epoch": 5.5634051571882726,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00023810812905630696,
      "loss": 5.0668,
      "mean_token_accuracy": 0.23958403021097183,
      "num_tokens": 117459268.0,
      "step": 63000
    },
    {
      "epoch": 5.5634051571882726,
      "eval_entropy": 5.704803832714042,
      "eval_loss": 5.300240516662598,
      "eval_mean_token_accuracy": 0.22902585296540406,
      "eval_num_tokens": 117459268.0,
      "eval_runtime": 20.243,
      "eval_samples_per_second": 1744.555,
      "eval_steps_per_second": 218.1,
      "step": 63000
    },
    {
      "entropy": 6.050472640991211,
      "epoch": 5.563846697280113,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023807706410679502,
      "loss": 5.2494,
      "mean_token_accuracy": 0.21633954495191574,
      "num_tokens": 117469965.0,
      "step": 63005
    },
    {
      "entropy": 6.087566566467285,
      "epoch": 5.5642882373719535,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00023804599988058234,
      "loss": 5.3095,
      "mean_token_accuracy": 0.21511188596487046,
      "num_tokens": 117479145.0,
      "step": 63010
    },
    {
      "entropy": 6.104937887191772,
      "epoch": 5.564729777463794,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023801493637827753,
      "loss": 5.2568,
      "mean_token_accuracy": 0.2253829598426819,
      "num_tokens": 117488288.0,
      "step": 63015
    },
    {
      "entropy": 6.117766189575195,
      "epoch": 5.5651713175556345,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002379838736004891,
      "loss": 5.2529,
      "mean_token_accuracy": 0.22111881971359254,
      "num_tokens": 117497569.0,
      "step": 63020
    },
    {
      "entropy": 6.010272693634033,
      "epoch": 5.565612857647475,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023795281154782562,
      "loss": 5.2305,
      "mean_token_accuracy": 0.22635496705770491,
      "num_tokens": 117506810.0,
      "step": 63025
    },
    {
      "entropy": 5.9815150737762455,
      "epoch": 5.566054397739315,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002379217502208954,
      "loss": 5.2266,
      "mean_token_accuracy": 0.22378915846347808,
      "num_tokens": 117517587.0,
      "step": 63030
    },
    {
      "entropy": 6.0989350318908695,
      "epoch": 5.566495937831155,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023789068962030692,
      "loss": 5.2402,
      "mean_token_accuracy": 0.21702072620391846,
      "num_tokens": 117526670.0,
      "step": 63035
    },
    {
      "entropy": 6.128538370132446,
      "epoch": 5.566937477922996,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023785962974666886,
      "loss": 5.3361,
      "mean_token_accuracy": 0.21950642317533492,
      "num_tokens": 117536051.0,
      "step": 63040
    },
    {
      "entropy": 5.9942872524261475,
      "epoch": 5.567379018014836,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002378285706005895,
      "loss": 5.1371,
      "mean_token_accuracy": 0.2320592910051346,
      "num_tokens": 117544990.0,
      "step": 63045
    },
    {
      "entropy": 6.062232589721679,
      "epoch": 5.567820558106676,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023779751218267725,
      "loss": 5.2518,
      "mean_token_accuracy": 0.2209369271993637,
      "num_tokens": 117554032.0,
      "step": 63050
    },
    {
      "entropy": 6.104014158248901,
      "epoch": 5.568262098198517,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002377664544935406,
      "loss": 5.3414,
      "mean_token_accuracy": 0.21474778950214385,
      "num_tokens": 117563788.0,
      "step": 63055
    },
    {
      "entropy": 6.010820960998535,
      "epoch": 5.568703638290357,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002377353975337879,
      "loss": 5.1416,
      "mean_token_accuracy": 0.2262490287423134,
      "num_tokens": 117573716.0,
      "step": 63060
    },
    {
      "entropy": 5.967405700683594,
      "epoch": 5.569145178382197,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00023770434130402762,
      "loss": 5.0736,
      "mean_token_accuracy": 0.2401696652173996,
      "num_tokens": 117583202.0,
      "step": 63065
    },
    {
      "entropy": 6.055924320220948,
      "epoch": 5.569586718474038,
      "grad_norm": 1.5,
      "learning_rate": 0.00023767328580486807,
      "loss": 5.2051,
      "mean_token_accuracy": 0.22205687910318375,
      "num_tokens": 117591228.0,
      "step": 63070
    },
    {
      "entropy": 6.003571081161499,
      "epoch": 5.570028258565878,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002376422310369176,
      "loss": 5.1926,
      "mean_token_accuracy": 0.22436946630477905,
      "num_tokens": 117600743.0,
      "step": 63075
    },
    {
      "entropy": 6.054128789901734,
      "epoch": 5.570469798657718,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002376111770007846,
      "loss": 5.2288,
      "mean_token_accuracy": 0.22850752621889114,
      "num_tokens": 117610104.0,
      "step": 63080
    },
    {
      "entropy": 6.053645992279053,
      "epoch": 5.570911338749559,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023758012369707744,
      "loss": 5.3123,
      "mean_token_accuracy": 0.21898682415485382,
      "num_tokens": 117619988.0,
      "step": 63085
    },
    {
      "entropy": 5.9093935012817385,
      "epoch": 5.571352878841399,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023754907112640434,
      "loss": 5.0384,
      "mean_token_accuracy": 0.23577670156955718,
      "num_tokens": 117628090.0,
      "step": 63090
    },
    {
      "entropy": 6.007919120788574,
      "epoch": 5.571794418933239,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002375180192893736,
      "loss": 5.1974,
      "mean_token_accuracy": 0.227288156747818,
      "num_tokens": 117637881.0,
      "step": 63095
    },
    {
      "entropy": 6.080968618392944,
      "epoch": 5.57223595902508,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002374869681865935,
      "loss": 5.2527,
      "mean_token_accuracy": 0.21528003364801407,
      "num_tokens": 117647056.0,
      "step": 63100
    },
    {
      "entropy": 6.0398400783538815,
      "epoch": 5.57267749911692,
      "grad_norm": 1.375,
      "learning_rate": 0.00023745591781867249,
      "loss": 5.1529,
      "mean_token_accuracy": 0.23247597068548204,
      "num_tokens": 117656077.0,
      "step": 63105
    },
    {
      "entropy": 6.10243034362793,
      "epoch": 5.57311903920876,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002374248681862186,
      "loss": 5.3344,
      "mean_token_accuracy": 0.21969058513641357,
      "num_tokens": 117664829.0,
      "step": 63110
    },
    {
      "entropy": 6.108450078964234,
      "epoch": 5.573560579300601,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00023739381928984018,
      "loss": 5.2524,
      "mean_token_accuracy": 0.2160043939948082,
      "num_tokens": 117673560.0,
      "step": 63115
    },
    {
      "entropy": 6.089063358306885,
      "epoch": 5.574002119392441,
      "grad_norm": 1.484375,
      "learning_rate": 0.00023736277113014544,
      "loss": 5.2313,
      "mean_token_accuracy": 0.22184016704559326,
      "num_tokens": 117682145.0,
      "step": 63120
    },
    {
      "entropy": 5.97505407333374,
      "epoch": 5.574443659484281,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002373317237077427,
      "loss": 5.1529,
      "mean_token_accuracy": 0.23056723922491074,
      "num_tokens": 117691018.0,
      "step": 63125
    },
    {
      "entropy": 6.007529258728027,
      "epoch": 5.574885199576121,
      "grad_norm": 1.515625,
      "learning_rate": 0.00023730067702323993,
      "loss": 5.2247,
      "mean_token_accuracy": 0.22623068392276763,
      "num_tokens": 117700121.0,
      "step": 63130
    },
    {
      "entropy": 5.921687173843384,
      "epoch": 5.575326739667962,
      "grad_norm": 1.515625,
      "learning_rate": 0.00023726963107724558,
      "loss": 5.1068,
      "mean_token_accuracy": 0.24116843342781066,
      "num_tokens": 117709225.0,
      "step": 63135
    },
    {
      "entropy": 6.057404232025147,
      "epoch": 5.575768279759802,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023723858587036757,
      "loss": 5.2679,
      "mean_token_accuracy": 0.2193283826112747,
      "num_tokens": 117720175.0,
      "step": 63140
    },
    {
      "entropy": 6.050176477432251,
      "epoch": 5.576209819851643,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023720754140321427,
      "loss": 5.1488,
      "mean_token_accuracy": 0.22490258663892745,
      "num_tokens": 117729408.0,
      "step": 63145
    },
    {
      "entropy": 6.089268493652344,
      "epoch": 5.576651359943483,
      "grad_norm": 1.375,
      "learning_rate": 0.00023717649767639377,
      "loss": 5.2122,
      "mean_token_accuracy": 0.2269898146390915,
      "num_tokens": 117738022.0,
      "step": 63150
    },
    {
      "entropy": 6.103833436965942,
      "epoch": 5.577092900035323,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023714545469051418,
      "loss": 5.2353,
      "mean_token_accuracy": 0.22377561628818513,
      "num_tokens": 117747124.0,
      "step": 63155
    },
    {
      "entropy": 6.095456981658936,
      "epoch": 5.577534440127163,
      "grad_norm": 1.5,
      "learning_rate": 0.0002371144124461836,
      "loss": 5.2616,
      "mean_token_accuracy": 0.21862053275108337,
      "num_tokens": 117755999.0,
      "step": 63160
    },
    {
      "entropy": 6.06402359008789,
      "epoch": 5.577975980219004,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023708337094401,
      "loss": 5.3221,
      "mean_token_accuracy": 0.21376770585775376,
      "num_tokens": 117766029.0,
      "step": 63165
    },
    {
      "entropy": 6.0186896324157715,
      "epoch": 5.578417520310844,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002370523301846017,
      "loss": 5.102,
      "mean_token_accuracy": 0.23697500228881835,
      "num_tokens": 117773710.0,
      "step": 63170
    },
    {
      "entropy": 5.995260620117188,
      "epoch": 5.578859060402684,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002370212901685667,
      "loss": 5.1666,
      "mean_token_accuracy": 0.22891333401203157,
      "num_tokens": 117783518.0,
      "step": 63175
    },
    {
      "entropy": 6.0617616176605225,
      "epoch": 5.579300600494525,
      "grad_norm": 1.40625,
      "learning_rate": 0.000236990250896513,
      "loss": 5.2073,
      "mean_token_accuracy": 0.23227176666259766,
      "num_tokens": 117792674.0,
      "step": 63180
    },
    {
      "entropy": 6.128083896636963,
      "epoch": 5.579742140586365,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002369592123690486,
      "loss": 5.2145,
      "mean_token_accuracy": 0.2251801460981369,
      "num_tokens": 117801815.0,
      "step": 63185
    },
    {
      "entropy": 6.034019231796265,
      "epoch": 5.580183680678205,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023692817458678163,
      "loss": 5.1834,
      "mean_token_accuracy": 0.23553949743509292,
      "num_tokens": 117810833.0,
      "step": 63190
    },
    {
      "entropy": 6.029193973541259,
      "epoch": 5.580625220770046,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002368971375503201,
      "loss": 5.1713,
      "mean_token_accuracy": 0.23678237795829774,
      "num_tokens": 117819476.0,
      "step": 63195
    },
    {
      "entropy": 6.067104148864746,
      "epoch": 5.581066760861886,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023686610126027202,
      "loss": 5.2417,
      "mean_token_accuracy": 0.22285948395729066,
      "num_tokens": 117829137.0,
      "step": 63200
    },
    {
      "entropy": 5.981983947753906,
      "epoch": 5.581508300953726,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023683506571724528,
      "loss": 5.1825,
      "mean_token_accuracy": 0.22628162503242494,
      "num_tokens": 117838750.0,
      "step": 63205
    },
    {
      "entropy": 6.090380525588989,
      "epoch": 5.581949841045567,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00023680403092184784,
      "loss": 5.2558,
      "mean_token_accuracy": 0.22402609139680862,
      "num_tokens": 117848879.0,
      "step": 63210
    },
    {
      "entropy": 6.0000317096710205,
      "epoch": 5.582391381137407,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023677299687468778,
      "loss": 5.2215,
      "mean_token_accuracy": 0.2251218482851982,
      "num_tokens": 117858495.0,
      "step": 63215
    },
    {
      "entropy": 6.009494400024414,
      "epoch": 5.582832921229247,
      "grad_norm": 1.5,
      "learning_rate": 0.00023674196357637296,
      "loss": 5.1755,
      "mean_token_accuracy": 0.22025329768657684,
      "num_tokens": 117868770.0,
      "step": 63220
    },
    {
      "entropy": 6.105563402175903,
      "epoch": 5.583274461321088,
      "grad_norm": 1.765625,
      "learning_rate": 0.00023671093102751135,
      "loss": 5.2906,
      "mean_token_accuracy": 0.22452817857265472,
      "num_tokens": 117878093.0,
      "step": 63225
    },
    {
      "entropy": 6.130505037307739,
      "epoch": 5.583716001412928,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002366798992287107,
      "loss": 5.2405,
      "mean_token_accuracy": 0.22183791399002076,
      "num_tokens": 117887247.0,
      "step": 63230
    },
    {
      "entropy": 6.0917219638824465,
      "epoch": 5.584157541504768,
      "grad_norm": 1.578125,
      "learning_rate": 0.00023664886818057913,
      "loss": 5.2408,
      "mean_token_accuracy": 0.2307007744908333,
      "num_tokens": 117895131.0,
      "step": 63235
    },
    {
      "entropy": 6.123511171340942,
      "epoch": 5.584599081596609,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023661783788372443,
      "loss": 5.3631,
      "mean_token_accuracy": 0.21466868817806245,
      "num_tokens": 117905644.0,
      "step": 63240
    },
    {
      "entropy": 6.083762454986572,
      "epoch": 5.585040621688449,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002365868083387544,
      "loss": 5.274,
      "mean_token_accuracy": 0.2222587525844574,
      "num_tokens": 117915002.0,
      "step": 63245
    },
    {
      "entropy": 6.106533861160278,
      "epoch": 5.5854821617802894,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023655577954627695,
      "loss": 5.3113,
      "mean_token_accuracy": 0.21592058837413788,
      "num_tokens": 117924792.0,
      "step": 63250
    },
    {
      "entropy": 6.0365479469299315,
      "epoch": 5.5859237018721295,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002365247515068999,
      "loss": 5.1656,
      "mean_token_accuracy": 0.2268784075975418,
      "num_tokens": 117935009.0,
      "step": 63255
    },
    {
      "entropy": 6.062637233734131,
      "epoch": 5.58636524196397,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023649372422123116,
      "loss": 5.2441,
      "mean_token_accuracy": 0.22355724424123763,
      "num_tokens": 117944052.0,
      "step": 63260
    },
    {
      "entropy": 6.080473566055298,
      "epoch": 5.5868067820558105,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023646269768987845,
      "loss": 5.1914,
      "mean_token_accuracy": 0.2251688614487648,
      "num_tokens": 117954359.0,
      "step": 63265
    },
    {
      "entropy": 6.08922758102417,
      "epoch": 5.587248322147651,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002364316719134496,
      "loss": 5.1998,
      "mean_token_accuracy": 0.23746152818202973,
      "num_tokens": 117963399.0,
      "step": 63270
    },
    {
      "entropy": 6.002683687210083,
      "epoch": 5.5876898622394915,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023640064689255225,
      "loss": 5.1545,
      "mean_token_accuracy": 0.2365868866443634,
      "num_tokens": 117972150.0,
      "step": 63275
    },
    {
      "entropy": 5.93069543838501,
      "epoch": 5.5881314023313315,
      "grad_norm": 1.375,
      "learning_rate": 0.00023636962262779437,
      "loss": 5.1495,
      "mean_token_accuracy": 0.23212896138429642,
      "num_tokens": 117981720.0,
      "step": 63280
    },
    {
      "entropy": 6.0691393375396725,
      "epoch": 5.588572942423172,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023633859911978357,
      "loss": 5.2456,
      "mean_token_accuracy": 0.22311742156744002,
      "num_tokens": 117991082.0,
      "step": 63285
    },
    {
      "entropy": 6.04100399017334,
      "epoch": 5.5890144825150125,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002363075763691277,
      "loss": 5.2468,
      "mean_token_accuracy": 0.21835048347711564,
      "num_tokens": 117999977.0,
      "step": 63290
    },
    {
      "entropy": 6.0780029296875,
      "epoch": 5.589456022606853,
      "grad_norm": 1.515625,
      "learning_rate": 0.00023627655437643424,
      "loss": 5.2581,
      "mean_token_accuracy": 0.23238690346479415,
      "num_tokens": 118008256.0,
      "step": 63295
    },
    {
      "entropy": 6.067503356933594,
      "epoch": 5.5898975626986935,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023624553314231117,
      "loss": 5.2495,
      "mean_token_accuracy": 0.22902667075395583,
      "num_tokens": 118018198.0,
      "step": 63300
    },
    {
      "entropy": 5.998247480392456,
      "epoch": 5.590339102790534,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002362145126673661,
      "loss": 5.1685,
      "mean_token_accuracy": 0.23289575576782226,
      "num_tokens": 118026846.0,
      "step": 63305
    },
    {
      "entropy": 6.039904403686523,
      "epoch": 5.590780642882374,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002361834929522066,
      "loss": 5.1895,
      "mean_token_accuracy": 0.22631138265132905,
      "num_tokens": 118036015.0,
      "step": 63310
    },
    {
      "entropy": 6.115890884399414,
      "epoch": 5.591222182974214,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002361524739974404,
      "loss": 5.3125,
      "mean_token_accuracy": 0.21857767105102538,
      "num_tokens": 118045714.0,
      "step": 63315
    },
    {
      "entropy": 6.004107141494751,
      "epoch": 5.591663723066055,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002361214558036751,
      "loss": 5.1805,
      "mean_token_accuracy": 0.22972283661365508,
      "num_tokens": 118055111.0,
      "step": 63320
    },
    {
      "entropy": 6.064022779464722,
      "epoch": 5.592105263157895,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002360904383715185,
      "loss": 5.1489,
      "mean_token_accuracy": 0.22745691984891891,
      "num_tokens": 118064415.0,
      "step": 63325
    },
    {
      "entropy": 6.081878185272217,
      "epoch": 5.592546803249735,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000236059421701578,
      "loss": 5.2544,
      "mean_token_accuracy": 0.21739021986722945,
      "num_tokens": 118073660.0,
      "step": 63330
    },
    {
      "entropy": 6.104978132247925,
      "epoch": 5.592988343341576,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023602840579446134,
      "loss": 5.304,
      "mean_token_accuracy": 0.2176912546157837,
      "num_tokens": 118083051.0,
      "step": 63335
    },
    {
      "entropy": 6.116115951538086,
      "epoch": 5.593429883433416,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002359973906507759,
      "loss": 5.3918,
      "mean_token_accuracy": 0.21278561353683473,
      "num_tokens": 118093044.0,
      "step": 63340
    },
    {
      "entropy": 6.030181741714477,
      "epoch": 5.593871423525256,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023596637627112954,
      "loss": 5.1826,
      "mean_token_accuracy": 0.2269764021039009,
      "num_tokens": 118102094.0,
      "step": 63345
    },
    {
      "entropy": 6.065448474884033,
      "epoch": 5.594312963617097,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023593536265612962,
      "loss": 5.2183,
      "mean_token_accuracy": 0.225397552549839,
      "num_tokens": 118111408.0,
      "step": 63350
    },
    {
      "entropy": 5.976472330093384,
      "epoch": 5.594754503708937,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002359043498063838,
      "loss": 5.0779,
      "mean_token_accuracy": 0.24075182378292084,
      "num_tokens": 118119409.0,
      "step": 63355
    },
    {
      "entropy": 6.0832664489746096,
      "epoch": 5.595196043800777,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002358733377224994,
      "loss": 5.24,
      "mean_token_accuracy": 0.2221383571624756,
      "num_tokens": 118128302.0,
      "step": 63360
    },
    {
      "entropy": 6.034033870697021,
      "epoch": 5.595637583892618,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00023584232640508414,
      "loss": 5.2147,
      "mean_token_accuracy": 0.22538751512765884,
      "num_tokens": 118136570.0,
      "step": 63365
    },
    {
      "entropy": 6.015045404434204,
      "epoch": 5.596079123984458,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023581131585474546,
      "loss": 5.16,
      "mean_token_accuracy": 0.22800385504961013,
      "num_tokens": 118145628.0,
      "step": 63370
    },
    {
      "entropy": 5.95561785697937,
      "epoch": 5.596520664076298,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002357803060720907,
      "loss": 5.1376,
      "mean_token_accuracy": 0.23369866609573364,
      "num_tokens": 118154486.0,
      "step": 63375
    },
    {
      "entropy": 6.104617071151734,
      "epoch": 5.596962204168139,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002357492970577275,
      "loss": 5.2799,
      "mean_token_accuracy": 0.22136005759239197,
      "num_tokens": 118165050.0,
      "step": 63380
    },
    {
      "entropy": 6.014575862884522,
      "epoch": 5.597403744259979,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023571828881226327,
      "loss": 5.2741,
      "mean_token_accuracy": 0.2262235015630722,
      "num_tokens": 118174335.0,
      "step": 63385
    },
    {
      "entropy": 6.040327072143555,
      "epoch": 5.597845284351819,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023568728133630546,
      "loss": 5.1624,
      "mean_token_accuracy": 0.2365814909338951,
      "num_tokens": 118184977.0,
      "step": 63390
    },
    {
      "entropy": 6.094600629806519,
      "epoch": 5.59828682444366,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002356562746304614,
      "loss": 5.1562,
      "mean_token_accuracy": 0.2297152653336525,
      "num_tokens": 118193954.0,
      "step": 63395
    },
    {
      "entropy": 6.114374160766602,
      "epoch": 5.5987283645355,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00023562526869533857,
      "loss": 5.2856,
      "mean_token_accuracy": 0.21850388050079345,
      "num_tokens": 118203368.0,
      "step": 63400
    },
    {
      "entropy": 6.047383737564087,
      "epoch": 5.59916990462734,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002355942635315443,
      "loss": 5.1921,
      "mean_token_accuracy": 0.224116051197052,
      "num_tokens": 118211869.0,
      "step": 63405
    },
    {
      "entropy": 6.013213348388672,
      "epoch": 5.59961144471918,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023556325913968597,
      "loss": 5.1757,
      "mean_token_accuracy": 0.2324913188815117,
      "num_tokens": 118221409.0,
      "step": 63410
    },
    {
      "entropy": 5.992879962921142,
      "epoch": 5.600052984811021,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023553225552037105,
      "loss": 5.2469,
      "mean_token_accuracy": 0.22597249895334243,
      "num_tokens": 118231104.0,
      "step": 63415
    },
    {
      "entropy": 5.986716794967651,
      "epoch": 5.600494524902861,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023550125267420675,
      "loss": 5.1437,
      "mean_token_accuracy": 0.23463860303163528,
      "num_tokens": 118240568.0,
      "step": 63420
    },
    {
      "entropy": 6.1302519798278805,
      "epoch": 5.600936064994702,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00023547025060180037,
      "loss": 5.2865,
      "mean_token_accuracy": 0.22394661605358124,
      "num_tokens": 118249252.0,
      "step": 63425
    },
    {
      "entropy": 6.131792402267456,
      "epoch": 5.601377605086542,
      "grad_norm": 1.375,
      "learning_rate": 0.00023543924930375943,
      "loss": 5.2703,
      "mean_token_accuracy": 0.2091729536652565,
      "num_tokens": 118258643.0,
      "step": 63430
    },
    {
      "entropy": 5.936943387985229,
      "epoch": 5.601819145178382,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023540824878069106,
      "loss": 5.0589,
      "mean_token_accuracy": 0.24135117679834367,
      "num_tokens": 118267042.0,
      "step": 63435
    },
    {
      "entropy": 5.971537685394287,
      "epoch": 5.602260685270222,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002353772490332025,
      "loss": 5.1044,
      "mean_token_accuracy": 0.23943561017513276,
      "num_tokens": 118276167.0,
      "step": 63440
    },
    {
      "entropy": 6.012282419204712,
      "epoch": 5.602702225362063,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002353462500619012,
      "loss": 5.2574,
      "mean_token_accuracy": 0.228677336871624,
      "num_tokens": 118286831.0,
      "step": 63445
    },
    {
      "entropy": 6.074276113510132,
      "epoch": 5.603143765453903,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023531525186739423,
      "loss": 5.2875,
      "mean_token_accuracy": 0.2172665774822235,
      "num_tokens": 118296520.0,
      "step": 63450
    },
    {
      "entropy": 6.065120649337769,
      "epoch": 5.603585305545743,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023528425445028896,
      "loss": 5.2585,
      "mean_token_accuracy": 0.2198134705424309,
      "num_tokens": 118304933.0,
      "step": 63455
    },
    {
      "entropy": 6.098453998565674,
      "epoch": 5.604026845637584,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023525325781119255,
      "loss": 5.2977,
      "mean_token_accuracy": 0.22315253019332887,
      "num_tokens": 118314217.0,
      "step": 63460
    },
    {
      "entropy": 6.123138475418091,
      "epoch": 5.604468385729424,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023522226195071223,
      "loss": 5.2551,
      "mean_token_accuracy": 0.21841692179441452,
      "num_tokens": 118323434.0,
      "step": 63465
    },
    {
      "entropy": 6.1260942935943605,
      "epoch": 5.604909925821264,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023519126686945507,
      "loss": 5.3146,
      "mean_token_accuracy": 0.21175136417150497,
      "num_tokens": 118333503.0,
      "step": 63470
    },
    {
      "entropy": 6.089923477172851,
      "epoch": 5.605351465913105,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023516027256802847,
      "loss": 5.2265,
      "mean_token_accuracy": 0.22780474573373793,
      "num_tokens": 118342959.0,
      "step": 63475
    },
    {
      "entropy": 6.062128210067749,
      "epoch": 5.605793006004945,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023512927904703945,
      "loss": 5.2281,
      "mean_token_accuracy": 0.21874128580093383,
      "num_tokens": 118351990.0,
      "step": 63480
    },
    {
      "entropy": 6.093557214736938,
      "epoch": 5.606234546096785,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023509828630709518,
      "loss": 5.2511,
      "mean_token_accuracy": 0.22133354395627974,
      "num_tokens": 118361136.0,
      "step": 63485
    },
    {
      "entropy": 6.056697177886963,
      "epoch": 5.606676086188626,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023506729434880276,
      "loss": 5.18,
      "mean_token_accuracy": 0.22957045137882232,
      "num_tokens": 118369959.0,
      "step": 63490
    },
    {
      "entropy": 5.970516967773437,
      "epoch": 5.607117626280466,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002350363031727693,
      "loss": 5.1959,
      "mean_token_accuracy": 0.23331305235624314,
      "num_tokens": 118378933.0,
      "step": 63495
    },
    {
      "entropy": 6.027913761138916,
      "epoch": 5.607559166372306,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023500531277960201,
      "loss": 5.284,
      "mean_token_accuracy": 0.22375576496124266,
      "num_tokens": 118388432.0,
      "step": 63500
    },
    {
      "entropy": 6.070449304580689,
      "epoch": 5.608000706464147,
      "grad_norm": 1.5625,
      "learning_rate": 0.00023497432316990784,
      "loss": 5.3018,
      "mean_token_accuracy": 0.2211626872420311,
      "num_tokens": 118398319.0,
      "step": 63505
    },
    {
      "entropy": 6.085250473022461,
      "epoch": 5.608442246555987,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000234943334344294,
      "loss": 5.2403,
      "mean_token_accuracy": 0.22158342599868774,
      "num_tokens": 118408634.0,
      "step": 63510
    },
    {
      "entropy": 6.001600503921509,
      "epoch": 5.608883786647827,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023491234630336732,
      "loss": 5.1339,
      "mean_token_accuracy": 0.23182208985090255,
      "num_tokens": 118417181.0,
      "step": 63515
    },
    {
      "entropy": 6.008557939529419,
      "epoch": 5.609325326739668,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023488135904773506,
      "loss": 5.1547,
      "mean_token_accuracy": 0.2318745031952858,
      "num_tokens": 118426459.0,
      "step": 63520
    },
    {
      "entropy": 5.964521598815918,
      "epoch": 5.609766866831508,
      "grad_norm": 1.5,
      "learning_rate": 0.0002348503725780042,
      "loss": 5.0796,
      "mean_token_accuracy": 0.2418392226099968,
      "num_tokens": 118435085.0,
      "step": 63525
    },
    {
      "entropy": 6.049523973464966,
      "epoch": 5.610208406923348,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023481938689478166,
      "loss": 5.2234,
      "mean_token_accuracy": 0.22855741828680037,
      "num_tokens": 118443876.0,
      "step": 63530
    },
    {
      "entropy": 6.076249074935913,
      "epoch": 5.6106499470151885,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002347884019986744,
      "loss": 5.2757,
      "mean_token_accuracy": 0.22536854445934296,
      "num_tokens": 118454023.0,
      "step": 63535
    },
    {
      "entropy": 6.045194387435913,
      "epoch": 5.611091487107029,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023475741789028958,
      "loss": 5.1993,
      "mean_token_accuracy": 0.22951222509145736,
      "num_tokens": 118462629.0,
      "step": 63540
    },
    {
      "entropy": 6.0152259349823,
      "epoch": 5.6115330271988695,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023472643457023406,
      "loss": 5.187,
      "mean_token_accuracy": 0.23529142141342163,
      "num_tokens": 118472300.0,
      "step": 63545
    },
    {
      "entropy": 6.032197284698486,
      "epoch": 5.61197456729071,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023469545203911473,
      "loss": 5.1947,
      "mean_token_accuracy": 0.23283754140138627,
      "num_tokens": 118481613.0,
      "step": 63550
    },
    {
      "entropy": 6.148652219772339,
      "epoch": 5.6124161073825505,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023466447029753856,
      "loss": 5.336,
      "mean_token_accuracy": 0.2198512852191925,
      "num_tokens": 118490057.0,
      "step": 63555
    },
    {
      "entropy": 6.097199296951294,
      "epoch": 5.6128576474743905,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002346334893461125,
      "loss": 5.2044,
      "mean_token_accuracy": 0.22560502737760543,
      "num_tokens": 118500221.0,
      "step": 63560
    },
    {
      "entropy": 6.09460563659668,
      "epoch": 5.613299187566231,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023460250918544347,
      "loss": 5.3046,
      "mean_token_accuracy": 0.21884082704782487,
      "num_tokens": 118509593.0,
      "step": 63565
    },
    {
      "entropy": 5.9610841274261475,
      "epoch": 5.6137407276580715,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023457152981613827,
      "loss": 5.1119,
      "mean_token_accuracy": 0.22977637201547624,
      "num_tokens": 118519655.0,
      "step": 63570
    },
    {
      "entropy": 6.0383768558502195,
      "epoch": 5.614182267749912,
      "grad_norm": 1.375,
      "learning_rate": 0.00023454055123880386,
      "loss": 5.1537,
      "mean_token_accuracy": 0.23285970240831375,
      "num_tokens": 118528855.0,
      "step": 63575
    },
    {
      "entropy": 6.009681034088135,
      "epoch": 5.6146238078417525,
      "grad_norm": 1.484375,
      "learning_rate": 0.00023450957345404688,
      "loss": 5.1904,
      "mean_token_accuracy": 0.2310081332921982,
      "num_tokens": 118537804.0,
      "step": 63580
    },
    {
      "entropy": 6.04622712135315,
      "epoch": 5.615065347933593,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002344785964624745,
      "loss": 5.2106,
      "mean_token_accuracy": 0.22844089120626448,
      "num_tokens": 118547104.0,
      "step": 63585
    },
    {
      "entropy": 5.933915328979492,
      "epoch": 5.615506888025433,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023444762026469335,
      "loss": 5.0565,
      "mean_token_accuracy": 0.24036093652248383,
      "num_tokens": 118556618.0,
      "step": 63590
    },
    {
      "entropy": 6.071360397338867,
      "epoch": 5.615948428117273,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002344166448613102,
      "loss": 5.2687,
      "mean_token_accuracy": 0.2207476869225502,
      "num_tokens": 118565538.0,
      "step": 63595
    },
    {
      "entropy": 6.080675935745239,
      "epoch": 5.616389968209114,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002343856702529319,
      "loss": 5.2319,
      "mean_token_accuracy": 0.21844878494739534,
      "num_tokens": 118574860.0,
      "step": 63600
    },
    {
      "entropy": 6.0300194263458256,
      "epoch": 5.616831508300954,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002343546964401652,
      "loss": 5.235,
      "mean_token_accuracy": 0.22965924739837645,
      "num_tokens": 118584660.0,
      "step": 63605
    },
    {
      "entropy": 6.030726480484009,
      "epoch": 5.617273048392794,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023432372342361698,
      "loss": 5.2019,
      "mean_token_accuracy": 0.23287438005208969,
      "num_tokens": 118595059.0,
      "step": 63610
    },
    {
      "entropy": 5.990171813964844,
      "epoch": 5.617714588484635,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023429275120389386,
      "loss": 5.1404,
      "mean_token_accuracy": 0.22355912923812865,
      "num_tokens": 118603901.0,
      "step": 63615
    },
    {
      "entropy": 6.057287549972534,
      "epoch": 5.618156128576475,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023426177978160245,
      "loss": 5.2369,
      "mean_token_accuracy": 0.22830778509378433,
      "num_tokens": 118613084.0,
      "step": 63620
    },
    {
      "entropy": 6.045939016342163,
      "epoch": 5.618597668668315,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00023423080915734972,
      "loss": 5.2062,
      "mean_token_accuracy": 0.2268682524561882,
      "num_tokens": 118621743.0,
      "step": 63625
    },
    {
      "entropy": 6.0365118980407715,
      "epoch": 5.619039208760156,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023419983933174222,
      "loss": 5.149,
      "mean_token_accuracy": 0.22542146742343902,
      "num_tokens": 118630881.0,
      "step": 63630
    },
    {
      "entropy": 6.102879619598388,
      "epoch": 5.619480748851996,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023416887030538674,
      "loss": 5.2543,
      "mean_token_accuracy": 0.22209425121545792,
      "num_tokens": 118639743.0,
      "step": 63635
    },
    {
      "entropy": 6.032015132904053,
      "epoch": 5.619922288943836,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023413790207888984,
      "loss": 5.2116,
      "mean_token_accuracy": 0.2269942820072174,
      "num_tokens": 118649293.0,
      "step": 63640
    },
    {
      "entropy": 6.045904731750488,
      "epoch": 5.620363829035677,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002341069346528581,
      "loss": 5.2405,
      "mean_token_accuracy": 0.22146255075931548,
      "num_tokens": 118658250.0,
      "step": 63645
    },
    {
      "entropy": 5.93848934173584,
      "epoch": 5.620805369127517,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023407596802789833,
      "loss": 5.1032,
      "mean_token_accuracy": 0.2419513151049614,
      "num_tokens": 118667745.0,
      "step": 63650
    },
    {
      "entropy": 6.023435258865357,
      "epoch": 5.621246909219357,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002340450022046171,
      "loss": 5.2869,
      "mean_token_accuracy": 0.22101123929023742,
      "num_tokens": 118678672.0,
      "step": 63655
    },
    {
      "entropy": 6.0012977600097654,
      "epoch": 5.621688449311198,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023401403718362095,
      "loss": 5.1019,
      "mean_token_accuracy": 0.22750217467546463,
      "num_tokens": 118687006.0,
      "step": 63660
    },
    {
      "entropy": 6.111174249649048,
      "epoch": 5.622129989403038,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023398307296551647,
      "loss": 5.2841,
      "mean_token_accuracy": 0.21203628033399582,
      "num_tokens": 118697597.0,
      "step": 63665
    },
    {
      "entropy": 6.005092334747315,
      "epoch": 5.622571529494878,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023395210955091028,
      "loss": 5.1052,
      "mean_token_accuracy": 0.233144873380661,
      "num_tokens": 118706665.0,
      "step": 63670
    },
    {
      "entropy": 6.018495702743531,
      "epoch": 5.623013069586719,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023392114694040895,
      "loss": 5.209,
      "mean_token_accuracy": 0.22124750167131424,
      "num_tokens": 118715697.0,
      "step": 63675
    },
    {
      "entropy": 5.9853991031646725,
      "epoch": 5.623454609678559,
      "grad_norm": 1.40625,
      "learning_rate": 0.000233890185134619,
      "loss": 5.1927,
      "mean_token_accuracy": 0.22934076488018035,
      "num_tokens": 118724080.0,
      "step": 63680
    },
    {
      "entropy": 6.008993625640869,
      "epoch": 5.623896149770399,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002338592241341469,
      "loss": 5.2043,
      "mean_token_accuracy": 0.21797374188899993,
      "num_tokens": 118734355.0,
      "step": 63685
    },
    {
      "entropy": 6.0272458553314205,
      "epoch": 5.624337689862239,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002338282639395992,
      "loss": 5.2471,
      "mean_token_accuracy": 0.2215880811214447,
      "num_tokens": 118744013.0,
      "step": 63690
    },
    {
      "entropy": 6.062645959854126,
      "epoch": 5.62477922995408,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023379730455158238,
      "loss": 5.2475,
      "mean_token_accuracy": 0.22433974891901015,
      "num_tokens": 118753748.0,
      "step": 63695
    },
    {
      "entropy": 6.016155529022217,
      "epoch": 5.62522077004592,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000233766345970703,
      "loss": 5.1849,
      "mean_token_accuracy": 0.2318340942263603,
      "num_tokens": 118762374.0,
      "step": 63700
    },
    {
      "entropy": 6.0805994987487795,
      "epoch": 5.625662310137761,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002337353881975674,
      "loss": 5.2393,
      "mean_token_accuracy": 0.22541681230068206,
      "num_tokens": 118771988.0,
      "step": 63705
    },
    {
      "entropy": 6.1284574508667,
      "epoch": 5.626103850229601,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023370443123278207,
      "loss": 5.1932,
      "mean_token_accuracy": 0.23104217946529387,
      "num_tokens": 118781285.0,
      "step": 63710
    },
    {
      "entropy": 6.068522596359253,
      "epoch": 5.626545390321441,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023367347507695347,
      "loss": 5.2345,
      "mean_token_accuracy": 0.23082933723926544,
      "num_tokens": 118789213.0,
      "step": 63715
    },
    {
      "entropy": 5.950120210647583,
      "epoch": 5.626986930413281,
      "grad_norm": 1.421875,
      "learning_rate": 0.000233642519730688,
      "loss": 5.2125,
      "mean_token_accuracy": 0.22396279275417327,
      "num_tokens": 118798378.0,
      "step": 63720
    },
    {
      "entropy": 6.038526439666748,
      "epoch": 5.627428470505122,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023361156519459203,
      "loss": 5.2775,
      "mean_token_accuracy": 0.21829622834920884,
      "num_tokens": 118808301.0,
      "step": 63725
    },
    {
      "entropy": 5.956276512145996,
      "epoch": 5.627870010596962,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002335806114692719,
      "loss": 5.1497,
      "mean_token_accuracy": 0.23507640063762664,
      "num_tokens": 118816662.0,
      "step": 63730
    },
    {
      "entropy": 6.050466108322143,
      "epoch": 5.628311550688802,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023354965855533412,
      "loss": 5.2257,
      "mean_token_accuracy": 0.22345592975616455,
      "num_tokens": 118826350.0,
      "step": 63735
    },
    {
      "entropy": 6.052295303344726,
      "epoch": 5.628753090780643,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023351870645338496,
      "loss": 5.2171,
      "mean_token_accuracy": 0.2231978878378868,
      "num_tokens": 118835338.0,
      "step": 63740
    },
    {
      "entropy": 5.995884704589844,
      "epoch": 5.629194630872483,
      "grad_norm": 1.375,
      "learning_rate": 0.00023348775516403075,
      "loss": 5.1463,
      "mean_token_accuracy": 0.23109084814786912,
      "num_tokens": 118844348.0,
      "step": 63745
    },
    {
      "entropy": 6.008705711364746,
      "epoch": 5.629636170964323,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023345680468787771,
      "loss": 5.1837,
      "mean_token_accuracy": 0.22627160847187042,
      "num_tokens": 118854426.0,
      "step": 63750
    },
    {
      "entropy": 6.187346744537353,
      "epoch": 5.630077711056164,
      "grad_norm": 1.578125,
      "learning_rate": 0.00023342585502553233,
      "loss": 5.3817,
      "mean_token_accuracy": 0.20651804208755492,
      "num_tokens": 118864814.0,
      "step": 63755
    },
    {
      "entropy": 6.086673164367676,
      "epoch": 5.630519251148004,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023339490617760073,
      "loss": 5.2015,
      "mean_token_accuracy": 0.2338154837489128,
      "num_tokens": 118874730.0,
      "step": 63760
    },
    {
      "entropy": 6.151106452941894,
      "epoch": 5.630960791239844,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023336395814468937,
      "loss": 5.3047,
      "mean_token_accuracy": 0.21911934167146682,
      "num_tokens": 118883791.0,
      "step": 63765
    },
    {
      "entropy": 6.110020732879638,
      "epoch": 5.631402331331685,
      "grad_norm": 1.484375,
      "learning_rate": 0.00023333301092740434,
      "loss": 5.2971,
      "mean_token_accuracy": 0.2201011896133423,
      "num_tokens": 118893274.0,
      "step": 63770
    },
    {
      "entropy": 6.072688245773316,
      "epoch": 5.631843871423525,
      "grad_norm": 1.375,
      "learning_rate": 0.00023330206452635183,
      "loss": 5.2574,
      "mean_token_accuracy": 0.22271660417318345,
      "num_tokens": 118902456.0,
      "step": 63775
    },
    {
      "entropy": 6.026181888580322,
      "epoch": 5.632285411515365,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023327111894213827,
      "loss": 5.1866,
      "mean_token_accuracy": 0.22537164390087128,
      "num_tokens": 118912275.0,
      "step": 63780
    },
    {
      "entropy": 5.970385122299194,
      "epoch": 5.632726951607206,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023324017417536975,
      "loss": 5.1727,
      "mean_token_accuracy": 0.23112685680389405,
      "num_tokens": 118921345.0,
      "step": 63785
    },
    {
      "entropy": 6.123282194137573,
      "epoch": 5.633168491699046,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023320923022665248,
      "loss": 5.3369,
      "mean_token_accuracy": 0.2184700235724449,
      "num_tokens": 118931266.0,
      "step": 63790
    },
    {
      "entropy": 6.001190137863159,
      "epoch": 5.633610031790886,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00023317828709659255,
      "loss": 5.1851,
      "mean_token_accuracy": 0.23038756251335143,
      "num_tokens": 118940849.0,
      "step": 63795
    },
    {
      "entropy": 6.021333646774292,
      "epoch": 5.634051571882727,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023314734478579625,
      "loss": 5.1852,
      "mean_token_accuracy": 0.2208709552884102,
      "num_tokens": 118950254.0,
      "step": 63800
    },
    {
      "entropy": 5.957528924942016,
      "epoch": 5.634493111974567,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023311640329486965,
      "loss": 5.073,
      "mean_token_accuracy": 0.24140063524246216,
      "num_tokens": 118959055.0,
      "step": 63805
    },
    {
      "entropy": 6.109123611450196,
      "epoch": 5.634934652066407,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023308546262441893,
      "loss": 5.3326,
      "mean_token_accuracy": 0.21006572544574736,
      "num_tokens": 118968202.0,
      "step": 63810
    },
    {
      "entropy": 6.060937929153442,
      "epoch": 5.6353761921582475,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023305452277505006,
      "loss": 5.1476,
      "mean_token_accuracy": 0.23980807960033418,
      "num_tokens": 118976970.0,
      "step": 63815
    },
    {
      "entropy": 6.105356168746948,
      "epoch": 5.635817732250088,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023302358374736927,
      "loss": 5.3198,
      "mean_token_accuracy": 0.2122856393456459,
      "num_tokens": 118986206.0,
      "step": 63820
    },
    {
      "entropy": 6.071893787384033,
      "epoch": 5.6362592723419285,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023299264554198262,
      "loss": 5.2868,
      "mean_token_accuracy": 0.21525031179189683,
      "num_tokens": 118996296.0,
      "step": 63825
    },
    {
      "entropy": 5.948290395736694,
      "epoch": 5.636700812433769,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002329617081594962,
      "loss": 5.1117,
      "mean_token_accuracy": 0.23574145436286925,
      "num_tokens": 119005332.0,
      "step": 63830
    },
    {
      "entropy": 6.081868505477905,
      "epoch": 5.6371423525256095,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023293077160051596,
      "loss": 5.2522,
      "mean_token_accuracy": 0.2203628584742546,
      "num_tokens": 119014370.0,
      "step": 63835
    },
    {
      "entropy": 6.106139898300171,
      "epoch": 5.6375838926174495,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002328998358656479,
      "loss": 5.2807,
      "mean_token_accuracy": 0.21422958076000215,
      "num_tokens": 119025237.0,
      "step": 63840
    },
    {
      "entropy": 6.115785217285156,
      "epoch": 5.63802543270929,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023286890095549823,
      "loss": 5.2506,
      "mean_token_accuracy": 0.22709749042987823,
      "num_tokens": 119034588.0,
      "step": 63845
    },
    {
      "entropy": 6.04126205444336,
      "epoch": 5.6384669728011305,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023283796687067283,
      "loss": 5.184,
      "mean_token_accuracy": 0.2237561523914337,
      "num_tokens": 119043079.0,
      "step": 63850
    },
    {
      "entropy": 6.014504432678223,
      "epoch": 5.638908512892971,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023280703361177764,
      "loss": 5.2137,
      "mean_token_accuracy": 0.2278675690293312,
      "num_tokens": 119052740.0,
      "step": 63855
    },
    {
      "entropy": 6.04075517654419,
      "epoch": 5.6393500529848115,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002327761011794186,
      "loss": 5.2565,
      "mean_token_accuracy": 0.22731961607933043,
      "num_tokens": 119061608.0,
      "step": 63860
    },
    {
      "entropy": 6.094745063781739,
      "epoch": 5.639791593076652,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0002327451695742018,
      "loss": 5.2313,
      "mean_token_accuracy": 0.23011157512664795,
      "num_tokens": 119070058.0,
      "step": 63865
    },
    {
      "entropy": 6.069575595855713,
      "epoch": 5.640233133168492,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002327142387967331,
      "loss": 5.2501,
      "mean_token_accuracy": 0.2257058024406433,
      "num_tokens": 119079275.0,
      "step": 63870
    },
    {
      "entropy": 6.022672033309936,
      "epoch": 5.640674673260332,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023268330884761845,
      "loss": 5.2842,
      "mean_token_accuracy": 0.2288698211312294,
      "num_tokens": 119088510.0,
      "step": 63875
    },
    {
      "entropy": 6.075184392929077,
      "epoch": 5.641116213352173,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002326523797274636,
      "loss": 5.2109,
      "mean_token_accuracy": 0.2217359408736229,
      "num_tokens": 119098190.0,
      "step": 63880
    },
    {
      "entropy": 6.03918137550354,
      "epoch": 5.641557753444013,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00023262145143687464,
      "loss": 5.1841,
      "mean_token_accuracy": 0.23226277083158492,
      "num_tokens": 119107159.0,
      "step": 63885
    },
    {
      "entropy": 6.045019149780273,
      "epoch": 5.641999293535853,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023259052397645724,
      "loss": 5.229,
      "mean_token_accuracy": 0.2260628268122673,
      "num_tokens": 119116978.0,
      "step": 63890
    },
    {
      "entropy": 6.022610092163086,
      "epoch": 5.642440833627694,
      "grad_norm": 1.515625,
      "learning_rate": 0.00023255959734681748,
      "loss": 5.2183,
      "mean_token_accuracy": 0.22984354496002196,
      "num_tokens": 119125791.0,
      "step": 63895
    },
    {
      "entropy": 6.013253021240234,
      "epoch": 5.642882373719534,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023252867154856106,
      "loss": 5.1917,
      "mean_token_accuracy": 0.22486760914325715,
      "num_tokens": 119135118.0,
      "step": 63900
    },
    {
      "entropy": 6.052192544937133,
      "epoch": 5.643323913811374,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023249774658229367,
      "loss": 5.1809,
      "mean_token_accuracy": 0.224118772149086,
      "num_tokens": 119143999.0,
      "step": 63905
    },
    {
      "entropy": 6.073197937011718,
      "epoch": 5.643765453903215,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023246682244862138,
      "loss": 5.2951,
      "mean_token_accuracy": 0.21806232929229735,
      "num_tokens": 119154170.0,
      "step": 63910
    },
    {
      "entropy": 6.143220138549805,
      "epoch": 5.644206993995055,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023243589914814984,
      "loss": 5.3205,
      "mean_token_accuracy": 0.21641191989183425,
      "num_tokens": 119163679.0,
      "step": 63915
    },
    {
      "entropy": 6.014334869384766,
      "epoch": 5.644648534086895,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002324049766814848,
      "loss": 5.1778,
      "mean_token_accuracy": 0.22156461030244828,
      "num_tokens": 119173615.0,
      "step": 63920
    },
    {
      "entropy": 6.106441640853882,
      "epoch": 5.645090074178736,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023237405504923208,
      "loss": 5.2562,
      "mean_token_accuracy": 0.22639849036931992,
      "num_tokens": 119183840.0,
      "step": 63925
    },
    {
      "entropy": 6.038314962387085,
      "epoch": 5.645531614270576,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00023234313425199733,
      "loss": 5.1955,
      "mean_token_accuracy": 0.22869047969579698,
      "num_tokens": 119194463.0,
      "step": 63930
    },
    {
      "entropy": 6.04985089302063,
      "epoch": 5.645973154362416,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023231221429038636,
      "loss": 5.2465,
      "mean_token_accuracy": 0.22328935265541078,
      "num_tokens": 119204792.0,
      "step": 63935
    },
    {
      "entropy": 6.005255079269409,
      "epoch": 5.646414694454257,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002322812951650048,
      "loss": 5.1306,
      "mean_token_accuracy": 0.23881800770759581,
      "num_tokens": 119213596.0,
      "step": 63940
    },
    {
      "entropy": 6.130972814559937,
      "epoch": 5.646856234546097,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023225037687645838,
      "loss": 5.242,
      "mean_token_accuracy": 0.22472829669713973,
      "num_tokens": 119223425.0,
      "step": 63945
    },
    {
      "entropy": 6.06179256439209,
      "epoch": 5.647297774637937,
      "grad_norm": 1.484375,
      "learning_rate": 0.00023221945942535272,
      "loss": 5.2136,
      "mean_token_accuracy": 0.22817239463329314,
      "num_tokens": 119232457.0,
      "step": 63950
    },
    {
      "entropy": 5.92307562828064,
      "epoch": 5.647739314729778,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023218854281229357,
      "loss": 5.0979,
      "mean_token_accuracy": 0.23795781433582305,
      "num_tokens": 119241517.0,
      "step": 63955
    },
    {
      "entropy": 6.046257019042969,
      "epoch": 5.648180854821618,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023215762703788652,
      "loss": 5.2321,
      "mean_token_accuracy": 0.22283237278461457,
      "num_tokens": 119250614.0,
      "step": 63960
    },
    {
      "entropy": 6.062768363952637,
      "epoch": 5.648622394913458,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023212671210273722,
      "loss": 5.22,
      "mean_token_accuracy": 0.219160957634449,
      "num_tokens": 119260506.0,
      "step": 63965
    },
    {
      "entropy": 6.078670597076416,
      "epoch": 5.649063935005298,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023209579800745111,
      "loss": 5.2579,
      "mean_token_accuracy": 0.22650308459997176,
      "num_tokens": 119270541.0,
      "step": 63970
    },
    {
      "entropy": 6.128918838500977,
      "epoch": 5.649505475097139,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023206488475263404,
      "loss": 5.3079,
      "mean_token_accuracy": 0.21417940258979798,
      "num_tokens": 119279490.0,
      "step": 63975
    },
    {
      "entropy": 6.010575437545777,
      "epoch": 5.649947015188979,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002320339723388915,
      "loss": 5.2356,
      "mean_token_accuracy": 0.22892124205827713,
      "num_tokens": 119289205.0,
      "step": 63980
    },
    {
      "entropy": 6.0207953453063965,
      "epoch": 5.65038855528082,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023200306076682898,
      "loss": 5.0881,
      "mean_token_accuracy": 0.23505179733037948,
      "num_tokens": 119297933.0,
      "step": 63985
    },
    {
      "entropy": 6.043267822265625,
      "epoch": 5.65083009537266,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023197215003705203,
      "loss": 5.1754,
      "mean_token_accuracy": 0.22670889347791673,
      "num_tokens": 119307178.0,
      "step": 63990
    },
    {
      "entropy": 6.124009847640991,
      "epoch": 5.6512716354645,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002319412401501661,
      "loss": 5.3545,
      "mean_token_accuracy": 0.21507305800914764,
      "num_tokens": 119317018.0,
      "step": 63995
    },
    {
      "entropy": 6.135951995849609,
      "epoch": 5.65171317555634,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023191033110677684,
      "loss": 5.2517,
      "mean_token_accuracy": 0.22041146010160445,
      "num_tokens": 119326407.0,
      "step": 64000
    },
    {
      "entropy": 6.0733436107635494,
      "epoch": 5.652154715648181,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002318794229074897,
      "loss": 5.2404,
      "mean_token_accuracy": 0.22103619426488877,
      "num_tokens": 119335948.0,
      "step": 64005
    },
    {
      "entropy": 5.98619065284729,
      "epoch": 5.652596255740021,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002318485155529102,
      "loss": 5.1458,
      "mean_token_accuracy": 0.22857077568769454,
      "num_tokens": 119345544.0,
      "step": 64010
    },
    {
      "entropy": 6.068209600448609,
      "epoch": 5.653037795831861,
      "grad_norm": 1.375,
      "learning_rate": 0.00023181760904364364,
      "loss": 5.2598,
      "mean_token_accuracy": 0.22305810451507568,
      "num_tokens": 119354429.0,
      "step": 64015
    },
    {
      "entropy": 6.06252760887146,
      "epoch": 5.653479335923702,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023178670338029557,
      "loss": 5.3016,
      "mean_token_accuracy": 0.21578405052423477,
      "num_tokens": 119363630.0,
      "step": 64020
    },
    {
      "entropy": 6.013606357574463,
      "epoch": 5.653920876015542,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023175579856347145,
      "loss": 5.201,
      "mean_token_accuracy": 0.22929309606552123,
      "num_tokens": 119373267.0,
      "step": 64025
    },
    {
      "entropy": 6.1198571681976315,
      "epoch": 5.654362416107382,
      "grad_norm": 1.25,
      "learning_rate": 0.00023172489459377665,
      "loss": 5.2905,
      "mean_token_accuracy": 0.22080962359905243,
      "num_tokens": 119382846.0,
      "step": 64030
    },
    {
      "entropy": 6.076379823684692,
      "epoch": 5.654803956199223,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023169399147181648,
      "loss": 5.2021,
      "mean_token_accuracy": 0.22083909213542938,
      "num_tokens": 119392633.0,
      "step": 64035
    },
    {
      "entropy": 6.108876466751099,
      "epoch": 5.655245496291063,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023166308919819644,
      "loss": 5.2246,
      "mean_token_accuracy": 0.22484949380159377,
      "num_tokens": 119401138.0,
      "step": 64040
    },
    {
      "entropy": 6.064746761322022,
      "epoch": 5.655687036382903,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023163218777352186,
      "loss": 5.226,
      "mean_token_accuracy": 0.23250896483659744,
      "num_tokens": 119411507.0,
      "step": 64045
    },
    {
      "entropy": 6.044063997268677,
      "epoch": 5.656128576474744,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023160128719839807,
      "loss": 5.2018,
      "mean_token_accuracy": 0.23142927438020705,
      "num_tokens": 119420607.0,
      "step": 64050
    },
    {
      "entropy": 6.082487487792969,
      "epoch": 5.656570116566584,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002315703874734304,
      "loss": 5.1977,
      "mean_token_accuracy": 0.230810709297657,
      "num_tokens": 119430667.0,
      "step": 64055
    },
    {
      "entropy": 5.982852554321289,
      "epoch": 5.657011656658424,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00023153948859922397,
      "loss": 5.1589,
      "mean_token_accuracy": 0.2399211198091507,
      "num_tokens": 119439769.0,
      "step": 64060
    },
    {
      "entropy": 6.051900196075439,
      "epoch": 5.657453196750265,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002315085905763844,
      "loss": 5.2471,
      "mean_token_accuracy": 0.22678424715995787,
      "num_tokens": 119448628.0,
      "step": 64065
    },
    {
      "entropy": 6.0384594917297365,
      "epoch": 5.657894736842105,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023147769340551671,
      "loss": 5.2511,
      "mean_token_accuracy": 0.2283065915107727,
      "num_tokens": 119457562.0,
      "step": 64070
    },
    {
      "entropy": 6.105910015106201,
      "epoch": 5.658336276933945,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023144679708722633,
      "loss": 5.3194,
      "mean_token_accuracy": 0.2158767983317375,
      "num_tokens": 119466310.0,
      "step": 64075
    },
    {
      "entropy": 6.060162210464478,
      "epoch": 5.658777817025786,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002314159016221184,
      "loss": 5.2207,
      "mean_token_accuracy": 0.2232363775372505,
      "num_tokens": 119475857.0,
      "step": 64080
    },
    {
      "entropy": 6.003444385528565,
      "epoch": 5.659219357117626,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023138500701079813,
      "loss": 5.1266,
      "mean_token_accuracy": 0.23082108795642853,
      "num_tokens": 119485167.0,
      "step": 64085
    },
    {
      "entropy": 6.138174724578858,
      "epoch": 5.659660897209466,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023135411325387085,
      "loss": 5.3373,
      "mean_token_accuracy": 0.21164874583482743,
      "num_tokens": 119495520.0,
      "step": 64090
    },
    {
      "entropy": 5.9983421802520756,
      "epoch": 5.6601024373013065,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00023132322035194163,
      "loss": 5.1898,
      "mean_token_accuracy": 0.22661590725183486,
      "num_tokens": 119504762.0,
      "step": 64095
    },
    {
      "entropy": 6.084429550170898,
      "epoch": 5.660543977393147,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002312923283056156,
      "loss": 5.2981,
      "mean_token_accuracy": 0.22452393621206285,
      "num_tokens": 119514280.0,
      "step": 64100
    },
    {
      "entropy": 6.084187698364258,
      "epoch": 5.6609855174849875,
      "grad_norm": 1.5,
      "learning_rate": 0.00023126143711549814,
      "loss": 5.2135,
      "mean_token_accuracy": 0.21603336930274963,
      "num_tokens": 119523479.0,
      "step": 64105
    },
    {
      "entropy": 6.072469091415405,
      "epoch": 5.661427057576828,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002312305467821942,
      "loss": 5.1771,
      "mean_token_accuracy": 0.23361113667488098,
      "num_tokens": 119532348.0,
      "step": 64110
    },
    {
      "entropy": 6.0721519947052,
      "epoch": 5.6618685976686685,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023119965730630898,
      "loss": 5.2142,
      "mean_token_accuracy": 0.22940575629472731,
      "num_tokens": 119541370.0,
      "step": 64115
    },
    {
      "entropy": 5.996736907958985,
      "epoch": 5.6623101377605085,
      "grad_norm": 1.5625,
      "learning_rate": 0.00023116876868844755,
      "loss": 5.1711,
      "mean_token_accuracy": 0.23294861614704132,
      "num_tokens": 119551658.0,
      "step": 64120
    },
    {
      "entropy": 6.079365062713623,
      "epoch": 5.662751677852349,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002311378809292149,
      "loss": 5.212,
      "mean_token_accuracy": 0.23160435408353805,
      "num_tokens": 119560253.0,
      "step": 64125
    },
    {
      "entropy": 6.058285140991211,
      "epoch": 5.6631932179441895,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002311069940292163,
      "loss": 5.2482,
      "mean_token_accuracy": 0.2177794858813286,
      "num_tokens": 119569733.0,
      "step": 64130
    },
    {
      "entropy": 6.0884967803955075,
      "epoch": 5.66363475803603,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002310761079890567,
      "loss": 5.2387,
      "mean_token_accuracy": 0.22425989657640458,
      "num_tokens": 119578366.0,
      "step": 64135
    },
    {
      "entropy": 6.020830821990967,
      "epoch": 5.6640762981278705,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023104522280934127,
      "loss": 5.2193,
      "mean_token_accuracy": 0.2239591583609581,
      "num_tokens": 119587498.0,
      "step": 64140
    },
    {
      "entropy": 6.036419010162353,
      "epoch": 5.6645178382197106,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023101433849067481,
      "loss": 5.207,
      "mean_token_accuracy": 0.22798559665679932,
      "num_tokens": 119596935.0,
      "step": 64145
    },
    {
      "entropy": 5.94940595626831,
      "epoch": 5.664959378311551,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023098345503366248,
      "loss": 5.1096,
      "mean_token_accuracy": 0.23639025837182998,
      "num_tokens": 119606120.0,
      "step": 64150
    },
    {
      "entropy": 6.067109775543213,
      "epoch": 5.665400918403391,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023095257243890922,
      "loss": 5.2813,
      "mean_token_accuracy": 0.22366981208324432,
      "num_tokens": 119615702.0,
      "step": 64155
    },
    {
      "entropy": 6.0558253765106205,
      "epoch": 5.665842458495232,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023092169070702007,
      "loss": 5.1812,
      "mean_token_accuracy": 0.22631529718637466,
      "num_tokens": 119624788.0,
      "step": 64160
    },
    {
      "entropy": 6.056080341339111,
      "epoch": 5.666283998587072,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023089080983859994,
      "loss": 5.2224,
      "mean_token_accuracy": 0.22709659039974212,
      "num_tokens": 119634473.0,
      "step": 64165
    },
    {
      "entropy": 6.075476741790771,
      "epoch": 5.666725538678912,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023085992983425363,
      "loss": 5.2535,
      "mean_token_accuracy": 0.22373798489570618,
      "num_tokens": 119644206.0,
      "step": 64170
    },
    {
      "entropy": 6.03076171875,
      "epoch": 5.667167078770753,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023082905069458627,
      "loss": 5.1005,
      "mean_token_accuracy": 0.23481155931949615,
      "num_tokens": 119653938.0,
      "step": 64175
    },
    {
      "entropy": 6.0620259761810305,
      "epoch": 5.667608618862593,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023079817242020273,
      "loss": 5.2874,
      "mean_token_accuracy": 0.22334303855895996,
      "num_tokens": 119663700.0,
      "step": 64180
    },
    {
      "entropy": 6.057231044769287,
      "epoch": 5.668050158954433,
      "grad_norm": 1.375,
      "learning_rate": 0.00023076729501170785,
      "loss": 5.2619,
      "mean_token_accuracy": 0.2201816827058792,
      "num_tokens": 119672701.0,
      "step": 64185
    },
    {
      "entropy": 5.941901159286499,
      "epoch": 5.668491699046274,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002307364184697064,
      "loss": 5.068,
      "mean_token_accuracy": 0.24450452029705047,
      "num_tokens": 119681348.0,
      "step": 64190
    },
    {
      "entropy": 6.059583616256714,
      "epoch": 5.668933239138114,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023070554279480343,
      "loss": 5.2741,
      "mean_token_accuracy": 0.22223344445228577,
      "num_tokens": 119690749.0,
      "step": 64195
    },
    {
      "entropy": 5.952998256683349,
      "epoch": 5.669374779229954,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023067466798760362,
      "loss": 5.0697,
      "mean_token_accuracy": 0.235493965446949,
      "num_tokens": 119699900.0,
      "step": 64200
    },
    {
      "entropy": 5.988610935211182,
      "epoch": 5.669816319321795,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023064379404871195,
      "loss": 5.2735,
      "mean_token_accuracy": 0.22080667614936828,
      "num_tokens": 119709499.0,
      "step": 64205
    },
    {
      "entropy": 5.992025995254517,
      "epoch": 5.670257859413635,
      "grad_norm": 1.453125,
      "learning_rate": 0.000230612920978733,
      "loss": 5.165,
      "mean_token_accuracy": 0.23046010434627534,
      "num_tokens": 119718935.0,
      "step": 64210
    },
    {
      "entropy": 6.055011844635009,
      "epoch": 5.670699399505475,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023058204877827177,
      "loss": 5.1912,
      "mean_token_accuracy": 0.23086427748203278,
      "num_tokens": 119728775.0,
      "step": 64215
    },
    {
      "entropy": 6.0867198467254635,
      "epoch": 5.671140939597316,
      "grad_norm": 1.375,
      "learning_rate": 0.0002305511774479329,
      "loss": 5.2028,
      "mean_token_accuracy": 0.22535354346036912,
      "num_tokens": 119738235.0,
      "step": 64220
    },
    {
      "entropy": 6.12337555885315,
      "epoch": 5.671582479689156,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023052030698832127,
      "loss": 5.2358,
      "mean_token_accuracy": 0.22138860672712327,
      "num_tokens": 119746983.0,
      "step": 64225
    },
    {
      "entropy": 5.968088626861572,
      "epoch": 5.672024019780996,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023048943740004146,
      "loss": 5.154,
      "mean_token_accuracy": 0.23398730307817459,
      "num_tokens": 119756448.0,
      "step": 64230
    },
    {
      "entropy": 5.994933128356934,
      "epoch": 5.672465559872837,
      "grad_norm": 1.546875,
      "learning_rate": 0.00023045856868369823,
      "loss": 5.1628,
      "mean_token_accuracy": 0.2296753630042076,
      "num_tokens": 119765018.0,
      "step": 64235
    },
    {
      "entropy": 6.055355548858643,
      "epoch": 5.672907099964677,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023042770083989632,
      "loss": 5.2346,
      "mean_token_accuracy": 0.2271884337067604,
      "num_tokens": 119774917.0,
      "step": 64240
    },
    {
      "entropy": 6.023053216934204,
      "epoch": 5.673348640056517,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023039683386924048,
      "loss": 5.1579,
      "mean_token_accuracy": 0.23179528415203093,
      "num_tokens": 119784295.0,
      "step": 64245
    },
    {
      "entropy": 6.00919246673584,
      "epoch": 5.673790180148357,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00023036596777233526,
      "loss": 5.1233,
      "mean_token_accuracy": 0.2385295018553734,
      "num_tokens": 119793649.0,
      "step": 64250
    },
    {
      "entropy": 6.036426830291748,
      "epoch": 5.674231720240198,
      "grad_norm": 1.53125,
      "learning_rate": 0.00023033510254978528,
      "loss": 5.1297,
      "mean_token_accuracy": 0.23228838294744492,
      "num_tokens": 119802185.0,
      "step": 64255
    },
    {
      "entropy": 6.065694904327392,
      "epoch": 5.674673260332038,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023030423820219532,
      "loss": 5.2318,
      "mean_token_accuracy": 0.22557660192251205,
      "num_tokens": 119811240.0,
      "step": 64260
    },
    {
      "entropy": 6.00711874961853,
      "epoch": 5.675114800423879,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00023027337473016985,
      "loss": 5.1415,
      "mean_token_accuracy": 0.22782740890979766,
      "num_tokens": 119819395.0,
      "step": 64265
    },
    {
      "entropy": 6.064270973205566,
      "epoch": 5.675556340515719,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023024251213431358,
      "loss": 5.267,
      "mean_token_accuracy": 0.22024231404066086,
      "num_tokens": 119829294.0,
      "step": 64270
    },
    {
      "entropy": 6.052889776229859,
      "epoch": 5.675997880607559,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00023021165041523095,
      "loss": 5.2241,
      "mean_token_accuracy": 0.2209206625819206,
      "num_tokens": 119838248.0,
      "step": 64275
    },
    {
      "entropy": 6.070553779602051,
      "epoch": 5.676439420699399,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00023018078957352673,
      "loss": 5.2488,
      "mean_token_accuracy": 0.2240344688296318,
      "num_tokens": 119847577.0,
      "step": 64280
    },
    {
      "entropy": 6.06378207206726,
      "epoch": 5.67688096079124,
      "grad_norm": 1.375,
      "learning_rate": 0.00023014992960980536,
      "loss": 5.3086,
      "mean_token_accuracy": 0.21891925483942032,
      "num_tokens": 119858511.0,
      "step": 64285
    },
    {
      "entropy": 6.027767705917358,
      "epoch": 5.67732250088308,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023011907052467135,
      "loss": 5.2033,
      "mean_token_accuracy": 0.23129819482564926,
      "num_tokens": 119867322.0,
      "step": 64290
    },
    {
      "entropy": 6.019578742980957,
      "epoch": 5.67776404097492,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023008821231872918,
      "loss": 5.2013,
      "mean_token_accuracy": 0.23192041516304016,
      "num_tokens": 119876931.0,
      "step": 64295
    },
    {
      "entropy": 6.002920436859131,
      "epoch": 5.678205581066761,
      "grad_norm": 1.734375,
      "learning_rate": 0.00023005735499258335,
      "loss": 5.1358,
      "mean_token_accuracy": 0.24362531155347825,
      "num_tokens": 119885459.0,
      "step": 64300
    },
    {
      "entropy": 6.006323051452637,
      "epoch": 5.678647121158601,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023002649854683842,
      "loss": 5.1655,
      "mean_token_accuracy": 0.23558349311351776,
      "num_tokens": 119894998.0,
      "step": 64305
    },
    {
      "entropy": 6.097140455245972,
      "epoch": 5.679088661250441,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022999564298209886,
      "loss": 5.3009,
      "mean_token_accuracy": 0.2214801087975502,
      "num_tokens": 119903745.0,
      "step": 64310
    },
    {
      "entropy": 6.0712854862213135,
      "epoch": 5.679530201342282,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000229964788298969,
      "loss": 5.2654,
      "mean_token_accuracy": 0.22651578933000566,
      "num_tokens": 119912678.0,
      "step": 64315
    },
    {
      "entropy": 6.0444886684417725,
      "epoch": 5.679971741434122,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00022993393449805328,
      "loss": 5.2373,
      "mean_token_accuracy": 0.21972679793834687,
      "num_tokens": 119921276.0,
      "step": 64320
    },
    {
      "entropy": 6.043264102935791,
      "epoch": 5.680413281525962,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022990308157995622,
      "loss": 5.1923,
      "mean_token_accuracy": 0.22702050656080247,
      "num_tokens": 119929724.0,
      "step": 64325
    },
    {
      "entropy": 6.053033018112183,
      "epoch": 5.680854821617803,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022987222954528214,
      "loss": 5.2337,
      "mean_token_accuracy": 0.21851613521575927,
      "num_tokens": 119939276.0,
      "step": 64330
    },
    {
      "entropy": 6.0254356384277346,
      "epoch": 5.681296361709643,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022984137839463542,
      "loss": 5.2175,
      "mean_token_accuracy": 0.23186353147029876,
      "num_tokens": 119948106.0,
      "step": 64335
    },
    {
      "entropy": 6.03131332397461,
      "epoch": 5.681737901801483,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002298105281286203,
      "loss": 5.1862,
      "mean_token_accuracy": 0.22545782029628753,
      "num_tokens": 119957870.0,
      "step": 64340
    },
    {
      "entropy": 6.09497299194336,
      "epoch": 5.682179441893324,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022977967874784135,
      "loss": 5.2847,
      "mean_token_accuracy": 0.21621648520231246,
      "num_tokens": 119967992.0,
      "step": 64345
    },
    {
      "entropy": 6.080926561355591,
      "epoch": 5.682620981985164,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022974883025290277,
      "loss": 5.222,
      "mean_token_accuracy": 0.22340450584888458,
      "num_tokens": 119978518.0,
      "step": 64350
    },
    {
      "entropy": 6.125198888778686,
      "epoch": 5.683062522077004,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002297179826444089,
      "loss": 5.214,
      "mean_token_accuracy": 0.2230077490210533,
      "num_tokens": 119987824.0,
      "step": 64355
    },
    {
      "entropy": 6.080354499816894,
      "epoch": 5.683504062168845,
      "grad_norm": 1.515625,
      "learning_rate": 0.00022968713592296403,
      "loss": 5.3429,
      "mean_token_accuracy": 0.21780598014593125,
      "num_tokens": 119996913.0,
      "step": 64360
    },
    {
      "entropy": 6.065709829330444,
      "epoch": 5.683945602260685,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002296562900891722,
      "loss": 5.2908,
      "mean_token_accuracy": 0.2223339080810547,
      "num_tokens": 120006242.0,
      "step": 64365
    },
    {
      "entropy": 6.0573255062103275,
      "epoch": 5.684387142352525,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00022962544514363803,
      "loss": 5.2221,
      "mean_token_accuracy": 0.2243541195988655,
      "num_tokens": 120015525.0,
      "step": 64370
    },
    {
      "entropy": 6.033401393890381,
      "epoch": 5.6848286824443655,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022959460108696556,
      "loss": 5.2339,
      "mean_token_accuracy": 0.22258369475603104,
      "num_tokens": 120025075.0,
      "step": 64375
    },
    {
      "entropy": 6.115711879730225,
      "epoch": 5.685270222536206,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022956375791975897,
      "loss": 5.3159,
      "mean_token_accuracy": 0.21585023403167725,
      "num_tokens": 120033882.0,
      "step": 64380
    },
    {
      "entropy": 6.035103225708008,
      "epoch": 5.6857117626280465,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022953291564262252,
      "loss": 5.1914,
      "mean_token_accuracy": 0.22622996270656587,
      "num_tokens": 120043710.0,
      "step": 64385
    },
    {
      "entropy": 6.055241250991822,
      "epoch": 5.686153302719887,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022950207425616044,
      "loss": 5.284,
      "mean_token_accuracy": 0.2232613444328308,
      "num_tokens": 120053064.0,
      "step": 64390
    },
    {
      "entropy": 6.016466665267944,
      "epoch": 5.6865948428117274,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00022947123376097683,
      "loss": 5.1184,
      "mean_token_accuracy": 0.2291925698518753,
      "num_tokens": 120061095.0,
      "step": 64395
    },
    {
      "entropy": 6.070451450347901,
      "epoch": 5.6870363829035675,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00022944039415767593,
      "loss": 5.1752,
      "mean_token_accuracy": 0.22909417599439622,
      "num_tokens": 120069230.0,
      "step": 64400
    },
    {
      "entropy": 6.004730892181397,
      "epoch": 5.6874779229954076,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002294095554468617,
      "loss": 5.2549,
      "mean_token_accuracy": 0.22026984840631486,
      "num_tokens": 120079079.0,
      "step": 64405
    },
    {
      "entropy": 6.086025094985962,
      "epoch": 5.6879194630872485,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022937871762913832,
      "loss": 5.2976,
      "mean_token_accuracy": 0.22099459618330003,
      "num_tokens": 120087924.0,
      "step": 64410
    },
    {
      "entropy": 6.031489515304566,
      "epoch": 5.6883610031790885,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022934788070511005,
      "loss": 5.2423,
      "mean_token_accuracy": 0.22237399369478225,
      "num_tokens": 120097271.0,
      "step": 64415
    },
    {
      "entropy": 6.084116268157959,
      "epoch": 5.6888025432709295,
      "grad_norm": 1.375,
      "learning_rate": 0.00022931704467538079,
      "loss": 5.2073,
      "mean_token_accuracy": 0.22506510466337204,
      "num_tokens": 120106377.0,
      "step": 64420
    },
    {
      "entropy": 6.107606029510498,
      "epoch": 5.6892440833627695,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022928620954055463,
      "loss": 5.2773,
      "mean_token_accuracy": 0.2202618956565857,
      "num_tokens": 120115603.0,
      "step": 64425
    },
    {
      "entropy": 6.071761178970337,
      "epoch": 5.68968562345461,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022925537530123552,
      "loss": 5.2355,
      "mean_token_accuracy": 0.2209542438387871,
      "num_tokens": 120125622.0,
      "step": 64430
    },
    {
      "entropy": 6.0883628845214846,
      "epoch": 5.69012716354645,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022922454195802773,
      "loss": 5.1977,
      "mean_token_accuracy": 0.21914686411619186,
      "num_tokens": 120133594.0,
      "step": 64435
    },
    {
      "entropy": 6.042026567459106,
      "epoch": 5.690568703638291,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022919370951153507,
      "loss": 5.1926,
      "mean_token_accuracy": 0.22834144681692123,
      "num_tokens": 120143628.0,
      "step": 64440
    },
    {
      "entropy": 6.074388647079468,
      "epoch": 5.691010243730131,
      "grad_norm": 1.375,
      "learning_rate": 0.00022916287796236157,
      "loss": 5.2884,
      "mean_token_accuracy": 0.2147987052798271,
      "num_tokens": 120152497.0,
      "step": 64445
    },
    {
      "entropy": 6.023843812942505,
      "epoch": 5.691451783821971,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022913204731111125,
      "loss": 5.2471,
      "mean_token_accuracy": 0.2240262895822525,
      "num_tokens": 120161611.0,
      "step": 64450
    },
    {
      "entropy": 6.063724565505981,
      "epoch": 5.691893323913812,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022910121755838808,
      "loss": 5.2429,
      "mean_token_accuracy": 0.2199820473790169,
      "num_tokens": 120170814.0,
      "step": 64455
    },
    {
      "entropy": 6.051152563095092,
      "epoch": 5.692334864005652,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022907038870479586,
      "loss": 5.1812,
      "mean_token_accuracy": 0.22721161395311357,
      "num_tokens": 120181236.0,
      "step": 64460
    },
    {
      "entropy": 6.133476495742798,
      "epoch": 5.692776404097492,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022903956075093868,
      "loss": 5.3161,
      "mean_token_accuracy": 0.21656529158353804,
      "num_tokens": 120190178.0,
      "step": 64465
    },
    {
      "entropy": 6.02775239944458,
      "epoch": 5.693217944189333,
      "grad_norm": 1.5,
      "learning_rate": 0.00022900873369742036,
      "loss": 5.216,
      "mean_token_accuracy": 0.22202563285827637,
      "num_tokens": 120199521.0,
      "step": 64470
    },
    {
      "entropy": 6.003298282623291,
      "epoch": 5.693659484281173,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022897790754484478,
      "loss": 5.1844,
      "mean_token_accuracy": 0.2266395002603531,
      "num_tokens": 120208874.0,
      "step": 64475
    },
    {
      "entropy": 6.043277311325073,
      "epoch": 5.694101024373013,
      "grad_norm": 1.4375,
      "learning_rate": 0.00022894708229381577,
      "loss": 5.1626,
      "mean_token_accuracy": 0.22689494639635086,
      "num_tokens": 120217943.0,
      "step": 64480
    },
    {
      "entropy": 6.036267328262329,
      "epoch": 5.694542564464854,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022891625794493732,
      "loss": 5.2058,
      "mean_token_accuracy": 0.23075441271066666,
      "num_tokens": 120227238.0,
      "step": 64485
    },
    {
      "entropy": 6.153110933303833,
      "epoch": 5.694984104556694,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022888543449881316,
      "loss": 5.2816,
      "mean_token_accuracy": 0.21752803325653075,
      "num_tokens": 120237097.0,
      "step": 64490
    },
    {
      "entropy": 5.9745691299438475,
      "epoch": 5.695425644648534,
      "grad_norm": 1.203125,
      "learning_rate": 0.00022885461195604694,
      "loss": 5.1531,
      "mean_token_accuracy": 0.23403184115886688,
      "num_tokens": 120246244.0,
      "step": 64495
    },
    {
      "entropy": 6.009371280670166,
      "epoch": 5.695867184740375,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022882379031724282,
      "loss": 5.1729,
      "mean_token_accuracy": 0.23463405221700667,
      "num_tokens": 120256395.0,
      "step": 64500
    },
    {
      "entropy": 6.055130338668823,
      "epoch": 5.696308724832215,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022879296958300434,
      "loss": 5.2881,
      "mean_token_accuracy": 0.21884698569774627,
      "num_tokens": 120265686.0,
      "step": 64505
    },
    {
      "entropy": 6.030555868148804,
      "epoch": 5.696750264924055,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022876214975393528,
      "loss": 5.1833,
      "mean_token_accuracy": 0.22866660952568055,
      "num_tokens": 120275730.0,
      "step": 64510
    },
    {
      "entropy": 6.016760683059692,
      "epoch": 5.697191805015896,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022873133083063936,
      "loss": 5.1054,
      "mean_token_accuracy": 0.23661316633224488,
      "num_tokens": 120284595.0,
      "step": 64515
    },
    {
      "entropy": 6.092371463775635,
      "epoch": 5.697633345107736,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022870051281372045,
      "loss": 5.2394,
      "mean_token_accuracy": 0.22327384650707244,
      "num_tokens": 120294905.0,
      "step": 64520
    },
    {
      "entropy": 6.075908374786377,
      "epoch": 5.698074885199576,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002286696957037821,
      "loss": 5.2896,
      "mean_token_accuracy": 0.214640249311924,
      "num_tokens": 120303533.0,
      "step": 64525
    },
    {
      "entropy": 5.9011537551879885,
      "epoch": 5.698516425291416,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002286388795014281,
      "loss": 5.1138,
      "mean_token_accuracy": 0.2392855241894722,
      "num_tokens": 120313095.0,
      "step": 64530
    },
    {
      "entropy": 5.968422174453735,
      "epoch": 5.698957965383257,
      "grad_norm": 1.5,
      "learning_rate": 0.0002286080642072621,
      "loss": 5.1649,
      "mean_token_accuracy": 0.22887583523988725,
      "num_tokens": 120322501.0,
      "step": 64535
    },
    {
      "entropy": 5.975747442245483,
      "epoch": 5.699399505475097,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022857724982188761,
      "loss": 5.1706,
      "mean_token_accuracy": 0.22688186466693877,
      "num_tokens": 120331012.0,
      "step": 64540
    },
    {
      "entropy": 6.098837852478027,
      "epoch": 5.699841045566938,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002285464363459085,
      "loss": 5.2608,
      "mean_token_accuracy": 0.22788869738578796,
      "num_tokens": 120339585.0,
      "step": 64545
    },
    {
      "entropy": 6.083356332778931,
      "epoch": 5.700282585658778,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002285156237799283,
      "loss": 5.2088,
      "mean_token_accuracy": 0.22372882068157196,
      "num_tokens": 120349460.0,
      "step": 64550
    },
    {
      "entropy": 6.086279106140137,
      "epoch": 5.700724125750618,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022848481212455062,
      "loss": 5.1668,
      "mean_token_accuracy": 0.2291752114892006,
      "num_tokens": 120359064.0,
      "step": 64555
    },
    {
      "entropy": 6.098154306411743,
      "epoch": 5.701165665842458,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022845400138037892,
      "loss": 5.258,
      "mean_token_accuracy": 0.22007008045911788,
      "num_tokens": 120368244.0,
      "step": 64560
    },
    {
      "entropy": 6.019584369659424,
      "epoch": 5.701607205934299,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022842319154801687,
      "loss": 5.1713,
      "mean_token_accuracy": 0.2256794661283493,
      "num_tokens": 120376946.0,
      "step": 64565
    },
    {
      "entropy": 6.042963027954102,
      "epoch": 5.702048746026139,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002283923826280681,
      "loss": 5.2882,
      "mean_token_accuracy": 0.22108257710933685,
      "num_tokens": 120385971.0,
      "step": 64570
    },
    {
      "entropy": 6.08293890953064,
      "epoch": 5.702490286117979,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022836157462113594,
      "loss": 5.3063,
      "mean_token_accuracy": 0.21717901527881622,
      "num_tokens": 120395891.0,
      "step": 64575
    },
    {
      "entropy": 6.048801898956299,
      "epoch": 5.70293182620982,
      "grad_norm": 1.5,
      "learning_rate": 0.00022833076752782407,
      "loss": 5.2575,
      "mean_token_accuracy": 0.22366945892572404,
      "num_tokens": 120404374.0,
      "step": 64580
    },
    {
      "entropy": 6.0654518604278564,
      "epoch": 5.70337336630166,
      "grad_norm": 1.5625,
      "learning_rate": 0.00022829996134873586,
      "loss": 5.1279,
      "mean_token_accuracy": 0.23343202024698256,
      "num_tokens": 120414436.0,
      "step": 64585
    },
    {
      "entropy": 5.980738401412964,
      "epoch": 5.7038149063935,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002282691560844749,
      "loss": 5.0912,
      "mean_token_accuracy": 0.23696849644184112,
      "num_tokens": 120423921.0,
      "step": 64590
    },
    {
      "entropy": 5.986513376235962,
      "epoch": 5.704256446485341,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022823835173564462,
      "loss": 5.2083,
      "mean_token_accuracy": 0.2256706699728966,
      "num_tokens": 120433087.0,
      "step": 64595
    },
    {
      "entropy": 5.992766380310059,
      "epoch": 5.704697986577181,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022820754830284846,
      "loss": 5.1957,
      "mean_token_accuracy": 0.23012888729572295,
      "num_tokens": 120441965.0,
      "step": 64600
    },
    {
      "entropy": 6.028047752380371,
      "epoch": 5.705139526669021,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022817674578668967,
      "loss": 5.1741,
      "mean_token_accuracy": 0.22525211572647094,
      "num_tokens": 120451099.0,
      "step": 64605
    },
    {
      "entropy": 6.046709966659546,
      "epoch": 5.705581066760862,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002281459441877719,
      "loss": 5.2619,
      "mean_token_accuracy": 0.22607458382844925,
      "num_tokens": 120461124.0,
      "step": 64610
    },
    {
      "entropy": 6.066010141372681,
      "epoch": 5.706022606852702,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022811514350669848,
      "loss": 5.2599,
      "mean_token_accuracy": 0.22052325010299684,
      "num_tokens": 120470591.0,
      "step": 64615
    },
    {
      "entropy": 6.022689628601074,
      "epoch": 5.706464146944542,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022808434374407266,
      "loss": 5.1394,
      "mean_token_accuracy": 0.22992742210626602,
      "num_tokens": 120479577.0,
      "step": 64620
    },
    {
      "entropy": 6.0372570037841795,
      "epoch": 5.706905687036383,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00022805354490049786,
      "loss": 5.2159,
      "mean_token_accuracy": 0.2282417595386505,
      "num_tokens": 120488486.0,
      "step": 64625
    },
    {
      "entropy": 6.051890897750854,
      "epoch": 5.707347227128223,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022802274697657744,
      "loss": 5.244,
      "mean_token_accuracy": 0.2231922924518585,
      "num_tokens": 120498205.0,
      "step": 64630
    },
    {
      "entropy": 6.0653845310211185,
      "epoch": 5.707788767220063,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022799194997291474,
      "loss": 5.2076,
      "mean_token_accuracy": 0.22613613605499266,
      "num_tokens": 120507757.0,
      "step": 64635
    },
    {
      "entropy": 5.991436338424682,
      "epoch": 5.708230307311904,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022796115389011291,
      "loss": 5.135,
      "mean_token_accuracy": 0.23701711297035216,
      "num_tokens": 120516866.0,
      "step": 64640
    },
    {
      "entropy": 5.923629665374756,
      "epoch": 5.708671847403744,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022793035872877538,
      "loss": 5.0651,
      "mean_token_accuracy": 0.23973822444677353,
      "num_tokens": 120525849.0,
      "step": 64645
    },
    {
      "entropy": 6.093776512145996,
      "epoch": 5.709113387495584,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022789956448950532,
      "loss": 5.3522,
      "mean_token_accuracy": 0.20760591924190522,
      "num_tokens": 120536097.0,
      "step": 64650
    },
    {
      "entropy": 6.039714193344116,
      "epoch": 5.7095549275874244,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000227868771172906,
      "loss": 5.2319,
      "mean_token_accuracy": 0.21647463738918304,
      "num_tokens": 120545628.0,
      "step": 64655
    },
    {
      "entropy": 6.07888560295105,
      "epoch": 5.709996467679265,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022783797877958068,
      "loss": 5.2544,
      "mean_token_accuracy": 0.21592862755060196,
      "num_tokens": 120554166.0,
      "step": 64660
    },
    {
      "entropy": 6.080185985565185,
      "epoch": 5.710438007771105,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022780718731013256,
      "loss": 5.2864,
      "mean_token_accuracy": 0.2205996945500374,
      "num_tokens": 120563755.0,
      "step": 64665
    },
    {
      "entropy": 6.073114681243896,
      "epoch": 5.710879547862946,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002277763967651647,
      "loss": 5.2341,
      "mean_token_accuracy": 0.2228602647781372,
      "num_tokens": 120573416.0,
      "step": 64670
    },
    {
      "entropy": 5.991053628921509,
      "epoch": 5.711321087954786,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022774560714528048,
      "loss": 5.1991,
      "mean_token_accuracy": 0.23011545091867447,
      "num_tokens": 120583213.0,
      "step": 64675
    },
    {
      "entropy": 6.06988034248352,
      "epoch": 5.7117626280466265,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022771481845108293,
      "loss": 5.2917,
      "mean_token_accuracy": 0.21353797614574432,
      "num_tokens": 120592561.0,
      "step": 64680
    },
    {
      "entropy": 6.060604619979858,
      "epoch": 5.7122041681384665,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00022768403068317518,
      "loss": 5.2047,
      "mean_token_accuracy": 0.22380769848823548,
      "num_tokens": 120602185.0,
      "step": 64685
    },
    {
      "entropy": 5.971560478210449,
      "epoch": 5.7126457082303075,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022765324384216025,
      "loss": 5.1767,
      "mean_token_accuracy": 0.23494547009468078,
      "num_tokens": 120611660.0,
      "step": 64690
    },
    {
      "entropy": 6.026274871826172,
      "epoch": 5.7130872483221475,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022762245792864146,
      "loss": 5.2099,
      "mean_token_accuracy": 0.22779621928930283,
      "num_tokens": 120621071.0,
      "step": 64695
    },
    {
      "entropy": 6.012297248840332,
      "epoch": 5.7135287884139885,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022759167294322176,
      "loss": 5.1999,
      "mean_token_accuracy": 0.23182999342679977,
      "num_tokens": 120630576.0,
      "step": 64700
    },
    {
      "entropy": 6.1118364334106445,
      "epoch": 5.7139703285058285,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022756088888650422,
      "loss": 5.355,
      "mean_token_accuracy": 0.22196165323257447,
      "num_tokens": 120640783.0,
      "step": 64705
    },
    {
      "entropy": 6.142512845993042,
      "epoch": 5.714411868597669,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002275301057590919,
      "loss": 5.3063,
      "mean_token_accuracy": 0.21621533930301667,
      "num_tokens": 120649201.0,
      "step": 64710
    },
    {
      "entropy": 6.179221677780151,
      "epoch": 5.714853408689509,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002274993235615877,
      "loss": 5.297,
      "mean_token_accuracy": 0.22058450132608415,
      "num_tokens": 120658641.0,
      "step": 64715
    },
    {
      "entropy": 6.061686611175537,
      "epoch": 5.71529494878135,
      "grad_norm": 1.515625,
      "learning_rate": 0.00022746854229459485,
      "loss": 5.2484,
      "mean_token_accuracy": 0.21802153140306474,
      "num_tokens": 120667720.0,
      "step": 64720
    },
    {
      "entropy": 5.9768754005432125,
      "epoch": 5.71573648887319,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022743776195871624,
      "loss": 5.2494,
      "mean_token_accuracy": 0.2229917377233505,
      "num_tokens": 120676822.0,
      "step": 64725
    },
    {
      "entropy": 6.008035326004029,
      "epoch": 5.71617802896503,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00022740698255455482,
      "loss": 5.206,
      "mean_token_accuracy": 0.22551426887512208,
      "num_tokens": 120685588.0,
      "step": 64730
    },
    {
      "entropy": 6.053620433807373,
      "epoch": 5.716619569056871,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022737620408271343,
      "loss": 5.1666,
      "mean_token_accuracy": 0.2294063851237297,
      "num_tokens": 120694532.0,
      "step": 64735
    },
    {
      "entropy": 6.0658129215240475,
      "epoch": 5.717061109148711,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00022734542654379518,
      "loss": 5.226,
      "mean_token_accuracy": 0.22475415468215942,
      "num_tokens": 120703530.0,
      "step": 64740
    },
    {
      "entropy": 6.031164073944092,
      "epoch": 5.717502649240551,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000227314649938403,
      "loss": 5.2027,
      "mean_token_accuracy": 0.22662367969751357,
      "num_tokens": 120712583.0,
      "step": 64745
    },
    {
      "entropy": 5.979190540313721,
      "epoch": 5.717944189332392,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022728387426713964,
      "loss": 5.1414,
      "mean_token_accuracy": 0.23273580521345139,
      "num_tokens": 120722093.0,
      "step": 64750
    },
    {
      "entropy": 6.044593715667725,
      "epoch": 5.718385729424232,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00022725309953060797,
      "loss": 5.2624,
      "mean_token_accuracy": 0.22261636704206467,
      "num_tokens": 120731513.0,
      "step": 64755
    },
    {
      "entropy": 6.053572511672973,
      "epoch": 5.718827269516072,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022722232572941104,
      "loss": 5.2688,
      "mean_token_accuracy": 0.22068918496370316,
      "num_tokens": 120740256.0,
      "step": 64760
    },
    {
      "entropy": 6.093364953994751,
      "epoch": 5.719268809607913,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002271915528641516,
      "loss": 5.251,
      "mean_token_accuracy": 0.22606363296508789,
      "num_tokens": 120750094.0,
      "step": 64765
    },
    {
      "entropy": 6.087379789352417,
      "epoch": 5.719710349699753,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022716078093543236,
      "loss": 5.245,
      "mean_token_accuracy": 0.22405577301979065,
      "num_tokens": 120759913.0,
      "step": 64770
    },
    {
      "entropy": 6.1120216846466064,
      "epoch": 5.720151889791593,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022713000994385623,
      "loss": 5.3345,
      "mean_token_accuracy": 0.21218059659004213,
      "num_tokens": 120769787.0,
      "step": 64775
    },
    {
      "entropy": 6.02663106918335,
      "epoch": 5.720593429883434,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022709923989002595,
      "loss": 5.1752,
      "mean_token_accuracy": 0.23120566457509995,
      "num_tokens": 120778834.0,
      "step": 64780
    },
    {
      "entropy": 6.10505747795105,
      "epoch": 5.721034969975274,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002270684707745444,
      "loss": 5.3043,
      "mean_token_accuracy": 0.22181327641010284,
      "num_tokens": 120789099.0,
      "step": 64785
    },
    {
      "entropy": 5.968724298477173,
      "epoch": 5.721476510067114,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002270377025980142,
      "loss": 5.1359,
      "mean_token_accuracy": 0.23153723478317262,
      "num_tokens": 120798292.0,
      "step": 64790
    },
    {
      "entropy": 5.955918598175049,
      "epoch": 5.721918050158955,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022700693536103816,
      "loss": 5.0784,
      "mean_token_accuracy": 0.23840800672769547,
      "num_tokens": 120806084.0,
      "step": 64795
    },
    {
      "entropy": 6.093039703369141,
      "epoch": 5.722359590250795,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002269761690642189,
      "loss": 5.3094,
      "mean_token_accuracy": 0.2215263158082962,
      "num_tokens": 120815112.0,
      "step": 64800
    },
    {
      "entropy": 6.046083450317383,
      "epoch": 5.722801130342635,
      "grad_norm": 1.609375,
      "learning_rate": 0.00022694540370815926,
      "loss": 5.2067,
      "mean_token_accuracy": 0.22363868951797486,
      "num_tokens": 120824620.0,
      "step": 64805
    },
    {
      "entropy": 6.033472204208374,
      "epoch": 5.723242670434475,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002269146392934618,
      "loss": 5.2036,
      "mean_token_accuracy": 0.22517645061016084,
      "num_tokens": 120834041.0,
      "step": 64810
    },
    {
      "entropy": 5.959655046463013,
      "epoch": 5.723684210526316,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022688387582072922,
      "loss": 5.1461,
      "mean_token_accuracy": 0.2301218956708908,
      "num_tokens": 120844316.0,
      "step": 64815
    },
    {
      "entropy": 5.9237782001495365,
      "epoch": 5.724125750618156,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022685311329056403,
      "loss": 5.1793,
      "mean_token_accuracy": 0.22753156274557113,
      "num_tokens": 120853933.0,
      "step": 64820
    },
    {
      "entropy": 6.112920665740967,
      "epoch": 5.724567290709997,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022682235170356913,
      "loss": 5.27,
      "mean_token_accuracy": 0.22272626161575318,
      "num_tokens": 120863779.0,
      "step": 64825
    },
    {
      "entropy": 6.065090847015381,
      "epoch": 5.725008830801837,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022679159106034686,
      "loss": 5.2788,
      "mean_token_accuracy": 0.22520652562379836,
      "num_tokens": 120873593.0,
      "step": 64830
    },
    {
      "entropy": 6.072163200378418,
      "epoch": 5.725450370893677,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022676083136149994,
      "loss": 5.2193,
      "mean_token_accuracy": 0.222272689640522,
      "num_tokens": 120884908.0,
      "step": 64835
    },
    {
      "entropy": 6.045508956909179,
      "epoch": 5.725891910985517,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002267300726076309,
      "loss": 5.2308,
      "mean_token_accuracy": 0.22099415957927704,
      "num_tokens": 120893588.0,
      "step": 64840
    },
    {
      "entropy": 6.066355800628662,
      "epoch": 5.726333451077358,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022669931479934224,
      "loss": 5.2044,
      "mean_token_accuracy": 0.22517113536596298,
      "num_tokens": 120902801.0,
      "step": 64845
    },
    {
      "entropy": 6.04302453994751,
      "epoch": 5.726774991169198,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022666855793723655,
      "loss": 5.1808,
      "mean_token_accuracy": 0.23301465809345245,
      "num_tokens": 120911338.0,
      "step": 64850
    },
    {
      "entropy": 6.031581020355224,
      "epoch": 5.727216531261038,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022663780202191637,
      "loss": 5.2449,
      "mean_token_accuracy": 0.22574107348918915,
      "num_tokens": 120919864.0,
      "step": 64855
    },
    {
      "entropy": 5.981935739517212,
      "epoch": 5.727658071352879,
      "grad_norm": 1.25,
      "learning_rate": 0.00022660704705398417,
      "loss": 5.219,
      "mean_token_accuracy": 0.22228154987096788,
      "num_tokens": 120929544.0,
      "step": 64860
    },
    {
      "entropy": 6.054442882537842,
      "epoch": 5.728099611444719,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022657629303404227,
      "loss": 5.246,
      "mean_token_accuracy": 0.21524803936481476,
      "num_tokens": 120939953.0,
      "step": 64865
    },
    {
      "entropy": 6.042230653762817,
      "epoch": 5.728541151536559,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022654553996269333,
      "loss": 5.1091,
      "mean_token_accuracy": 0.24028144180774688,
      "num_tokens": 120948319.0,
      "step": 64870
    },
    {
      "entropy": 6.112628555297851,
      "epoch": 5.7289826916284,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022651478784053975,
      "loss": 5.2724,
      "mean_token_accuracy": 0.22554864287376403,
      "num_tokens": 120957230.0,
      "step": 64875
    },
    {
      "entropy": 6.0460387706756595,
      "epoch": 5.72942423172024,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022648403666818384,
      "loss": 5.2001,
      "mean_token_accuracy": 0.2252305790781975,
      "num_tokens": 120966827.0,
      "step": 64880
    },
    {
      "entropy": 6.0565179824829105,
      "epoch": 5.72986577181208,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022645328644622803,
      "loss": 5.2505,
      "mean_token_accuracy": 0.2176957055926323,
      "num_tokens": 120976941.0,
      "step": 64885
    },
    {
      "entropy": 5.990104150772095,
      "epoch": 5.730307311903921,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022642253717527472,
      "loss": 5.1329,
      "mean_token_accuracy": 0.23566804975271224,
      "num_tokens": 120986500.0,
      "step": 64890
    },
    {
      "entropy": 5.971258687973022,
      "epoch": 5.730748851995761,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022639178885592638,
      "loss": 5.1854,
      "mean_token_accuracy": 0.2366543710231781,
      "num_tokens": 120996246.0,
      "step": 64895
    },
    {
      "entropy": 6.010472297668457,
      "epoch": 5.731190392087601,
      "grad_norm": 1.515625,
      "learning_rate": 0.00022636104148878518,
      "loss": 5.1885,
      "mean_token_accuracy": 0.23122541159391402,
      "num_tokens": 121005302.0,
      "step": 64900
    },
    {
      "entropy": 6.103404474258423,
      "epoch": 5.731631932179442,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022633029507445357,
      "loss": 5.3116,
      "mean_token_accuracy": 0.21865862160921096,
      "num_tokens": 121015236.0,
      "step": 64905
    },
    {
      "entropy": 6.100571155548096,
      "epoch": 5.732073472271282,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022629954961353365,
      "loss": 5.2602,
      "mean_token_accuracy": 0.21578036546707152,
      "num_tokens": 121024827.0,
      "step": 64910
    },
    {
      "entropy": 6.023182582855225,
      "epoch": 5.732515012363122,
      "grad_norm": 1.390625,
      "learning_rate": 0.000226268805106628,
      "loss": 5.2386,
      "mean_token_accuracy": 0.22773851603269576,
      "num_tokens": 121034438.0,
      "step": 64915
    },
    {
      "entropy": 6.008780765533447,
      "epoch": 5.732956552454963,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022623806155433875,
      "loss": 5.1871,
      "mean_token_accuracy": 0.2304384246468544,
      "num_tokens": 121043176.0,
      "step": 64920
    },
    {
      "entropy": 5.993228101730347,
      "epoch": 5.733398092546803,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022620731895726815,
      "loss": 5.1548,
      "mean_token_accuracy": 0.2299407809972763,
      "num_tokens": 121053023.0,
      "step": 64925
    },
    {
      "entropy": 6.13562388420105,
      "epoch": 5.733839632638643,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022617657731601826,
      "loss": 5.3211,
      "mean_token_accuracy": 0.21374329924583435,
      "num_tokens": 121063000.0,
      "step": 64930
    },
    {
      "entropy": 6.092815685272217,
      "epoch": 5.734281172730483,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002261458366311916,
      "loss": 5.2562,
      "mean_token_accuracy": 0.2185105249285698,
      "num_tokens": 121072612.0,
      "step": 64935
    },
    {
      "entropy": 6.135110664367676,
      "epoch": 5.734722712822324,
      "grad_norm": 1.375,
      "learning_rate": 0.00022611509690339022,
      "loss": 5.2386,
      "mean_token_accuracy": 0.22207489609718323,
      "num_tokens": 121081254.0,
      "step": 64940
    },
    {
      "entropy": 6.118426513671875,
      "epoch": 5.735164252914164,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022608435813321626,
      "loss": 5.1989,
      "mean_token_accuracy": 0.2289925619959831,
      "num_tokens": 121090519.0,
      "step": 64945
    },
    {
      "entropy": 6.088303518295288,
      "epoch": 5.735605793006005,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022605362032127185,
      "loss": 5.2116,
      "mean_token_accuracy": 0.23023901879787445,
      "num_tokens": 121099520.0,
      "step": 64950
    },
    {
      "entropy": 6.005373430252075,
      "epoch": 5.736047333097845,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022602288346815924,
      "loss": 5.1594,
      "mean_token_accuracy": 0.22711270302534103,
      "num_tokens": 121108442.0,
      "step": 64955
    },
    {
      "entropy": 6.0069167137146,
      "epoch": 5.7364888731896855,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002259921475744805,
      "loss": 5.1889,
      "mean_token_accuracy": 0.22746557891368865,
      "num_tokens": 121118500.0,
      "step": 64960
    },
    {
      "entropy": 6.036863851547241,
      "epoch": 5.7369304132815255,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022596141264083765,
      "loss": 5.2078,
      "mean_token_accuracy": 0.22513587474823,
      "num_tokens": 121127598.0,
      "step": 64965
    },
    {
      "entropy": 6.031845331192017,
      "epoch": 5.7373719533733665,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002259306786678329,
      "loss": 5.2779,
      "mean_token_accuracy": 0.22278582453727722,
      "num_tokens": 121136612.0,
      "step": 64970
    },
    {
      "entropy": 5.944968509674072,
      "epoch": 5.7378134934652065,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022589994565606813,
      "loss": 5.1659,
      "mean_token_accuracy": 0.23524857461452484,
      "num_tokens": 121145529.0,
      "step": 64975
    },
    {
      "entropy": 6.058869409561157,
      "epoch": 5.7382550335570475,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00022586921360614565,
      "loss": 5.2434,
      "mean_token_accuracy": 0.21255634278059005,
      "num_tokens": 121155542.0,
      "step": 64980
    },
    {
      "entropy": 6.051764726638794,
      "epoch": 5.7386965736488875,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002258384825186673,
      "loss": 5.2216,
      "mean_token_accuracy": 0.22813447415828705,
      "num_tokens": 121164677.0,
      "step": 64985
    },
    {
      "entropy": 6.08813066482544,
      "epoch": 5.739138113740728,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022580775239423512,
      "loss": 5.2051,
      "mean_token_accuracy": 0.22325177490711212,
      "num_tokens": 121173971.0,
      "step": 64990
    },
    {
      "entropy": 6.123931550979615,
      "epoch": 5.739579653832568,
      "grad_norm": 1.4375,
      "learning_rate": 0.00022577702323345095,
      "loss": 5.3008,
      "mean_token_accuracy": 0.21782266944646836,
      "num_tokens": 121183332.0,
      "step": 64995
    },
    {
      "entropy": 6.025940179824829,
      "epoch": 5.740021193924409,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00022574629503691707,
      "loss": 5.1661,
      "mean_token_accuracy": 0.23312825411558152,
      "num_tokens": 121191236.0,
      "step": 65000
    },
    {
      "entropy": 6.023004388809204,
      "epoch": 5.740462734016249,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022571556780523523,
      "loss": 5.2506,
      "mean_token_accuracy": 0.22439238280057908,
      "num_tokens": 121200858.0,
      "step": 65005
    },
    {
      "entropy": 5.9922479629516605,
      "epoch": 5.740904274108089,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022568484153900743,
      "loss": 5.1642,
      "mean_token_accuracy": 0.2315322145819664,
      "num_tokens": 121209335.0,
      "step": 65010
    },
    {
      "entropy": 6.007657432556153,
      "epoch": 5.74134581419993,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022565411623883542,
      "loss": 5.2307,
      "mean_token_accuracy": 0.22799715548753738,
      "num_tokens": 121219109.0,
      "step": 65015
    },
    {
      "entropy": 6.052299976348877,
      "epoch": 5.74178735429177,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022562339190532123,
      "loss": 5.2457,
      "mean_token_accuracy": 0.22791391015052795,
      "num_tokens": 121228793.0,
      "step": 65020
    },
    {
      "entropy": 6.140032386779785,
      "epoch": 5.74222889438361,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022559266853906673,
      "loss": 5.2604,
      "mean_token_accuracy": 0.21425240933895112,
      "num_tokens": 121238213.0,
      "step": 65025
    },
    {
      "entropy": 6.088366603851318,
      "epoch": 5.742670434475451,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022556194614067377,
      "loss": 5.1732,
      "mean_token_accuracy": 0.22733699679374694,
      "num_tokens": 121247167.0,
      "step": 65030
    },
    {
      "entropy": 6.025928401947022,
      "epoch": 5.743111974567291,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022553122471074416,
      "loss": 5.1896,
      "mean_token_accuracy": 0.22738633006811143,
      "num_tokens": 121256794.0,
      "step": 65035
    },
    {
      "entropy": 6.053484296798706,
      "epoch": 5.743553514659131,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022550050424987968,
      "loss": 5.2796,
      "mean_token_accuracy": 0.21979994475841522,
      "num_tokens": 121266093.0,
      "step": 65040
    },
    {
      "entropy": 6.059653186798096,
      "epoch": 5.743995054750972,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022546978475868225,
      "loss": 5.2364,
      "mean_token_accuracy": 0.22430995106697083,
      "num_tokens": 121275425.0,
      "step": 65045
    },
    {
      "entropy": 6.055868291854859,
      "epoch": 5.744436594842812,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022543906623775356,
      "loss": 5.2018,
      "mean_token_accuracy": 0.22884510606527328,
      "num_tokens": 121284894.0,
      "step": 65050
    },
    {
      "entropy": 6.064989709854126,
      "epoch": 5.744878134934652,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00022540834868769533,
      "loss": 5.3179,
      "mean_token_accuracy": 0.2189380019903183,
      "num_tokens": 121293233.0,
      "step": 65055
    },
    {
      "entropy": 6.028264045715332,
      "epoch": 5.745319675026493,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00022537763210910933,
      "loss": 5.2428,
      "mean_token_accuracy": 0.22260084599256516,
      "num_tokens": 121302095.0,
      "step": 65060
    },
    {
      "entropy": 6.102111291885376,
      "epoch": 5.745761215118333,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022534691650259736,
      "loss": 5.288,
      "mean_token_accuracy": 0.21930871158838272,
      "num_tokens": 121312540.0,
      "step": 65065
    },
    {
      "entropy": 6.054708099365234,
      "epoch": 5.746202755210173,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022531620186876106,
      "loss": 5.2384,
      "mean_token_accuracy": 0.22896224707365037,
      "num_tokens": 121321583.0,
      "step": 65070
    },
    {
      "entropy": 5.985583257675171,
      "epoch": 5.746644295302014,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022528548820820204,
      "loss": 5.1183,
      "mean_token_accuracy": 0.24135449528694153,
      "num_tokens": 121331734.0,
      "step": 65075
    },
    {
      "entropy": 5.9907526016235355,
      "epoch": 5.747085835393854,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00022525477552152206,
      "loss": 5.1407,
      "mean_token_accuracy": 0.232337786257267,
      "num_tokens": 121340784.0,
      "step": 65080
    },
    {
      "entropy": 6.032756996154785,
      "epoch": 5.747527375485694,
      "grad_norm": 1.375,
      "learning_rate": 0.00022522406380932276,
      "loss": 5.2069,
      "mean_token_accuracy": 0.22643100172281266,
      "num_tokens": 121350412.0,
      "step": 65085
    },
    {
      "entropy": 6.042453956604004,
      "epoch": 5.747968915577534,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022519335307220567,
      "loss": 5.2167,
      "mean_token_accuracy": 0.22613301128149033,
      "num_tokens": 121359290.0,
      "step": 65090
    },
    {
      "entropy": 5.968304634094238,
      "epoch": 5.748410455669375,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022516264331077257,
      "loss": 5.0966,
      "mean_token_accuracy": 0.23719628304243087,
      "num_tokens": 121368493.0,
      "step": 65095
    },
    {
      "entropy": 6.027510929107666,
      "epoch": 5.748851995761215,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00022513193452562485,
      "loss": 5.1791,
      "mean_token_accuracy": 0.23055585622787475,
      "num_tokens": 121377768.0,
      "step": 65100
    },
    {
      "entropy": 6.070351505279541,
      "epoch": 5.749293535853056,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022510122671736416,
      "loss": 5.2606,
      "mean_token_accuracy": 0.21972150057554246,
      "num_tokens": 121386018.0,
      "step": 65105
    },
    {
      "entropy": 6.084107112884522,
      "epoch": 5.749735075944896,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022507051988659218,
      "loss": 5.2815,
      "mean_token_accuracy": 0.2237027868628502,
      "num_tokens": 121395664.0,
      "step": 65110
    },
    {
      "entropy": 6.100950908660889,
      "epoch": 5.750176616036736,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022503981403391028,
      "loss": 5.1981,
      "mean_token_accuracy": 0.23269277215003967,
      "num_tokens": 121403683.0,
      "step": 65115
    },
    {
      "entropy": 5.9901073455810545,
      "epoch": 5.750618156128576,
      "grad_norm": 1.4375,
      "learning_rate": 0.00022500910915991995,
      "loss": 5.2007,
      "mean_token_accuracy": 0.23113679885864258,
      "num_tokens": 121413664.0,
      "step": 65120
    },
    {
      "entropy": 6.080886173248291,
      "epoch": 5.751059696220417,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022497840526522283,
      "loss": 5.2819,
      "mean_token_accuracy": 0.22288938760757446,
      "num_tokens": 121423205.0,
      "step": 65125
    },
    {
      "entropy": 6.009397411346436,
      "epoch": 5.751501236312257,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022494770235042017,
      "loss": 5.1344,
      "mean_token_accuracy": 0.23640809804201127,
      "num_tokens": 121432258.0,
      "step": 65130
    },
    {
      "entropy": 6.075330018997192,
      "epoch": 5.751942776404097,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002249170004161137,
      "loss": 5.273,
      "mean_token_accuracy": 0.2273866355419159,
      "num_tokens": 121441189.0,
      "step": 65135
    },
    {
      "entropy": 6.0495201587677006,
      "epoch": 5.752384316495938,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022488629946290463,
      "loss": 5.2084,
      "mean_token_accuracy": 0.22676659375429153,
      "num_tokens": 121450940.0,
      "step": 65140
    },
    {
      "entropy": 6.030805826187134,
      "epoch": 5.752825856587778,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002248555994913944,
      "loss": 5.1149,
      "mean_token_accuracy": 0.2444385588169098,
      "num_tokens": 121459797.0,
      "step": 65145
    },
    {
      "entropy": 6.016961526870728,
      "epoch": 5.753267396679618,
      "grad_norm": 1.546875,
      "learning_rate": 0.00022482490050218453,
      "loss": 5.1857,
      "mean_token_accuracy": 0.22583237886428834,
      "num_tokens": 121468927.0,
      "step": 65150
    },
    {
      "entropy": 6.036112117767334,
      "epoch": 5.753708936771459,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002247942024958763,
      "loss": 5.1608,
      "mean_token_accuracy": 0.2322193831205368,
      "num_tokens": 121478075.0,
      "step": 65155
    },
    {
      "entropy": 6.033099794387818,
      "epoch": 5.754150476863299,
      "grad_norm": 1.5625,
      "learning_rate": 0.00022476350547307106,
      "loss": 5.1435,
      "mean_token_accuracy": 0.2301931843161583,
      "num_tokens": 121486683.0,
      "step": 65160
    },
    {
      "entropy": 5.973668003082276,
      "epoch": 5.754592016955139,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00022473280943437025,
      "loss": 5.158,
      "mean_token_accuracy": 0.22908213138580322,
      "num_tokens": 121495906.0,
      "step": 65165
    },
    {
      "entropy": 6.039243268966675,
      "epoch": 5.75503355704698,
      "grad_norm": 1.4375,
      "learning_rate": 0.00022470211438037504,
      "loss": 5.2623,
      "mean_token_accuracy": 0.21631310880184174,
      "num_tokens": 121505586.0,
      "step": 65170
    },
    {
      "entropy": 6.073486280441284,
      "epoch": 5.75547509713882,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022467142031168687,
      "loss": 5.2645,
      "mean_token_accuracy": 0.22260105907917022,
      "num_tokens": 121515241.0,
      "step": 65175
    },
    {
      "entropy": 6.126099157333374,
      "epoch": 5.75591663723066,
      "grad_norm": 1.25,
      "learning_rate": 0.00022464072722890694,
      "loss": 5.2664,
      "mean_token_accuracy": 0.21139838248491288,
      "num_tokens": 121525010.0,
      "step": 65180
    },
    {
      "entropy": 6.044147634506226,
      "epoch": 5.756358177322501,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022461003513263655,
      "loss": 5.1366,
      "mean_token_accuracy": 0.22630207538604735,
      "num_tokens": 121533914.0,
      "step": 65185
    },
    {
      "entropy": 5.986064529418945,
      "epoch": 5.756799717414341,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002245793440234769,
      "loss": 5.1337,
      "mean_token_accuracy": 0.23250947892665863,
      "num_tokens": 121543041.0,
      "step": 65190
    },
    {
      "entropy": 6.0010247230529785,
      "epoch": 5.757241257506181,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002245486539020291,
      "loss": 5.1277,
      "mean_token_accuracy": 0.23250033408403398,
      "num_tokens": 121551776.0,
      "step": 65195
    },
    {
      "entropy": 5.992952108383179,
      "epoch": 5.757682797598022,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022451796476889458,
      "loss": 5.2002,
      "mean_token_accuracy": 0.2329029843211174,
      "num_tokens": 121560928.0,
      "step": 65200
    },
    {
      "entropy": 6.037503051757812,
      "epoch": 5.758124337689862,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00022448727662467444,
      "loss": 5.1901,
      "mean_token_accuracy": 0.23055570870637893,
      "num_tokens": 121569739.0,
      "step": 65205
    },
    {
      "entropy": 6.072567939758301,
      "epoch": 5.758565877781702,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022445658946996982,
      "loss": 5.2939,
      "mean_token_accuracy": 0.2243887573480606,
      "num_tokens": 121579389.0,
      "step": 65210
    },
    {
      "entropy": 6.150631904602051,
      "epoch": 5.759007417873542,
      "grad_norm": 1.375,
      "learning_rate": 0.00022442590330538182,
      "loss": 5.3064,
      "mean_token_accuracy": 0.2138887494802475,
      "num_tokens": 121588631.0,
      "step": 65215
    },
    {
      "entropy": 6.095701599121094,
      "epoch": 5.759448957965383,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002243952181315116,
      "loss": 5.3009,
      "mean_token_accuracy": 0.2228161796927452,
      "num_tokens": 121599219.0,
      "step": 65220
    },
    {
      "entropy": 6.016352939605713,
      "epoch": 5.759890498057223,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022436453394896033,
      "loss": 5.1841,
      "mean_token_accuracy": 0.23187300860881804,
      "num_tokens": 121608731.0,
      "step": 65225
    },
    {
      "entropy": 6.052864456176758,
      "epoch": 5.760332038149064,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022433385075832908,
      "loss": 5.2021,
      "mean_token_accuracy": 0.21942168474197388,
      "num_tokens": 121617772.0,
      "step": 65230
    },
    {
      "entropy": 6.077276182174683,
      "epoch": 5.760773578240904,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002243031685602187,
      "loss": 5.2725,
      "mean_token_accuracy": 0.21978256851434708,
      "num_tokens": 121626969.0,
      "step": 65235
    },
    {
      "entropy": 6.0047181129455565,
      "epoch": 5.7612151183327445,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00022427248735523055,
      "loss": 5.2484,
      "mean_token_accuracy": 0.22825865149497987,
      "num_tokens": 121635639.0,
      "step": 65240
    },
    {
      "entropy": 6.0616843700408936,
      "epoch": 5.7616566584245845,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022424180714396554,
      "loss": 5.2003,
      "mean_token_accuracy": 0.22543377578258514,
      "num_tokens": 121644383.0,
      "step": 65245
    },
    {
      "entropy": 6.005344820022583,
      "epoch": 5.7620981985164255,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022421112792702464,
      "loss": 5.1596,
      "mean_token_accuracy": 0.23410073071718215,
      "num_tokens": 121654143.0,
      "step": 65250
    },
    {
      "entropy": 6.024462556838989,
      "epoch": 5.7625397386082655,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002241804497050089,
      "loss": 5.2004,
      "mean_token_accuracy": 0.2164280503988266,
      "num_tokens": 121662572.0,
      "step": 65255
    },
    {
      "entropy": 6.021202659606933,
      "epoch": 5.7629812787001065,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002241497724785191,
      "loss": 5.2556,
      "mean_token_accuracy": 0.2290619969367981,
      "num_tokens": 121672029.0,
      "step": 65260
    },
    {
      "entropy": 6.126464796066284,
      "epoch": 5.7634228187919465,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022411909624815647,
      "loss": 5.2881,
      "mean_token_accuracy": 0.2231197953224182,
      "num_tokens": 121682377.0,
      "step": 65265
    },
    {
      "entropy": 6.004750442504883,
      "epoch": 5.763864358883787,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002240884210145218,
      "loss": 5.153,
      "mean_token_accuracy": 0.23844006210565566,
      "num_tokens": 121691722.0,
      "step": 65270
    },
    {
      "entropy": 6.011303091049195,
      "epoch": 5.764305898975627,
      "grad_norm": 1.3125,
      "learning_rate": 0.000224057746778216,
      "loss": 5.1731,
      "mean_token_accuracy": 0.2316318780183792,
      "num_tokens": 121701731.0,
      "step": 65275
    },
    {
      "entropy": 6.038801622390747,
      "epoch": 5.764747439067468,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00022402707353983987,
      "loss": 5.1448,
      "mean_token_accuracy": 0.23158996999263765,
      "num_tokens": 121711055.0,
      "step": 65280
    },
    {
      "entropy": 5.974018239974976,
      "epoch": 5.765188979159308,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022399640129999444,
      "loss": 5.1055,
      "mean_token_accuracy": 0.23312443792819976,
      "num_tokens": 121720298.0,
      "step": 65285
    },
    {
      "entropy": 6.061288356781006,
      "epoch": 5.765630519251148,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022396573005928057,
      "loss": 5.2259,
      "mean_token_accuracy": 0.22648661732673644,
      "num_tokens": 121729763.0,
      "step": 65290
    },
    {
      "entropy": 5.961727476119995,
      "epoch": 5.766072059342989,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022393505981829897,
      "loss": 5.185,
      "mean_token_accuracy": 0.23272374123334885,
      "num_tokens": 121738855.0,
      "step": 65295
    },
    {
      "entropy": 6.0794442653656,
      "epoch": 5.766513599434829,
      "grad_norm": 1.4375,
      "learning_rate": 0.00022390439057765044,
      "loss": 5.2632,
      "mean_token_accuracy": 0.21933842748403548,
      "num_tokens": 121748229.0,
      "step": 65300
    },
    {
      "entropy": 6.069154167175293,
      "epoch": 5.766955139526669,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022387372233793591,
      "loss": 5.1973,
      "mean_token_accuracy": 0.22691559046506882,
      "num_tokens": 121758421.0,
      "step": 65305
    },
    {
      "entropy": 6.036192131042481,
      "epoch": 5.76739667961851,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002238430550997561,
      "loss": 5.2059,
      "mean_token_accuracy": 0.22861550450325013,
      "num_tokens": 121767666.0,
      "step": 65310
    },
    {
      "entropy": 5.995635175704956,
      "epoch": 5.76783821971035,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022381238886371176,
      "loss": 5.1706,
      "mean_token_accuracy": 0.2301442340016365,
      "num_tokens": 121776053.0,
      "step": 65315
    },
    {
      "entropy": 6.0045140266418455,
      "epoch": 5.76827975980219,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002237817236304036,
      "loss": 5.2129,
      "mean_token_accuracy": 0.22222837805747986,
      "num_tokens": 121785309.0,
      "step": 65320
    },
    {
      "entropy": 5.983602237701416,
      "epoch": 5.768721299894031,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022375105940043222,
      "loss": 5.116,
      "mean_token_accuracy": 0.23204842060804368,
      "num_tokens": 121794212.0,
      "step": 65325
    },
    {
      "entropy": 6.058159875869751,
      "epoch": 5.769162839985871,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002237203961743986,
      "loss": 5.2635,
      "mean_token_accuracy": 0.22338917553424836,
      "num_tokens": 121804685.0,
      "step": 65330
    },
    {
      "entropy": 6.099439525604248,
      "epoch": 5.769604380077711,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002236897339529032,
      "loss": 5.2978,
      "mean_token_accuracy": 0.22114662379026412,
      "num_tokens": 121814695.0,
      "step": 65335
    },
    {
      "entropy": 6.019290256500244,
      "epoch": 5.770045920169552,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022365907273654668,
      "loss": 5.1917,
      "mean_token_accuracy": 0.23132004737854003,
      "num_tokens": 121823914.0,
      "step": 65340
    },
    {
      "entropy": 6.051368951797485,
      "epoch": 5.770487460261392,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022362841252592975,
      "loss": 5.1845,
      "mean_token_accuracy": 0.23567629903554915,
      "num_tokens": 121833185.0,
      "step": 65345
    },
    {
      "entropy": 6.036080646514892,
      "epoch": 5.770929000353232,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000223597753321653,
      "loss": 5.2217,
      "mean_token_accuracy": 0.22684987634420395,
      "num_tokens": 121842586.0,
      "step": 65350
    },
    {
      "entropy": 6.09366135597229,
      "epoch": 5.771370540445073,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022356709512431704,
      "loss": 5.2545,
      "mean_token_accuracy": 0.22154962718486787,
      "num_tokens": 121852141.0,
      "step": 65355
    },
    {
      "entropy": 6.092366886138916,
      "epoch": 5.771812080536913,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002235364379345225,
      "loss": 5.2476,
      "mean_token_accuracy": 0.22960733622312546,
      "num_tokens": 121862349.0,
      "step": 65360
    },
    {
      "entropy": 6.127330732345581,
      "epoch": 5.772253620628753,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00022350578175286983,
      "loss": 5.2561,
      "mean_token_accuracy": 0.22206326872110366,
      "num_tokens": 121872213.0,
      "step": 65365
    },
    {
      "entropy": 6.058110141754151,
      "epoch": 5.772695160720593,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022347512657995956,
      "loss": 5.2219,
      "mean_token_accuracy": 0.22006965577602386,
      "num_tokens": 121881447.0,
      "step": 65370
    },
    {
      "entropy": 6.0387287616729735,
      "epoch": 5.773136700812434,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022344447241639233,
      "loss": 5.2528,
      "mean_token_accuracy": 0.22250719517469406,
      "num_tokens": 121890541.0,
      "step": 65375
    },
    {
      "entropy": 6.08551139831543,
      "epoch": 5.773578240904274,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002234138192627685,
      "loss": 5.2873,
      "mean_token_accuracy": 0.21207962185144424,
      "num_tokens": 121899640.0,
      "step": 65380
    },
    {
      "entropy": 6.01145224571228,
      "epoch": 5.774019780996115,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002233831671196887,
      "loss": 5.1564,
      "mean_token_accuracy": 0.23218727856874466,
      "num_tokens": 121908730.0,
      "step": 65385
    },
    {
      "entropy": 6.064224624633789,
      "epoch": 5.774461321087955,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022335251598775313,
      "loss": 5.1939,
      "mean_token_accuracy": 0.23062646985054017,
      "num_tokens": 121917658.0,
      "step": 65390
    },
    {
      "entropy": 6.013734006881714,
      "epoch": 5.774902861179795,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022332186586756253,
      "loss": 5.1473,
      "mean_token_accuracy": 0.23467841744422913,
      "num_tokens": 121926794.0,
      "step": 65395
    },
    {
      "entropy": 6.013387680053711,
      "epoch": 5.775344401271635,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002232912167597172,
      "loss": 5.2285,
      "mean_token_accuracy": 0.22557155191898345,
      "num_tokens": 121936486.0,
      "step": 65400
    },
    {
      "entropy": 6.108391904830933,
      "epoch": 5.775785941363476,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002232605686648174,
      "loss": 5.347,
      "mean_token_accuracy": 0.21174100041389465,
      "num_tokens": 121946322.0,
      "step": 65405
    },
    {
      "entropy": 6.097451639175415,
      "epoch": 5.776227481455316,
      "grad_norm": 1.53125,
      "learning_rate": 0.00022322992158346373,
      "loss": 5.1763,
      "mean_token_accuracy": 0.22832812517881393,
      "num_tokens": 121954801.0,
      "step": 65410
    },
    {
      "entropy": 5.9492762088775635,
      "epoch": 5.776669021547156,
      "grad_norm": 1.5,
      "learning_rate": 0.00022319927551625647,
      "loss": 5.1457,
      "mean_token_accuracy": 0.23432580530643463,
      "num_tokens": 121962821.0,
      "step": 65415
    },
    {
      "entropy": 6.035027599334716,
      "epoch": 5.777110561638997,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022316863046379588,
      "loss": 5.2251,
      "mean_token_accuracy": 0.2221979171037674,
      "num_tokens": 121972114.0,
      "step": 65420
    },
    {
      "entropy": 5.989419651031494,
      "epoch": 5.777552101730837,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022313798642668243,
      "loss": 5.1093,
      "mean_token_accuracy": 0.23420485407114028,
      "num_tokens": 121980507.0,
      "step": 65425
    },
    {
      "entropy": 6.015398597717285,
      "epoch": 5.777993641822677,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002231073434055163,
      "loss": 5.1693,
      "mean_token_accuracy": 0.22813682556152343,
      "num_tokens": 121989368.0,
      "step": 65430
    },
    {
      "entropy": 6.0959876537322994,
      "epoch": 5.778435181914518,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002230767014008977,
      "loss": 5.2177,
      "mean_token_accuracy": 0.22422831356525422,
      "num_tokens": 121997844.0,
      "step": 65435
    },
    {
      "entropy": 6.0360020160675045,
      "epoch": 5.778876722006358,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022304606041342707,
      "loss": 5.1863,
      "mean_token_accuracy": 0.23401260375976562,
      "num_tokens": 122006890.0,
      "step": 65440
    },
    {
      "entropy": 6.025902223587036,
      "epoch": 5.779318262098198,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022301542044370466,
      "loss": 5.1581,
      "mean_token_accuracy": 0.22589788883924483,
      "num_tokens": 122015179.0,
      "step": 65445
    },
    {
      "entropy": 6.028558158874512,
      "epoch": 5.779759802190039,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022298478149233048,
      "loss": 5.2073,
      "mean_token_accuracy": 0.22611669898033143,
      "num_tokens": 122024147.0,
      "step": 65450
    },
    {
      "entropy": 5.960100173950195,
      "epoch": 5.780201342281879,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022295414355990477,
      "loss": 5.1875,
      "mean_token_accuracy": 0.23193652033805848,
      "num_tokens": 122033239.0,
      "step": 65455
    },
    {
      "entropy": 6.0662048816680905,
      "epoch": 5.780642882373719,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00022292350664702794,
      "loss": 5.2552,
      "mean_token_accuracy": 0.218912073969841,
      "num_tokens": 122042568.0,
      "step": 65460
    },
    {
      "entropy": 6.009271192550659,
      "epoch": 5.78108442246556,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022289287075429992,
      "loss": 5.1175,
      "mean_token_accuracy": 0.23355114609003066,
      "num_tokens": 122051955.0,
      "step": 65465
    },
    {
      "entropy": 6.192168188095093,
      "epoch": 5.7815259625574,
      "grad_norm": 1.375,
      "learning_rate": 0.00022286223588232102,
      "loss": 5.3522,
      "mean_token_accuracy": 0.21070128083229064,
      "num_tokens": 122060742.0,
      "step": 65470
    },
    {
      "entropy": 6.065824222564697,
      "epoch": 5.78196750264924,
      "grad_norm": 1.4375,
      "learning_rate": 0.00022283160203169112,
      "loss": 5.1786,
      "mean_token_accuracy": 0.22376721650362014,
      "num_tokens": 122069614.0,
      "step": 65475
    },
    {
      "entropy": 6.0310547828674315,
      "epoch": 5.782409042741081,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022280096920301057,
      "loss": 5.2121,
      "mean_token_accuracy": 0.22569530606269836,
      "num_tokens": 122079242.0,
      "step": 65480
    },
    {
      "entropy": 5.994255781173706,
      "epoch": 5.782850582832921,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022277033739687935,
      "loss": 5.1562,
      "mean_token_accuracy": 0.23374653458595276,
      "num_tokens": 122088820.0,
      "step": 65485
    },
    {
      "entropy": 6.029172801971436,
      "epoch": 5.783292122924761,
      "grad_norm": 1.25,
      "learning_rate": 0.00022273970661389748,
      "loss": 5.26,
      "mean_token_accuracy": 0.21870932430028917,
      "num_tokens": 122098918.0,
      "step": 65490
    },
    {
      "entropy": 5.943043994903564,
      "epoch": 5.783733663016601,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00022270907685466507,
      "loss": 5.0939,
      "mean_token_accuracy": 0.23774440884590148,
      "num_tokens": 122107985.0,
      "step": 65495
    },
    {
      "entropy": 6.019787693023682,
      "epoch": 5.784175203108442,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000222678448119782,
      "loss": 5.2511,
      "mean_token_accuracy": 0.2189606711268425,
      "num_tokens": 122117082.0,
      "step": 65500
    },
    {
      "entropy": 6.063082122802735,
      "epoch": 5.784616743200282,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022264782040984848,
      "loss": 5.1564,
      "mean_token_accuracy": 0.2307637080550194,
      "num_tokens": 122126558.0,
      "step": 65505
    },
    {
      "entropy": 6.008072137832642,
      "epoch": 5.785058283292123,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002226171937254644,
      "loss": 5.1678,
      "mean_token_accuracy": 0.22854497879743577,
      "num_tokens": 122135441.0,
      "step": 65510
    },
    {
      "entropy": 6.013346529006958,
      "epoch": 5.785499823383963,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022258656806722966,
      "loss": 5.2337,
      "mean_token_accuracy": 0.2319600388407707,
      "num_tokens": 122145106.0,
      "step": 65515
    },
    {
      "entropy": 5.988863515853882,
      "epoch": 5.7859413634758035,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002225559434357442,
      "loss": 5.1869,
      "mean_token_accuracy": 0.23680416196584703,
      "num_tokens": 122155855.0,
      "step": 65520
    },
    {
      "entropy": 6.038463163375854,
      "epoch": 5.7863829035676435,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000222525319831608,
      "loss": 5.2228,
      "mean_token_accuracy": 0.2230675473809242,
      "num_tokens": 122164997.0,
      "step": 65525
    },
    {
      "entropy": 6.065632295608521,
      "epoch": 5.7868244436594845,
      "grad_norm": 1.296875,
      "learning_rate": 0.000222494697255421,
      "loss": 5.2679,
      "mean_token_accuracy": 0.22589389383792877,
      "num_tokens": 122174800.0,
      "step": 65530
    },
    {
      "entropy": 6.023795557022095,
      "epoch": 5.7872659837513245,
      "grad_norm": 1.515625,
      "learning_rate": 0.000222464075707783,
      "loss": 5.1267,
      "mean_token_accuracy": 0.23670027852058412,
      "num_tokens": 122184456.0,
      "step": 65535
    },
    {
      "entropy": 6.066815280914307,
      "epoch": 5.7877075238431654,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002224334551892938,
      "loss": 5.2354,
      "mean_token_accuracy": 0.22513245940208435,
      "num_tokens": 122194780.0,
      "step": 65540
    },
    {
      "entropy": 5.969586038589478,
      "epoch": 5.7881490639350055,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022240283570055337,
      "loss": 5.1954,
      "mean_token_accuracy": 0.22928360998630523,
      "num_tokens": 122206527.0,
      "step": 65545
    },
    {
      "entropy": 6.076727724075317,
      "epoch": 5.7885906040268456,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002223722172421615,
      "loss": 5.2318,
      "mean_token_accuracy": 0.22701606899499893,
      "num_tokens": 122215491.0,
      "step": 65550
    },
    {
      "entropy": 6.015161180496216,
      "epoch": 5.789032144118686,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022234159981471803,
      "loss": 5.1745,
      "mean_token_accuracy": 0.22791612446308135,
      "num_tokens": 122225786.0,
      "step": 65555
    },
    {
      "entropy": 6.0663234233856205,
      "epoch": 5.7894736842105265,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022231098341882256,
      "loss": 5.236,
      "mean_token_accuracy": 0.22160804122686387,
      "num_tokens": 122234482.0,
      "step": 65560
    },
    {
      "entropy": 6.077876091003418,
      "epoch": 5.789915224302367,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002222803680550749,
      "loss": 5.2587,
      "mean_token_accuracy": 0.21715339571237563,
      "num_tokens": 122243821.0,
      "step": 65565
    },
    {
      "entropy": 6.1317219734191895,
      "epoch": 5.790356764394207,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000222249753724075,
      "loss": 5.2407,
      "mean_token_accuracy": 0.22085647135972977,
      "num_tokens": 122252682.0,
      "step": 65570
    },
    {
      "entropy": 5.993988132476806,
      "epoch": 5.790798304486048,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022221914042642237,
      "loss": 5.1138,
      "mean_token_accuracy": 0.23889248073101044,
      "num_tokens": 122261608.0,
      "step": 65575
    },
    {
      "entropy": 6.048919439315796,
      "epoch": 5.791239844577888,
      "grad_norm": 1.4375,
      "learning_rate": 0.00022218852816271678,
      "loss": 5.2099,
      "mean_token_accuracy": 0.2212964877486229,
      "num_tokens": 122271073.0,
      "step": 65580
    },
    {
      "entropy": 6.082578420639038,
      "epoch": 5.791681384669728,
      "grad_norm": 1.5,
      "learning_rate": 0.0002221579169335578,
      "loss": 5.2459,
      "mean_token_accuracy": 0.22278513312339782,
      "num_tokens": 122279709.0,
      "step": 65585
    },
    {
      "entropy": 6.016443824768066,
      "epoch": 5.792122924761569,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022212730673954522,
      "loss": 5.1722,
      "mean_token_accuracy": 0.23264989554882048,
      "num_tokens": 122289748.0,
      "step": 65590
    },
    {
      "entropy": 5.931319665908814,
      "epoch": 5.792564464853409,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022209669758127865,
      "loss": 5.0143,
      "mean_token_accuracy": 0.24819647520780563,
      "num_tokens": 122299300.0,
      "step": 65595
    },
    {
      "entropy": 6.012723112106324,
      "epoch": 5.793006004945249,
      "grad_norm": 1.375,
      "learning_rate": 0.00022206608945935768,
      "loss": 5.2091,
      "mean_token_accuracy": 0.2308603808283806,
      "num_tokens": 122308299.0,
      "step": 65600
    },
    {
      "entropy": 6.041140460968018,
      "epoch": 5.79344754503709,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022203548237438193,
      "loss": 5.1892,
      "mean_token_accuracy": 0.22450470328330993,
      "num_tokens": 122317357.0,
      "step": 65605
    },
    {
      "entropy": 6.073525333404541,
      "epoch": 5.79388908512893,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022200487632695088,
      "loss": 5.2601,
      "mean_token_accuracy": 0.22194407731294633,
      "num_tokens": 122327426.0,
      "step": 65610
    },
    {
      "entropy": 6.07096381187439,
      "epoch": 5.79433062522077,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00022197427131766424,
      "loss": 5.2677,
      "mean_token_accuracy": 0.2172692060470581,
      "num_tokens": 122336086.0,
      "step": 65615
    },
    {
      "entropy": 6.017029237747193,
      "epoch": 5.794772165312611,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00022194366734712146,
      "loss": 5.2018,
      "mean_token_accuracy": 0.23243207037448882,
      "num_tokens": 122345687.0,
      "step": 65620
    },
    {
      "entropy": 6.023472023010254,
      "epoch": 5.795213705404451,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022191306441592207,
      "loss": 5.1858,
      "mean_token_accuracy": 0.22716357707977294,
      "num_tokens": 122354927.0,
      "step": 65625
    },
    {
      "entropy": 6.058358478546142,
      "epoch": 5.795655245496291,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00022188246252466547,
      "loss": 5.0752,
      "mean_token_accuracy": 0.24111326932907104,
      "num_tokens": 122362652.0,
      "step": 65630
    },
    {
      "entropy": 5.951130962371826,
      "epoch": 5.796096785588132,
      "grad_norm": 1.515625,
      "learning_rate": 0.00022185186167395134,
      "loss": 5.155,
      "mean_token_accuracy": 0.22426010370254518,
      "num_tokens": 122371442.0,
      "step": 65635
    },
    {
      "entropy": 6.04016342163086,
      "epoch": 5.796538325679972,
      "grad_norm": 1.4375,
      "learning_rate": 0.00022182126186437895,
      "loss": 5.3196,
      "mean_token_accuracy": 0.2098372295498848,
      "num_tokens": 122381674.0,
      "step": 65640
    },
    {
      "entropy": 6.086938858032227,
      "epoch": 5.796979865771812,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002217906630965479,
      "loss": 5.2164,
      "mean_token_accuracy": 0.2218410700559616,
      "num_tokens": 122391909.0,
      "step": 65645
    },
    {
      "entropy": 6.123226690292358,
      "epoch": 5.797421405863652,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002217600653710573,
      "loss": 5.2867,
      "mean_token_accuracy": 0.21994948536157607,
      "num_tokens": 122400937.0,
      "step": 65650
    },
    {
      "entropy": 6.077358293533325,
      "epoch": 5.797862945955493,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022172946868850686,
      "loss": 5.232,
      "mean_token_accuracy": 0.2242426335811615,
      "num_tokens": 122410016.0,
      "step": 65655
    },
    {
      "entropy": 6.107325172424316,
      "epoch": 5.798304486047333,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022169887304949575,
      "loss": 5.2534,
      "mean_token_accuracy": 0.2222396418452263,
      "num_tokens": 122419415.0,
      "step": 65660
    },
    {
      "entropy": 5.9848250389099125,
      "epoch": 5.798746026139174,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022166827845462351,
      "loss": 5.1632,
      "mean_token_accuracy": 0.2351755380630493,
      "num_tokens": 122428630.0,
      "step": 65665
    },
    {
      "entropy": 5.985037803649902,
      "epoch": 5.799187566231014,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022163768490448938,
      "loss": 5.1791,
      "mean_token_accuracy": 0.22525254786014556,
      "num_tokens": 122437711.0,
      "step": 65670
    },
    {
      "entropy": 6.038557338714599,
      "epoch": 5.799629106322854,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022160709239969252,
      "loss": 5.1789,
      "mean_token_accuracy": 0.2269619345664978,
      "num_tokens": 122447380.0,
      "step": 65675
    },
    {
      "entropy": 6.083459043502808,
      "epoch": 5.800070646414694,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022157650094083248,
      "loss": 5.2154,
      "mean_token_accuracy": 0.22745247930288315,
      "num_tokens": 122456536.0,
      "step": 65680
    },
    {
      "entropy": 5.945775651931763,
      "epoch": 5.800512186506535,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002215459105285083,
      "loss": 5.1259,
      "mean_token_accuracy": 0.23846030235290527,
      "num_tokens": 122466608.0,
      "step": 65685
    },
    {
      "entropy": 6.040573453903198,
      "epoch": 5.800953726598375,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00022151532116331947,
      "loss": 5.1613,
      "mean_token_accuracy": 0.2258485108613968,
      "num_tokens": 122475833.0,
      "step": 65690
    },
    {
      "entropy": 6.070053195953369,
      "epoch": 5.801395266690215,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002214847328458649,
      "loss": 5.2581,
      "mean_token_accuracy": 0.2259420558810234,
      "num_tokens": 122485221.0,
      "step": 65695
    },
    {
      "entropy": 6.025947427749633,
      "epoch": 5.801836806782056,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002214541455767441,
      "loss": 5.1855,
      "mean_token_accuracy": 0.22797148376703263,
      "num_tokens": 122494277.0,
      "step": 65700
    },
    {
      "entropy": 5.93333592414856,
      "epoch": 5.802278346873896,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002214235593565561,
      "loss": 5.1146,
      "mean_token_accuracy": 0.23908071368932723,
      "num_tokens": 122503172.0,
      "step": 65705
    },
    {
      "entropy": 6.023181056976318,
      "epoch": 5.802719886965736,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022139297418590016,
      "loss": 5.1999,
      "mean_token_accuracy": 0.22860633283853532,
      "num_tokens": 122512107.0,
      "step": 65710
    },
    {
      "entropy": 6.031009006500244,
      "epoch": 5.803161427057577,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022136239006537522,
      "loss": 5.2033,
      "mean_token_accuracy": 0.2295200228691101,
      "num_tokens": 122521586.0,
      "step": 65715
    },
    {
      "entropy": 6.082115507125854,
      "epoch": 5.803602967149417,
      "grad_norm": 1.375,
      "learning_rate": 0.0002213318069955807,
      "loss": 5.2974,
      "mean_token_accuracy": 0.21291587203741075,
      "num_tokens": 122530738.0,
      "step": 65720
    },
    {
      "entropy": 6.084195518493653,
      "epoch": 5.804044507241257,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022130122497711547,
      "loss": 5.246,
      "mean_token_accuracy": 0.22302550375461577,
      "num_tokens": 122539688.0,
      "step": 65725
    },
    {
      "entropy": 6.012630033493042,
      "epoch": 5.804486047333098,
      "grad_norm": 1.5,
      "learning_rate": 0.00022127064401057878,
      "loss": 5.195,
      "mean_token_accuracy": 0.22410021126270294,
      "num_tokens": 122548844.0,
      "step": 65730
    },
    {
      "entropy": 6.0516222476959225,
      "epoch": 5.804927587424938,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002212400640965696,
      "loss": 5.2924,
      "mean_token_accuracy": 0.22270189821720124,
      "num_tokens": 122558302.0,
      "step": 65735
    },
    {
      "entropy": 6.116660118103027,
      "epoch": 5.805369127516778,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002212094852356869,
      "loss": 5.2821,
      "mean_token_accuracy": 0.2209528461098671,
      "num_tokens": 122567353.0,
      "step": 65740
    },
    {
      "entropy": 6.0353821277618405,
      "epoch": 5.805810667608619,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022117890742852988,
      "loss": 5.2705,
      "mean_token_accuracy": 0.22087520062923433,
      "num_tokens": 122576133.0,
      "step": 65745
    },
    {
      "entropy": 6.13278112411499,
      "epoch": 5.806252207700459,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00022114833067569744,
      "loss": 5.2868,
      "mean_token_accuracy": 0.21577527672052382,
      "num_tokens": 122585327.0,
      "step": 65750
    },
    {
      "entropy": 6.092634487152099,
      "epoch": 5.806693747792299,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00022111775497778862,
      "loss": 5.2046,
      "mean_token_accuracy": 0.2314775288105011,
      "num_tokens": 122594472.0,
      "step": 65755
    },
    {
      "entropy": 6.122159194946289,
      "epoch": 5.80713528788414,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002210871803354022,
      "loss": 5.2961,
      "mean_token_accuracy": 0.21685751676559448,
      "num_tokens": 122604768.0,
      "step": 65760
    },
    {
      "entropy": 5.999444007873535,
      "epoch": 5.80757682797598,
      "grad_norm": 1.375,
      "learning_rate": 0.00022105660674913736,
      "loss": 5.2041,
      "mean_token_accuracy": 0.22423696964979173,
      "num_tokens": 122613692.0,
      "step": 65765
    },
    {
      "entropy": 5.952839326858521,
      "epoch": 5.80801836806782,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022102603421959293,
      "loss": 5.1516,
      "mean_token_accuracy": 0.23451646864414216,
      "num_tokens": 122622838.0,
      "step": 65770
    },
    {
      "entropy": 6.084170484542847,
      "epoch": 5.80845990815966,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022099546274736775,
      "loss": 5.2934,
      "mean_token_accuracy": 0.2155238062143326,
      "num_tokens": 122633007.0,
      "step": 65775
    },
    {
      "entropy": 6.064469861984253,
      "epoch": 5.808901448251501,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022096489233306072,
      "loss": 5.2199,
      "mean_token_accuracy": 0.22401026785373687,
      "num_tokens": 122641791.0,
      "step": 65780
    },
    {
      "entropy": 6.001987504959106,
      "epoch": 5.809342988343341,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00022093432297727074,
      "loss": 5.1821,
      "mean_token_accuracy": 0.23356907516717912,
      "num_tokens": 122651996.0,
      "step": 65785
    },
    {
      "entropy": 6.026666069030762,
      "epoch": 5.809784528435182,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002209037546805966,
      "loss": 5.2632,
      "mean_token_accuracy": 0.21624922454357148,
      "num_tokens": 122661521.0,
      "step": 65790
    },
    {
      "entropy": 6.108872985839843,
      "epoch": 5.810226068527022,
      "grad_norm": 1.375,
      "learning_rate": 0.0002208731874436372,
      "loss": 5.2633,
      "mean_token_accuracy": 0.22312331646680833,
      "num_tokens": 122670567.0,
      "step": 65795
    },
    {
      "entropy": 6.08818097114563,
      "epoch": 5.8106676086188624,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022084262126699123,
      "loss": 5.2828,
      "mean_token_accuracy": 0.2215433105826378,
      "num_tokens": 122679805.0,
      "step": 65800
    },
    {
      "entropy": 6.009017372131348,
      "epoch": 5.8111091487107025,
      "grad_norm": 1.375,
      "learning_rate": 0.0002208120561512575,
      "loss": 5.1052,
      "mean_token_accuracy": 0.24078028053045272,
      "num_tokens": 122689257.0,
      "step": 65805
    },
    {
      "entropy": 6.032030534744263,
      "epoch": 5.811550688802543,
      "grad_norm": 1.375,
      "learning_rate": 0.0002207814920970348,
      "loss": 5.1617,
      "mean_token_accuracy": 0.22906910479068757,
      "num_tokens": 122698534.0,
      "step": 65810
    },
    {
      "entropy": 6.013185024261475,
      "epoch": 5.8119922288943835,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022075092910492185,
      "loss": 5.2426,
      "mean_token_accuracy": 0.22673576325178146,
      "num_tokens": 122708122.0,
      "step": 65815
    },
    {
      "entropy": 6.109366703033447,
      "epoch": 5.812433768986224,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022072036717551735,
      "loss": 5.3486,
      "mean_token_accuracy": 0.21177763342857361,
      "num_tokens": 122717743.0,
      "step": 65820
    },
    {
      "entropy": 6.04230375289917,
      "epoch": 5.8128753090780645,
      "grad_norm": 2.703125,
      "learning_rate": 0.00022068980630941988,
      "loss": 5.1781,
      "mean_token_accuracy": 0.22964750230312347,
      "num_tokens": 122726331.0,
      "step": 65825
    },
    {
      "entropy": 6.111379861831665,
      "epoch": 5.8133168491699045,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002206592465072283,
      "loss": 5.2538,
      "mean_token_accuracy": 0.21977420747280121,
      "num_tokens": 122735919.0,
      "step": 65830
    },
    {
      "entropy": 6.078382205963135,
      "epoch": 5.813758389261745,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002206286877695412,
      "loss": 5.2079,
      "mean_token_accuracy": 0.22499728202819824,
      "num_tokens": 122744596.0,
      "step": 65835
    },
    {
      "entropy": 6.075895214080811,
      "epoch": 5.8141999293535855,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022059813009695718,
      "loss": 5.2345,
      "mean_token_accuracy": 0.22206606715917587,
      "num_tokens": 122753153.0,
      "step": 65840
    },
    {
      "entropy": 6.055168676376343,
      "epoch": 5.814641469445426,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022056757349007478,
      "loss": 5.2088,
      "mean_token_accuracy": 0.2233305126428604,
      "num_tokens": 122761663.0,
      "step": 65845
    },
    {
      "entropy": 6.030634832382202,
      "epoch": 5.815083009537266,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002205370179494927,
      "loss": 5.1732,
      "mean_token_accuracy": 0.2246103122830391,
      "num_tokens": 122770165.0,
      "step": 65850
    },
    {
      "entropy": 6.102433872222901,
      "epoch": 5.815524549629107,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022050646347580944,
      "loss": 5.3322,
      "mean_token_accuracy": 0.2154688537120819,
      "num_tokens": 122780193.0,
      "step": 65855
    },
    {
      "entropy": 6.0582273483276365,
      "epoch": 5.815966089720947,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022047591006962364,
      "loss": 5.1697,
      "mean_token_accuracy": 0.22783711105585097,
      "num_tokens": 122789222.0,
      "step": 65860
    },
    {
      "entropy": 6.05830192565918,
      "epoch": 5.816407629812787,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00022044535773153373,
      "loss": 5.1784,
      "mean_token_accuracy": 0.23426368534564973,
      "num_tokens": 122797568.0,
      "step": 65865
    },
    {
      "entropy": 6.096519804000854,
      "epoch": 5.816849169904628,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022041480646213814,
      "loss": 5.2261,
      "mean_token_accuracy": 0.22660828083753587,
      "num_tokens": 122806290.0,
      "step": 65870
    },
    {
      "entropy": 5.946173286437988,
      "epoch": 5.817290709996468,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022038425626203555,
      "loss": 5.1202,
      "mean_token_accuracy": 0.23788751661777496,
      "num_tokens": 122815607.0,
      "step": 65875
    },
    {
      "entropy": 6.016677045822144,
      "epoch": 5.817732250088308,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022035370713182435,
      "loss": 5.146,
      "mean_token_accuracy": 0.2285651832818985,
      "num_tokens": 122824306.0,
      "step": 65880
    },
    {
      "entropy": 6.047721433639526,
      "epoch": 5.818173790180149,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022032315907210287,
      "loss": 5.2002,
      "mean_token_accuracy": 0.22969649434089662,
      "num_tokens": 122833435.0,
      "step": 65885
    },
    {
      "entropy": 6.1230788230896,
      "epoch": 5.818615330271989,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022029261208346957,
      "loss": 5.3547,
      "mean_token_accuracy": 0.21427324563264846,
      "num_tokens": 122843002.0,
      "step": 65890
    },
    {
      "entropy": 6.066595602035522,
      "epoch": 5.819056870363829,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022026206616652296,
      "loss": 5.1849,
      "mean_token_accuracy": 0.2265809565782547,
      "num_tokens": 122852762.0,
      "step": 65895
    },
    {
      "entropy": 6.0171441555023195,
      "epoch": 5.81949841045567,
      "grad_norm": 1.515625,
      "learning_rate": 0.00022023152132186132,
      "loss": 5.2058,
      "mean_token_accuracy": 0.22878989428281785,
      "num_tokens": 122862927.0,
      "step": 65900
    },
    {
      "entropy": 5.917736577987671,
      "epoch": 5.81993995054751,
      "grad_norm": 1.390625,
      "learning_rate": 0.000220200977550083,
      "loss": 5.1138,
      "mean_token_accuracy": 0.24125254899263382,
      "num_tokens": 122872328.0,
      "step": 65905
    },
    {
      "entropy": 6.120899772644043,
      "epoch": 5.82038149063935,
      "grad_norm": 1.546875,
      "learning_rate": 0.00022017043485178644,
      "loss": 5.1997,
      "mean_token_accuracy": 0.23526770919561385,
      "num_tokens": 122881336.0,
      "step": 65910
    },
    {
      "entropy": 6.026816558837891,
      "epoch": 5.820823030731191,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022013989322756976,
      "loss": 5.1962,
      "mean_token_accuracy": 0.22462122589349748,
      "num_tokens": 122890568.0,
      "step": 65915
    },
    {
      "entropy": 6.053133726119995,
      "epoch": 5.821264570823031,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022010935267803146,
      "loss": 5.1885,
      "mean_token_accuracy": 0.23325590342283248,
      "num_tokens": 122900865.0,
      "step": 65920
    },
    {
      "entropy": 6.074100065231323,
      "epoch": 5.821706110914871,
      "grad_norm": 1.375,
      "learning_rate": 0.00022007881320376977,
      "loss": 5.2724,
      "mean_token_accuracy": 0.2227628469467163,
      "num_tokens": 122909713.0,
      "step": 65925
    },
    {
      "entropy": 6.014522504806519,
      "epoch": 5.822147651006711,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002200482748053828,
      "loss": 5.1934,
      "mean_token_accuracy": 0.2205938220024109,
      "num_tokens": 122918261.0,
      "step": 65930
    },
    {
      "entropy": 6.067411851882935,
      "epoch": 5.822589191098552,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022001773748346888,
      "loss": 5.3165,
      "mean_token_accuracy": 0.21473681181669235,
      "num_tokens": 122928108.0,
      "step": 65935
    },
    {
      "entropy": 6.055249929428101,
      "epoch": 5.823030731190392,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00021998720123862626,
      "loss": 5.1848,
      "mean_token_accuracy": 0.2283680856227875,
      "num_tokens": 122937758.0,
      "step": 65940
    },
    {
      "entropy": 6.059260034561158,
      "epoch": 5.823472271282233,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021995666607145303,
      "loss": 5.1583,
      "mean_token_accuracy": 0.23201967030763626,
      "num_tokens": 122946892.0,
      "step": 65945
    },
    {
      "entropy": 6.006383323669434,
      "epoch": 5.823913811374073,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00021992613198254746,
      "loss": 5.1106,
      "mean_token_accuracy": 0.23245998322963715,
      "num_tokens": 122956169.0,
      "step": 65950
    },
    {
      "entropy": 6.011411046981811,
      "epoch": 5.824355351465913,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021989559897250761,
      "loss": 5.1991,
      "mean_token_accuracy": 0.2259521320462227,
      "num_tokens": 122965725.0,
      "step": 65955
    },
    {
      "entropy": 6.027888393402099,
      "epoch": 5.824796891557753,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021986506704193165,
      "loss": 5.2237,
      "mean_token_accuracy": 0.22401978075504303,
      "num_tokens": 122975458.0,
      "step": 65960
    },
    {
      "entropy": 6.069184017181397,
      "epoch": 5.825238431649594,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021983453619141768,
      "loss": 5.229,
      "mean_token_accuracy": 0.2266518920660019,
      "num_tokens": 122985691.0,
      "step": 65965
    },
    {
      "entropy": 6.053991651535034,
      "epoch": 5.825679971741434,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021980400642156373,
      "loss": 5.2157,
      "mean_token_accuracy": 0.22287560403347015,
      "num_tokens": 122993714.0,
      "step": 65970
    },
    {
      "entropy": 6.046225547790527,
      "epoch": 5.826121511833274,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021977347773296797,
      "loss": 5.1713,
      "mean_token_accuracy": 0.23185481280088424,
      "num_tokens": 123002968.0,
      "step": 65975
    },
    {
      "entropy": 6.028983640670776,
      "epoch": 5.826563051925115,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002197429501262283,
      "loss": 5.1798,
      "mean_token_accuracy": 0.22595978528261185,
      "num_tokens": 123012158.0,
      "step": 65980
    },
    {
      "entropy": 6.029480981826782,
      "epoch": 5.827004592016955,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00021971242360194277,
      "loss": 5.1862,
      "mean_token_accuracy": 0.22936026602983475,
      "num_tokens": 123021915.0,
      "step": 65985
    },
    {
      "entropy": 6.019189262390137,
      "epoch": 5.827446132108795,
      "grad_norm": 1.515625,
      "learning_rate": 0.00021968189816070954,
      "loss": 5.17,
      "mean_token_accuracy": 0.22914963066577912,
      "num_tokens": 123031263.0,
      "step": 65990
    },
    {
      "entropy": 6.065674066543579,
      "epoch": 5.827887672200636,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021965137380312643,
      "loss": 5.3155,
      "mean_token_accuracy": 0.21452075988054276,
      "num_tokens": 123040057.0,
      "step": 65995
    },
    {
      "entropy": 6.125188589096069,
      "epoch": 5.828329212292476,
      "grad_norm": 1.578125,
      "learning_rate": 0.00021962085052979136,
      "loss": 5.3131,
      "mean_token_accuracy": 0.21379150897264482,
      "num_tokens": 123050338.0,
      "step": 66000
    },
    {
      "epoch": 5.828329212292476,
      "eval_entropy": 5.783321963296201,
      "eval_loss": 5.292881965637207,
      "eval_mean_token_accuracy": 0.2298081892938063,
      "eval_num_tokens": 123050338.0,
      "eval_runtime": 20.2328,
      "eval_samples_per_second": 1745.43,
      "eval_steps_per_second": 218.21,
      "step": 66000
    },
    {
      "entropy": 6.0357976913452145,
      "epoch": 5.828770752384316,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002195903283413024,
      "loss": 5.1653,
      "mean_token_accuracy": 0.23384419232606887,
      "num_tokens": 123059201.0,
      "step": 66005
    },
    {
      "entropy": 6.162852239608765,
      "epoch": 5.829212292476157,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021955980723825742,
      "loss": 5.2972,
      "mean_token_accuracy": 0.22074667662382125,
      "num_tokens": 123069040.0,
      "step": 66010
    },
    {
      "entropy": 6.094892835617065,
      "epoch": 5.829653832567997,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021952928722125426,
      "loss": 5.262,
      "mean_token_accuracy": 0.22137294262647628,
      "num_tokens": 123077647.0,
      "step": 66015
    },
    {
      "entropy": 5.959438133239746,
      "epoch": 5.830095372659837,
      "grad_norm": 1.28125,
      "learning_rate": 0.00021949876829089072,
      "loss": 5.1382,
      "mean_token_accuracy": 0.23985061347484588,
      "num_tokens": 123087205.0,
      "step": 66020
    },
    {
      "entropy": 6.051990699768067,
      "epoch": 5.830536912751678,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021946825044776486,
      "loss": 5.2164,
      "mean_token_accuracy": 0.21847254931926727,
      "num_tokens": 123096932.0,
      "step": 66025
    },
    {
      "entropy": 6.088017749786377,
      "epoch": 5.830978452843518,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021943773369247438,
      "loss": 5.3015,
      "mean_token_accuracy": 0.21277981251478195,
      "num_tokens": 123106959.0,
      "step": 66030
    },
    {
      "entropy": 6.011731719970703,
      "epoch": 5.831419992935358,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00021940721802561703,
      "loss": 5.0614,
      "mean_token_accuracy": 0.22908889204263688,
      "num_tokens": 123115047.0,
      "step": 66035
    },
    {
      "entropy": 6.063280820846558,
      "epoch": 5.831861533027199,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021937670344779076,
      "loss": 5.245,
      "mean_token_accuracy": 0.21920513808727266,
      "num_tokens": 123124297.0,
      "step": 66040
    },
    {
      "entropy": 6.058444213867188,
      "epoch": 5.832303073119039,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002193461899595931,
      "loss": 5.2678,
      "mean_token_accuracy": 0.22309852093458177,
      "num_tokens": 123134452.0,
      "step": 66045
    },
    {
      "entropy": 6.040267610549927,
      "epoch": 5.832744613210879,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00021931567756162203,
      "loss": 5.2672,
      "mean_token_accuracy": 0.2217488929629326,
      "num_tokens": 123143662.0,
      "step": 66050
    },
    {
      "entropy": 6.071365880966186,
      "epoch": 5.833186153302719,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021928516625447514,
      "loss": 5.3385,
      "mean_token_accuracy": 0.21165213435888292,
      "num_tokens": 123153451.0,
      "step": 66055
    },
    {
      "entropy": 6.066800975799561,
      "epoch": 5.83362769339456,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021925465603875018,
      "loss": 5.1951,
      "mean_token_accuracy": 0.23064442574977875,
      "num_tokens": 123163534.0,
      "step": 66060
    },
    {
      "entropy": 6.118885946273804,
      "epoch": 5.8340692334864,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00021922414691504473,
      "loss": 5.311,
      "mean_token_accuracy": 0.2159489870071411,
      "num_tokens": 123171873.0,
      "step": 66065
    },
    {
      "entropy": 6.081696319580078,
      "epoch": 5.834510773578241,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002191936388839566,
      "loss": 5.2999,
      "mean_token_accuracy": 0.21814502477645875,
      "num_tokens": 123182295.0,
      "step": 66070
    },
    {
      "entropy": 6.000941944122315,
      "epoch": 5.834952313670081,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021916313194608335,
      "loss": 5.1685,
      "mean_token_accuracy": 0.22966379523277283,
      "num_tokens": 123191655.0,
      "step": 66075
    },
    {
      "entropy": 6.063952970504761,
      "epoch": 5.835393853761921,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021913262610202255,
      "loss": 5.2084,
      "mean_token_accuracy": 0.22992117404937745,
      "num_tokens": 123200582.0,
      "step": 66080
    },
    {
      "entropy": 6.048808908462524,
      "epoch": 5.8358353938537615,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002191021213523719,
      "loss": 5.2822,
      "mean_token_accuracy": 0.22258710265159606,
      "num_tokens": 123210093.0,
      "step": 66085
    },
    {
      "entropy": 6.028319883346557,
      "epoch": 5.836276933945602,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002190716176977287,
      "loss": 5.1459,
      "mean_token_accuracy": 0.24059638977050782,
      "num_tokens": 123218942.0,
      "step": 66090
    },
    {
      "entropy": 6.053065061569214,
      "epoch": 5.8367184740374425,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002190411151386909,
      "loss": 5.29,
      "mean_token_accuracy": 0.22198963165283203,
      "num_tokens": 123228662.0,
      "step": 66095
    },
    {
      "entropy": 6.019006872177124,
      "epoch": 5.837160014129283,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021901061367585574,
      "loss": 5.1844,
      "mean_token_accuracy": 0.22741109728813172,
      "num_tokens": 123238567.0,
      "step": 66100
    },
    {
      "entropy": 6.042810583114624,
      "epoch": 5.8376015542211235,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021898011330982088,
      "loss": 5.1674,
      "mean_token_accuracy": 0.22445442527532578,
      "num_tokens": 123246877.0,
      "step": 66105
    },
    {
      "entropy": 6.035333585739136,
      "epoch": 5.8380430943129635,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002189496140411837,
      "loss": 5.1903,
      "mean_token_accuracy": 0.23166944682598115,
      "num_tokens": 123256381.0,
      "step": 66110
    },
    {
      "entropy": 6.06522307395935,
      "epoch": 5.838484634404804,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021891911587054163,
      "loss": 5.275,
      "mean_token_accuracy": 0.2168948844075203,
      "num_tokens": 123266184.0,
      "step": 66115
    },
    {
      "entropy": 5.984441328048706,
      "epoch": 5.8389261744966445,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002188886187984923,
      "loss": 5.1223,
      "mean_token_accuracy": 0.23421044498682023,
      "num_tokens": 123275396.0,
      "step": 66120
    },
    {
      "entropy": 6.061817264556884,
      "epoch": 5.839367714588485,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00021885812282563304,
      "loss": 5.198,
      "mean_token_accuracy": 0.22253057807683946,
      "num_tokens": 123284716.0,
      "step": 66125
    },
    {
      "entropy": 6.103953886032104,
      "epoch": 5.839809254680325,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021882762795256107,
      "loss": 5.2271,
      "mean_token_accuracy": 0.22594136297702788,
      "num_tokens": 123293734.0,
      "step": 66130
    },
    {
      "entropy": 5.97386474609375,
      "epoch": 5.840250794772166,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000218797134179874,
      "loss": 5.1685,
      "mean_token_accuracy": 0.23325068801641463,
      "num_tokens": 123303228.0,
      "step": 66135
    },
    {
      "entropy": 6.054828405380249,
      "epoch": 5.840692334864006,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00021876664150816922,
      "loss": 5.2696,
      "mean_token_accuracy": 0.21515384465456008,
      "num_tokens": 123311837.0,
      "step": 66140
    },
    {
      "entropy": 6.063238477706909,
      "epoch": 5.841133874955846,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021873614993804386,
      "loss": 5.2247,
      "mean_token_accuracy": 0.22575586438179016,
      "num_tokens": 123320635.0,
      "step": 66145
    },
    {
      "entropy": 6.013072395324707,
      "epoch": 5.841575415047687,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021870565947009529,
      "loss": 5.1607,
      "mean_token_accuracy": 0.23241674304008483,
      "num_tokens": 123330024.0,
      "step": 66150
    },
    {
      "entropy": 6.109173536300659,
      "epoch": 5.842016955139527,
      "grad_norm": 1.5625,
      "learning_rate": 0.00021867517010492083,
      "loss": 5.281,
      "mean_token_accuracy": 0.22221680134534835,
      "num_tokens": 123338535.0,
      "step": 66155
    },
    {
      "entropy": 6.042670345306396,
      "epoch": 5.842458495231367,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00021864468184311782,
      "loss": 5.2792,
      "mean_token_accuracy": 0.22114483118057252,
      "num_tokens": 123346686.0,
      "step": 66160
    },
    {
      "entropy": 6.103736019134521,
      "epoch": 5.842900035323208,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021861419468528336,
      "loss": 5.2613,
      "mean_token_accuracy": 0.21457065492868424,
      "num_tokens": 123356220.0,
      "step": 66165
    },
    {
      "entropy": 6.030423212051391,
      "epoch": 5.843341575415048,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021858370863201483,
      "loss": 5.2523,
      "mean_token_accuracy": 0.23112713098526,
      "num_tokens": 123367091.0,
      "step": 66170
    },
    {
      "entropy": 6.117994117736816,
      "epoch": 5.843783115506888,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021855322368390928,
      "loss": 5.3087,
      "mean_token_accuracy": 0.2182541996240616,
      "num_tokens": 123377129.0,
      "step": 66175
    },
    {
      "entropy": 6.082968759536743,
      "epoch": 5.844224655598729,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021852273984156406,
      "loss": 5.2042,
      "mean_token_accuracy": 0.22707828730344773,
      "num_tokens": 123386337.0,
      "step": 66180
    },
    {
      "entropy": 6.011449766159058,
      "epoch": 5.844666195690569,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021849225710557623,
      "loss": 5.276,
      "mean_token_accuracy": 0.22232780009508132,
      "num_tokens": 123397375.0,
      "step": 66185
    },
    {
      "entropy": 6.078810453414917,
      "epoch": 5.845107735782409,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021846177547654293,
      "loss": 5.2169,
      "mean_token_accuracy": 0.2263839915394783,
      "num_tokens": 123405739.0,
      "step": 66190
    },
    {
      "entropy": 6.048076581954956,
      "epoch": 5.84554927587425,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00021843129495506127,
      "loss": 5.172,
      "mean_token_accuracy": 0.23885810226202012,
      "num_tokens": 123415361.0,
      "step": 66195
    },
    {
      "entropy": 6.090105581283569,
      "epoch": 5.84599081596609,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002184008155417284,
      "loss": 5.3232,
      "mean_token_accuracy": 0.21209834814071654,
      "num_tokens": 123425060.0,
      "step": 66200
    },
    {
      "entropy": 6.006353664398193,
      "epoch": 5.84643235605793,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021837033723714138,
      "loss": 5.2643,
      "mean_token_accuracy": 0.22388163208961487,
      "num_tokens": 123435354.0,
      "step": 66205
    },
    {
      "entropy": 6.106024694442749,
      "epoch": 5.84687389614977,
      "grad_norm": 1.375,
      "learning_rate": 0.00021833986004189728,
      "loss": 5.2156,
      "mean_token_accuracy": 0.22820402383804322,
      "num_tokens": 123444450.0,
      "step": 66210
    },
    {
      "entropy": 6.1047811031341555,
      "epoch": 5.847315436241611,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00021830938395659305,
      "loss": 5.1849,
      "mean_token_accuracy": 0.23344203233718872,
      "num_tokens": 123452879.0,
      "step": 66215
    },
    {
      "entropy": 6.0516749858856205,
      "epoch": 5.847756976333451,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021827890898182568,
      "loss": 5.2349,
      "mean_token_accuracy": 0.2227979928255081,
      "num_tokens": 123461675.0,
      "step": 66220
    },
    {
      "entropy": 6.070217180252075,
      "epoch": 5.848198516425292,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002182484351181923,
      "loss": 5.2849,
      "mean_token_accuracy": 0.21754326820373535,
      "num_tokens": 123470357.0,
      "step": 66225
    },
    {
      "entropy": 6.05859694480896,
      "epoch": 5.848640056517132,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021821796236628975,
      "loss": 5.2572,
      "mean_token_accuracy": 0.22807794511318208,
      "num_tokens": 123479965.0,
      "step": 66230
    },
    {
      "entropy": 6.040908861160278,
      "epoch": 5.849081596608972,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021818749072671513,
      "loss": 5.2399,
      "mean_token_accuracy": 0.22578494846820832,
      "num_tokens": 123489404.0,
      "step": 66235
    },
    {
      "entropy": 6.001973056793213,
      "epoch": 5.849523136700812,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0002181570202000651,
      "loss": 5.176,
      "mean_token_accuracy": 0.22389434278011322,
      "num_tokens": 123498141.0,
      "step": 66240
    },
    {
      "entropy": 6.005120182037354,
      "epoch": 5.849964676792653,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00021812655078693683,
      "loss": 5.2184,
      "mean_token_accuracy": 0.23197101503610612,
      "num_tokens": 123507176.0,
      "step": 66245
    },
    {
      "entropy": 6.029007005691528,
      "epoch": 5.850406216884493,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00021809608248792706,
      "loss": 5.1974,
      "mean_token_accuracy": 0.2321378618478775,
      "num_tokens": 123516626.0,
      "step": 66250
    },
    {
      "entropy": 6.067557716369629,
      "epoch": 5.850847756976333,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021806561530363268,
      "loss": 5.2404,
      "mean_token_accuracy": 0.22378725707530975,
      "num_tokens": 123525789.0,
      "step": 66255
    },
    {
      "entropy": 6.046704387664795,
      "epoch": 5.851289297068174,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021803514923465046,
      "loss": 5.1905,
      "mean_token_accuracy": 0.22476017028093337,
      "num_tokens": 123536062.0,
      "step": 66260
    },
    {
      "entropy": 6.043331098556519,
      "epoch": 5.851730837160014,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021800468428157733,
      "loss": 5.2223,
      "mean_token_accuracy": 0.2212025746703148,
      "num_tokens": 123545050.0,
      "step": 66265
    },
    {
      "entropy": 6.08393874168396,
      "epoch": 5.852172377251854,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021797422044501002,
      "loss": 5.2552,
      "mean_token_accuracy": 0.2252253770828247,
      "num_tokens": 123555029.0,
      "step": 66270
    },
    {
      "entropy": 6.021966934204102,
      "epoch": 5.852613917343695,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021794375772554532,
      "loss": 5.1542,
      "mean_token_accuracy": 0.23468369692564012,
      "num_tokens": 123564612.0,
      "step": 66275
    },
    {
      "entropy": 6.066075563430786,
      "epoch": 5.853055457435535,
      "grad_norm": 1.578125,
      "learning_rate": 0.00021791329612377985,
      "loss": 5.2046,
      "mean_token_accuracy": 0.22071075439453125,
      "num_tokens": 123573290.0,
      "step": 66280
    },
    {
      "entropy": 6.064444828033447,
      "epoch": 5.853496997527375,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021788283564031053,
      "loss": 5.2491,
      "mean_token_accuracy": 0.2230787083506584,
      "num_tokens": 123582863.0,
      "step": 66285
    },
    {
      "entropy": 6.044376039505005,
      "epoch": 5.853938537619216,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021785237627573395,
      "loss": 5.1808,
      "mean_token_accuracy": 0.23234131336212158,
      "num_tokens": 123591507.0,
      "step": 66290
    },
    {
      "entropy": 6.064584636688233,
      "epoch": 5.854380077711056,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00021782191803064682,
      "loss": 5.2263,
      "mean_token_accuracy": 0.22389415204524993,
      "num_tokens": 123600712.0,
      "step": 66295
    },
    {
      "entropy": 6.09212646484375,
      "epoch": 5.854821617802896,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021779146090564585,
      "loss": 5.3053,
      "mean_token_accuracy": 0.21647993177175523,
      "num_tokens": 123609686.0,
      "step": 66300
    },
    {
      "entropy": 6.033782577514648,
      "epoch": 5.855263157894737,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002177610049013275,
      "loss": 5.1416,
      "mean_token_accuracy": 0.23508764207363128,
      "num_tokens": 123619526.0,
      "step": 66305
    },
    {
      "entropy": 6.048025989532471,
      "epoch": 5.855704697986577,
      "grad_norm": 1.515625,
      "learning_rate": 0.00021773055001828855,
      "loss": 5.195,
      "mean_token_accuracy": 0.22531006932258607,
      "num_tokens": 123627881.0,
      "step": 66310
    },
    {
      "entropy": 6.03445839881897,
      "epoch": 5.856146238078417,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021770009625712562,
      "loss": 5.1636,
      "mean_token_accuracy": 0.23259273618459703,
      "num_tokens": 123637396.0,
      "step": 66315
    },
    {
      "entropy": 6.067575931549072,
      "epoch": 5.856587778170258,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021766964361843522,
      "loss": 5.2582,
      "mean_token_accuracy": 0.22389478236436844,
      "num_tokens": 123646209.0,
      "step": 66320
    },
    {
      "entropy": 6.034580421447754,
      "epoch": 5.857029318262098,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002176391921028139,
      "loss": 5.2552,
      "mean_token_accuracy": 0.21855818778276442,
      "num_tokens": 123656216.0,
      "step": 66325
    },
    {
      "entropy": 6.087258052825928,
      "epoch": 5.857470858353938,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021760874171085804,
      "loss": 5.2033,
      "mean_token_accuracy": 0.2262930303812027,
      "num_tokens": 123665259.0,
      "step": 66330
    },
    {
      "entropy": 6.002837800979615,
      "epoch": 5.857912398445778,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002175782924431644,
      "loss": 5.1951,
      "mean_token_accuracy": 0.2293803334236145,
      "num_tokens": 123674183.0,
      "step": 66335
    },
    {
      "entropy": 6.0317164897918705,
      "epoch": 5.858353938537619,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021754784430032942,
      "loss": 5.1955,
      "mean_token_accuracy": 0.23170504868030548,
      "num_tokens": 123683610.0,
      "step": 66340
    },
    {
      "entropy": 6.059613227844238,
      "epoch": 5.858795478629459,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002175173972829494,
      "loss": 5.2081,
      "mean_token_accuracy": 0.22776435911655427,
      "num_tokens": 123693521.0,
      "step": 66345
    },
    {
      "entropy": 6.100270700454712,
      "epoch": 5.8592370187213,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021748695139162084,
      "loss": 5.1858,
      "mean_token_accuracy": 0.22305103093385698,
      "num_tokens": 123702127.0,
      "step": 66350
    },
    {
      "entropy": 6.02467188835144,
      "epoch": 5.85967855881314,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021745650662694028,
      "loss": 5.1854,
      "mean_token_accuracy": 0.22917559891939163,
      "num_tokens": 123711907.0,
      "step": 66355
    },
    {
      "entropy": 5.997778844833374,
      "epoch": 5.86012009890498,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021742606298950402,
      "loss": 5.1144,
      "mean_token_accuracy": 0.23297643065452575,
      "num_tokens": 123721240.0,
      "step": 66360
    },
    {
      "entropy": 5.9358738422393795,
      "epoch": 5.8605616389968205,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002173956204799084,
      "loss": 5.0684,
      "mean_token_accuracy": 0.23937147110700607,
      "num_tokens": 123729981.0,
      "step": 66365
    },
    {
      "entropy": 6.070351791381836,
      "epoch": 5.861003179088661,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021736517909874976,
      "loss": 5.248,
      "mean_token_accuracy": 0.2207145407795906,
      "num_tokens": 123739371.0,
      "step": 66370
    },
    {
      "entropy": 6.010837984085083,
      "epoch": 5.8614447191805015,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021733473884662458,
      "loss": 5.1323,
      "mean_token_accuracy": 0.22850898653268814,
      "num_tokens": 123748445.0,
      "step": 66375
    },
    {
      "entropy": 6.053656959533692,
      "epoch": 5.861886259272342,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021730429972412907,
      "loss": 5.2817,
      "mean_token_accuracy": 0.22394300401210784,
      "num_tokens": 123757360.0,
      "step": 66380
    },
    {
      "entropy": 6.0419921875,
      "epoch": 5.8623277993641825,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021727386173185954,
      "loss": 5.2039,
      "mean_token_accuracy": 0.2240697681903839,
      "num_tokens": 123766527.0,
      "step": 66385
    },
    {
      "entropy": 6.111005878448486,
      "epoch": 5.8627693394560225,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002172434248704122,
      "loss": 5.2801,
      "mean_token_accuracy": 0.21730123609304428,
      "num_tokens": 123776136.0,
      "step": 66390
    },
    {
      "entropy": 6.082100248336792,
      "epoch": 5.863210879547863,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021721298914038323,
      "loss": 5.2136,
      "mean_token_accuracy": 0.22679590433835983,
      "num_tokens": 123785558.0,
      "step": 66395
    },
    {
      "entropy": 6.034625911712647,
      "epoch": 5.8636524196397035,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000217182554542369,
      "loss": 5.1306,
      "mean_token_accuracy": 0.22859421372413635,
      "num_tokens": 123794982.0,
      "step": 66400
    },
    {
      "entropy": 6.062983798980713,
      "epoch": 5.864093959731544,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002171521210769657,
      "loss": 5.2551,
      "mean_token_accuracy": 0.22108689993619918,
      "num_tokens": 123804220.0,
      "step": 66405
    },
    {
      "entropy": 6.122010803222656,
      "epoch": 5.864535499823384,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021712168874476933,
      "loss": 5.3092,
      "mean_token_accuracy": 0.22625817060470582,
      "num_tokens": 123813141.0,
      "step": 66410
    },
    {
      "entropy": 6.131035327911377,
      "epoch": 5.864977039915225,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021709125754637622,
      "loss": 5.2358,
      "mean_token_accuracy": 0.2243868112564087,
      "num_tokens": 123822829.0,
      "step": 66415
    },
    {
      "entropy": 6.007540035247803,
      "epoch": 5.865418580007065,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021706082748238249,
      "loss": 5.1767,
      "mean_token_accuracy": 0.23599663376808167,
      "num_tokens": 123832178.0,
      "step": 66420
    },
    {
      "entropy": 5.954244470596313,
      "epoch": 5.865860120098905,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021703039855338414,
      "loss": 5.0694,
      "mean_token_accuracy": 0.24050210565328597,
      "num_tokens": 123841878.0,
      "step": 66425
    },
    {
      "entropy": 6.019831752777099,
      "epoch": 5.866301660190746,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021699997075997734,
      "loss": 5.1931,
      "mean_token_accuracy": 0.22542103379964828,
      "num_tokens": 123851962.0,
      "step": 66430
    },
    {
      "entropy": 6.099584484100342,
      "epoch": 5.866743200282586,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021696954410275804,
      "loss": 5.2584,
      "mean_token_accuracy": 0.2232063129544258,
      "num_tokens": 123861552.0,
      "step": 66435
    },
    {
      "entropy": 6.036844396591187,
      "epoch": 5.867184740374426,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00021693911858232251,
      "loss": 5.1929,
      "mean_token_accuracy": 0.22982268780469894,
      "num_tokens": 123870457.0,
      "step": 66440
    },
    {
      "entropy": 6.057239627838134,
      "epoch": 5.867626280466267,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021690869419926657,
      "loss": 5.31,
      "mean_token_accuracy": 0.22683042734861375,
      "num_tokens": 123880806.0,
      "step": 66445
    },
    {
      "entropy": 6.060159206390381,
      "epoch": 5.868067820558107,
      "grad_norm": 1.375,
      "learning_rate": 0.00021687827095418627,
      "loss": 5.2166,
      "mean_token_accuracy": 0.22147864550352098,
      "num_tokens": 123890630.0,
      "step": 66450
    },
    {
      "entropy": 6.000669527053833,
      "epoch": 5.868509360649947,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0002168478488476776,
      "loss": 5.2061,
      "mean_token_accuracy": 0.23193343728780746,
      "num_tokens": 123900142.0,
      "step": 66455
    },
    {
      "entropy": 6.07431116104126,
      "epoch": 5.868950900741788,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021681742788033642,
      "loss": 5.1922,
      "mean_token_accuracy": 0.22659846544265747,
      "num_tokens": 123908905.0,
      "step": 66460
    },
    {
      "entropy": 6.164862155914307,
      "epoch": 5.869392440833628,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021678700805275881,
      "loss": 5.3521,
      "mean_token_accuracy": 0.21056763529777528,
      "num_tokens": 123919194.0,
      "step": 66465
    },
    {
      "entropy": 6.117703819274903,
      "epoch": 5.869833980925468,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002167565893655406,
      "loss": 5.2217,
      "mean_token_accuracy": 0.22337541431188584,
      "num_tokens": 123928041.0,
      "step": 66470
    },
    {
      "entropy": 6.026450872421265,
      "epoch": 5.870275521017309,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021672617181927768,
      "loss": 5.1492,
      "mean_token_accuracy": 0.23237048834562302,
      "num_tokens": 123936756.0,
      "step": 66475
    },
    {
      "entropy": 6.01060996055603,
      "epoch": 5.870717061109149,
      "grad_norm": 1.375,
      "learning_rate": 0.00021669575541456582,
      "loss": 5.1735,
      "mean_token_accuracy": 0.23069096058607103,
      "num_tokens": 123945528.0,
      "step": 66480
    },
    {
      "entropy": 6.1197926044464115,
      "epoch": 5.871158601200989,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021666534015200108,
      "loss": 5.3313,
      "mean_token_accuracy": 0.21374238282442093,
      "num_tokens": 123956615.0,
      "step": 66485
    },
    {
      "entropy": 6.11544451713562,
      "epoch": 5.871600141292829,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002166349260321791,
      "loss": 5.2689,
      "mean_token_accuracy": 0.2136554628610611,
      "num_tokens": 123965867.0,
      "step": 66490
    },
    {
      "entropy": 6.163018465042114,
      "epoch": 5.87204168138467,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021660451305569573,
      "loss": 5.2559,
      "mean_token_accuracy": 0.22534130066633223,
      "num_tokens": 123975232.0,
      "step": 66495
    },
    {
      "entropy": 5.989037418365479,
      "epoch": 5.87248322147651,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021657410122314665,
      "loss": 5.1543,
      "mean_token_accuracy": 0.2284847617149353,
      "num_tokens": 123984393.0,
      "step": 66500
    },
    {
      "entropy": 6.061715984344483,
      "epoch": 5.872924761568351,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021654369053512783,
      "loss": 5.3753,
      "mean_token_accuracy": 0.21206070482730865,
      "num_tokens": 123995257.0,
      "step": 66505
    },
    {
      "entropy": 6.064369773864746,
      "epoch": 5.873366301660191,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021651328099223482,
      "loss": 5.2295,
      "mean_token_accuracy": 0.22586586326360703,
      "num_tokens": 124003996.0,
      "step": 66510
    },
    {
      "entropy": 6.004315614700317,
      "epoch": 5.873807841752031,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021648287259506344,
      "loss": 5.1052,
      "mean_token_accuracy": 0.2382934257388115,
      "num_tokens": 124012760.0,
      "step": 66515
    },
    {
      "entropy": 6.153611564636231,
      "epoch": 5.874249381843871,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021645246534420927,
      "loss": 5.3132,
      "mean_token_accuracy": 0.20860440880060196,
      "num_tokens": 124023083.0,
      "step": 66520
    },
    {
      "entropy": 6.111908960342407,
      "epoch": 5.874690921935712,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021642205924026788,
      "loss": 5.2443,
      "mean_token_accuracy": 0.21408459544181824,
      "num_tokens": 124032152.0,
      "step": 66525
    },
    {
      "entropy": 6.1144062042236325,
      "epoch": 5.875132462027552,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021639165428383517,
      "loss": 5.3132,
      "mean_token_accuracy": 0.21804259121418,
      "num_tokens": 124041394.0,
      "step": 66530
    },
    {
      "entropy": 5.9958498001098635,
      "epoch": 5.875574002119392,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021636125047550658,
      "loss": 5.117,
      "mean_token_accuracy": 0.2326888248324394,
      "num_tokens": 124049738.0,
      "step": 66535
    },
    {
      "entropy": 6.064697313308716,
      "epoch": 5.876015542211233,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002163308478158777,
      "loss": 5.2565,
      "mean_token_accuracy": 0.2215771809220314,
      "num_tokens": 124058396.0,
      "step": 66540
    },
    {
      "entropy": 6.139511346817017,
      "epoch": 5.876457082303073,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002163004463055442,
      "loss": 5.3799,
      "mean_token_accuracy": 0.20985639989376068,
      "num_tokens": 124068260.0,
      "step": 66545
    },
    {
      "entropy": 6.083827018737793,
      "epoch": 5.876898622394913,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021627004594510152,
      "loss": 5.1949,
      "mean_token_accuracy": 0.22660510390996932,
      "num_tokens": 124077206.0,
      "step": 66550
    },
    {
      "entropy": 6.040755319595337,
      "epoch": 5.877340162486754,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002162396467351453,
      "loss": 5.1748,
      "mean_token_accuracy": 0.22114403843879699,
      "num_tokens": 124085832.0,
      "step": 66555
    },
    {
      "entropy": 6.024070739746094,
      "epoch": 5.877781702578594,
      "grad_norm": 1.53125,
      "learning_rate": 0.00021620924867627096,
      "loss": 5.2565,
      "mean_token_accuracy": 0.22577573657035827,
      "num_tokens": 124095522.0,
      "step": 66560
    },
    {
      "entropy": 6.000883674621582,
      "epoch": 5.878223242670434,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021617885176907403,
      "loss": 5.1851,
      "mean_token_accuracy": 0.2331830620765686,
      "num_tokens": 124105635.0,
      "step": 66565
    },
    {
      "entropy": 5.9839334964752195,
      "epoch": 5.878664782762275,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00021614845601414985,
      "loss": 5.0777,
      "mean_token_accuracy": 0.23738474547863006,
      "num_tokens": 124113249.0,
      "step": 66570
    },
    {
      "entropy": 6.083881568908692,
      "epoch": 5.879106322854115,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021611806141209406,
      "loss": 5.2537,
      "mean_token_accuracy": 0.21966308504343032,
      "num_tokens": 124122727.0,
      "step": 66575
    },
    {
      "entropy": 6.044032573699951,
      "epoch": 5.879547862945955,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021608766796350193,
      "loss": 5.2158,
      "mean_token_accuracy": 0.226665860414505,
      "num_tokens": 124131403.0,
      "step": 66580
    },
    {
      "entropy": 5.96666989326477,
      "epoch": 5.879989403037796,
      "grad_norm": 1.28125,
      "learning_rate": 0.00021605727566896888,
      "loss": 5.0734,
      "mean_token_accuracy": 0.24153183847665788,
      "num_tokens": 124140913.0,
      "step": 66585
    },
    {
      "entropy": 6.036703443527221,
      "epoch": 5.880430943129636,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021602688452909019,
      "loss": 5.2502,
      "mean_token_accuracy": 0.22814295142889024,
      "num_tokens": 124150136.0,
      "step": 66590
    },
    {
      "entropy": 5.932095384597778,
      "epoch": 5.880872483221476,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021599649454446136,
      "loss": 5.1149,
      "mean_token_accuracy": 0.24277760237455367,
      "num_tokens": 124159890.0,
      "step": 66595
    },
    {
      "entropy": 6.068654346466064,
      "epoch": 5.881314023313317,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00021596610571567767,
      "loss": 5.1942,
      "mean_token_accuracy": 0.22706115543842315,
      "num_tokens": 124168502.0,
      "step": 66600
    },
    {
      "entropy": 6.048100328445434,
      "epoch": 5.881755563405157,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021593571804333434,
      "loss": 5.2221,
      "mean_token_accuracy": 0.2227315053343773,
      "num_tokens": 124177531.0,
      "step": 66605
    },
    {
      "entropy": 6.009530639648437,
      "epoch": 5.882197103496997,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021590533152802673,
      "loss": 5.2139,
      "mean_token_accuracy": 0.2260904148221016,
      "num_tokens": 124187437.0,
      "step": 66610
    },
    {
      "entropy": 6.049399518966675,
      "epoch": 5.882638643588837,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021587494617034998,
      "loss": 5.2153,
      "mean_token_accuracy": 0.23262002170085908,
      "num_tokens": 124196162.0,
      "step": 66615
    },
    {
      "entropy": 6.067045402526856,
      "epoch": 5.883080183680678,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021584456197089953,
      "loss": 5.2352,
      "mean_token_accuracy": 0.2222118392586708,
      "num_tokens": 124205496.0,
      "step": 66620
    },
    {
      "entropy": 6.06107759475708,
      "epoch": 5.883521723772518,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021581417893027044,
      "loss": 5.1695,
      "mean_token_accuracy": 0.22761785387992858,
      "num_tokens": 124214814.0,
      "step": 66625
    },
    {
      "entropy": 6.124405097961426,
      "epoch": 5.883963263864359,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021578379704905798,
      "loss": 5.2597,
      "mean_token_accuracy": 0.22275041788816452,
      "num_tokens": 124224039.0,
      "step": 66630
    },
    {
      "entropy": 6.096902370452881,
      "epoch": 5.884404803956199,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021575341632785712,
      "loss": 5.2799,
      "mean_token_accuracy": 0.22432964593172072,
      "num_tokens": 124233805.0,
      "step": 66635
    },
    {
      "entropy": 6.059509658813477,
      "epoch": 5.884846344048039,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002157230367672632,
      "loss": 5.2291,
      "mean_token_accuracy": 0.22912947088479996,
      "num_tokens": 124244130.0,
      "step": 66640
    },
    {
      "entropy": 6.077831649780274,
      "epoch": 5.8852878841398795,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021569265836787134,
      "loss": 5.2071,
      "mean_token_accuracy": 0.22865078449249268,
      "num_tokens": 124253901.0,
      "step": 66645
    },
    {
      "entropy": 6.029167652130127,
      "epoch": 5.88572942423172,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021566228113027654,
      "loss": 5.1573,
      "mean_token_accuracy": 0.23090304732322692,
      "num_tokens": 124262543.0,
      "step": 66650
    },
    {
      "entropy": 6.041242218017578,
      "epoch": 5.8861709643235605,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002156319050550738,
      "loss": 5.2662,
      "mean_token_accuracy": 0.22136572301387786,
      "num_tokens": 124272041.0,
      "step": 66655
    },
    {
      "entropy": 6.108542394638062,
      "epoch": 5.886612504415401,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021560153014285838,
      "loss": 5.3327,
      "mean_token_accuracy": 0.21542533934116365,
      "num_tokens": 124283042.0,
      "step": 66660
    },
    {
      "entropy": 6.006865978240967,
      "epoch": 5.8870540445072415,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002155711563942252,
      "loss": 5.179,
      "mean_token_accuracy": 0.2289328917860985,
      "num_tokens": 124292617.0,
      "step": 66665
    },
    {
      "entropy": 6.095309591293335,
      "epoch": 5.8874955845990815,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00021554078380976927,
      "loss": 5.1747,
      "mean_token_accuracy": 0.2286830350756645,
      "num_tokens": 124301862.0,
      "step": 66670
    },
    {
      "entropy": 6.053347539901734,
      "epoch": 5.887937124690922,
      "grad_norm": 1.578125,
      "learning_rate": 0.00021551041239008552,
      "loss": 5.1486,
      "mean_token_accuracy": 0.23354995250701904,
      "num_tokens": 124309993.0,
      "step": 66675
    },
    {
      "entropy": 5.994001388549805,
      "epoch": 5.8883786647827625,
      "grad_norm": 1.5,
      "learning_rate": 0.00021548004213576893,
      "loss": 5.1952,
      "mean_token_accuracy": 0.2323823779821396,
      "num_tokens": 124318630.0,
      "step": 66680
    },
    {
      "entropy": 6.042727327346801,
      "epoch": 5.888820204874603,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021544967304741458,
      "loss": 5.2106,
      "mean_token_accuracy": 0.22048808485269547,
      "num_tokens": 124327275.0,
      "step": 66685
    },
    {
      "entropy": 6.039126825332642,
      "epoch": 5.889261744966443,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002154193051256172,
      "loss": 5.2571,
      "mean_token_accuracy": 0.22500807642936707,
      "num_tokens": 124336655.0,
      "step": 66690
    },
    {
      "entropy": 6.07209005355835,
      "epoch": 5.8897032850582836,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00021538893837097178,
      "loss": 5.2123,
      "mean_token_accuracy": 0.22467538267374038,
      "num_tokens": 124345866.0,
      "step": 66695
    },
    {
      "entropy": 6.104927206039429,
      "epoch": 5.890144825150124,
      "grad_norm": 1.59375,
      "learning_rate": 0.00021535857278407307,
      "loss": 5.1898,
      "mean_token_accuracy": 0.2319646805524826,
      "num_tokens": 124354889.0,
      "step": 66700
    },
    {
      "entropy": 6.122549772262573,
      "epoch": 5.890586365241964,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002153282083655161,
      "loss": 5.3362,
      "mean_token_accuracy": 0.2165471076965332,
      "num_tokens": 124363963.0,
      "step": 66705
    },
    {
      "entropy": 6.049868202209472,
      "epoch": 5.891027905333805,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00021529784511589557,
      "loss": 5.2099,
      "mean_token_accuracy": 0.22432421445846557,
      "num_tokens": 124373266.0,
      "step": 66710
    },
    {
      "entropy": 6.024956130981446,
      "epoch": 5.891469445425645,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021526748303580628,
      "loss": 5.1444,
      "mean_token_accuracy": 0.2352724090218544,
      "num_tokens": 124381805.0,
      "step": 66715
    },
    {
      "entropy": 6.052096271514893,
      "epoch": 5.891910985517485,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00021523712212584295,
      "loss": 5.2065,
      "mean_token_accuracy": 0.22418080270290375,
      "num_tokens": 124390744.0,
      "step": 66720
    },
    {
      "entropy": 6.0540690422058105,
      "epoch": 5.892352525609326,
      "grad_norm": 1.5,
      "learning_rate": 0.00021520676238660053,
      "loss": 5.1953,
      "mean_token_accuracy": 0.22684589624404908,
      "num_tokens": 124399386.0,
      "step": 66725
    },
    {
      "entropy": 6.085961484909058,
      "epoch": 5.892794065701166,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002151764038186736,
      "loss": 5.2919,
      "mean_token_accuracy": 0.20928631573915482,
      "num_tokens": 124408579.0,
      "step": 66730
    },
    {
      "entropy": 6.101934289932251,
      "epoch": 5.893235605793006,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00021514604642265683,
      "loss": 5.289,
      "mean_token_accuracy": 0.21398652344942093,
      "num_tokens": 124416604.0,
      "step": 66735
    },
    {
      "entropy": 6.157070541381836,
      "epoch": 5.893677145884847,
      "grad_norm": 1.296875,
      "learning_rate": 0.000215115690199145,
      "loss": 5.3491,
      "mean_token_accuracy": 0.21219803392887115,
      "num_tokens": 124427485.0,
      "step": 66740
    },
    {
      "entropy": 6.004391622543335,
      "epoch": 5.894118685976687,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021508533514873275,
      "loss": 5.1217,
      "mean_token_accuracy": 0.23199364095926284,
      "num_tokens": 124435971.0,
      "step": 66745
    },
    {
      "entropy": 5.99807915687561,
      "epoch": 5.894560226068527,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021505498127201474,
      "loss": 5.1717,
      "mean_token_accuracy": 0.22977714538574218,
      "num_tokens": 124444809.0,
      "step": 66750
    },
    {
      "entropy": 6.044422769546509,
      "epoch": 5.895001766160368,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021502462856958553,
      "loss": 5.2061,
      "mean_token_accuracy": 0.22008273303508757,
      "num_tokens": 124454052.0,
      "step": 66755
    },
    {
      "entropy": 6.033158254623413,
      "epoch": 5.895443306252208,
      "grad_norm": 1.375,
      "learning_rate": 0.0002149942770420398,
      "loss": 5.1562,
      "mean_token_accuracy": 0.2366922304034233,
      "num_tokens": 124462968.0,
      "step": 66760
    },
    {
      "entropy": 6.021048212051392,
      "epoch": 5.895884846344048,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002149639266899719,
      "loss": 5.2058,
      "mean_token_accuracy": 0.22412120848894118,
      "num_tokens": 124472269.0,
      "step": 66765
    },
    {
      "entropy": 6.038842058181762,
      "epoch": 5.896326386435888,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002149335775139767,
      "loss": 5.2769,
      "mean_token_accuracy": 0.22113879173994064,
      "num_tokens": 124482046.0,
      "step": 66770
    },
    {
      "entropy": 6.050476312637329,
      "epoch": 5.896767926527729,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021490322951464846,
      "loss": 5.3102,
      "mean_token_accuracy": 0.21469053924083709,
      "num_tokens": 124491868.0,
      "step": 66775
    },
    {
      "entropy": 6.045475006103516,
      "epoch": 5.897209466619569,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021487288269258183,
      "loss": 5.2116,
      "mean_token_accuracy": 0.22260419577360152,
      "num_tokens": 124501007.0,
      "step": 66780
    },
    {
      "entropy": 5.9893604755401615,
      "epoch": 5.89765100671141,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021484253704837108,
      "loss": 5.1798,
      "mean_token_accuracy": 0.230080446600914,
      "num_tokens": 124510473.0,
      "step": 66785
    },
    {
      "entropy": 6.082938098907471,
      "epoch": 5.89809254680325,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00021481219258261098,
      "loss": 5.2483,
      "mean_token_accuracy": 0.22505534440279007,
      "num_tokens": 124520357.0,
      "step": 66790
    },
    {
      "entropy": 6.0566717147827145,
      "epoch": 5.89853408689509,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021478184929589578,
      "loss": 5.1554,
      "mean_token_accuracy": 0.2332124337553978,
      "num_tokens": 124529919.0,
      "step": 66795
    },
    {
      "entropy": 6.010277175903321,
      "epoch": 5.89897562698693,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00021475150718881986,
      "loss": 5.1738,
      "mean_token_accuracy": 0.2316194847226143,
      "num_tokens": 124538615.0,
      "step": 66800
    },
    {
      "entropy": 6.017087459564209,
      "epoch": 5.899417167078771,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021472116626197768,
      "loss": 5.2668,
      "mean_token_accuracy": 0.22633512169122696,
      "num_tokens": 124549483.0,
      "step": 66805
    },
    {
      "entropy": 6.083030939102173,
      "epoch": 5.899858707170611,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002146908265159635,
      "loss": 5.1652,
      "mean_token_accuracy": 0.23011651486158372,
      "num_tokens": 124558323.0,
      "step": 66810
    },
    {
      "entropy": 6.045943927764893,
      "epoch": 5.900300247262451,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021466048795137183,
      "loss": 5.2184,
      "mean_token_accuracy": 0.21961380988359452,
      "num_tokens": 124567599.0,
      "step": 66815
    },
    {
      "entropy": 6.0413086891174315,
      "epoch": 5.900741787354292,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021463015056879682,
      "loss": 5.2461,
      "mean_token_accuracy": 0.2187975153326988,
      "num_tokens": 124577970.0,
      "step": 66820
    },
    {
      "entropy": 6.145997858047485,
      "epoch": 5.901183327446132,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021459981436883292,
      "loss": 5.3261,
      "mean_token_accuracy": 0.21370811462402345,
      "num_tokens": 124587708.0,
      "step": 66825
    },
    {
      "entropy": 6.077152919769287,
      "epoch": 5.901624867537972,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021456947935207416,
      "loss": 5.2957,
      "mean_token_accuracy": 0.22157383412122728,
      "num_tokens": 124598120.0,
      "step": 66830
    },
    {
      "entropy": 6.04451961517334,
      "epoch": 5.902066407629813,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021453914551911496,
      "loss": 5.2425,
      "mean_token_accuracy": 0.217770953476429,
      "num_tokens": 124609127.0,
      "step": 66835
    },
    {
      "entropy": 6.138224124908447,
      "epoch": 5.902507947721653,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021450881287054963,
      "loss": 5.2932,
      "mean_token_accuracy": 0.21373989880084993,
      "num_tokens": 124619336.0,
      "step": 66840
    },
    {
      "entropy": 6.117512321472168,
      "epoch": 5.902949487813493,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002144784814069722,
      "loss": 5.2697,
      "mean_token_accuracy": 0.22320457100868224,
      "num_tokens": 124629002.0,
      "step": 66845
    },
    {
      "entropy": 5.99872579574585,
      "epoch": 5.903391027905334,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021444815112897682,
      "loss": 5.1278,
      "mean_token_accuracy": 0.2320953130722046,
      "num_tokens": 124638178.0,
      "step": 66850
    },
    {
      "entropy": 6.031063747406006,
      "epoch": 5.903832567997174,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021441782203715775,
      "loss": 5.2043,
      "mean_token_accuracy": 0.22519214153289796,
      "num_tokens": 124647504.0,
      "step": 66855
    },
    {
      "entropy": 5.991258907318115,
      "epoch": 5.904274108089014,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002143874941321091,
      "loss": 5.1311,
      "mean_token_accuracy": 0.23575061112642287,
      "num_tokens": 124656472.0,
      "step": 66860
    },
    {
      "entropy": 6.02931489944458,
      "epoch": 5.904715648180855,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021435716741442502,
      "loss": 5.2243,
      "mean_token_accuracy": 0.23558271825313568,
      "num_tokens": 124665313.0,
      "step": 66865
    },
    {
      "entropy": 6.034744691848755,
      "epoch": 5.905157188272695,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021432684188469952,
      "loss": 5.2705,
      "mean_token_accuracy": 0.2233613148331642,
      "num_tokens": 124674040.0,
      "step": 66870
    },
    {
      "entropy": 5.978275394439697,
      "epoch": 5.905598728364535,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021429651754352659,
      "loss": 5.1891,
      "mean_token_accuracy": 0.2251008316874504,
      "num_tokens": 124682482.0,
      "step": 66875
    },
    {
      "entropy": 6.017651033401489,
      "epoch": 5.906040268456376,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021426619439150047,
      "loss": 5.2287,
      "mean_token_accuracy": 0.23127209395170212,
      "num_tokens": 124691977.0,
      "step": 66880
    },
    {
      "entropy": 6.010209369659424,
      "epoch": 5.906481808548216,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00021423587242921506,
      "loss": 5.097,
      "mean_token_accuracy": 0.2255975514650345,
      "num_tokens": 124700084.0,
      "step": 66885
    },
    {
      "entropy": 5.935938549041748,
      "epoch": 5.906923348640056,
      "grad_norm": 1.53125,
      "learning_rate": 0.00021420555165726436,
      "loss": 5.084,
      "mean_token_accuracy": 0.23230768144130706,
      "num_tokens": 124708546.0,
      "step": 66890
    },
    {
      "entropy": 6.049429035186767,
      "epoch": 5.907364888731896,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021417523207624216,
      "loss": 5.2523,
      "mean_token_accuracy": 0.22663129568099977,
      "num_tokens": 124717231.0,
      "step": 66895
    },
    {
      "entropy": 6.052717781066894,
      "epoch": 5.907806428823737,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002141449136867427,
      "loss": 5.202,
      "mean_token_accuracy": 0.22581994384527207,
      "num_tokens": 124725872.0,
      "step": 66900
    },
    {
      "entropy": 6.078418397903443,
      "epoch": 5.908247968915577,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021411459648935983,
      "loss": 5.2456,
      "mean_token_accuracy": 0.21897562742233276,
      "num_tokens": 124736489.0,
      "step": 66905
    },
    {
      "entropy": 6.016872501373291,
      "epoch": 5.908689509007418,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021408428048468736,
      "loss": 5.2196,
      "mean_token_accuracy": 0.21809112876653672,
      "num_tokens": 124745515.0,
      "step": 66910
    },
    {
      "entropy": 5.979565382003784,
      "epoch": 5.909131049099258,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021405396567331908,
      "loss": 5.2077,
      "mean_token_accuracy": 0.22627214938402176,
      "num_tokens": 124754890.0,
      "step": 66915
    },
    {
      "entropy": 6.069076776504517,
      "epoch": 5.909572589191098,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021402365205584907,
      "loss": 5.2333,
      "mean_token_accuracy": 0.22527813017368317,
      "num_tokens": 124764104.0,
      "step": 66920
    },
    {
      "entropy": 6.0182352542877195,
      "epoch": 5.9100141292829385,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021399333963287098,
      "loss": 5.1499,
      "mean_token_accuracy": 0.23465195298194885,
      "num_tokens": 124773790.0,
      "step": 66925
    },
    {
      "entropy": 6.047946739196777,
      "epoch": 5.910455669374779,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021396302840497866,
      "loss": 5.2492,
      "mean_token_accuracy": 0.2252206802368164,
      "num_tokens": 124782994.0,
      "step": 66930
    },
    {
      "entropy": 6.034108352661133,
      "epoch": 5.9108972094666195,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021393271837276596,
      "loss": 5.199,
      "mean_token_accuracy": 0.2311555787920952,
      "num_tokens": 124792762.0,
      "step": 66935
    },
    {
      "entropy": 5.994521188735962,
      "epoch": 5.91133874955846,
      "grad_norm": 1.234375,
      "learning_rate": 0.00021390240953682643,
      "loss": 5.172,
      "mean_token_accuracy": 0.2282037541270256,
      "num_tokens": 124802667.0,
      "step": 66940
    },
    {
      "entropy": 6.036450958251953,
      "epoch": 5.9117802896503004,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021387210189775408,
      "loss": 5.2021,
      "mean_token_accuracy": 0.21788398921489716,
      "num_tokens": 124812532.0,
      "step": 66945
    },
    {
      "entropy": 6.1141993522644045,
      "epoch": 5.9122218297421405,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00021384179545614246,
      "loss": 5.2275,
      "mean_token_accuracy": 0.21969768404960632,
      "num_tokens": 124822567.0,
      "step": 66950
    },
    {
      "entropy": 5.9691397190094,
      "epoch": 5.9126633698339806,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021381149021258534,
      "loss": 5.1037,
      "mean_token_accuracy": 0.23889748603105546,
      "num_tokens": 124832316.0,
      "step": 66955
    },
    {
      "entropy": 5.985900163650513,
      "epoch": 5.9131049099258215,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002137811861676762,
      "loss": 5.141,
      "mean_token_accuracy": 0.23540243208408357,
      "num_tokens": 124841077.0,
      "step": 66960
    },
    {
      "entropy": 6.105058002471924,
      "epoch": 5.9135464500176615,
      "grad_norm": 1.28125,
      "learning_rate": 0.00021375088332200893,
      "loss": 5.2802,
      "mean_token_accuracy": 0.22385354936122895,
      "num_tokens": 124850663.0,
      "step": 66965
    },
    {
      "entropy": 6.046379280090332,
      "epoch": 5.913987990109502,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021372058167617704,
      "loss": 5.225,
      "mean_token_accuracy": 0.23126835227012635,
      "num_tokens": 124860329.0,
      "step": 66970
    },
    {
      "entropy": 6.088943719863892,
      "epoch": 5.9144295302013425,
      "grad_norm": 1.25,
      "learning_rate": 0.000213690281230774,
      "loss": 5.3187,
      "mean_token_accuracy": 0.20976796746253967,
      "num_tokens": 124870698.0,
      "step": 66975
    },
    {
      "entropy": 6.055007362365723,
      "epoch": 5.914871070293183,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021365998198639348,
      "loss": 5.2207,
      "mean_token_accuracy": 0.22277279794216157,
      "num_tokens": 124879921.0,
      "step": 66980
    },
    {
      "entropy": 6.070971822738647,
      "epoch": 5.915312610385023,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002136296839436291,
      "loss": 5.2474,
      "mean_token_accuracy": 0.21491543352603912,
      "num_tokens": 124889403.0,
      "step": 66985
    },
    {
      "entropy": 6.031211709976196,
      "epoch": 5.915754150476864,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021359938710307426,
      "loss": 5.2014,
      "mean_token_accuracy": 0.2281779021024704,
      "num_tokens": 124899060.0,
      "step": 66990
    },
    {
      "entropy": 6.096860885620117,
      "epoch": 5.916195690568704,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002135690914653225,
      "loss": 5.2359,
      "mean_token_accuracy": 0.2219248265028,
      "num_tokens": 124908735.0,
      "step": 66995
    },
    {
      "entropy": 5.995317506790161,
      "epoch": 5.916637230660544,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00021353879703096734,
      "loss": 5.1639,
      "mean_token_accuracy": 0.22667192220687865,
      "num_tokens": 124918101.0,
      "step": 67000
    },
    {
      "entropy": 6.03596715927124,
      "epoch": 5.917078770752385,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002135085038006021,
      "loss": 5.208,
      "mean_token_accuracy": 0.22276535034179687,
      "num_tokens": 124927218.0,
      "step": 67005
    },
    {
      "entropy": 6.037216424942017,
      "epoch": 5.917520310844225,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002134782117748204,
      "loss": 5.2328,
      "mean_token_accuracy": 0.22423741221427917,
      "num_tokens": 124936391.0,
      "step": 67010
    },
    {
      "entropy": 5.9454539775848385,
      "epoch": 5.917961850936065,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021344792095421555,
      "loss": 5.0882,
      "mean_token_accuracy": 0.23186294883489608,
      "num_tokens": 124946074.0,
      "step": 67015
    },
    {
      "entropy": 6.052266454696655,
      "epoch": 5.918403391027906,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002134176313393809,
      "loss": 5.1966,
      "mean_token_accuracy": 0.2274580642580986,
      "num_tokens": 124955145.0,
      "step": 67020
    },
    {
      "entropy": 5.959078741073609,
      "epoch": 5.918844931119746,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002133873429309097,
      "loss": 5.1189,
      "mean_token_accuracy": 0.23649072647094727,
      "num_tokens": 124963354.0,
      "step": 67025
    },
    {
      "entropy": 6.0433107852935795,
      "epoch": 5.919286471211586,
      "grad_norm": 1.25,
      "learning_rate": 0.0002133570557293956,
      "loss": 5.2207,
      "mean_token_accuracy": 0.22541569024324418,
      "num_tokens": 124973613.0,
      "step": 67030
    },
    {
      "entropy": 6.018110656738282,
      "epoch": 5.919728011303427,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00021332676973543163,
      "loss": 5.1865,
      "mean_token_accuracy": 0.23305855840444564,
      "num_tokens": 124983116.0,
      "step": 67035
    },
    {
      "entropy": 6.07292423248291,
      "epoch": 5.920169551395267,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021329648494961123,
      "loss": 5.2389,
      "mean_token_accuracy": 0.22022756934165955,
      "num_tokens": 124992591.0,
      "step": 67040
    },
    {
      "entropy": 5.977809238433838,
      "epoch": 5.920611091487107,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002132662013725275,
      "loss": 5.1106,
      "mean_token_accuracy": 0.2315235987305641,
      "num_tokens": 125000788.0,
      "step": 67045
    },
    {
      "entropy": 6.046761178970337,
      "epoch": 5.921052631578947,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021323591900477385,
      "loss": 5.2316,
      "mean_token_accuracy": 0.21494332253932952,
      "num_tokens": 125009322.0,
      "step": 67050
    },
    {
      "entropy": 6.094365882873535,
      "epoch": 5.921494171670788,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021320563784694334,
      "loss": 5.2455,
      "mean_token_accuracy": 0.21661664843559264,
      "num_tokens": 125019332.0,
      "step": 67055
    },
    {
      "entropy": 6.07942304611206,
      "epoch": 5.921935711762628,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021317535789962933,
      "loss": 5.22,
      "mean_token_accuracy": 0.22493966072797775,
      "num_tokens": 125029164.0,
      "step": 67060
    },
    {
      "entropy": 6.002233982086182,
      "epoch": 5.922377251854469,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002131450791634249,
      "loss": 5.1457,
      "mean_token_accuracy": 0.2340474396944046,
      "num_tokens": 125037970.0,
      "step": 67065
    },
    {
      "entropy": 6.132065963745117,
      "epoch": 5.922818791946309,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021311480163892305,
      "loss": 5.355,
      "mean_token_accuracy": 0.22354182004928588,
      "num_tokens": 125048186.0,
      "step": 67070
    },
    {
      "entropy": 6.032468223571778,
      "epoch": 5.923260332038149,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00021308452532671718,
      "loss": 5.1826,
      "mean_token_accuracy": 0.23711497187614441,
      "num_tokens": 125058265.0,
      "step": 67075
    },
    {
      "entropy": 6.0100983619689945,
      "epoch": 5.923701872129989,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021305425022740023,
      "loss": 5.1933,
      "mean_token_accuracy": 0.23090488612651824,
      "num_tokens": 125067155.0,
      "step": 67080
    },
    {
      "entropy": 6.02728304862976,
      "epoch": 5.92414341222183,
      "grad_norm": 1.5,
      "learning_rate": 0.00021302397634156528,
      "loss": 5.1879,
      "mean_token_accuracy": 0.224807445704937,
      "num_tokens": 125077082.0,
      "step": 67085
    },
    {
      "entropy": 6.0852587699890135,
      "epoch": 5.92458495231367,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002129937036698053,
      "loss": 5.2832,
      "mean_token_accuracy": 0.22001894265413285,
      "num_tokens": 125086048.0,
      "step": 67090
    },
    {
      "entropy": 6.100051641464233,
      "epoch": 5.92502649240551,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021296343221271347,
      "loss": 5.3278,
      "mean_token_accuracy": 0.21375354379415512,
      "num_tokens": 125095997.0,
      "step": 67095
    },
    {
      "entropy": 5.989481210708618,
      "epoch": 5.925468032497351,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021293316197088275,
      "loss": 5.1923,
      "mean_token_accuracy": 0.2286098137497902,
      "num_tokens": 125105684.0,
      "step": 67100
    },
    {
      "entropy": 5.952549314498901,
      "epoch": 5.925909572589191,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021290289294490612,
      "loss": 5.0936,
      "mean_token_accuracy": 0.2321358412504196,
      "num_tokens": 125114497.0,
      "step": 67105
    },
    {
      "entropy": 6.009115600585938,
      "epoch": 5.926351112681031,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021287262513537637,
      "loss": 5.1884,
      "mean_token_accuracy": 0.22274185717105865,
      "num_tokens": 125124069.0,
      "step": 67110
    },
    {
      "entropy": 6.0226030349731445,
      "epoch": 5.926792652772872,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021284235854288658,
      "loss": 5.2161,
      "mean_token_accuracy": 0.2257998451590538,
      "num_tokens": 125133364.0,
      "step": 67115
    },
    {
      "entropy": 5.98338942527771,
      "epoch": 5.927234192864712,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021281209316802968,
      "loss": 5.1792,
      "mean_token_accuracy": 0.23038516491651534,
      "num_tokens": 125142796.0,
      "step": 67120
    },
    {
      "entropy": 6.030766868591309,
      "epoch": 5.927675732956552,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021278182901139853,
      "loss": 5.1761,
      "mean_token_accuracy": 0.23373415917158127,
      "num_tokens": 125152541.0,
      "step": 67125
    },
    {
      "entropy": 6.042885208129883,
      "epoch": 5.928117273048393,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021275156607358597,
      "loss": 5.2055,
      "mean_token_accuracy": 0.22893495857715607,
      "num_tokens": 125161825.0,
      "step": 67130
    },
    {
      "entropy": 6.025340890884399,
      "epoch": 5.928558813140233,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021272130435518472,
      "loss": 5.1442,
      "mean_token_accuracy": 0.23628265112638475,
      "num_tokens": 125171494.0,
      "step": 67135
    },
    {
      "entropy": 6.064283561706543,
      "epoch": 5.929000353232073,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021269104385678784,
      "loss": 5.2021,
      "mean_token_accuracy": 0.22722953855991362,
      "num_tokens": 125180909.0,
      "step": 67140
    },
    {
      "entropy": 6.053762912750244,
      "epoch": 5.929441893323914,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021266078457898792,
      "loss": 5.2289,
      "mean_token_accuracy": 0.22413670271635056,
      "num_tokens": 125189549.0,
      "step": 67145
    },
    {
      "entropy": 6.027833747863769,
      "epoch": 5.929883433415754,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0002126305265223778,
      "loss": 5.1902,
      "mean_token_accuracy": 0.22765109837055206,
      "num_tokens": 125198708.0,
      "step": 67150
    },
    {
      "entropy": 6.015199899673462,
      "epoch": 5.930324973507594,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002126002696875501,
      "loss": 5.1318,
      "mean_token_accuracy": 0.22954574823379517,
      "num_tokens": 125207976.0,
      "step": 67155
    },
    {
      "entropy": 6.009912014007568,
      "epoch": 5.930766513599435,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021257001407509767,
      "loss": 5.1276,
      "mean_token_accuracy": 0.2327559307217598,
      "num_tokens": 125217178.0,
      "step": 67160
    },
    {
      "entropy": 5.918415307998657,
      "epoch": 5.931208053691275,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00021253975968561322,
      "loss": 5.0527,
      "mean_token_accuracy": 0.23995050340890883,
      "num_tokens": 125226284.0,
      "step": 67165
    },
    {
      "entropy": 6.013046979904175,
      "epoch": 5.931649593783115,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021250950651968936,
      "loss": 5.2337,
      "mean_token_accuracy": 0.22920115292072296,
      "num_tokens": 125235951.0,
      "step": 67170
    },
    {
      "entropy": 6.039224433898926,
      "epoch": 5.932091133874955,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021247925457791868,
      "loss": 5.2531,
      "mean_token_accuracy": 0.2254711851477623,
      "num_tokens": 125245925.0,
      "step": 67175
    },
    {
      "entropy": 6.012128019332886,
      "epoch": 5.932532673966796,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021244900386089383,
      "loss": 5.1129,
      "mean_token_accuracy": 0.233999764919281,
      "num_tokens": 125255473.0,
      "step": 67180
    },
    {
      "entropy": 6.170004940032959,
      "epoch": 5.932974214058636,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021241875436920737,
      "loss": 5.3225,
      "mean_token_accuracy": 0.22009024620056153,
      "num_tokens": 125265110.0,
      "step": 67185
    },
    {
      "entropy": 6.106123733520508,
      "epoch": 5.933415754150477,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000212388506103452,
      "loss": 5.2824,
      "mean_token_accuracy": 0.21510667353868484,
      "num_tokens": 125274732.0,
      "step": 67190
    },
    {
      "entropy": 6.023903703689575,
      "epoch": 5.933857294242317,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021235825906422025,
      "loss": 5.2065,
      "mean_token_accuracy": 0.22874386012554168,
      "num_tokens": 125284562.0,
      "step": 67195
    },
    {
      "entropy": 5.995973014831543,
      "epoch": 5.934298834334157,
      "grad_norm": 1.375,
      "learning_rate": 0.00021232801325210437,
      "loss": 5.1164,
      "mean_token_accuracy": 0.23150293976068498,
      "num_tokens": 125294385.0,
      "step": 67200
    },
    {
      "entropy": 6.038051795959473,
      "epoch": 5.9347403744259974,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021229776866769725,
      "loss": 5.2471,
      "mean_token_accuracy": 0.2219327822327614,
      "num_tokens": 125303029.0,
      "step": 67205
    },
    {
      "entropy": 6.017214250564575,
      "epoch": 5.935181914517838,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021226752531159116,
      "loss": 5.1973,
      "mean_token_accuracy": 0.23106400966644286,
      "num_tokens": 125311487.0,
      "step": 67210
    },
    {
      "entropy": 6.045269441604614,
      "epoch": 5.935623454609678,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00021223728318437858,
      "loss": 5.2161,
      "mean_token_accuracy": 0.22665150016546248,
      "num_tokens": 125319926.0,
      "step": 67215
    },
    {
      "entropy": 6.067590761184692,
      "epoch": 5.936064994701519,
      "grad_norm": 1.375,
      "learning_rate": 0.00021220704228665178,
      "loss": 5.2125,
      "mean_token_accuracy": 0.21766325384378432,
      "num_tokens": 125329362.0,
      "step": 67220
    },
    {
      "entropy": 6.016321134567261,
      "epoch": 5.936506534793359,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021217680261900347,
      "loss": 5.1627,
      "mean_token_accuracy": 0.22847800254821776,
      "num_tokens": 125338622.0,
      "step": 67225
    },
    {
      "entropy": 6.0665935516357425,
      "epoch": 5.9369480748851995,
      "grad_norm": 1.5625,
      "learning_rate": 0.00021214656418202583,
      "loss": 5.2073,
      "mean_token_accuracy": 0.22773438245058059,
      "num_tokens": 125347656.0,
      "step": 67230
    },
    {
      "entropy": 5.994075155258178,
      "epoch": 5.9373896149770395,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021211632697631122,
      "loss": 5.1389,
      "mean_token_accuracy": 0.23326073288917543,
      "num_tokens": 125356848.0,
      "step": 67235
    },
    {
      "entropy": 6.040908288955689,
      "epoch": 5.9378311550688805,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021208609100245202,
      "loss": 5.2194,
      "mean_token_accuracy": 0.22404717952013015,
      "num_tokens": 125366346.0,
      "step": 67240
    },
    {
      "entropy": 5.952464771270752,
      "epoch": 5.9382726951607205,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002120558562610405,
      "loss": 5.0528,
      "mean_token_accuracy": 0.24363954812288285,
      "num_tokens": 125375901.0,
      "step": 67245
    },
    {
      "entropy": 6.0802003860473635,
      "epoch": 5.938714235252561,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021202562275266895,
      "loss": 5.2945,
      "mean_token_accuracy": 0.2177878811955452,
      "num_tokens": 125384299.0,
      "step": 67250
    },
    {
      "entropy": 6.061047029495239,
      "epoch": 5.9391557753444015,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021199539047792965,
      "loss": 5.2056,
      "mean_token_accuracy": 0.2301390826702118,
      "num_tokens": 125393345.0,
      "step": 67255
    },
    {
      "entropy": 6.006500196456909,
      "epoch": 5.939597315436242,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021196515943741484,
      "loss": 5.2363,
      "mean_token_accuracy": 0.22530437111854554,
      "num_tokens": 125401978.0,
      "step": 67260
    },
    {
      "entropy": 6.0660735130310055,
      "epoch": 5.940038855528082,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021193492963171668,
      "loss": 5.1919,
      "mean_token_accuracy": 0.22169471085071563,
      "num_tokens": 125411560.0,
      "step": 67265
    },
    {
      "entropy": 6.069657659530639,
      "epoch": 5.940480395619923,
      "grad_norm": 1.5,
      "learning_rate": 0.0002119047010614274,
      "loss": 5.3058,
      "mean_token_accuracy": 0.21362620890140532,
      "num_tokens": 125421071.0,
      "step": 67270
    },
    {
      "entropy": 6.082340574264526,
      "epoch": 5.940921935711763,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021187447372713914,
      "loss": 5.3035,
      "mean_token_accuracy": 0.22164987921714782,
      "num_tokens": 125430894.0,
      "step": 67275
    },
    {
      "entropy": 6.100770282745361,
      "epoch": 5.941363475803603,
      "grad_norm": 1.5,
      "learning_rate": 0.0002118442476294441,
      "loss": 5.2432,
      "mean_token_accuracy": 0.2154076799750328,
      "num_tokens": 125440281.0,
      "step": 67280
    },
    {
      "entropy": 6.070390367507935,
      "epoch": 5.941805015895444,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021181402276893426,
      "loss": 5.1421,
      "mean_token_accuracy": 0.22879059314727784,
      "num_tokens": 125450220.0,
      "step": 67285
    },
    {
      "entropy": 6.008889818191529,
      "epoch": 5.942246555987284,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021178379914620172,
      "loss": 5.2209,
      "mean_token_accuracy": 0.22092297673225403,
      "num_tokens": 125459505.0,
      "step": 67290
    },
    {
      "entropy": 6.069889831542969,
      "epoch": 5.942688096079124,
      "grad_norm": 1.53125,
      "learning_rate": 0.00021175357676183866,
      "loss": 5.1727,
      "mean_token_accuracy": 0.2331426963210106,
      "num_tokens": 125468403.0,
      "step": 67295
    },
    {
      "entropy": 5.957492303848267,
      "epoch": 5.943129636170965,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021172335561643707,
      "loss": 5.1872,
      "mean_token_accuracy": 0.22544364035129547,
      "num_tokens": 125478684.0,
      "step": 67300
    },
    {
      "entropy": 6.049700736999512,
      "epoch": 5.943571176262805,
      "grad_norm": 1.546875,
      "learning_rate": 0.00021169313571058895,
      "loss": 5.2882,
      "mean_token_accuracy": 0.2211044803261757,
      "num_tokens": 125487854.0,
      "step": 67305
    },
    {
      "entropy": 6.012013006210327,
      "epoch": 5.944012716354645,
      "grad_norm": 1.546875,
      "learning_rate": 0.00021166291704488628,
      "loss": 5.1633,
      "mean_token_accuracy": 0.2338678225874901,
      "num_tokens": 125496842.0,
      "step": 67310
    },
    {
      "entropy": 6.051923656463623,
      "epoch": 5.944454256446486,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021163269961992103,
      "loss": 5.2338,
      "mean_token_accuracy": 0.224807907640934,
      "num_tokens": 125507416.0,
      "step": 67315
    },
    {
      "entropy": 6.037640619277954,
      "epoch": 5.944895796538326,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002116024834362852,
      "loss": 5.1703,
      "mean_token_accuracy": 0.23106762319803237,
      "num_tokens": 125515905.0,
      "step": 67320
    },
    {
      "entropy": 6.046088075637817,
      "epoch": 5.945337336630166,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021157226849457067,
      "loss": 5.1745,
      "mean_token_accuracy": 0.2298075869679451,
      "num_tokens": 125525165.0,
      "step": 67325
    },
    {
      "entropy": 6.002207469940186,
      "epoch": 5.945778876722006,
      "grad_norm": 5.0,
      "learning_rate": 0.00021154205479536925,
      "loss": 5.2681,
      "mean_token_accuracy": 0.22252507656812667,
      "num_tokens": 125534362.0,
      "step": 67330
    },
    {
      "entropy": 6.100546646118164,
      "epoch": 5.946220416813847,
      "grad_norm": 1.515625,
      "learning_rate": 0.00021151184233927296,
      "loss": 5.3108,
      "mean_token_accuracy": 0.21642959415912627,
      "num_tokens": 125543528.0,
      "step": 67335
    },
    {
      "entropy": 5.99236798286438,
      "epoch": 5.946661956905687,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021148163112687357,
      "loss": 5.1874,
      "mean_token_accuracy": 0.23155421614646912,
      "num_tokens": 125553132.0,
      "step": 67340
    },
    {
      "entropy": 6.007138586044311,
      "epoch": 5.947103496997528,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00021145142115876293,
      "loss": 5.1669,
      "mean_token_accuracy": 0.23122894167900085,
      "num_tokens": 125563085.0,
      "step": 67345
    },
    {
      "entropy": 6.027173662185669,
      "epoch": 5.947545037089368,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021142121243553275,
      "loss": 5.1647,
      "mean_token_accuracy": 0.22418125718832016,
      "num_tokens": 125572071.0,
      "step": 67350
    },
    {
      "entropy": 6.043410778045654,
      "epoch": 5.947986577181208,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021139100495777481,
      "loss": 5.2459,
      "mean_token_accuracy": 0.22116029411554336,
      "num_tokens": 125581842.0,
      "step": 67355
    },
    {
      "entropy": 6.05499119758606,
      "epoch": 5.948428117273048,
      "grad_norm": 1.390625,
      "learning_rate": 0.000211360798726081,
      "loss": 5.2854,
      "mean_token_accuracy": 0.2220860794186592,
      "num_tokens": 125590525.0,
      "step": 67360
    },
    {
      "entropy": 6.056475734710693,
      "epoch": 5.948869657364889,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002113305937410429,
      "loss": 5.2491,
      "mean_token_accuracy": 0.22140901386737824,
      "num_tokens": 125599192.0,
      "step": 67365
    },
    {
      "entropy": 6.008730936050415,
      "epoch": 5.949311197456729,
      "grad_norm": 1.5,
      "learning_rate": 0.0002113003900032523,
      "loss": 5.1618,
      "mean_token_accuracy": 0.2291609451174736,
      "num_tokens": 125608088.0,
      "step": 67370
    },
    {
      "entropy": 6.101672792434693,
      "epoch": 5.949752737548569,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00021127018751330073,
      "loss": 5.2358,
      "mean_token_accuracy": 0.22385910004377366,
      "num_tokens": 125617752.0,
      "step": 67375
    },
    {
      "entropy": 6.099934482574463,
      "epoch": 5.95019427764041,
      "grad_norm": 1.5,
      "learning_rate": 0.00021123998627178004,
      "loss": 5.2729,
      "mean_token_accuracy": 0.22268106490373613,
      "num_tokens": 125627531.0,
      "step": 67380
    },
    {
      "entropy": 5.999693441390991,
      "epoch": 5.95063581773225,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00021120978627928177,
      "loss": 5.162,
      "mean_token_accuracy": 0.23376044034957885,
      "num_tokens": 125636668.0,
      "step": 67385
    },
    {
      "entropy": 6.110735130310059,
      "epoch": 5.95107735782409,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002111795875363975,
      "loss": 5.226,
      "mean_token_accuracy": 0.2237336128950119,
      "num_tokens": 125645872.0,
      "step": 67390
    },
    {
      "entropy": 5.939136123657226,
      "epoch": 5.951518897915931,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002111493900437187,
      "loss": 5.0694,
      "mean_token_accuracy": 0.22522523552179335,
      "num_tokens": 125654695.0,
      "step": 67395
    },
    {
      "entropy": 5.962715435028076,
      "epoch": 5.951960438007771,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021111919380183714,
      "loss": 5.0964,
      "mean_token_accuracy": 0.23678121119737625,
      "num_tokens": 125664569.0,
      "step": 67400
    },
    {
      "entropy": 6.0971949100494385,
      "epoch": 5.952401978099611,
      "grad_norm": 1.5625,
      "learning_rate": 0.00021108899881134424,
      "loss": 5.322,
      "mean_token_accuracy": 0.21960391104221344,
      "num_tokens": 125673203.0,
      "step": 67405
    },
    {
      "entropy": 6.0832396984100345,
      "epoch": 5.952843518191452,
      "grad_norm": 1.53125,
      "learning_rate": 0.00021105880507283153,
      "loss": 5.2725,
      "mean_token_accuracy": 0.22625081092119217,
      "num_tokens": 125681570.0,
      "step": 67410
    },
    {
      "entropy": 6.054014062881469,
      "epoch": 5.953285058283292,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021102861258689043,
      "loss": 5.2093,
      "mean_token_accuracy": 0.22693924754858016,
      "num_tokens": 125690724.0,
      "step": 67415
    },
    {
      "entropy": 6.072931337356567,
      "epoch": 5.953726598375132,
      "grad_norm": 1.28125,
      "learning_rate": 0.00021099842135411236,
      "loss": 5.1645,
      "mean_token_accuracy": 0.22778947055339813,
      "num_tokens": 125700047.0,
      "step": 67420
    },
    {
      "entropy": 6.114666557312011,
      "epoch": 5.954168138466973,
      "grad_norm": 1.5,
      "learning_rate": 0.00021096823137508886,
      "loss": 5.3463,
      "mean_token_accuracy": 0.21349935978651047,
      "num_tokens": 125708758.0,
      "step": 67425
    },
    {
      "entropy": 6.06107988357544,
      "epoch": 5.954609678558813,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021093804265041128,
      "loss": 5.2164,
      "mean_token_accuracy": 0.22940165549516678,
      "num_tokens": 125718101.0,
      "step": 67430
    },
    {
      "entropy": 5.8262310981750485,
      "epoch": 5.955051218650653,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021090785518067107,
      "loss": 5.0144,
      "mean_token_accuracy": 0.2450176864862442,
      "num_tokens": 125727308.0,
      "step": 67435
    },
    {
      "entropy": 5.939604234695435,
      "epoch": 5.955492758742494,
      "grad_norm": 1.59375,
      "learning_rate": 0.00021087766896645948,
      "loss": 5.0562,
      "mean_token_accuracy": 0.24894948303699493,
      "num_tokens": 125735784.0,
      "step": 67440
    },
    {
      "entropy": 6.022879981994629,
      "epoch": 5.955934298834334,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002108474840083679,
      "loss": 5.2477,
      "mean_token_accuracy": 0.23289978355169297,
      "num_tokens": 125745584.0,
      "step": 67445
    },
    {
      "entropy": 6.061662101745606,
      "epoch": 5.956375838926174,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021081730030698764,
      "loss": 5.2402,
      "mean_token_accuracy": 0.22250514328479767,
      "num_tokens": 125754176.0,
      "step": 67450
    },
    {
      "entropy": 6.094346141815185,
      "epoch": 5.956817379018015,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021078711786290993,
      "loss": 5.3299,
      "mean_token_accuracy": 0.21392184048891066,
      "num_tokens": 125763529.0,
      "step": 67455
    },
    {
      "entropy": 6.022844362258911,
      "epoch": 5.957258919109855,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021075693667672602,
      "loss": 5.0932,
      "mean_token_accuracy": 0.2342488780617714,
      "num_tokens": 125772532.0,
      "step": 67460
    },
    {
      "entropy": 6.019479465484619,
      "epoch": 5.957700459201695,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021072675674902724,
      "loss": 5.1539,
      "mean_token_accuracy": 0.22806539088487626,
      "num_tokens": 125782131.0,
      "step": 67465
    },
    {
      "entropy": 5.984171390533447,
      "epoch": 5.958141999293536,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021069657808040482,
      "loss": 5.2269,
      "mean_token_accuracy": 0.22944531291723252,
      "num_tokens": 125792796.0,
      "step": 67470
    },
    {
      "entropy": 6.1135626316070555,
      "epoch": 5.958583539385376,
      "grad_norm": 1.375,
      "learning_rate": 0.00021066640067144977,
      "loss": 5.3368,
      "mean_token_accuracy": 0.2146704077720642,
      "num_tokens": 125803466.0,
      "step": 67475
    },
    {
      "entropy": 6.050587511062622,
      "epoch": 5.959025079477216,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021063622452275338,
      "loss": 5.2205,
      "mean_token_accuracy": 0.2319306656718254,
      "num_tokens": 125813086.0,
      "step": 67480
    },
    {
      "entropy": 6.0740560531616214,
      "epoch": 5.959466619569056,
      "grad_norm": 1.53125,
      "learning_rate": 0.00021060604963490665,
      "loss": 5.2491,
      "mean_token_accuracy": 0.2229975402355194,
      "num_tokens": 125822293.0,
      "step": 67485
    },
    {
      "entropy": 6.108096742630005,
      "epoch": 5.959908159660897,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002105758760085009,
      "loss": 5.366,
      "mean_token_accuracy": 0.20536169707775115,
      "num_tokens": 125832395.0,
      "step": 67490
    },
    {
      "entropy": 6.148461103439331,
      "epoch": 5.960349699752737,
      "grad_norm": 1.390625,
      "learning_rate": 0.000210545703644127,
      "loss": 5.2832,
      "mean_token_accuracy": 0.21606906205415727,
      "num_tokens": 125841919.0,
      "step": 67495
    },
    {
      "entropy": 6.020512580871582,
      "epoch": 5.960791239844578,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002105155325423762,
      "loss": 5.2092,
      "mean_token_accuracy": 0.22686816304922103,
      "num_tokens": 125851402.0,
      "step": 67500
    },
    {
      "entropy": 6.038963270187378,
      "epoch": 5.961232779936418,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00021048536270383938,
      "loss": 5.2309,
      "mean_token_accuracy": 0.22002965062856675,
      "num_tokens": 125860182.0,
      "step": 67505
    },
    {
      "entropy": 6.098501205444336,
      "epoch": 5.9616743200282585,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002104551941291077,
      "loss": 5.2136,
      "mean_token_accuracy": 0.2296711191534996,
      "num_tokens": 125868879.0,
      "step": 67510
    },
    {
      "entropy": 6.032116460800171,
      "epoch": 5.9621158601200985,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021042502681877195,
      "loss": 5.1834,
      "mean_token_accuracy": 0.23345595151185988,
      "num_tokens": 125878112.0,
      "step": 67515
    },
    {
      "entropy": 6.092068815231324,
      "epoch": 5.9625574002119395,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002103948607734233,
      "loss": 5.341,
      "mean_token_accuracy": 0.21355045586824417,
      "num_tokens": 125887985.0,
      "step": 67520
    },
    {
      "entropy": 6.019366931915283,
      "epoch": 5.9629989403037795,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00021036469599365254,
      "loss": 5.2866,
      "mean_token_accuracy": 0.2247445300221443,
      "num_tokens": 125897161.0,
      "step": 67525
    },
    {
      "entropy": 6.094869375228882,
      "epoch": 5.96344048039562,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021033453248005057,
      "loss": 5.2411,
      "mean_token_accuracy": 0.21983564049005508,
      "num_tokens": 125906324.0,
      "step": 67530
    },
    {
      "entropy": 6.022242259979248,
      "epoch": 5.9638820204874605,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00021030437023320843,
      "loss": 5.2006,
      "mean_token_accuracy": 0.22985812723636628,
      "num_tokens": 125915450.0,
      "step": 67535
    },
    {
      "entropy": 6.079084873199463,
      "epoch": 5.964323560579301,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00021027420925371683,
      "loss": 5.2324,
      "mean_token_accuracy": 0.22244890928268432,
      "num_tokens": 125924064.0,
      "step": 67540
    },
    {
      "entropy": 6.115929365158081,
      "epoch": 5.964765100671141,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002102440495421667,
      "loss": 5.2136,
      "mean_token_accuracy": 0.2314988076686859,
      "num_tokens": 125933573.0,
      "step": 67545
    },
    {
      "entropy": 6.057673406600952,
      "epoch": 5.965206640762982,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021021389109914868,
      "loss": 5.2095,
      "mean_token_accuracy": 0.2223620221018791,
      "num_tokens": 125942316.0,
      "step": 67550
    },
    {
      "entropy": 6.01735258102417,
      "epoch": 5.965648180854822,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002101837339252538,
      "loss": 5.2049,
      "mean_token_accuracy": 0.23070079684257508,
      "num_tokens": 125951885.0,
      "step": 67555
    },
    {
      "entropy": 5.945865631103516,
      "epoch": 5.966089720946662,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00021015357802107266,
      "loss": 5.1267,
      "mean_token_accuracy": 0.23457015603780745,
      "num_tokens": 125960569.0,
      "step": 67560
    },
    {
      "entropy": 6.002192735671997,
      "epoch": 5.966531261038503,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002101234233871961,
      "loss": 5.1997,
      "mean_token_accuracy": 0.23349427580833435,
      "num_tokens": 125969758.0,
      "step": 67565
    },
    {
      "entropy": 6.0150322914123535,
      "epoch": 5.966972801130343,
      "grad_norm": 1.515625,
      "learning_rate": 0.00021009327002421475,
      "loss": 5.1872,
      "mean_token_accuracy": 0.22886717021465303,
      "num_tokens": 125978851.0,
      "step": 67570
    },
    {
      "entropy": 6.0282618522644045,
      "epoch": 5.967414341222183,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002100631179327193,
      "loss": 5.1844,
      "mean_token_accuracy": 0.2235430583357811,
      "num_tokens": 125987798.0,
      "step": 67575
    },
    {
      "entropy": 6.035247135162353,
      "epoch": 5.967855881314024,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002100329671133005,
      "loss": 5.2437,
      "mean_token_accuracy": 0.22373770773410798,
      "num_tokens": 125997094.0,
      "step": 67580
    },
    {
      "entropy": 6.04066162109375,
      "epoch": 5.968297421405864,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021000281756654892,
      "loss": 5.2074,
      "mean_token_accuracy": 0.22840662598609923,
      "num_tokens": 126007480.0,
      "step": 67585
    },
    {
      "entropy": 6.137254095077514,
      "epoch": 5.968738961497704,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002099726692930552,
      "loss": 5.3439,
      "mean_token_accuracy": 0.21637780368328094,
      "num_tokens": 126016600.0,
      "step": 67590
    },
    {
      "entropy": 6.019260692596435,
      "epoch": 5.969180501589545,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020994252229340976,
      "loss": 5.1574,
      "mean_token_accuracy": 0.22959651201963424,
      "num_tokens": 126025996.0,
      "step": 67595
    },
    {
      "entropy": 6.0563356399536135,
      "epoch": 5.969622041681385,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020991237656820344,
      "loss": 5.1755,
      "mean_token_accuracy": 0.2339174285531044,
      "num_tokens": 126034554.0,
      "step": 67600
    },
    {
      "entropy": 6.080195713043213,
      "epoch": 5.970063581773225,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002098822321180266,
      "loss": 5.1924,
      "mean_token_accuracy": 0.2345072478055954,
      "num_tokens": 126043847.0,
      "step": 67605
    },
    {
      "entropy": 6.039619970321655,
      "epoch": 5.970505121865065,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020985208894346987,
      "loss": 5.3087,
      "mean_token_accuracy": 0.20964305698871613,
      "num_tokens": 126053019.0,
      "step": 67610
    },
    {
      "entropy": 6.047851657867431,
      "epoch": 5.970946661956906,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020982194704512354,
      "loss": 5.198,
      "mean_token_accuracy": 0.22473544776439666,
      "num_tokens": 126062279.0,
      "step": 67615
    },
    {
      "entropy": 6.069180536270141,
      "epoch": 5.971388202048746,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002097918064235783,
      "loss": 5.22,
      "mean_token_accuracy": 0.2233905702829361,
      "num_tokens": 126071827.0,
      "step": 67620
    },
    {
      "entropy": 6.08094630241394,
      "epoch": 5.971829742140587,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002097616670794244,
      "loss": 5.2497,
      "mean_token_accuracy": 0.2272494152188301,
      "num_tokens": 126081853.0,
      "step": 67625
    },
    {
      "entropy": 6.014058780670166,
      "epoch": 5.972271282232427,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020973152901325242,
      "loss": 5.1464,
      "mean_token_accuracy": 0.2310980573296547,
      "num_tokens": 126091084.0,
      "step": 67630
    },
    {
      "entropy": 5.951195621490479,
      "epoch": 5.972712822324267,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002097013922256526,
      "loss": 5.1193,
      "mean_token_accuracy": 0.24008859544992447,
      "num_tokens": 126100584.0,
      "step": 67635
    },
    {
      "entropy": 6.0842094898223875,
      "epoch": 5.973154362416107,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020967125671721536,
      "loss": 5.3259,
      "mean_token_accuracy": 0.21551880985498428,
      "num_tokens": 126110866.0,
      "step": 67640
    },
    {
      "entropy": 6.073423290252686,
      "epoch": 5.973595902507948,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020964112248853117,
      "loss": 5.3405,
      "mean_token_accuracy": 0.21800926923751832,
      "num_tokens": 126119948.0,
      "step": 67645
    },
    {
      "entropy": 6.150590085983277,
      "epoch": 5.974037442599788,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020961098954019015,
      "loss": 5.308,
      "mean_token_accuracy": 0.21315153688192368,
      "num_tokens": 126128916.0,
      "step": 67650
    },
    {
      "entropy": 6.12711272239685,
      "epoch": 5.974478982691628,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002095808578727827,
      "loss": 5.2334,
      "mean_token_accuracy": 0.21675649285316467,
      "num_tokens": 126137917.0,
      "step": 67655
    },
    {
      "entropy": 6.067298221588135,
      "epoch": 5.974920522783469,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002095507274868989,
      "loss": 5.238,
      "mean_token_accuracy": 0.2243778556585312,
      "num_tokens": 126147419.0,
      "step": 67660
    },
    {
      "entropy": 6.066959714889526,
      "epoch": 5.975362062875309,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002095205983831292,
      "loss": 5.2567,
      "mean_token_accuracy": 0.22025093138217927,
      "num_tokens": 126156945.0,
      "step": 67665
    },
    {
      "entropy": 6.022540473937989,
      "epoch": 5.975803602967149,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020949047056206376,
      "loss": 5.1777,
      "mean_token_accuracy": 0.23173590153455734,
      "num_tokens": 126166307.0,
      "step": 67670
    },
    {
      "entropy": 6.052701854705811,
      "epoch": 5.97624514305899,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020946034402429282,
      "loss": 5.2027,
      "mean_token_accuracy": 0.22490302920341493,
      "num_tokens": 126175117.0,
      "step": 67675
    },
    {
      "entropy": 6.129659748077392,
      "epoch": 5.97668668315083,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002094302187704063,
      "loss": 5.3635,
      "mean_token_accuracy": 0.21826937198638915,
      "num_tokens": 126183693.0,
      "step": 67680
    },
    {
      "entropy": 6.127540826797485,
      "epoch": 5.97712822324267,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020940009480099458,
      "loss": 5.3143,
      "mean_token_accuracy": 0.21409834623336793,
      "num_tokens": 126193718.0,
      "step": 67685
    },
    {
      "entropy": 6.035837459564209,
      "epoch": 5.977569763334511,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002093699721166477,
      "loss": 5.1196,
      "mean_token_accuracy": 0.23587304800748826,
      "num_tokens": 126203833.0,
      "step": 67690
    },
    {
      "entropy": 6.045893430709839,
      "epoch": 5.978011303426351,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020933985071795575,
      "loss": 5.2087,
      "mean_token_accuracy": 0.22522293627262116,
      "num_tokens": 126213636.0,
      "step": 67695
    },
    {
      "entropy": 6.115929222106933,
      "epoch": 5.978452843518191,
      "grad_norm": 1.375,
      "learning_rate": 0.00020930973060550868,
      "loss": 5.3402,
      "mean_token_accuracy": 0.21180927902460098,
      "num_tokens": 126223624.0,
      "step": 67700
    },
    {
      "entropy": 6.065137577056885,
      "epoch": 5.978894383610032,
      "grad_norm": 1.453125,
      "learning_rate": 0.00020927961177989675,
      "loss": 5.2242,
      "mean_token_accuracy": 0.22359518855810165,
      "num_tokens": 126232852.0,
      "step": 67705
    },
    {
      "entropy": 6.092230176925659,
      "epoch": 5.979335923701872,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020924949424170982,
      "loss": 5.3049,
      "mean_token_accuracy": 0.22026660442352294,
      "num_tokens": 126242294.0,
      "step": 67710
    },
    {
      "entropy": 6.042815923690796,
      "epoch": 5.979777463793712,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020921937799153796,
      "loss": 5.1816,
      "mean_token_accuracy": 0.22742441147565842,
      "num_tokens": 126251347.0,
      "step": 67715
    },
    {
      "entropy": 6.025171947479248,
      "epoch": 5.980219003885553,
      "grad_norm": 1.375,
      "learning_rate": 0.000209189263029971,
      "loss": 5.2524,
      "mean_token_accuracy": 0.22681159973144532,
      "num_tokens": 126260157.0,
      "step": 67720
    },
    {
      "entropy": 6.0994486808776855,
      "epoch": 5.980660543977393,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002091591493575989,
      "loss": 5.2308,
      "mean_token_accuracy": 0.22740215957164764,
      "num_tokens": 126268653.0,
      "step": 67725
    },
    {
      "entropy": 6.131260347366333,
      "epoch": 5.981102084069233,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020912903697501169,
      "loss": 5.3115,
      "mean_token_accuracy": 0.21992556303739547,
      "num_tokens": 126278398.0,
      "step": 67730
    },
    {
      "entropy": 6.094602870941162,
      "epoch": 5.981543624161074,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002090989258827992,
      "loss": 5.2545,
      "mean_token_accuracy": 0.22347903102636338,
      "num_tokens": 126288120.0,
      "step": 67735
    },
    {
      "entropy": 5.943430376052857,
      "epoch": 5.981985164252914,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002090688160815512,
      "loss": 5.13,
      "mean_token_accuracy": 0.22524355798959733,
      "num_tokens": 126297288.0,
      "step": 67740
    },
    {
      "entropy": 6.039941358566284,
      "epoch": 5.982426704344754,
      "grad_norm": 1.28125,
      "learning_rate": 0.00020903870757185767,
      "loss": 5.2486,
      "mean_token_accuracy": 0.22498960345983504,
      "num_tokens": 126306578.0,
      "step": 67745
    },
    {
      "entropy": 6.012269115447998,
      "epoch": 5.982868244436595,
      "grad_norm": 1.375,
      "learning_rate": 0.0002090086003543083,
      "loss": 5.1836,
      "mean_token_accuracy": 0.23133722692728043,
      "num_tokens": 126316074.0,
      "step": 67750
    },
    {
      "entropy": 6.077460145950317,
      "epoch": 5.983309784528435,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020897849442949292,
      "loss": 5.2357,
      "mean_token_accuracy": 0.218405120074749,
      "num_tokens": 126325953.0,
      "step": 67755
    },
    {
      "entropy": 6.00268235206604,
      "epoch": 5.983751324620275,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020894838979800134,
      "loss": 5.1152,
      "mean_token_accuracy": 0.23228270709514617,
      "num_tokens": 126334615.0,
      "step": 67760
    },
    {
      "entropy": 6.142542934417724,
      "epoch": 5.984192864712115,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020891828646042317,
      "loss": 5.3167,
      "mean_token_accuracy": 0.2128770485520363,
      "num_tokens": 126343583.0,
      "step": 67765
    },
    {
      "entropy": 6.034371709823608,
      "epoch": 5.984634404803956,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020888818441734813,
      "loss": 5.1999,
      "mean_token_accuracy": 0.2251880168914795,
      "num_tokens": 126353252.0,
      "step": 67770
    },
    {
      "entropy": 6.044303894042969,
      "epoch": 5.985075944895796,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020885808366936608,
      "loss": 5.2088,
      "mean_token_accuracy": 0.2285463348031044,
      "num_tokens": 126362840.0,
      "step": 67775
    },
    {
      "entropy": 6.062043905258179,
      "epoch": 5.985517484987637,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020882798421706657,
      "loss": 5.1385,
      "mean_token_accuracy": 0.23342652320861818,
      "num_tokens": 126371501.0,
      "step": 67780
    },
    {
      "entropy": 6.140240144729614,
      "epoch": 5.985959025079477,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00020879788606103917,
      "loss": 5.2792,
      "mean_token_accuracy": 0.22023598551750184,
      "num_tokens": 126381700.0,
      "step": 67785
    },
    {
      "entropy": 6.088572025299072,
      "epoch": 5.9864005651713175,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020876778920187345,
      "loss": 5.2428,
      "mean_token_accuracy": 0.22370459884405136,
      "num_tokens": 126391159.0,
      "step": 67790
    },
    {
      "entropy": 6.1106232643127445,
      "epoch": 5.9868421052631575,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020873769364015916,
      "loss": 5.2952,
      "mean_token_accuracy": 0.2184867814183235,
      "num_tokens": 126400302.0,
      "step": 67795
    },
    {
      "entropy": 5.9785926818847654,
      "epoch": 5.9872836453549985,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020870759937648576,
      "loss": 5.0939,
      "mean_token_accuracy": 0.23790573626756667,
      "num_tokens": 126409594.0,
      "step": 67800
    },
    {
      "entropy": 6.145056915283203,
      "epoch": 5.9877251854468385,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020867750641144272,
      "loss": 5.2164,
      "mean_token_accuracy": 0.22248342931270598,
      "num_tokens": 126418572.0,
      "step": 67805
    },
    {
      "entropy": 6.047995662689209,
      "epoch": 5.988166725538679,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020864741474561967,
      "loss": 5.1432,
      "mean_token_accuracy": 0.23048193454742433,
      "num_tokens": 126427515.0,
      "step": 67810
    },
    {
      "entropy": 6.064563035964966,
      "epoch": 5.9886082656305195,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020861732437960607,
      "loss": 5.256,
      "mean_token_accuracy": 0.21567910313606262,
      "num_tokens": 126436684.0,
      "step": 67815
    },
    {
      "entropy": 5.979529857635498,
      "epoch": 5.98904980572236,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002085872353139912,
      "loss": 5.1899,
      "mean_token_accuracy": 0.2317643240094185,
      "num_tokens": 126446272.0,
      "step": 67820
    },
    {
      "entropy": 6.052448844909668,
      "epoch": 5.9894913458142,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002085571475493648,
      "loss": 5.2051,
      "mean_token_accuracy": 0.2267799198627472,
      "num_tokens": 126455372.0,
      "step": 67825
    },
    {
      "entropy": 6.111840724945068,
      "epoch": 5.989932885906041,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020852706108631598,
      "loss": 5.2697,
      "mean_token_accuracy": 0.21790335923433304,
      "num_tokens": 126465387.0,
      "step": 67830
    },
    {
      "entropy": 6.142586708068848,
      "epoch": 5.990374425997881,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020849697592543415,
      "loss": 5.3658,
      "mean_token_accuracy": 0.21167051047086716,
      "num_tokens": 126476188.0,
      "step": 67835
    },
    {
      "entropy": 6.0762206554412845,
      "epoch": 5.990815966089721,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020846689206730884,
      "loss": 5.2223,
      "mean_token_accuracy": 0.2286590412259102,
      "num_tokens": 126485183.0,
      "step": 67840
    },
    {
      "entropy": 6.063307809829712,
      "epoch": 5.991257506181562,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002084368095125293,
      "loss": 5.1611,
      "mean_token_accuracy": 0.2276762306690216,
      "num_tokens": 126494444.0,
      "step": 67845
    },
    {
      "entropy": 6.007392883300781,
      "epoch": 5.991699046273402,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020840672826168477,
      "loss": 5.1393,
      "mean_token_accuracy": 0.2318261221051216,
      "num_tokens": 126503840.0,
      "step": 67850
    },
    {
      "entropy": 6.077895164489746,
      "epoch": 5.992140586365242,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020837664831536453,
      "loss": 5.2197,
      "mean_token_accuracy": 0.2249983325600624,
      "num_tokens": 126512939.0,
      "step": 67855
    },
    {
      "entropy": 6.006305122375489,
      "epoch": 5.992582126457083,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020834656967415788,
      "loss": 5.1664,
      "mean_token_accuracy": 0.2267332434654236,
      "num_tokens": 126522070.0,
      "step": 67860
    },
    {
      "entropy": 6.1032304763793945,
      "epoch": 5.993023666548923,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020831649233865406,
      "loss": 5.2991,
      "mean_token_accuracy": 0.22194590419530869,
      "num_tokens": 126531036.0,
      "step": 67865
    },
    {
      "entropy": 5.958023118972778,
      "epoch": 5.993465206640763,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020828641630944218,
      "loss": 5.1778,
      "mean_token_accuracy": 0.2246905818581581,
      "num_tokens": 126540709.0,
      "step": 67870
    },
    {
      "entropy": 6.016808748245239,
      "epoch": 5.993906746732604,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020825634158711144,
      "loss": 5.1719,
      "mean_token_accuracy": 0.2320873811841011,
      "num_tokens": 126550055.0,
      "step": 67875
    },
    {
      "entropy": 5.990831136703491,
      "epoch": 5.994348286824444,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002082262681722511,
      "loss": 5.1285,
      "mean_token_accuracy": 0.23493027687072754,
      "num_tokens": 126559287.0,
      "step": 67880
    },
    {
      "entropy": 6.131988954544068,
      "epoch": 5.994789826916284,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002081961960654501,
      "loss": 5.3279,
      "mean_token_accuracy": 0.2135391876101494,
      "num_tokens": 126569054.0,
      "step": 67885
    },
    {
      "entropy": 6.087100028991699,
      "epoch": 5.995231367008124,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020816612526729773,
      "loss": 5.2618,
      "mean_token_accuracy": 0.22409045845270156,
      "num_tokens": 126578800.0,
      "step": 67890
    },
    {
      "entropy": 6.045039129257202,
      "epoch": 5.995672907099965,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002081360557783829,
      "loss": 5.229,
      "mean_token_accuracy": 0.2255491018295288,
      "num_tokens": 126588861.0,
      "step": 67895
    },
    {
      "entropy": 6.107744550704956,
      "epoch": 5.996114447191805,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020810598759929462,
      "loss": 5.2837,
      "mean_token_accuracy": 0.21602724492549896,
      "num_tokens": 126598091.0,
      "step": 67900
    },
    {
      "entropy": 6.0432703495025635,
      "epoch": 5.996555987283646,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002080759207306221,
      "loss": 5.1554,
      "mean_token_accuracy": 0.23324740380048753,
      "num_tokens": 126606243.0,
      "step": 67905
    },
    {
      "entropy": 6.048876857757568,
      "epoch": 5.996997527375486,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020804585517295428,
      "loss": 5.2344,
      "mean_token_accuracy": 0.2184666484594345,
      "num_tokens": 126616222.0,
      "step": 67910
    },
    {
      "entropy": 6.109847068786621,
      "epoch": 5.997439067467326,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020801579092688,
      "loss": 5.3088,
      "mean_token_accuracy": 0.21570498943328859,
      "num_tokens": 126626165.0,
      "step": 67915
    },
    {
      "entropy": 6.093162393569946,
      "epoch": 5.997880607559166,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002079857279929882,
      "loss": 5.2571,
      "mean_token_accuracy": 0.22210447937250138,
      "num_tokens": 126634267.0,
      "step": 67920
    },
    {
      "entropy": 6.12260251045227,
      "epoch": 5.998322147651007,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020795566637186802,
      "loss": 5.2687,
      "mean_token_accuracy": 0.22337078154087067,
      "num_tokens": 126644718.0,
      "step": 67925
    },
    {
      "entropy": 6.075731515884399,
      "epoch": 5.998763687742847,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020792560606410816,
      "loss": 5.1614,
      "mean_token_accuracy": 0.23674359619617463,
      "num_tokens": 126653495.0,
      "step": 67930
    },
    {
      "entropy": 6.0936798572540285,
      "epoch": 5.999205227834687,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002078955470702975,
      "loss": 5.2665,
      "mean_token_accuracy": 0.22156008332967758,
      "num_tokens": 126663316.0,
      "step": 67935
    },
    {
      "entropy": 6.0158350467681885,
      "epoch": 5.999646767926528,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002078654893910249,
      "loss": 5.1782,
      "mean_token_accuracy": 0.23129387199878693,
      "num_tokens": 126672606.0,
      "step": 67940
    },
    {
      "entropy": 6.069815158843994,
      "epoch": 6.000088308018368,
      "grad_norm": 1.453125,
      "learning_rate": 0.00020783543302687924,
      "loss": 5.1825,
      "mean_token_accuracy": 0.22823770940303803,
      "num_tokens": 126680306.0,
      "step": 67945
    },
    {
      "entropy": 6.032050609588623,
      "epoch": 6.000529848110208,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00020780537797844917,
      "loss": 5.2163,
      "mean_token_accuracy": 0.23180420249700545,
      "num_tokens": 126689500.0,
      "step": 67950
    },
    {
      "entropy": 6.052932357788086,
      "epoch": 6.000971388202049,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00020777532424632366,
      "loss": 5.1513,
      "mean_token_accuracy": 0.2326548933982849,
      "num_tokens": 126698642.0,
      "step": 67955
    },
    {
      "entropy": 6.037668323516845,
      "epoch": 6.001412928293889,
      "grad_norm": 1.5,
      "learning_rate": 0.00020774527183109126,
      "loss": 5.1229,
      "mean_token_accuracy": 0.2332853779196739,
      "num_tokens": 126707713.0,
      "step": 67960
    },
    {
      "entropy": 6.094077062606812,
      "epoch": 6.001854468385729,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002077152207333406,
      "loss": 5.1398,
      "mean_token_accuracy": 0.2350960686802864,
      "num_tokens": 126717952.0,
      "step": 67965
    },
    {
      "entropy": 6.055530118942261,
      "epoch": 6.00229600847757,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00020768517095366064,
      "loss": 5.2034,
      "mean_token_accuracy": 0.23212554156780243,
      "num_tokens": 126727702.0,
      "step": 67970
    },
    {
      "entropy": 6.096760702133179,
      "epoch": 6.00273754856941,
      "grad_norm": 1.5,
      "learning_rate": 0.0002076551224926399,
      "loss": 5.1468,
      "mean_token_accuracy": 0.23952223658561705,
      "num_tokens": 126736283.0,
      "step": 67975
    },
    {
      "entropy": 6.0543187141418455,
      "epoch": 6.00317908866125,
      "grad_norm": 1.390625,
      "learning_rate": 0.000207625075350867,
      "loss": 5.2012,
      "mean_token_accuracy": 0.22499166131019593,
      "num_tokens": 126745352.0,
      "step": 67980
    },
    {
      "entropy": 6.0221946239471436,
      "epoch": 6.003620628753091,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00020759502952893044,
      "loss": 5.1918,
      "mean_token_accuracy": 0.22829943001270295,
      "num_tokens": 126754280.0,
      "step": 67985
    },
    {
      "entropy": 5.987802600860595,
      "epoch": 6.004062168844931,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020756498502741904,
      "loss": 5.1643,
      "mean_token_accuracy": 0.22124979943037032,
      "num_tokens": 126763583.0,
      "step": 67990
    },
    {
      "entropy": 6.03144063949585,
      "epoch": 6.004503708936771,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020753494184692122,
      "loss": 5.2151,
      "mean_token_accuracy": 0.2275724396109581,
      "num_tokens": 126771708.0,
      "step": 67995
    },
    {
      "entropy": 5.9723186016082765,
      "epoch": 6.004945249028612,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020750489998802547,
      "loss": 5.118,
      "mean_token_accuracy": 0.23736486881971358,
      "num_tokens": 126780833.0,
      "step": 68000
    },
    {
      "entropy": 6.112297868728637,
      "epoch": 6.005386789120452,
      "grad_norm": 1.5,
      "learning_rate": 0.0002074748594513204,
      "loss": 5.2667,
      "mean_token_accuracy": 0.2175741821527481,
      "num_tokens": 126790836.0,
      "step": 68005
    },
    {
      "entropy": 5.996189212799072,
      "epoch": 6.005828329212292,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002074448202373944,
      "loss": 5.1448,
      "mean_token_accuracy": 0.23454396426677704,
      "num_tokens": 126800158.0,
      "step": 68010
    },
    {
      "entropy": 5.999194526672364,
      "epoch": 6.006269869304133,
      "grad_norm": 1.375,
      "learning_rate": 0.00020741478234683592,
      "loss": 5.1562,
      "mean_token_accuracy": 0.2355152651667595,
      "num_tokens": 126809374.0,
      "step": 68015
    },
    {
      "entropy": 6.0569299221038815,
      "epoch": 6.006711409395973,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020738474578023349,
      "loss": 5.2338,
      "mean_token_accuracy": 0.2303272530436516,
      "num_tokens": 126819688.0,
      "step": 68020
    },
    {
      "entropy": 6.112611103057861,
      "epoch": 6.007152949487813,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002073547105381754,
      "loss": 5.22,
      "mean_token_accuracy": 0.21946553140878677,
      "num_tokens": 126828975.0,
      "step": 68025
    },
    {
      "entropy": 6.0242808818817135,
      "epoch": 6.007594489579654,
      "grad_norm": 1.53125,
      "learning_rate": 0.00020732467662125,
      "loss": 5.161,
      "mean_token_accuracy": 0.22727785259485245,
      "num_tokens": 126837537.0,
      "step": 68030
    },
    {
      "entropy": 6.03239951133728,
      "epoch": 6.008036029671494,
      "grad_norm": 1.375,
      "learning_rate": 0.00020729464403004572,
      "loss": 5.2043,
      "mean_token_accuracy": 0.22434238195419312,
      "num_tokens": 126847030.0,
      "step": 68035
    },
    {
      "entropy": 5.980032444000244,
      "epoch": 6.008477569763334,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002072646127651509,
      "loss": 5.1151,
      "mean_token_accuracy": 0.23579881489276885,
      "num_tokens": 126856361.0,
      "step": 68040
    },
    {
      "entropy": 6.033471870422363,
      "epoch": 6.008919109855174,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002072345828271538,
      "loss": 5.1111,
      "mean_token_accuracy": 0.23470052182674409,
      "num_tokens": 126865363.0,
      "step": 68045
    },
    {
      "entropy": 5.9871619701385494,
      "epoch": 6.009360649947015,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020720455421664263,
      "loss": 5.1473,
      "mean_token_accuracy": 0.23921744376420975,
      "num_tokens": 126874934.0,
      "step": 68050
    },
    {
      "entropy": 6.031972646713257,
      "epoch": 6.009802190038855,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020717452693420573,
      "loss": 5.1374,
      "mean_token_accuracy": 0.229043547809124,
      "num_tokens": 126883456.0,
      "step": 68055
    },
    {
      "entropy": 6.043228435516357,
      "epoch": 6.0102437301306955,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00020714450098043125,
      "loss": 5.1796,
      "mean_token_accuracy": 0.23901190608739853,
      "num_tokens": 126892794.0,
      "step": 68060
    },
    {
      "entropy": 6.038504362106323,
      "epoch": 6.010685270222536,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0002071144763559074,
      "loss": 5.2017,
      "mean_token_accuracy": 0.2234383627772331,
      "num_tokens": 126901550.0,
      "step": 68065
    },
    {
      "entropy": 6.113057708740234,
      "epoch": 6.0111268103143765,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00020708445306122242,
      "loss": 5.2922,
      "mean_token_accuracy": 0.2178441882133484,
      "num_tokens": 126911635.0,
      "step": 68070
    },
    {
      "entropy": 6.0660346984863285,
      "epoch": 6.0115683504062165,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002070544310969642,
      "loss": 5.185,
      "mean_token_accuracy": 0.23154344856739045,
      "num_tokens": 126921086.0,
      "step": 68075
    },
    {
      "entropy": 5.982894659042358,
      "epoch": 6.0120098904980575,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00020702441046372116,
      "loss": 5.118,
      "mean_token_accuracy": 0.23719456046819687,
      "num_tokens": 126930781.0,
      "step": 68080
    },
    {
      "entropy": 5.9716063976287845,
      "epoch": 6.0124514305898975,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002069943911620813,
      "loss": 5.1161,
      "mean_token_accuracy": 0.23875040858983992,
      "num_tokens": 126940289.0,
      "step": 68085
    },
    {
      "entropy": 6.003564834594727,
      "epoch": 6.012892970681738,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002069643731926326,
      "loss": 5.17,
      "mean_token_accuracy": 0.2272258847951889,
      "num_tokens": 126949432.0,
      "step": 68090
    },
    {
      "entropy": 6.047569942474365,
      "epoch": 6.0133345107735785,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002069343565559631,
      "loss": 5.1896,
      "mean_token_accuracy": 0.23470554798841475,
      "num_tokens": 126958404.0,
      "step": 68095
    },
    {
      "entropy": 6.0899986743927,
      "epoch": 6.0137760508654186,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002069043412526609,
      "loss": 5.2915,
      "mean_token_accuracy": 0.2252212256193161,
      "num_tokens": 126968730.0,
      "step": 68100
    },
    {
      "entropy": 6.018439197540284,
      "epoch": 6.014217590957259,
      "grad_norm": 1.375,
      "learning_rate": 0.0002068743272833139,
      "loss": 5.0712,
      "mean_token_accuracy": 0.24111240804195405,
      "num_tokens": 126978271.0,
      "step": 68105
    },
    {
      "entropy": 6.003443241119385,
      "epoch": 6.0146591310490995,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002068443146485101,
      "loss": 5.136,
      "mean_token_accuracy": 0.23213237524032593,
      "num_tokens": 126986594.0,
      "step": 68110
    },
    {
      "entropy": 6.045358276367187,
      "epoch": 6.01510067114094,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020681430334883733,
      "loss": 5.1888,
      "mean_token_accuracy": 0.2229669749736786,
      "num_tokens": 126996803.0,
      "step": 68115
    },
    {
      "entropy": 5.9991552352905275,
      "epoch": 6.01554221123278,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002067842933848837,
      "loss": 5.1294,
      "mean_token_accuracy": 0.23693986237049103,
      "num_tokens": 127005956.0,
      "step": 68120
    },
    {
      "entropy": 6.069217491149902,
      "epoch": 6.015983751324621,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002067542847572369,
      "loss": 5.2712,
      "mean_token_accuracy": 0.22057513892650604,
      "num_tokens": 127016354.0,
      "step": 68125
    },
    {
      "entropy": 6.020267677307129,
      "epoch": 6.016425291416461,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020672427746648485,
      "loss": 5.1459,
      "mean_token_accuracy": 0.23187618404626847,
      "num_tokens": 127024824.0,
      "step": 68130
    },
    {
      "entropy": 6.195035600662232,
      "epoch": 6.016866831508301,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002066942715132154,
      "loss": 5.3118,
      "mean_token_accuracy": 0.22394858300685883,
      "num_tokens": 127035482.0,
      "step": 68135
    },
    {
      "entropy": 6.126449728012085,
      "epoch": 6.017308371600142,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002066642668980162,
      "loss": 5.2107,
      "mean_token_accuracy": 0.2247169002890587,
      "num_tokens": 127044096.0,
      "step": 68140
    },
    {
      "entropy": 6.032310485839844,
      "epoch": 6.017749911691982,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020663426362147532,
      "loss": 5.268,
      "mean_token_accuracy": 0.22660193592309952,
      "num_tokens": 127053817.0,
      "step": 68145
    },
    {
      "entropy": 5.989174509048462,
      "epoch": 6.018191451783822,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020660426168418028,
      "loss": 5.138,
      "mean_token_accuracy": 0.23099831789731978,
      "num_tokens": 127063125.0,
      "step": 68150
    },
    {
      "entropy": 6.111903810501099,
      "epoch": 6.018632991875663,
      "grad_norm": 1.375,
      "learning_rate": 0.0002065742610867189,
      "loss": 5.2272,
      "mean_token_accuracy": 0.22787250429391862,
      "num_tokens": 127071988.0,
      "step": 68155
    },
    {
      "entropy": 6.08404541015625,
      "epoch": 6.019074531967503,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00020654426182967867,
      "loss": 5.1833,
      "mean_token_accuracy": 0.23211510330438614,
      "num_tokens": 127080846.0,
      "step": 68160
    },
    {
      "entropy": 6.108543491363525,
      "epoch": 6.019516072059343,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020651426391364754,
      "loss": 5.1985,
      "mean_token_accuracy": 0.23140741735696793,
      "num_tokens": 127089725.0,
      "step": 68165
    },
    {
      "entropy": 6.114470911026001,
      "epoch": 6.019957612151184,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00020648426733921305,
      "loss": 5.1992,
      "mean_token_accuracy": 0.2272478684782982,
      "num_tokens": 127098220.0,
      "step": 68170
    },
    {
      "entropy": 6.022489309310913,
      "epoch": 6.020399152243024,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00020645427210696276,
      "loss": 5.1788,
      "mean_token_accuracy": 0.23258122950792312,
      "num_tokens": 127107168.0,
      "step": 68175
    },
    {
      "entropy": 6.001578760147095,
      "epoch": 6.020840692334864,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002064242782174842,
      "loss": 5.1549,
      "mean_token_accuracy": 0.2376064747571945,
      "num_tokens": 127116452.0,
      "step": 68180
    },
    {
      "entropy": 6.045471477508545,
      "epoch": 6.021282232426704,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020639428567136513,
      "loss": 5.1934,
      "mean_token_accuracy": 0.22602097392082215,
      "num_tokens": 127125804.0,
      "step": 68185
    },
    {
      "entropy": 5.969919919967651,
      "epoch": 6.021723772518545,
      "grad_norm": 1.453125,
      "learning_rate": 0.000206364294469193,
      "loss": 5.1325,
      "mean_token_accuracy": 0.2374950796365738,
      "num_tokens": 127134405.0,
      "step": 68190
    },
    {
      "entropy": 6.083370733261108,
      "epoch": 6.022165312610385,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002063343046115553,
      "loss": 5.2389,
      "mean_token_accuracy": 0.21903052181005478,
      "num_tokens": 127144746.0,
      "step": 68195
    },
    {
      "entropy": 6.083693361282348,
      "epoch": 6.022606852702225,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002063043160990395,
      "loss": 5.1953,
      "mean_token_accuracy": 0.2326147198677063,
      "num_tokens": 127153616.0,
      "step": 68200
    },
    {
      "entropy": 6.0934367179870605,
      "epoch": 6.023048392794066,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000206274328932233,
      "loss": 5.2886,
      "mean_token_accuracy": 0.22189740240573883,
      "num_tokens": 127163541.0,
      "step": 68205
    },
    {
      "entropy": 6.0255211353302,
      "epoch": 6.023489932885906,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002062443431117234,
      "loss": 5.1901,
      "mean_token_accuracy": 0.22641877681016923,
      "num_tokens": 127172652.0,
      "step": 68210
    },
    {
      "entropy": 5.923278331756592,
      "epoch": 6.023931472977746,
      "grad_norm": 1.375,
      "learning_rate": 0.00020621435863809806,
      "loss": 5.0465,
      "mean_token_accuracy": 0.24148911088705063,
      "num_tokens": 127181960.0,
      "step": 68215
    },
    {
      "entropy": 5.999250602722168,
      "epoch": 6.024373013069587,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020618437551194426,
      "loss": 5.1789,
      "mean_token_accuracy": 0.23882705569267274,
      "num_tokens": 127191440.0,
      "step": 68220
    },
    {
      "entropy": 6.009984254837036,
      "epoch": 6.024814553161427,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020615439373384938,
      "loss": 5.1175,
      "mean_token_accuracy": 0.23854991495609285,
      "num_tokens": 127199948.0,
      "step": 68225
    },
    {
      "entropy": 6.067723417282105,
      "epoch": 6.025256093253267,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002061244133044008,
      "loss": 5.2054,
      "mean_token_accuracy": 0.22366244792938234,
      "num_tokens": 127208817.0,
      "step": 68230
    },
    {
      "entropy": 6.066022443771362,
      "epoch": 6.025697633345108,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020609443422418578,
      "loss": 5.1597,
      "mean_token_accuracy": 0.22266893535852433,
      "num_tokens": 127217560.0,
      "step": 68235
    },
    {
      "entropy": 6.014392375946045,
      "epoch": 6.026139173436948,
      "grad_norm": 1.546875,
      "learning_rate": 0.00020606445649379163,
      "loss": 5.1342,
      "mean_token_accuracy": 0.23312620371580123,
      "num_tokens": 127228354.0,
      "step": 68240
    },
    {
      "entropy": 6.012200021743775,
      "epoch": 6.026580713528788,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020603448011380555,
      "loss": 5.1312,
      "mean_token_accuracy": 0.23684362769126893,
      "num_tokens": 127237250.0,
      "step": 68245
    },
    {
      "entropy": 6.044633626937866,
      "epoch": 6.027022253620629,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020600450508481477,
      "loss": 5.1844,
      "mean_token_accuracy": 0.23510342985391616,
      "num_tokens": 127246668.0,
      "step": 68250
    },
    {
      "entropy": 6.000413990020752,
      "epoch": 6.027463793712469,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002059745314074065,
      "loss": 5.1247,
      "mean_token_accuracy": 0.23658910989761353,
      "num_tokens": 127255884.0,
      "step": 68255
    },
    {
      "entropy": 5.962347793579101,
      "epoch": 6.027905333804309,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002059445590821679,
      "loss": 5.1658,
      "mean_token_accuracy": 0.22735383212566376,
      "num_tokens": 127265596.0,
      "step": 68260
    },
    {
      "entropy": 6.114947175979614,
      "epoch": 6.02834687389615,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020591458810968616,
      "loss": 5.2847,
      "mean_token_accuracy": 0.21998170912265777,
      "num_tokens": 127274584.0,
      "step": 68265
    },
    {
      "entropy": 6.074293375015259,
      "epoch": 6.02878841398799,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020588461849054823,
      "loss": 5.213,
      "mean_token_accuracy": 0.23393332958221436,
      "num_tokens": 127284982.0,
      "step": 68270
    },
    {
      "entropy": 6.0539467334747314,
      "epoch": 6.02922995407983,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020585465022534145,
      "loss": 5.1644,
      "mean_token_accuracy": 0.22883380949497223,
      "num_tokens": 127294173.0,
      "step": 68275
    },
    {
      "entropy": 6.034250926971436,
      "epoch": 6.029671494171671,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002058246833146527,
      "loss": 5.1662,
      "mean_token_accuracy": 0.22856459021568298,
      "num_tokens": 127303908.0,
      "step": 68280
    },
    {
      "entropy": 6.021513938903809,
      "epoch": 6.030113034263511,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020579471775906911,
      "loss": 5.1892,
      "mean_token_accuracy": 0.2245924949645996,
      "num_tokens": 127313279.0,
      "step": 68285
    },
    {
      "entropy": 6.034915256500244,
      "epoch": 6.030554574355351,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002057647535591775,
      "loss": 5.1423,
      "mean_token_accuracy": 0.23740660399198532,
      "num_tokens": 127321602.0,
      "step": 68290
    },
    {
      "entropy": 5.992421388626099,
      "epoch": 6.030996114447192,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002057347907155651,
      "loss": 5.0629,
      "mean_token_accuracy": 0.24042427390813828,
      "num_tokens": 127329816.0,
      "step": 68295
    },
    {
      "entropy": 6.035216999053955,
      "epoch": 6.031437654539032,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020570482922881877,
      "loss": 5.2229,
      "mean_token_accuracy": 0.22637438774108887,
      "num_tokens": 127339408.0,
      "step": 68300
    },
    {
      "entropy": 6.0214362144470215,
      "epoch": 6.031879194630872,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020567486909952543,
      "loss": 5.2221,
      "mean_token_accuracy": 0.22748701870441437,
      "num_tokens": 127349310.0,
      "step": 68305
    },
    {
      "entropy": 6.051425743103027,
      "epoch": 6.032320734722713,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020564491032827194,
      "loss": 5.1587,
      "mean_token_accuracy": 0.23037364929914475,
      "num_tokens": 127358306.0,
      "step": 68310
    },
    {
      "entropy": 6.063794136047363,
      "epoch": 6.032762274814553,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002056149529156452,
      "loss": 5.2308,
      "mean_token_accuracy": 0.22622910588979722,
      "num_tokens": 127368832.0,
      "step": 68315
    },
    {
      "entropy": 6.018553066253662,
      "epoch": 6.033203814906393,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020558499686223209,
      "loss": 5.1823,
      "mean_token_accuracy": 0.23201112002134322,
      "num_tokens": 127378797.0,
      "step": 68320
    },
    {
      "entropy": 6.002047061920166,
      "epoch": 6.033645354998233,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020555504216861938,
      "loss": 5.118,
      "mean_token_accuracy": 0.23253829181194305,
      "num_tokens": 127387916.0,
      "step": 68325
    },
    {
      "entropy": 5.999107503890992,
      "epoch": 6.034086895090074,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020552508883539395,
      "loss": 5.2294,
      "mean_token_accuracy": 0.22009876668453215,
      "num_tokens": 127397870.0,
      "step": 68330
    },
    {
      "entropy": 5.923299360275268,
      "epoch": 6.034528435181914,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002054951368631424,
      "loss": 5.0148,
      "mean_token_accuracy": 0.25034678876399996,
      "num_tokens": 127408683.0,
      "step": 68335
    },
    {
      "entropy": 6.078214120864868,
      "epoch": 6.0349699752737545,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002054651862524517,
      "loss": 5.2805,
      "mean_token_accuracy": 0.2213786870241165,
      "num_tokens": 127418910.0,
      "step": 68340
    },
    {
      "entropy": 6.039480400085449,
      "epoch": 6.035411515365595,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020543523700390848,
      "loss": 5.1474,
      "mean_token_accuracy": 0.2330731749534607,
      "num_tokens": 127428064.0,
      "step": 68345
    },
    {
      "entropy": 6.120043087005615,
      "epoch": 6.0358530554574354,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020540528911809936,
      "loss": 5.3053,
      "mean_token_accuracy": 0.22754499316215515,
      "num_tokens": 127438782.0,
      "step": 68350
    },
    {
      "entropy": 6.113601732254028,
      "epoch": 6.0362945955492755,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020537534259561102,
      "loss": 5.2136,
      "mean_token_accuracy": 0.21490430682897568,
      "num_tokens": 127447994.0,
      "step": 68355
    },
    {
      "entropy": 6.099669313430786,
      "epoch": 6.036736135641116,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020534539743703012,
      "loss": 5.2438,
      "mean_token_accuracy": 0.22272645831108093,
      "num_tokens": 127456910.0,
      "step": 68360
    },
    {
      "entropy": 6.013243961334228,
      "epoch": 6.0371776757329565,
      "grad_norm": 1.640625,
      "learning_rate": 0.00020531545364294335,
      "loss": 5.1692,
      "mean_token_accuracy": 0.23222353905439377,
      "num_tokens": 127466248.0,
      "step": 68365
    },
    {
      "entropy": 6.107167053222656,
      "epoch": 6.0376192158247965,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002052855112139372,
      "loss": 5.2505,
      "mean_token_accuracy": 0.22297683507204055,
      "num_tokens": 127475256.0,
      "step": 68370
    },
    {
      "entropy": 6.072658538818359,
      "epoch": 6.0380607559166375,
      "grad_norm": 1.453125,
      "learning_rate": 0.00020525557015059815,
      "loss": 5.2138,
      "mean_token_accuracy": 0.2207595646381378,
      "num_tokens": 127483939.0,
      "step": 68375
    },
    {
      "entropy": 6.062266635894775,
      "epoch": 6.0385022960084775,
      "grad_norm": 1.375,
      "learning_rate": 0.00020522563045351284,
      "loss": 5.2986,
      "mean_token_accuracy": 0.21610200107097627,
      "num_tokens": 127493711.0,
      "step": 68380
    },
    {
      "entropy": 6.007531452178955,
      "epoch": 6.038943836100318,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020519569212326777,
      "loss": 5.1156,
      "mean_token_accuracy": 0.23688159435987471,
      "num_tokens": 127503610.0,
      "step": 68385
    },
    {
      "entropy": 5.979927825927734,
      "epoch": 6.0393853761921585,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020516575516044938,
      "loss": 5.114,
      "mean_token_accuracy": 0.23918548971414566,
      "num_tokens": 127513035.0,
      "step": 68390
    },
    {
      "entropy": 6.034308815002442,
      "epoch": 6.039826916283999,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020513581956564414,
      "loss": 5.219,
      "mean_token_accuracy": 0.23067689090967178,
      "num_tokens": 127522817.0,
      "step": 68395
    },
    {
      "entropy": 6.077077627182007,
      "epoch": 6.040268456375839,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020510588533943842,
      "loss": 5.2683,
      "mean_token_accuracy": 0.2191936954855919,
      "num_tokens": 127532229.0,
      "step": 68400
    },
    {
      "entropy": 6.02531681060791,
      "epoch": 6.04070999646768,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020507595248241866,
      "loss": 5.157,
      "mean_token_accuracy": 0.23083046674728394,
      "num_tokens": 127540970.0,
      "step": 68405
    },
    {
      "entropy": 5.997530794143676,
      "epoch": 6.04115153655952,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00020504602099517133,
      "loss": 5.1047,
      "mean_token_accuracy": 0.23809561282396316,
      "num_tokens": 127549189.0,
      "step": 68410
    },
    {
      "entropy": 5.935570192337036,
      "epoch": 6.04159307665136,
      "grad_norm": 1.375,
      "learning_rate": 0.0002050160908782826,
      "loss": 5.096,
      "mean_token_accuracy": 0.24111291617155076,
      "num_tokens": 127558780.0,
      "step": 68415
    },
    {
      "entropy": 6.056687307357788,
      "epoch": 6.042034616743201,
      "grad_norm": 1.578125,
      "learning_rate": 0.00020498616213233877,
      "loss": 5.1719,
      "mean_token_accuracy": 0.23213996589183808,
      "num_tokens": 127566904.0,
      "step": 68420
    },
    {
      "entropy": 6.1305443286895756,
      "epoch": 6.042476156835041,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002049562347579263,
      "loss": 5.2606,
      "mean_token_accuracy": 0.22298016995191575,
      "num_tokens": 127575838.0,
      "step": 68425
    },
    {
      "entropy": 5.952628660202026,
      "epoch": 6.042917696926881,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00020492630875563133,
      "loss": 5.1198,
      "mean_token_accuracy": 0.23679038733243943,
      "num_tokens": 127584251.0,
      "step": 68430
    },
    {
      "entropy": 5.975878810882568,
      "epoch": 6.043359237018722,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00020489638412604018,
      "loss": 5.1092,
      "mean_token_accuracy": 0.22971561104059218,
      "num_tokens": 127592654.0,
      "step": 68435
    },
    {
      "entropy": 6.04233570098877,
      "epoch": 6.043800777110562,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020486646086973886,
      "loss": 5.2936,
      "mean_token_accuracy": 0.21899272203445436,
      "num_tokens": 127601605.0,
      "step": 68440
    },
    {
      "entropy": 6.024550819396973,
      "epoch": 6.044242317202402,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00020483653898731375,
      "loss": 5.2308,
      "mean_token_accuracy": 0.2298437163233757,
      "num_tokens": 127610844.0,
      "step": 68445
    },
    {
      "entropy": 6.031311511993408,
      "epoch": 6.044683857294243,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020480661847935093,
      "loss": 5.1923,
      "mean_token_accuracy": 0.22894295901060105,
      "num_tokens": 127620239.0,
      "step": 68450
    },
    {
      "entropy": 6.0410693168640135,
      "epoch": 6.045125397386083,
      "grad_norm": 1.59375,
      "learning_rate": 0.00020477669934643655,
      "loss": 5.2441,
      "mean_token_accuracy": 0.2251530185341835,
      "num_tokens": 127629480.0,
      "step": 68455
    },
    {
      "entropy": 6.0792145252227785,
      "epoch": 6.045566937477923,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002047467815891567,
      "loss": 5.2042,
      "mean_token_accuracy": 0.22537995874881744,
      "num_tokens": 127638581.0,
      "step": 68460
    },
    {
      "entropy": 6.015585470199585,
      "epoch": 6.046008477569763,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020471686520809736,
      "loss": 5.1019,
      "mean_token_accuracy": 0.23409348130226135,
      "num_tokens": 127647375.0,
      "step": 68465
    },
    {
      "entropy": 6.032312870025635,
      "epoch": 6.046450017661604,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020468695020384476,
      "loss": 5.2071,
      "mean_token_accuracy": 0.21995486319065094,
      "num_tokens": 127657307.0,
      "step": 68470
    },
    {
      "entropy": 6.043305683135986,
      "epoch": 6.046891557753444,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002046570365769848,
      "loss": 5.197,
      "mean_token_accuracy": 0.22488270550966263,
      "num_tokens": 127666268.0,
      "step": 68475
    },
    {
      "entropy": 6.110389614105225,
      "epoch": 6.047333097845284,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020462712432810348,
      "loss": 5.2507,
      "mean_token_accuracy": 0.21953329741954802,
      "num_tokens": 127676113.0,
      "step": 68480
    },
    {
      "entropy": 6.124528551101685,
      "epoch": 6.047774637937125,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002045972134577868,
      "loss": 5.267,
      "mean_token_accuracy": 0.21735224425792693,
      "num_tokens": 127685565.0,
      "step": 68485
    },
    {
      "entropy": 6.0519959926605225,
      "epoch": 6.048216178028965,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020456730396662055,
      "loss": 5.1997,
      "mean_token_accuracy": 0.2230795443058014,
      "num_tokens": 127693925.0,
      "step": 68490
    },
    {
      "entropy": 6.044432210922241,
      "epoch": 6.048657718120805,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020453739585519088,
      "loss": 5.1769,
      "mean_token_accuracy": 0.22325901240110396,
      "num_tokens": 127703546.0,
      "step": 68495
    },
    {
      "entropy": 6.012968873977661,
      "epoch": 6.049099258212646,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002045074891240835,
      "loss": 5.1539,
      "mean_token_accuracy": 0.231265527009964,
      "num_tokens": 127713157.0,
      "step": 68500
    },
    {
      "entropy": 5.993966579437256,
      "epoch": 6.049540798304486,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00020447758377388432,
      "loss": 5.0868,
      "mean_token_accuracy": 0.23654120564460754,
      "num_tokens": 127722673.0,
      "step": 68505
    },
    {
      "entropy": 6.093763256072998,
      "epoch": 6.049982338396326,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002044476798051792,
      "loss": 5.1831,
      "mean_token_accuracy": 0.22702926993370057,
      "num_tokens": 127731814.0,
      "step": 68510
    },
    {
      "entropy": 6.007887649536133,
      "epoch": 6.050423878488167,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00020441777721855382,
      "loss": 5.199,
      "mean_token_accuracy": 0.22563063651323317,
      "num_tokens": 127740684.0,
      "step": 68515
    },
    {
      "entropy": 6.073254919052124,
      "epoch": 6.050865418580007,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020438787601459414,
      "loss": 5.2527,
      "mean_token_accuracy": 0.22271384298801422,
      "num_tokens": 127749832.0,
      "step": 68520
    },
    {
      "entropy": 6.0526996612548825,
      "epoch": 6.051306958671847,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0002043579761938858,
      "loss": 5.2324,
      "mean_token_accuracy": 0.22405664771795272,
      "num_tokens": 127759435.0,
      "step": 68525
    },
    {
      "entropy": 5.970176649093628,
      "epoch": 6.051748498763688,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020432807775701446,
      "loss": 5.1959,
      "mean_token_accuracy": 0.23001236617565154,
      "num_tokens": 127769053.0,
      "step": 68530
    },
    {
      "entropy": 6.036038160324097,
      "epoch": 6.052190038855528,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002042981807045659,
      "loss": 5.2136,
      "mean_token_accuracy": 0.22873952686786653,
      "num_tokens": 127778252.0,
      "step": 68535
    },
    {
      "entropy": 6.062402200698853,
      "epoch": 6.052631578947368,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002042682850371258,
      "loss": 5.2501,
      "mean_token_accuracy": 0.22149586528539658,
      "num_tokens": 127788603.0,
      "step": 68540
    },
    {
      "entropy": 6.0844504833221436,
      "epoch": 6.053073119039209,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020423839075527978,
      "loss": 5.2335,
      "mean_token_accuracy": 0.2287730947136879,
      "num_tokens": 127797660.0,
      "step": 68545
    },
    {
      "entropy": 6.126795291900635,
      "epoch": 6.053514659131049,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020420849785961347,
      "loss": 5.2252,
      "mean_token_accuracy": 0.22883328348398208,
      "num_tokens": 127807102.0,
      "step": 68550
    },
    {
      "entropy": 6.078522729873657,
      "epoch": 6.053956199222889,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020417860635071232,
      "loss": 5.2679,
      "mean_token_accuracy": 0.22274816036224365,
      "num_tokens": 127816358.0,
      "step": 68555
    },
    {
      "entropy": 6.057801198959351,
      "epoch": 6.05439773931473,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002041487162291621,
      "loss": 5.2227,
      "mean_token_accuracy": 0.22520684003829955,
      "num_tokens": 127826080.0,
      "step": 68560
    },
    {
      "entropy": 6.0367063045501705,
      "epoch": 6.05483927940657,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020411882749554817,
      "loss": 5.1679,
      "mean_token_accuracy": 0.22681148946285248,
      "num_tokens": 127835343.0,
      "step": 68565
    },
    {
      "entropy": 6.029353666305542,
      "epoch": 6.05528081949841,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002040889401504561,
      "loss": 5.1831,
      "mean_token_accuracy": 0.22840326130390168,
      "num_tokens": 127843470.0,
      "step": 68570
    },
    {
      "entropy": 6.016080093383789,
      "epoch": 6.055722359590251,
      "grad_norm": 1.5625,
      "learning_rate": 0.00020405905419447134,
      "loss": 5.2353,
      "mean_token_accuracy": 0.22325010895729064,
      "num_tokens": 127853440.0,
      "step": 68575
    },
    {
      "entropy": 6.049495887756348,
      "epoch": 6.056163899682091,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020402916962817942,
      "loss": 5.2039,
      "mean_token_accuracy": 0.22782406508922576,
      "num_tokens": 127862560.0,
      "step": 68580
    },
    {
      "entropy": 6.027737379074097,
      "epoch": 6.056605439773931,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002039992864521657,
      "loss": 5.224,
      "mean_token_accuracy": 0.22861989438533784,
      "num_tokens": 127871816.0,
      "step": 68585
    },
    {
      "entropy": 6.116654729843139,
      "epoch": 6.057046979865772,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020396940466701558,
      "loss": 5.2528,
      "mean_token_accuracy": 0.22066651731729509,
      "num_tokens": 127881687.0,
      "step": 68590
    },
    {
      "entropy": 6.082818222045899,
      "epoch": 6.057488519957612,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002039395242733143,
      "loss": 5.2121,
      "mean_token_accuracy": 0.22895529270172119,
      "num_tokens": 127891244.0,
      "step": 68595
    },
    {
      "entropy": 5.9542872428894045,
      "epoch": 6.057930060049452,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020390964527164745,
      "loss": 5.0172,
      "mean_token_accuracy": 0.24768128991127014,
      "num_tokens": 127900731.0,
      "step": 68600
    },
    {
      "entropy": 6.093715572357178,
      "epoch": 6.058371600141292,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020387976766260025,
      "loss": 5.2375,
      "mean_token_accuracy": 0.2241520792245865,
      "num_tokens": 127909594.0,
      "step": 68605
    },
    {
      "entropy": 5.974640226364135,
      "epoch": 6.058813140233133,
      "grad_norm": 1.546875,
      "learning_rate": 0.00020384989144675792,
      "loss": 5.1621,
      "mean_token_accuracy": 0.23057633489370347,
      "num_tokens": 127918552.0,
      "step": 68610
    },
    {
      "entropy": 6.04465970993042,
      "epoch": 6.059254680324973,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020382001662470577,
      "loss": 5.2576,
      "mean_token_accuracy": 0.2245867908000946,
      "num_tokens": 127929951.0,
      "step": 68615
    },
    {
      "entropy": 6.026757049560547,
      "epoch": 6.059696220416813,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020379014319702894,
      "loss": 5.142,
      "mean_token_accuracy": 0.23488947302103041,
      "num_tokens": 127939794.0,
      "step": 68620
    },
    {
      "entropy": 5.97967119216919,
      "epoch": 6.060137760508654,
      "grad_norm": 1.59375,
      "learning_rate": 0.00020376027116431273,
      "loss": 5.1032,
      "mean_token_accuracy": 0.22930233478546141,
      "num_tokens": 127948328.0,
      "step": 68625
    },
    {
      "entropy": 5.969474458694458,
      "epoch": 6.060579300600494,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00020373040052714233,
      "loss": 5.1012,
      "mean_token_accuracy": 0.2454376369714737,
      "num_tokens": 127958055.0,
      "step": 68630
    },
    {
      "entropy": 6.01692008972168,
      "epoch": 6.0610208406923345,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020370053128610278,
      "loss": 5.1201,
      "mean_token_accuracy": 0.2308169350028038,
      "num_tokens": 127967279.0,
      "step": 68635
    },
    {
      "entropy": 6.0256335735321045,
      "epoch": 6.061462380784175,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002036706634417793,
      "loss": 5.1993,
      "mean_token_accuracy": 0.22738422006368636,
      "num_tokens": 127975657.0,
      "step": 68640
    },
    {
      "entropy": 6.0281005859375,
      "epoch": 6.0619039208760155,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020364079699475697,
      "loss": 5.2632,
      "mean_token_accuracy": 0.21927433013916015,
      "num_tokens": 127985470.0,
      "step": 68645
    },
    {
      "entropy": 5.9704282760620115,
      "epoch": 6.0623454609678555,
      "grad_norm": 1.5,
      "learning_rate": 0.00020361093194562086,
      "loss": 5.0678,
      "mean_token_accuracy": 0.24285034835338593,
      "num_tokens": 127993789.0,
      "step": 68650
    },
    {
      "entropy": 6.015828084945679,
      "epoch": 6.0627870010596965,
      "grad_norm": 1.359375,
      "learning_rate": 0.000203581068294956,
      "loss": 5.1915,
      "mean_token_accuracy": 0.23184885829687119,
      "num_tokens": 128002912.0,
      "step": 68655
    },
    {
      "entropy": 6.160162353515625,
      "epoch": 6.0632285411515365,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020355120604334732,
      "loss": 5.2273,
      "mean_token_accuracy": 0.23165747970342637,
      "num_tokens": 128013171.0,
      "step": 68660
    },
    {
      "entropy": 6.14688458442688,
      "epoch": 6.063670081243377,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020352134519137995,
      "loss": 5.2772,
      "mean_token_accuracy": 0.21531120985746383,
      "num_tokens": 128022534.0,
      "step": 68665
    },
    {
      "entropy": 5.993050956726075,
      "epoch": 6.0641116213352175,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00020349148573963876,
      "loss": 5.1102,
      "mean_token_accuracy": 0.23415288627147673,
      "num_tokens": 128031621.0,
      "step": 68670
    },
    {
      "entropy": 6.0282392501831055,
      "epoch": 6.064553161427058,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020346162768870873,
      "loss": 5.1615,
      "mean_token_accuracy": 0.23530116975307463,
      "num_tokens": 128041595.0,
      "step": 68675
    },
    {
      "entropy": 6.09639253616333,
      "epoch": 6.064994701518898,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002034317710391747,
      "loss": 5.3414,
      "mean_token_accuracy": 0.2162230581045151,
      "num_tokens": 128051925.0,
      "step": 68680
    },
    {
      "entropy": 5.962364196777344,
      "epoch": 6.065436241610739,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002034019157916215,
      "loss": 5.0959,
      "mean_token_accuracy": 0.2406483232975006,
      "num_tokens": 128060272.0,
      "step": 68685
    },
    {
      "entropy": 5.896264743804932,
      "epoch": 6.065877781702579,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002033720619466341,
      "loss": 5.0226,
      "mean_token_accuracy": 0.2544210970401764,
      "num_tokens": 128070016.0,
      "step": 68690
    },
    {
      "entropy": 6.009323787689209,
      "epoch": 6.066319321794419,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020334220950479732,
      "loss": 5.2107,
      "mean_token_accuracy": 0.22600814551115037,
      "num_tokens": 128078357.0,
      "step": 68695
    },
    {
      "entropy": 6.065809440612793,
      "epoch": 6.06676086188626,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020331235846669581,
      "loss": 5.2317,
      "mean_token_accuracy": 0.2344508498907089,
      "num_tokens": 128087235.0,
      "step": 68700
    },
    {
      "entropy": 6.080423259735108,
      "epoch": 6.0672024019781,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002032825088329145,
      "loss": 5.2847,
      "mean_token_accuracy": 0.22311486899852753,
      "num_tokens": 128097070.0,
      "step": 68705
    },
    {
      "entropy": 6.011965322494507,
      "epoch": 6.06764394206994,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000203252660604038,
      "loss": 5.1663,
      "mean_token_accuracy": 0.23256437927484513,
      "num_tokens": 128106217.0,
      "step": 68710
    },
    {
      "entropy": 5.95961561203003,
      "epoch": 6.068085482161781,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002032228137806511,
      "loss": 5.1631,
      "mean_token_accuracy": 0.24224553108215333,
      "num_tokens": 128115553.0,
      "step": 68715
    },
    {
      "entropy": 6.121248054504394,
      "epoch": 6.068527022253621,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002031929683633385,
      "loss": 5.2703,
      "mean_token_accuracy": 0.22519951015710832,
      "num_tokens": 128125714.0,
      "step": 68720
    },
    {
      "entropy": 6.072005987167358,
      "epoch": 6.068968562345461,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002031631243526848,
      "loss": 5.1255,
      "mean_token_accuracy": 0.23090905845165252,
      "num_tokens": 128134407.0,
      "step": 68725
    },
    {
      "entropy": 6.008733224868775,
      "epoch": 6.069410102437302,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020313328174927453,
      "loss": 5.1826,
      "mean_token_accuracy": 0.23293524086475373,
      "num_tokens": 128143430.0,
      "step": 68730
    },
    {
      "entropy": 6.066218423843384,
      "epoch": 6.069851642529142,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002031034405536925,
      "loss": 5.2202,
      "mean_token_accuracy": 0.23224394470453263,
      "num_tokens": 128152121.0,
      "step": 68735
    },
    {
      "entropy": 5.993559503555298,
      "epoch": 6.070293182620982,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002030736007665232,
      "loss": 5.2135,
      "mean_token_accuracy": 0.22793974727392197,
      "num_tokens": 128160750.0,
      "step": 68740
    },
    {
      "entropy": 5.989493274688721,
      "epoch": 6.070734722712823,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00020304376238835115,
      "loss": 5.2075,
      "mean_token_accuracy": 0.23067173808813096,
      "num_tokens": 128170140.0,
      "step": 68745
    },
    {
      "entropy": 6.0348286628723145,
      "epoch": 6.071176262804663,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020301392541976078,
      "loss": 5.2686,
      "mean_token_accuracy": 0.2186100721359253,
      "num_tokens": 128180384.0,
      "step": 68750
    },
    {
      "entropy": 6.005724811553955,
      "epoch": 6.071617802896503,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002029840898613367,
      "loss": 5.1747,
      "mean_token_accuracy": 0.2291742205619812,
      "num_tokens": 128190086.0,
      "step": 68755
    },
    {
      "entropy": 6.028083992004395,
      "epoch": 6.072059342988343,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002029542557136634,
      "loss": 5.154,
      "mean_token_accuracy": 0.23219843208789825,
      "num_tokens": 128198627.0,
      "step": 68760
    },
    {
      "entropy": 6.001116752624512,
      "epoch": 6.072500883080184,
      "grad_norm": 1.375,
      "learning_rate": 0.00020292442297732523,
      "loss": 5.1583,
      "mean_token_accuracy": 0.23899993300437927,
      "num_tokens": 128208716.0,
      "step": 68765
    },
    {
      "entropy": 6.045260620117188,
      "epoch": 6.072942423172024,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020289459165290658,
      "loss": 5.1644,
      "mean_token_accuracy": 0.23068013191223144,
      "num_tokens": 128217899.0,
      "step": 68770
    },
    {
      "entropy": 5.985938978195191,
      "epoch": 6.073383963263864,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002028647617409919,
      "loss": 5.1656,
      "mean_token_accuracy": 0.2279597043991089,
      "num_tokens": 128227074.0,
      "step": 68775
    },
    {
      "entropy": 6.0549530506134035,
      "epoch": 6.073825503355705,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002028349332421655,
      "loss": 5.2294,
      "mean_token_accuracy": 0.22774786055088042,
      "num_tokens": 128236631.0,
      "step": 68780
    },
    {
      "entropy": 6.041823863983154,
      "epoch": 6.074267043447545,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020280510615701175,
      "loss": 5.2003,
      "mean_token_accuracy": 0.22229715287685395,
      "num_tokens": 128244921.0,
      "step": 68785
    },
    {
      "entropy": 6.064205026626587,
      "epoch": 6.074708583539385,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020277528048611492,
      "loss": 5.1564,
      "mean_token_accuracy": 0.22589980512857438,
      "num_tokens": 128253863.0,
      "step": 68790
    },
    {
      "entropy": 6.067108821868897,
      "epoch": 6.075150123631226,
      "grad_norm": 1.53125,
      "learning_rate": 0.00020274545623005918,
      "loss": 5.1894,
      "mean_token_accuracy": 0.22988293915987015,
      "num_tokens": 128263025.0,
      "step": 68795
    },
    {
      "entropy": 5.986789703369141,
      "epoch": 6.075591663723066,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020271563338942895,
      "loss": 5.1356,
      "mean_token_accuracy": 0.23553992062807083,
      "num_tokens": 128272311.0,
      "step": 68800
    },
    {
      "entropy": 6.021257829666138,
      "epoch": 6.076033203814906,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020268581196480835,
      "loss": 5.1679,
      "mean_token_accuracy": 0.21933160871267318,
      "num_tokens": 128281506.0,
      "step": 68805
    },
    {
      "entropy": 6.063224363327026,
      "epoch": 6.076474743906747,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020265599195678152,
      "loss": 5.2745,
      "mean_token_accuracy": 0.22034713625907898,
      "num_tokens": 128290477.0,
      "step": 68810
    },
    {
      "entropy": 6.050729703903198,
      "epoch": 6.076916283998587,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020262617336593263,
      "loss": 5.2299,
      "mean_token_accuracy": 0.2236555278301239,
      "num_tokens": 128301420.0,
      "step": 68815
    },
    {
      "entropy": 6.031196737289429,
      "epoch": 6.077357824090427,
      "grad_norm": 1.53125,
      "learning_rate": 0.00020259635619284595,
      "loss": 5.2107,
      "mean_token_accuracy": 0.22968092411756516,
      "num_tokens": 128310731.0,
      "step": 68820
    },
    {
      "entropy": 6.034160518646241,
      "epoch": 6.077799364182268,
      "grad_norm": 1.609375,
      "learning_rate": 0.0002025665404381054,
      "loss": 5.1616,
      "mean_token_accuracy": 0.22974581271409988,
      "num_tokens": 128319988.0,
      "step": 68825
    },
    {
      "entropy": 6.070870208740234,
      "epoch": 6.078240904274108,
      "grad_norm": 1.5,
      "learning_rate": 0.0002025367261022951,
      "loss": 5.2227,
      "mean_token_accuracy": 0.22310098111629487,
      "num_tokens": 128328897.0,
      "step": 68830
    },
    {
      "entropy": 5.972155332565308,
      "epoch": 6.078682444365948,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020250691318599918,
      "loss": 5.1714,
      "mean_token_accuracy": 0.2335825741291046,
      "num_tokens": 128337892.0,
      "step": 68835
    },
    {
      "entropy": 6.107429075241089,
      "epoch": 6.079123984457789,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002024771016898016,
      "loss": 5.2329,
      "mean_token_accuracy": 0.22121714055538177,
      "num_tokens": 128347036.0,
      "step": 68840
    },
    {
      "entropy": 6.095124864578247,
      "epoch": 6.079565524549629,
      "grad_norm": 1.375,
      "learning_rate": 0.00020244729161428632,
      "loss": 5.2231,
      "mean_token_accuracy": 0.22715286612510682,
      "num_tokens": 128356948.0,
      "step": 68845
    },
    {
      "entropy": 6.067704725265503,
      "epoch": 6.080007064641469,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020241748296003742,
      "loss": 5.2001,
      "mean_token_accuracy": 0.23003162443637848,
      "num_tokens": 128367003.0,
      "step": 68850
    },
    {
      "entropy": 6.0213847160339355,
      "epoch": 6.08044860473331,
      "grad_norm": 1.546875,
      "learning_rate": 0.00020238767572763872,
      "loss": 5.1327,
      "mean_token_accuracy": 0.23042566031217576,
      "num_tokens": 128375328.0,
      "step": 68855
    },
    {
      "entropy": 5.96631646156311,
      "epoch": 6.08089014482515,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020235786991767402,
      "loss": 5.1174,
      "mean_token_accuracy": 0.2386825606226921,
      "num_tokens": 128384306.0,
      "step": 68860
    },
    {
      "entropy": 6.017078447341919,
      "epoch": 6.08133168491699,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00020232806553072746,
      "loss": 5.2042,
      "mean_token_accuracy": 0.22900004237890242,
      "num_tokens": 128393514.0,
      "step": 68865
    },
    {
      "entropy": 6.013782644271851,
      "epoch": 6.081773225008831,
      "grad_norm": 1.28125,
      "learning_rate": 0.00020229826256738277,
      "loss": 5.1421,
      "mean_token_accuracy": 0.2381375402212143,
      "num_tokens": 128403319.0,
      "step": 68870
    },
    {
      "entropy": 6.071657991409301,
      "epoch": 6.082214765100671,
      "grad_norm": 1.5,
      "learning_rate": 0.00020226846102822378,
      "loss": 5.1733,
      "mean_token_accuracy": 0.2297625407576561,
      "num_tokens": 128413061.0,
      "step": 68875
    },
    {
      "entropy": 6.129362297058106,
      "epoch": 6.082656305192511,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00020223866091383414,
      "loss": 5.2827,
      "mean_token_accuracy": 0.21722349822521209,
      "num_tokens": 128422668.0,
      "step": 68880
    },
    {
      "entropy": 6.06793246269226,
      "epoch": 6.083097845284351,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002022088622247979,
      "loss": 5.2241,
      "mean_token_accuracy": 0.22279365062713624,
      "num_tokens": 128432375.0,
      "step": 68885
    },
    {
      "entropy": 6.091542291641235,
      "epoch": 6.083539385376192,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020217906496169857,
      "loss": 5.2608,
      "mean_token_accuracy": 0.22587971687316893,
      "num_tokens": 128442305.0,
      "step": 68890
    },
    {
      "entropy": 6.09002456665039,
      "epoch": 6.083980925468032,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020214926912512,
      "loss": 5.3086,
      "mean_token_accuracy": 0.223953452706337,
      "num_tokens": 128452481.0,
      "step": 68895
    },
    {
      "entropy": 6.101512670516968,
      "epoch": 6.084422465559872,
      "grad_norm": 1.5,
      "learning_rate": 0.0002021194747156458,
      "loss": 5.3096,
      "mean_token_accuracy": 0.2160175323486328,
      "num_tokens": 128462424.0,
      "step": 68900
    },
    {
      "entropy": 6.071499347686768,
      "epoch": 6.084864005651713,
      "grad_norm": 1.546875,
      "learning_rate": 0.00020208968173385955,
      "loss": 5.221,
      "mean_token_accuracy": 0.2232048287987709,
      "num_tokens": 128471748.0,
      "step": 68905
    },
    {
      "entropy": 6.044204902648926,
      "epoch": 6.085305545743553,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00020205989018034503,
      "loss": 5.1914,
      "mean_token_accuracy": 0.22875264585018157,
      "num_tokens": 128481090.0,
      "step": 68910
    },
    {
      "entropy": 6.0887538433074955,
      "epoch": 6.0857470858353935,
      "grad_norm": 1.625,
      "learning_rate": 0.0002020301000556859,
      "loss": 5.206,
      "mean_token_accuracy": 0.22518669962882995,
      "num_tokens": 128490051.0,
      "step": 68915
    },
    {
      "entropy": 6.089367008209228,
      "epoch": 6.086188625927234,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002020003113604655,
      "loss": 5.2643,
      "mean_token_accuracy": 0.22304270714521407,
      "num_tokens": 128499583.0,
      "step": 68920
    },
    {
      "entropy": 6.059062242507935,
      "epoch": 6.0866301660190745,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002019705240952674,
      "loss": 5.2389,
      "mean_token_accuracy": 0.22404682785272598,
      "num_tokens": 128509228.0,
      "step": 68925
    },
    {
      "entropy": 6.045724678039551,
      "epoch": 6.0870717061109145,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00020194073826067528,
      "loss": 5.1886,
      "mean_token_accuracy": 0.2253459244966507,
      "num_tokens": 128518586.0,
      "step": 68930
    },
    {
      "entropy": 6.096683120727539,
      "epoch": 6.0875132462027555,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002019109538572726,
      "loss": 5.2693,
      "mean_token_accuracy": 0.21669287532567977,
      "num_tokens": 128527987.0,
      "step": 68935
    },
    {
      "entropy": 6.083089303970337,
      "epoch": 6.0879547862945955,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020188117088564274,
      "loss": 5.2714,
      "mean_token_accuracy": 0.227237506210804,
      "num_tokens": 128537689.0,
      "step": 68940
    },
    {
      "entropy": 6.0602141380310055,
      "epoch": 6.088396326386436,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020185138934636904,
      "loss": 5.2116,
      "mean_token_accuracy": 0.2355748802423477,
      "num_tokens": 128546813.0,
      "step": 68945
    },
    {
      "entropy": 6.082528305053711,
      "epoch": 6.0888378664782765,
      "grad_norm": 1.375,
      "learning_rate": 0.0002018216092400352,
      "loss": 5.2518,
      "mean_token_accuracy": 0.2260381579399109,
      "num_tokens": 128555981.0,
      "step": 68950
    },
    {
      "entropy": 6.060858345031738,
      "epoch": 6.089279406570117,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020179183056722428,
      "loss": 5.2479,
      "mean_token_accuracy": 0.21780150085687638,
      "num_tokens": 128566545.0,
      "step": 68955
    },
    {
      "entropy": 6.0980078220367435,
      "epoch": 6.089720946661957,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020176205332851982,
      "loss": 5.2159,
      "mean_token_accuracy": 0.2137711003422737,
      "num_tokens": 128575189.0,
      "step": 68960
    },
    {
      "entropy": 6.009777641296386,
      "epoch": 6.090162486753798,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020173227752450513,
      "loss": 5.1425,
      "mean_token_accuracy": 0.23167340755462645,
      "num_tokens": 128584464.0,
      "step": 68965
    },
    {
      "entropy": 6.099998426437378,
      "epoch": 6.090604026845638,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020170250315576328,
      "loss": 5.3196,
      "mean_token_accuracy": 0.21396816670894622,
      "num_tokens": 128594488.0,
      "step": 68970
    },
    {
      "entropy": 6.072729206085205,
      "epoch": 6.091045566937478,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020167273022287784,
      "loss": 5.2343,
      "mean_token_accuracy": 0.2249274104833603,
      "num_tokens": 128605398.0,
      "step": 68975
    },
    {
      "entropy": 6.090210962295532,
      "epoch": 6.091487107029319,
      "grad_norm": 1.546875,
      "learning_rate": 0.00020164295872643196,
      "loss": 5.1841,
      "mean_token_accuracy": 0.2264674723148346,
      "num_tokens": 128614375.0,
      "step": 68980
    },
    {
      "entropy": 6.09049072265625,
      "epoch": 6.091928647121159,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020161318866700878,
      "loss": 5.2095,
      "mean_token_accuracy": 0.21696951389312744,
      "num_tokens": 128623706.0,
      "step": 68985
    },
    {
      "entropy": 6.025452375411987,
      "epoch": 6.092370187212999,
      "grad_norm": 1.453125,
      "learning_rate": 0.00020158342004519136,
      "loss": 5.168,
      "mean_token_accuracy": 0.23297056257724763,
      "num_tokens": 128631933.0,
      "step": 68990
    },
    {
      "entropy": 6.114924621582031,
      "epoch": 6.09281172730484,
      "grad_norm": 1.5,
      "learning_rate": 0.00020155365286156314,
      "loss": 5.3024,
      "mean_token_accuracy": 0.21213987320661545,
      "num_tokens": 128640996.0,
      "step": 68995
    },
    {
      "entropy": 6.078611755371094,
      "epoch": 6.09325326739668,
      "grad_norm": 1.4375,
      "learning_rate": 0.000201523887116707,
      "loss": 5.1865,
      "mean_token_accuracy": 0.23688741624355317,
      "num_tokens": 128650132.0,
      "step": 69000
    },
    {
      "epoch": 6.09325326739668,
      "eval_entropy": 5.770745272976626,
      "eval_loss": 5.287646293640137,
      "eval_mean_token_accuracy": 0.23069411096577602,
      "eval_num_tokens": 128650132.0,
      "eval_runtime": 20.231,
      "eval_samples_per_second": 1745.587,
      "eval_steps_per_second": 218.229,
      "step": 69000
    },
    {
      "entropy": 6.078699970245362,
      "epoch": 6.09369480748852,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020149412281120622,
      "loss": 5.2659,
      "mean_token_accuracy": 0.22380424439907073,
      "num_tokens": 128660013.0,
      "step": 69005
    },
    {
      "entropy": 6.0329474925994875,
      "epoch": 6.094136347580361,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002014643599456436,
      "loss": 5.1702,
      "mean_token_accuracy": 0.22686320692300796,
      "num_tokens": 128668791.0,
      "step": 69010
    },
    {
      "entropy": 6.070741033554077,
      "epoch": 6.094577887672201,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020143459852060247,
      "loss": 5.2098,
      "mean_token_accuracy": 0.2283696174621582,
      "num_tokens": 128677879.0,
      "step": 69015
    },
    {
      "entropy": 6.051577758789063,
      "epoch": 6.095019427764041,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002014048385366657,
      "loss": 5.1981,
      "mean_token_accuracy": 0.23673950135707855,
      "num_tokens": 128686500.0,
      "step": 69020
    },
    {
      "entropy": 6.02639741897583,
      "epoch": 6.095460967855882,
      "grad_norm": 1.609375,
      "learning_rate": 0.00020137507999441629,
      "loss": 5.239,
      "mean_token_accuracy": 0.22418809086084365,
      "num_tokens": 128695048.0,
      "step": 69025
    },
    {
      "entropy": 6.021706819534302,
      "epoch": 6.095902507947722,
      "grad_norm": 1.5,
      "learning_rate": 0.0002013453228944372,
      "loss": 5.1598,
      "mean_token_accuracy": 0.2311922714114189,
      "num_tokens": 128704672.0,
      "step": 69030
    },
    {
      "entropy": 6.050638437271118,
      "epoch": 6.096344048039562,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020131556723731125,
      "loss": 5.2533,
      "mean_token_accuracy": 0.22205458134412764,
      "num_tokens": 128714634.0,
      "step": 69035
    },
    {
      "entropy": 5.9755606174469,
      "epoch": 6.096785588131402,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020128581302362154,
      "loss": 5.1379,
      "mean_token_accuracy": 0.23331184983253478,
      "num_tokens": 128724284.0,
      "step": 69040
    },
    {
      "entropy": 6.1234130859375,
      "epoch": 6.097227128223243,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020125606025395083,
      "loss": 5.3294,
      "mean_token_accuracy": 0.21732159554958344,
      "num_tokens": 128735025.0,
      "step": 69045
    },
    {
      "entropy": 6.037827444076538,
      "epoch": 6.097668668315083,
      "grad_norm": 1.234375,
      "learning_rate": 0.00020122630892888195,
      "loss": 5.2331,
      "mean_token_accuracy": 0.2275708183646202,
      "num_tokens": 128744851.0,
      "step": 69050
    },
    {
      "entropy": 5.9658502578735355,
      "epoch": 6.098110208406923,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020119655904899764,
      "loss": 5.1411,
      "mean_token_accuracy": 0.2466131180524826,
      "num_tokens": 128753898.0,
      "step": 69055
    },
    {
      "entropy": 6.055512809753418,
      "epoch": 6.098551748498764,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020116681061488085,
      "loss": 5.2056,
      "mean_token_accuracy": 0.2340737834572792,
      "num_tokens": 128764289.0,
      "step": 69060
    },
    {
      "entropy": 6.082236003875733,
      "epoch": 6.098993288590604,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002011370636271142,
      "loss": 5.2431,
      "mean_token_accuracy": 0.22277358770370484,
      "num_tokens": 128773770.0,
      "step": 69065
    },
    {
      "entropy": 6.044987392425537,
      "epoch": 6.099434828682444,
      "grad_norm": 1.375,
      "learning_rate": 0.00020110731808628053,
      "loss": 5.1905,
      "mean_token_accuracy": 0.22163716107606887,
      "num_tokens": 128783439.0,
      "step": 69070
    },
    {
      "entropy": 6.06991982460022,
      "epoch": 6.099876368774285,
      "grad_norm": 1.453125,
      "learning_rate": 0.00020107757399296235,
      "loss": 5.2376,
      "mean_token_accuracy": 0.22379582971334458,
      "num_tokens": 128792796.0,
      "step": 69075
    },
    {
      "entropy": 6.0947503566741945,
      "epoch": 6.100317908866125,
      "grad_norm": 1.640625,
      "learning_rate": 0.00020104783134774256,
      "loss": 5.2477,
      "mean_token_accuracy": 0.218338343501091,
      "num_tokens": 128801579.0,
      "step": 69080
    },
    {
      "entropy": 6.059230327606201,
      "epoch": 6.100759448957965,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020101809015120366,
      "loss": 5.2101,
      "mean_token_accuracy": 0.23143284618854523,
      "num_tokens": 128811340.0,
      "step": 69085
    },
    {
      "entropy": 6.121455144882202,
      "epoch": 6.101200989049806,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020098835040392832,
      "loss": 5.3112,
      "mean_token_accuracy": 0.21342070251703263,
      "num_tokens": 128821284.0,
      "step": 69090
    },
    {
      "entropy": 6.0741979598999025,
      "epoch": 6.101642529141646,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020095861210649914,
      "loss": 5.2969,
      "mean_token_accuracy": 0.2218896508216858,
      "num_tokens": 128829387.0,
      "step": 69095
    },
    {
      "entropy": 6.135783720016479,
      "epoch": 6.102084069233486,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002009288752594985,
      "loss": 5.2656,
      "mean_token_accuracy": 0.21798748075962066,
      "num_tokens": 128838485.0,
      "step": 69100
    },
    {
      "entropy": 6.0930005550384525,
      "epoch": 6.102525609325327,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020089913986350916,
      "loss": 5.2115,
      "mean_token_accuracy": 0.22644826471805574,
      "num_tokens": 128847856.0,
      "step": 69105
    },
    {
      "entropy": 6.049187755584716,
      "epoch": 6.102967149417167,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020086940591911357,
      "loss": 5.2569,
      "mean_token_accuracy": 0.21788515746593476,
      "num_tokens": 128857785.0,
      "step": 69110
    },
    {
      "entropy": 6.1229820728302,
      "epoch": 6.103408689509007,
      "grad_norm": 1.375,
      "learning_rate": 0.00020083967342689413,
      "loss": 5.226,
      "mean_token_accuracy": 0.2210854783654213,
      "num_tokens": 128866733.0,
      "step": 69115
    },
    {
      "entropy": 6.055609464645386,
      "epoch": 6.103850229600848,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002008099423874332,
      "loss": 5.1904,
      "mean_token_accuracy": 0.22993861138820648,
      "num_tokens": 128875779.0,
      "step": 69120
    },
    {
      "entropy": 6.052991771697998,
      "epoch": 6.104291769692688,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020078021280131342,
      "loss": 5.188,
      "mean_token_accuracy": 0.22902821153402328,
      "num_tokens": 128885942.0,
      "step": 69125
    },
    {
      "entropy": 6.000657367706299,
      "epoch": 6.104733309784528,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002007504846691171,
      "loss": 5.2225,
      "mean_token_accuracy": 0.23389810174703599,
      "num_tokens": 128895948.0,
      "step": 69130
    },
    {
      "entropy": 5.961183834075928,
      "epoch": 6.105174849876369,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020072075799142648,
      "loss": 5.0759,
      "mean_token_accuracy": 0.23938021212816238,
      "num_tokens": 128905293.0,
      "step": 69135
    },
    {
      "entropy": 6.061864948272705,
      "epoch": 6.105616389968209,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002006910327688239,
      "loss": 5.2216,
      "mean_token_accuracy": 0.22357419580221177,
      "num_tokens": 128914152.0,
      "step": 69140
    },
    {
      "entropy": 6.050005483627319,
      "epoch": 6.106057930060049,
      "grad_norm": 1.609375,
      "learning_rate": 0.00020066130900189185,
      "loss": 5.1392,
      "mean_token_accuracy": 0.2252887025475502,
      "num_tokens": 128922696.0,
      "step": 69145
    },
    {
      "entropy": 6.061739349365235,
      "epoch": 6.10649947015189,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002006315866912124,
      "loss": 5.2153,
      "mean_token_accuracy": 0.2276334807276726,
      "num_tokens": 128932687.0,
      "step": 69150
    },
    {
      "entropy": 6.061402797698975,
      "epoch": 6.10694101024373,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020060186583736793,
      "loss": 5.2541,
      "mean_token_accuracy": 0.2266904056072235,
      "num_tokens": 128942524.0,
      "step": 69155
    },
    {
      "entropy": 6.073107385635376,
      "epoch": 6.10738255033557,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020057214644094064,
      "loss": 5.2173,
      "mean_token_accuracy": 0.22884777188301086,
      "num_tokens": 128952451.0,
      "step": 69160
    },
    {
      "entropy": 6.0784814834594725,
      "epoch": 6.10782409042741,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020054242850251248,
      "loss": 5.2428,
      "mean_token_accuracy": 0.22002427875995637,
      "num_tokens": 128962048.0,
      "step": 69165
    },
    {
      "entropy": 6.0226949691772464,
      "epoch": 6.108265630519251,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00020051271202266597,
      "loss": 5.1484,
      "mean_token_accuracy": 0.23381472378969193,
      "num_tokens": 128971149.0,
      "step": 69170
    },
    {
      "entropy": 6.08123812675476,
      "epoch": 6.108707170611091,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020048299700198302,
      "loss": 5.2215,
      "mean_token_accuracy": 0.2242137685418129,
      "num_tokens": 128980058.0,
      "step": 69175
    },
    {
      "entropy": 6.046914291381836,
      "epoch": 6.109148710702931,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002004532834410458,
      "loss": 5.2716,
      "mean_token_accuracy": 0.21887463331222534,
      "num_tokens": 128990640.0,
      "step": 69180
    },
    {
      "entropy": 5.991315078735352,
      "epoch": 6.109590250794772,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020042357134043625,
      "loss": 5.151,
      "mean_token_accuracy": 0.23641414642333985,
      "num_tokens": 128999564.0,
      "step": 69185
    },
    {
      "entropy": 6.08000373840332,
      "epoch": 6.110031790886612,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0002003938607007366,
      "loss": 5.1948,
      "mean_token_accuracy": 0.22883374392986297,
      "num_tokens": 129009198.0,
      "step": 69190
    },
    {
      "entropy": 6.034385824203492,
      "epoch": 6.1104733309784525,
      "grad_norm": 1.375,
      "learning_rate": 0.0002003641515225289,
      "loss": 5.0978,
      "mean_token_accuracy": 0.23937806040048598,
      "num_tokens": 129019657.0,
      "step": 69195
    },
    {
      "entropy": 6.1082282066345215,
      "epoch": 6.110914871070293,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00020033444380639494,
      "loss": 5.2461,
      "mean_token_accuracy": 0.22597662806510926,
      "num_tokens": 129028096.0,
      "step": 69200
    },
    {
      "entropy": 5.99617486000061,
      "epoch": 6.1113564111621335,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020030473755291672,
      "loss": 5.2219,
      "mean_token_accuracy": 0.22975639700889589,
      "num_tokens": 129037553.0,
      "step": 69205
    },
    {
      "entropy": 5.996911525726318,
      "epoch": 6.1117979512539735,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00020027503276267623,
      "loss": 5.1632,
      "mean_token_accuracy": 0.2275244876742363,
      "num_tokens": 129046734.0,
      "step": 69210
    },
    {
      "entropy": 6.046722793579102,
      "epoch": 6.1122394913458145,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020024532943625528,
      "loss": 5.2438,
      "mean_token_accuracy": 0.2263082206249237,
      "num_tokens": 129055608.0,
      "step": 69215
    },
    {
      "entropy": 6.083708429336548,
      "epoch": 6.1126810314376545,
      "grad_norm": 1.765625,
      "learning_rate": 0.00020021562757423588,
      "loss": 5.1752,
      "mean_token_accuracy": 0.23175135254859924,
      "num_tokens": 129065031.0,
      "step": 69220
    },
    {
      "entropy": 6.083673191070557,
      "epoch": 6.113122571529495,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020018592717719987,
      "loss": 5.2815,
      "mean_token_accuracy": 0.22127961814403535,
      "num_tokens": 129074146.0,
      "step": 69225
    },
    {
      "entropy": 5.98807373046875,
      "epoch": 6.1135641116213355,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020015622824572888,
      "loss": 5.0988,
      "mean_token_accuracy": 0.2406974509358406,
      "num_tokens": 129083958.0,
      "step": 69230
    },
    {
      "entropy": 6.068387603759765,
      "epoch": 6.114005651713176,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020012653078040489,
      "loss": 5.163,
      "mean_token_accuracy": 0.22853826582431794,
      "num_tokens": 129093234.0,
      "step": 69235
    },
    {
      "entropy": 6.099830150604248,
      "epoch": 6.114447191805016,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020009683478180957,
      "loss": 5.2513,
      "mean_token_accuracy": 0.22378182411193848,
      "num_tokens": 129102547.0,
      "step": 69240
    },
    {
      "entropy": 6.01290078163147,
      "epoch": 6.1148887318968566,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002000671402505247,
      "loss": 5.138,
      "mean_token_accuracy": 0.23251679837703704,
      "num_tokens": 129112222.0,
      "step": 69245
    },
    {
      "entropy": 5.988444519042969,
      "epoch": 6.115330271988697,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002000374471871318,
      "loss": 5.0985,
      "mean_token_accuracy": 0.23881348669528962,
      "num_tokens": 129121646.0,
      "step": 69250
    },
    {
      "entropy": 5.986944723129272,
      "epoch": 6.115771812080537,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002000077555922128,
      "loss": 5.214,
      "mean_token_accuracy": 0.22841069251298904,
      "num_tokens": 129130959.0,
      "step": 69255
    },
    {
      "entropy": 6.035964155197144,
      "epoch": 6.116213352172378,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019997806546634918,
      "loss": 5.1953,
      "mean_token_accuracy": 0.2195817157626152,
      "num_tokens": 129140095.0,
      "step": 69260
    },
    {
      "entropy": 6.0757654190063475,
      "epoch": 6.116654892264218,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019994837681012258,
      "loss": 5.1861,
      "mean_token_accuracy": 0.23021625280380248,
      "num_tokens": 129148644.0,
      "step": 69265
    },
    {
      "entropy": 5.917458438873291,
      "epoch": 6.117096432356058,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001999186896241146,
      "loss": 5.1163,
      "mean_token_accuracy": 0.2419154554605484,
      "num_tokens": 129158180.0,
      "step": 69270
    },
    {
      "entropy": 5.9677736282348635,
      "epoch": 6.117537972447899,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019988900390890675,
      "loss": 5.0888,
      "mean_token_accuracy": 0.2397029757499695,
      "num_tokens": 129167269.0,
      "step": 69275
    },
    {
      "entropy": 6.0557982444763185,
      "epoch": 6.117979512539739,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019985931966508058,
      "loss": 5.2558,
      "mean_token_accuracy": 0.22224128544330596,
      "num_tokens": 129176350.0,
      "step": 69280
    },
    {
      "entropy": 6.073530387878418,
      "epoch": 6.118421052631579,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001998296368932177,
      "loss": 5.2655,
      "mean_token_accuracy": 0.2221576154232025,
      "num_tokens": 129187094.0,
      "step": 69285
    },
    {
      "entropy": 6.070757675170898,
      "epoch": 6.11886259272342,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019979995559389952,
      "loss": 5.195,
      "mean_token_accuracy": 0.22388286888599396,
      "num_tokens": 129195757.0,
      "step": 69290
    },
    {
      "entropy": 6.071588087081909,
      "epoch": 6.11930413281526,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001997702757677073,
      "loss": 5.2222,
      "mean_token_accuracy": 0.22716862708330154,
      "num_tokens": 129204401.0,
      "step": 69295
    },
    {
      "entropy": 6.0236616134643555,
      "epoch": 6.1197456729071,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019974059741522266,
      "loss": 5.2315,
      "mean_token_accuracy": 0.22612731754779816,
      "num_tokens": 129214026.0,
      "step": 69300
    },
    {
      "entropy": 6.076118516921997,
      "epoch": 6.120187212998941,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001997109205370269,
      "loss": 5.2173,
      "mean_token_accuracy": 0.22241157293319702,
      "num_tokens": 129223398.0,
      "step": 69305
    },
    {
      "entropy": 6.102945852279663,
      "epoch": 6.120628753090781,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019968124513370145,
      "loss": 5.2589,
      "mean_token_accuracy": 0.22646500021219254,
      "num_tokens": 129233493.0,
      "step": 69310
    },
    {
      "entropy": 6.136575031280517,
      "epoch": 6.121070293182621,
      "grad_norm": 1.375,
      "learning_rate": 0.00019965157120582745,
      "loss": 5.2778,
      "mean_token_accuracy": 0.22098317593336106,
      "num_tokens": 129243337.0,
      "step": 69315
    },
    {
      "entropy": 6.07387146949768,
      "epoch": 6.121511833274461,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019962189875398639,
      "loss": 5.2253,
      "mean_token_accuracy": 0.22733054459095,
      "num_tokens": 129253232.0,
      "step": 69320
    },
    {
      "entropy": 6.0419083595275875,
      "epoch": 6.121953373366302,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00019959222777875946,
      "loss": 5.1372,
      "mean_token_accuracy": 0.23057866096496582,
      "num_tokens": 129262402.0,
      "step": 69325
    },
    {
      "entropy": 6.006045484542847,
      "epoch": 6.122394913458142,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019956255828072788,
      "loss": 5.1821,
      "mean_token_accuracy": 0.2264582023024559,
      "num_tokens": 129271889.0,
      "step": 69330
    },
    {
      "entropy": 6.058817529678345,
      "epoch": 6.122836453549982,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019953289026047294,
      "loss": 5.1364,
      "mean_token_accuracy": 0.23152782917022705,
      "num_tokens": 129281100.0,
      "step": 69335
    },
    {
      "entropy": 6.057643747329712,
      "epoch": 6.123277993641823,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00019950322371857572,
      "loss": 5.2515,
      "mean_token_accuracy": 0.22190921604633332,
      "num_tokens": 129289785.0,
      "step": 69340
    },
    {
      "entropy": 5.944434642791748,
      "epoch": 6.123719533733663,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019947355865561735,
      "loss": 5.13,
      "mean_token_accuracy": 0.23436102420091628,
      "num_tokens": 129298658.0,
      "step": 69345
    },
    {
      "entropy": 6.091066503524781,
      "epoch": 6.124161073825503,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019944389507217908,
      "loss": 5.248,
      "mean_token_accuracy": 0.22519028931856155,
      "num_tokens": 129309231.0,
      "step": 69350
    },
    {
      "entropy": 6.032761192321777,
      "epoch": 6.124602613917344,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019941423296884197,
      "loss": 5.1981,
      "mean_token_accuracy": 0.22541468292474748,
      "num_tokens": 129318359.0,
      "step": 69355
    },
    {
      "entropy": 6.006273078918457,
      "epoch": 6.125044154009184,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001993845723461869,
      "loss": 5.1875,
      "mean_token_accuracy": 0.2307191953063011,
      "num_tokens": 129326523.0,
      "step": 69360
    },
    {
      "entropy": 6.02092227935791,
      "epoch": 6.125485694101024,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019935491320479515,
      "loss": 5.2075,
      "mean_token_accuracy": 0.2275170236825943,
      "num_tokens": 129336079.0,
      "step": 69365
    },
    {
      "entropy": 5.993955469131469,
      "epoch": 6.125927234192865,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019932525554524773,
      "loss": 5.1587,
      "mean_token_accuracy": 0.23454984426498413,
      "num_tokens": 129345431.0,
      "step": 69370
    },
    {
      "entropy": 6.0058534145355225,
      "epoch": 6.126368774284705,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001992955993681254,
      "loss": 5.1672,
      "mean_token_accuracy": 0.23103209286928178,
      "num_tokens": 129354484.0,
      "step": 69375
    },
    {
      "entropy": 6.050739336013794,
      "epoch": 6.126810314376545,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019926594467400917,
      "loss": 5.2545,
      "mean_token_accuracy": 0.22344498187303544,
      "num_tokens": 129363361.0,
      "step": 69380
    },
    {
      "entropy": 6.1038978576660154,
      "epoch": 6.127251854468386,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001992362914634801,
      "loss": 5.2717,
      "mean_token_accuracy": 0.2144702598452568,
      "num_tokens": 129371988.0,
      "step": 69385
    },
    {
      "entropy": 6.085813236236572,
      "epoch": 6.127693394560226,
      "grad_norm": 1.296875,
      "learning_rate": 0.000199206639737119,
      "loss": 5.274,
      "mean_token_accuracy": 0.22680974155664443,
      "num_tokens": 129382115.0,
      "step": 69390
    },
    {
      "entropy": 6.0249292850494385,
      "epoch": 6.128134934652066,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001991769894955067,
      "loss": 5.1269,
      "mean_token_accuracy": 0.23382241725921632,
      "num_tokens": 129390851.0,
      "step": 69395
    },
    {
      "entropy": 6.005634450912476,
      "epoch": 6.128576474743907,
      "grad_norm": 1.578125,
      "learning_rate": 0.00019914734073922408,
      "loss": 5.1086,
      "mean_token_accuracy": 0.2297893777489662,
      "num_tokens": 129399646.0,
      "step": 69400
    },
    {
      "entropy": 6.039033222198486,
      "epoch": 6.129018014835747,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019911769346885183,
      "loss": 5.222,
      "mean_token_accuracy": 0.2217525616288185,
      "num_tokens": 129409220.0,
      "step": 69405
    },
    {
      "entropy": 6.016296529769898,
      "epoch": 6.129459554927587,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019908804768497084,
      "loss": 5.1901,
      "mean_token_accuracy": 0.22531058341264726,
      "num_tokens": 129419803.0,
      "step": 69410
    },
    {
      "entropy": 6.015670967102051,
      "epoch": 6.129901095019428,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019905840338816188,
      "loss": 5.1515,
      "mean_token_accuracy": 0.22790796011686326,
      "num_tokens": 129428970.0,
      "step": 69415
    },
    {
      "entropy": 6.138033533096314,
      "epoch": 6.130342635111268,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019902876057900558,
      "loss": 5.292,
      "mean_token_accuracy": 0.22234242409467697,
      "num_tokens": 129438543.0,
      "step": 69420
    },
    {
      "entropy": 6.003952884674073,
      "epoch": 6.130784175203108,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019899911925808257,
      "loss": 5.1463,
      "mean_token_accuracy": 0.23131344467401505,
      "num_tokens": 129448499.0,
      "step": 69425
    },
    {
      "entropy": 5.970882415771484,
      "epoch": 6.131225715294949,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001989694794259737,
      "loss": 5.221,
      "mean_token_accuracy": 0.23162365108728408,
      "num_tokens": 129457671.0,
      "step": 69430
    },
    {
      "entropy": 5.992248106002807,
      "epoch": 6.131667255386789,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00019893984108325946,
      "loss": 5.1848,
      "mean_token_accuracy": 0.23221516758203506,
      "num_tokens": 129467248.0,
      "step": 69435
    },
    {
      "entropy": 6.061936473846435,
      "epoch": 6.132108795478629,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019891020423052052,
      "loss": 5.1762,
      "mean_token_accuracy": 0.22673390209674835,
      "num_tokens": 129476476.0,
      "step": 69440
    },
    {
      "entropy": 6.141303062438965,
      "epoch": 6.132550335570469,
      "grad_norm": 1.671875,
      "learning_rate": 0.00019888056886833738,
      "loss": 5.2554,
      "mean_token_accuracy": 0.22363197952508926,
      "num_tokens": 129485731.0,
      "step": 69445
    },
    {
      "entropy": 6.0247057437896725,
      "epoch": 6.13299187566231,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00019885093499729046,
      "loss": 5.1842,
      "mean_token_accuracy": 0.23781138211488723,
      "num_tokens": 129494692.0,
      "step": 69450
    },
    {
      "entropy": 6.083323764801025,
      "epoch": 6.13343341575415,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001988213026179606,
      "loss": 5.1785,
      "mean_token_accuracy": 0.2329763278365135,
      "num_tokens": 129503010.0,
      "step": 69455
    },
    {
      "entropy": 6.081404352188111,
      "epoch": 6.13387495584599,
      "grad_norm": 1.375,
      "learning_rate": 0.000198791671730928,
      "loss": 5.2987,
      "mean_token_accuracy": 0.2161840096116066,
      "num_tokens": 129513213.0,
      "step": 69460
    },
    {
      "entropy": 5.99488091468811,
      "epoch": 6.134316495937831,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019876204233677325,
      "loss": 5.1397,
      "mean_token_accuracy": 0.23505301177501678,
      "num_tokens": 129521972.0,
      "step": 69465
    },
    {
      "entropy": 6.078088331222534,
      "epoch": 6.134758036029671,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001987324144360767,
      "loss": 5.2613,
      "mean_token_accuracy": 0.22179182767868041,
      "num_tokens": 129531096.0,
      "step": 69470
    },
    {
      "entropy": 6.040374231338501,
      "epoch": 6.1351995761215115,
      "grad_norm": 1.6875,
      "learning_rate": 0.00019870278802941877,
      "loss": 5.1905,
      "mean_token_accuracy": 0.23293566554784775,
      "num_tokens": 129539858.0,
      "step": 69475
    },
    {
      "entropy": 6.090621089935302,
      "epoch": 6.135641116213352,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001986731631173799,
      "loss": 5.2469,
      "mean_token_accuracy": 0.22563634663820267,
      "num_tokens": 129549018.0,
      "step": 69480
    },
    {
      "entropy": 6.07129487991333,
      "epoch": 6.1360826563051925,
      "grad_norm": 1.5625,
      "learning_rate": 0.00019864353970054034,
      "loss": 5.3006,
      "mean_token_accuracy": 0.21544111967086793,
      "num_tokens": 129557875.0,
      "step": 69485
    },
    {
      "entropy": 6.035914373397827,
      "epoch": 6.1365241963970325,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00019861391777948028,
      "loss": 5.165,
      "mean_token_accuracy": 0.23328231126070023,
      "num_tokens": 129566953.0,
      "step": 69490
    },
    {
      "entropy": 6.047666120529175,
      "epoch": 6.1369657364888734,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019858429735478022,
      "loss": 5.2306,
      "mean_token_accuracy": 0.22934730350971222,
      "num_tokens": 129575815.0,
      "step": 69495
    },
    {
      "entropy": 5.971358060836792,
      "epoch": 6.1374072765807135,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001985546784270204,
      "loss": 5.1832,
      "mean_token_accuracy": 0.23513871133327485,
      "num_tokens": 129585272.0,
      "step": 69500
    },
    {
      "entropy": 6.031457185745239,
      "epoch": 6.1378488166725536,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019852506099678086,
      "loss": 5.1948,
      "mean_token_accuracy": 0.23336841017007828,
      "num_tokens": 129594699.0,
      "step": 69505
    },
    {
      "entropy": 6.040505409240723,
      "epoch": 6.1382903567643945,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019849544506464196,
      "loss": 5.1083,
      "mean_token_accuracy": 0.23394395262002946,
      "num_tokens": 129603180.0,
      "step": 69510
    },
    {
      "entropy": 6.016110229492187,
      "epoch": 6.1387318968562345,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019846583063118362,
      "loss": 5.1977,
      "mean_token_accuracy": 0.22828310281038283,
      "num_tokens": 129612976.0,
      "step": 69515
    },
    {
      "entropy": 5.990418148040772,
      "epoch": 6.139173436948075,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019843621769698627,
      "loss": 5.1836,
      "mean_token_accuracy": 0.23120299279689788,
      "num_tokens": 129622766.0,
      "step": 69520
    },
    {
      "entropy": 6.002102708816528,
      "epoch": 6.1396149770399155,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001984066062626298,
      "loss": 5.1903,
      "mean_token_accuracy": 0.23305777460336685,
      "num_tokens": 129633061.0,
      "step": 69525
    },
    {
      "entropy": 6.019413566589355,
      "epoch": 6.140056517131756,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019837699632869438,
      "loss": 5.1539,
      "mean_token_accuracy": 0.22664216607809068,
      "num_tokens": 129642452.0,
      "step": 69530
    },
    {
      "entropy": 6.072464084625244,
      "epoch": 6.140498057223596,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019834738789576,
      "loss": 5.2723,
      "mean_token_accuracy": 0.22398590296506882,
      "num_tokens": 129651152.0,
      "step": 69535
    },
    {
      "entropy": 6.1066807270050045,
      "epoch": 6.140939597315437,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019831778096440672,
      "loss": 5.2124,
      "mean_token_accuracy": 0.2224976658821106,
      "num_tokens": 129660355.0,
      "step": 69540
    },
    {
      "entropy": 6.06947603225708,
      "epoch": 6.141381137407277,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001982881755352145,
      "loss": 5.2316,
      "mean_token_accuracy": 0.2228003352880478,
      "num_tokens": 129670392.0,
      "step": 69545
    },
    {
      "entropy": 6.0115903377532955,
      "epoch": 6.141822677499117,
      "grad_norm": 1.578125,
      "learning_rate": 0.00019825857160876332,
      "loss": 5.1106,
      "mean_token_accuracy": 0.2382293462753296,
      "num_tokens": 129679317.0,
      "step": 69550
    },
    {
      "entropy": 6.057297134399414,
      "epoch": 6.142264217590958,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00019822896918563298,
      "loss": 5.2008,
      "mean_token_accuracy": 0.22506831586360931,
      "num_tokens": 129688933.0,
      "step": 69555
    },
    {
      "entropy": 5.992380809783936,
      "epoch": 6.142705757682798,
      "grad_norm": 1.5,
      "learning_rate": 0.00019819936826640362,
      "loss": 5.0599,
      "mean_token_accuracy": 0.2470710903406143,
      "num_tokens": 129697840.0,
      "step": 69560
    },
    {
      "entropy": 6.0558984756469725,
      "epoch": 6.143147297774638,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001981697688516549,
      "loss": 5.2195,
      "mean_token_accuracy": 0.22467308044433593,
      "num_tokens": 129706613.0,
      "step": 69565
    },
    {
      "entropy": 6.007953357696533,
      "epoch": 6.143588837866479,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001981401709419667,
      "loss": 5.1881,
      "mean_token_accuracy": 0.22471406906843186,
      "num_tokens": 129716144.0,
      "step": 69570
    },
    {
      "entropy": 6.019940900802612,
      "epoch": 6.144030377958319,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001981105745379188,
      "loss": 5.0961,
      "mean_token_accuracy": 0.2331131637096405,
      "num_tokens": 129724639.0,
      "step": 69575
    },
    {
      "entropy": 5.989075374603272,
      "epoch": 6.144471918050159,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00019808097964009103,
      "loss": 5.1484,
      "mean_token_accuracy": 0.2330273434519768,
      "num_tokens": 129733084.0,
      "step": 69580
    },
    {
      "entropy": 6.017346048355103,
      "epoch": 6.144913458142,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019805138624906315,
      "loss": 5.1306,
      "mean_token_accuracy": 0.23363954871892928,
      "num_tokens": 129742196.0,
      "step": 69585
    },
    {
      "entropy": 6.060224294662476,
      "epoch": 6.14535499823384,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019802179436541483,
      "loss": 5.2001,
      "mean_token_accuracy": 0.2257113963365555,
      "num_tokens": 129751294.0,
      "step": 69590
    },
    {
      "entropy": 5.944929933547973,
      "epoch": 6.14579653832568,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019799220398972577,
      "loss": 5.0978,
      "mean_token_accuracy": 0.23844099193811416,
      "num_tokens": 129760414.0,
      "step": 69595
    },
    {
      "entropy": 5.974719476699829,
      "epoch": 6.14623807841752,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019796261512257564,
      "loss": 5.1406,
      "mean_token_accuracy": 0.23478762656450272,
      "num_tokens": 129769205.0,
      "step": 69600
    },
    {
      "entropy": 6.0615668296813965,
      "epoch": 6.146679618509361,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019793302776454407,
      "loss": 5.2254,
      "mean_token_accuracy": 0.22416753470897674,
      "num_tokens": 129779278.0,
      "step": 69605
    },
    {
      "entropy": 6.0594010829925535,
      "epoch": 6.147121158601201,
      "grad_norm": 1.671875,
      "learning_rate": 0.0001979034419162107,
      "loss": 5.2337,
      "mean_token_accuracy": 0.22665760517120362,
      "num_tokens": 129788268.0,
      "step": 69610
    },
    {
      "entropy": 6.028836917877197,
      "epoch": 6.147562698693041,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019787385757815508,
      "loss": 5.2098,
      "mean_token_accuracy": 0.22686734944581985,
      "num_tokens": 129797491.0,
      "step": 69615
    },
    {
      "entropy": 6.076826763153076,
      "epoch": 6.148004238784882,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019784427475095658,
      "loss": 5.1956,
      "mean_token_accuracy": 0.23041169941425324,
      "num_tokens": 129806920.0,
      "step": 69620
    },
    {
      "entropy": 6.02366271018982,
      "epoch": 6.148445778876722,
      "grad_norm": 1.484375,
      "learning_rate": 0.000197814693435195,
      "loss": 5.2104,
      "mean_token_accuracy": 0.23477280437946318,
      "num_tokens": 129816206.0,
      "step": 69625
    },
    {
      "entropy": 6.067210531234741,
      "epoch": 6.148887318968562,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019778511363144962,
      "loss": 5.2089,
      "mean_token_accuracy": 0.2259989634156227,
      "num_tokens": 129825799.0,
      "step": 69630
    },
    {
      "entropy": 5.976857328414917,
      "epoch": 6.149328859060403,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019775553534029993,
      "loss": 5.1852,
      "mean_token_accuracy": 0.22930634766817093,
      "num_tokens": 129835635.0,
      "step": 69635
    },
    {
      "entropy": 5.980532264709472,
      "epoch": 6.149770399152243,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001977259585623254,
      "loss": 5.1449,
      "mean_token_accuracy": 0.23766281008720397,
      "num_tokens": 129845023.0,
      "step": 69640
    },
    {
      "entropy": 6.025374841690064,
      "epoch": 6.150211939244083,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019769638329810536,
      "loss": 5.2336,
      "mean_token_accuracy": 0.22584450095891953,
      "num_tokens": 129853483.0,
      "step": 69645
    },
    {
      "entropy": 6.01048002243042,
      "epoch": 6.150653479335924,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019766680954821925,
      "loss": 5.1787,
      "mean_token_accuracy": 0.2350684031844139,
      "num_tokens": 129863320.0,
      "step": 69650
    },
    {
      "entropy": 6.127118444442749,
      "epoch": 6.151095019427764,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019763723731324633,
      "loss": 5.2259,
      "mean_token_accuracy": 0.22805556654930115,
      "num_tokens": 129872071.0,
      "step": 69655
    },
    {
      "entropy": 6.052942371368408,
      "epoch": 6.151536559519604,
      "grad_norm": 1.4375,
      "learning_rate": 0.000197607666593766,
      "loss": 5.1881,
      "mean_token_accuracy": 0.22610499411821366,
      "num_tokens": 129880821.0,
      "step": 69660
    },
    {
      "entropy": 6.10198483467102,
      "epoch": 6.151978099611445,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00019757809739035732,
      "loss": 5.2706,
      "mean_token_accuracy": 0.21895581036806105,
      "num_tokens": 129890260.0,
      "step": 69665
    },
    {
      "entropy": 6.036958503723144,
      "epoch": 6.152419639703285,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019754852970359978,
      "loss": 5.2077,
      "mean_token_accuracy": 0.22655934989452362,
      "num_tokens": 129899635.0,
      "step": 69670
    },
    {
      "entropy": 5.965312576293945,
      "epoch": 6.152861179795125,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00019751896353407256,
      "loss": 5.1161,
      "mean_token_accuracy": 0.23666429221630098,
      "num_tokens": 129909798.0,
      "step": 69675
    },
    {
      "entropy": 6.0205635070800785,
      "epoch": 6.153302719886966,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019748939888235474,
      "loss": 5.1109,
      "mean_token_accuracy": 0.23277928084135055,
      "num_tokens": 129919530.0,
      "step": 69680
    },
    {
      "entropy": 5.995714139938355,
      "epoch": 6.153744259978806,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019745983574902544,
      "loss": 5.1367,
      "mean_token_accuracy": 0.2359289601445198,
      "num_tokens": 129928382.0,
      "step": 69685
    },
    {
      "entropy": 6.024072360992432,
      "epoch": 6.154185800070646,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019743027413466385,
      "loss": 5.2103,
      "mean_token_accuracy": 0.2252338007092476,
      "num_tokens": 129938010.0,
      "step": 69690
    },
    {
      "entropy": 6.1013483047485355,
      "epoch": 6.154627340162487,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001974007140398491,
      "loss": 5.2892,
      "mean_token_accuracy": 0.21968634873628617,
      "num_tokens": 129948490.0,
      "step": 69695
    },
    {
      "entropy": 6.1106733798980715,
      "epoch": 6.155068880254327,
      "grad_norm": 1.5,
      "learning_rate": 0.00019737115546516034,
      "loss": 5.2845,
      "mean_token_accuracy": 0.21802412420511247,
      "num_tokens": 129957983.0,
      "step": 69700
    },
    {
      "entropy": 6.025829935073853,
      "epoch": 6.155510420346167,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019734159841117637,
      "loss": 5.1772,
      "mean_token_accuracy": 0.2326575353741646,
      "num_tokens": 129966625.0,
      "step": 69705
    },
    {
      "entropy": 6.057508277893066,
      "epoch": 6.155951960438008,
      "grad_norm": 1.703125,
      "learning_rate": 0.00019731204287847632,
      "loss": 5.2072,
      "mean_token_accuracy": 0.22683483213186265,
      "num_tokens": 129975207.0,
      "step": 69710
    },
    {
      "entropy": 5.9559561252594,
      "epoch": 6.156393500529848,
      "grad_norm": 1.53125,
      "learning_rate": 0.00019728248886763923,
      "loss": 5.0549,
      "mean_token_accuracy": 0.24847566336393356,
      "num_tokens": 129984401.0,
      "step": 69715
    },
    {
      "entropy": 6.106728506088257,
      "epoch": 6.156835040621688,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019725293637924392,
      "loss": 5.2526,
      "mean_token_accuracy": 0.22355511486530305,
      "num_tokens": 129993354.0,
      "step": 69720
    },
    {
      "entropy": 5.992809200286866,
      "epoch": 6.157276580713528,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019722338541386942,
      "loss": 5.1163,
      "mean_token_accuracy": 0.2382863610982895,
      "num_tokens": 130002293.0,
      "step": 69725
    },
    {
      "entropy": 6.01588077545166,
      "epoch": 6.157718120805369,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019719383597209444,
      "loss": 5.0866,
      "mean_token_accuracy": 0.23834459185600282,
      "num_tokens": 130011295.0,
      "step": 69730
    },
    {
      "entropy": 5.993757629394532,
      "epoch": 6.158159660897209,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019716428805449814,
      "loss": 5.1586,
      "mean_token_accuracy": 0.23045815378427506,
      "num_tokens": 130021165.0,
      "step": 69735
    },
    {
      "entropy": 6.081446886062622,
      "epoch": 6.158601200989049,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019713474166165913,
      "loss": 5.1851,
      "mean_token_accuracy": 0.2254504680633545,
      "num_tokens": 130031246.0,
      "step": 69740
    },
    {
      "entropy": 6.002603673934937,
      "epoch": 6.15904274108089,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019710519679415624,
      "loss": 5.1305,
      "mean_token_accuracy": 0.23777461796998978,
      "num_tokens": 130040387.0,
      "step": 69745
    },
    {
      "entropy": 5.96879243850708,
      "epoch": 6.15948428117273,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019707565345256824,
      "loss": 5.1514,
      "mean_token_accuracy": 0.2381306082010269,
      "num_tokens": 130049931.0,
      "step": 69750
    },
    {
      "entropy": 5.98319935798645,
      "epoch": 6.1599258212645704,
      "grad_norm": 1.59375,
      "learning_rate": 0.00019704611163747377,
      "loss": 5.121,
      "mean_token_accuracy": 0.22741734534502028,
      "num_tokens": 130058152.0,
      "step": 69755
    },
    {
      "entropy": 6.0265381813049315,
      "epoch": 6.160367361356411,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019701657134945172,
      "loss": 5.1704,
      "mean_token_accuracy": 0.22686414867639543,
      "num_tokens": 130066815.0,
      "step": 69760
    },
    {
      "entropy": 6.028636455535889,
      "epoch": 6.160808901448251,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00019698703258908074,
      "loss": 5.2628,
      "mean_token_accuracy": 0.221487158536911,
      "num_tokens": 130075838.0,
      "step": 69765
    },
    {
      "entropy": 6.027356576919556,
      "epoch": 6.1612504415400915,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019695749535693935,
      "loss": 5.1389,
      "mean_token_accuracy": 0.2416561186313629,
      "num_tokens": 130084941.0,
      "step": 69770
    },
    {
      "entropy": 5.993278551101684,
      "epoch": 6.161691981631932,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001969279596536062,
      "loss": 5.1337,
      "mean_token_accuracy": 0.23323857933282852,
      "num_tokens": 130093510.0,
      "step": 69775
    },
    {
      "entropy": 6.097537517547607,
      "epoch": 6.1621335217237725,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019689842547966003,
      "loss": 5.1943,
      "mean_token_accuracy": 0.22579023391008377,
      "num_tokens": 130102977.0,
      "step": 69780
    },
    {
      "entropy": 6.00074200630188,
      "epoch": 6.1625750618156125,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00019686889283567917,
      "loss": 5.089,
      "mean_token_accuracy": 0.2370801404118538,
      "num_tokens": 130111164.0,
      "step": 69785
    },
    {
      "entropy": 6.025646352767945,
      "epoch": 6.1630166019074535,
      "grad_norm": 1.59375,
      "learning_rate": 0.00019683936172224236,
      "loss": 5.1836,
      "mean_token_accuracy": 0.2279091462492943,
      "num_tokens": 130120061.0,
      "step": 69790
    },
    {
      "entropy": 6.052109479904175,
      "epoch": 6.1634581419992935,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00019680983213992792,
      "loss": 5.2266,
      "mean_token_accuracy": 0.2248185768723488,
      "num_tokens": 130129355.0,
      "step": 69795
    },
    {
      "entropy": 6.006750440597534,
      "epoch": 6.163899682091134,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00019678030408931446,
      "loss": 5.1112,
      "mean_token_accuracy": 0.23793639093637467,
      "num_tokens": 130138397.0,
      "step": 69800
    },
    {
      "entropy": 6.1230292320251465,
      "epoch": 6.1643412221829745,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019675077757098035,
      "loss": 5.2152,
      "mean_token_accuracy": 0.2273486614227295,
      "num_tokens": 130147849.0,
      "step": 69805
    },
    {
      "entropy": 6.043188190460205,
      "epoch": 6.164782762274815,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019672125258550404,
      "loss": 5.2221,
      "mean_token_accuracy": 0.2308845266699791,
      "num_tokens": 130157550.0,
      "step": 69810
    },
    {
      "entropy": 6.091934585571289,
      "epoch": 6.165224302366655,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019669172913346384,
      "loss": 5.1571,
      "mean_token_accuracy": 0.23047646284103393,
      "num_tokens": 130165875.0,
      "step": 69815
    },
    {
      "entropy": 6.027383947372437,
      "epoch": 6.165665842458496,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001966622072154381,
      "loss": 5.3029,
      "mean_token_accuracy": 0.2139770969748497,
      "num_tokens": 130175206.0,
      "step": 69820
    },
    {
      "entropy": 5.9678154468536375,
      "epoch": 6.166107382550336,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001966326868320052,
      "loss": 5.1065,
      "mean_token_accuracy": 0.23691409230232238,
      "num_tokens": 130184322.0,
      "step": 69825
    },
    {
      "entropy": 6.07364068031311,
      "epoch": 6.166548922642176,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001966031679837434,
      "loss": 5.2435,
      "mean_token_accuracy": 0.22758951932191848,
      "num_tokens": 130193519.0,
      "step": 69830
    },
    {
      "entropy": 6.042526483535767,
      "epoch": 6.166990462734017,
      "grad_norm": 1.5,
      "learning_rate": 0.0001965736506712309,
      "loss": 5.1453,
      "mean_token_accuracy": 0.23235390782356263,
      "num_tokens": 130202120.0,
      "step": 69835
    },
    {
      "entropy": 6.009621572494507,
      "epoch": 6.167432002825857,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019654413489504603,
      "loss": 5.1113,
      "mean_token_accuracy": 0.24001331776380538,
      "num_tokens": 130212113.0,
      "step": 69840
    },
    {
      "entropy": 6.04964804649353,
      "epoch": 6.167873542917697,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019651462065576692,
      "loss": 5.2602,
      "mean_token_accuracy": 0.2206118807196617,
      "num_tokens": 130221227.0,
      "step": 69845
    },
    {
      "entropy": 6.1216248035430905,
      "epoch": 6.168315083009538,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00019648510795397173,
      "loss": 5.2528,
      "mean_token_accuracy": 0.22525147199630738,
      "num_tokens": 130230097.0,
      "step": 69850
    },
    {
      "entropy": 6.112734651565551,
      "epoch": 6.168756623101378,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019645559679023868,
      "loss": 5.2762,
      "mean_token_accuracy": 0.22559633553028108,
      "num_tokens": 130239430.0,
      "step": 69855
    },
    {
      "entropy": 6.0200423240661625,
      "epoch": 6.169198163193218,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001964260871651457,
      "loss": 5.1558,
      "mean_token_accuracy": 0.22716599255800246,
      "num_tokens": 130248155.0,
      "step": 69860
    },
    {
      "entropy": 5.993584203720093,
      "epoch": 6.169639703285059,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001963965790792711,
      "loss": 5.149,
      "mean_token_accuracy": 0.23488881587982177,
      "num_tokens": 130258782.0,
      "step": 69865
    },
    {
      "entropy": 6.005341958999634,
      "epoch": 6.170081243376899,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019636707253319286,
      "loss": 5.2156,
      "mean_token_accuracy": 0.22926658540964126,
      "num_tokens": 130267638.0,
      "step": 69870
    },
    {
      "entropy": 6.079413175582886,
      "epoch": 6.170522783468739,
      "grad_norm": 1.59375,
      "learning_rate": 0.00019633756752748889,
      "loss": 5.2379,
      "mean_token_accuracy": 0.2251259207725525,
      "num_tokens": 130277370.0,
      "step": 69875
    },
    {
      "entropy": 6.036379814147949,
      "epoch": 6.170964323560579,
      "grad_norm": 1.375,
      "learning_rate": 0.0001963080640627372,
      "loss": 5.1655,
      "mean_token_accuracy": 0.22464330345392228,
      "num_tokens": 130286761.0,
      "step": 69880
    },
    {
      "entropy": 6.022111272811889,
      "epoch": 6.17140586365242,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00019627856213951577,
      "loss": 5.2225,
      "mean_token_accuracy": 0.2258414477109909,
      "num_tokens": 130294740.0,
      "step": 69885
    },
    {
      "entropy": 6.125735902786255,
      "epoch": 6.17184740374426,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019624906175840258,
      "loss": 5.2769,
      "mean_token_accuracy": 0.2233436107635498,
      "num_tokens": 130304757.0,
      "step": 69890
    },
    {
      "entropy": 6.071077585220337,
      "epoch": 6.1722889438361,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019621956291997544,
      "loss": 5.2825,
      "mean_token_accuracy": 0.22298537492752074,
      "num_tokens": 130314596.0,
      "step": 69895
    },
    {
      "entropy": 6.140512323379516,
      "epoch": 6.172730483927941,
      "grad_norm": 1.28125,
      "learning_rate": 0.00019619006562481228,
      "loss": 5.3222,
      "mean_token_accuracy": 0.21739897429943084,
      "num_tokens": 130324454.0,
      "step": 69900
    },
    {
      "entropy": 6.081811141967774,
      "epoch": 6.173172024019781,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001961605698734909,
      "loss": 5.2487,
      "mean_token_accuracy": 0.22695103138685227,
      "num_tokens": 130334204.0,
      "step": 69905
    },
    {
      "entropy": 6.065596628189087,
      "epoch": 6.173613564111621,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019613107566658917,
      "loss": 5.2025,
      "mean_token_accuracy": 0.232583649456501,
      "num_tokens": 130343312.0,
      "step": 69910
    },
    {
      "entropy": 6.012376070022583,
      "epoch": 6.174055104203462,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019610158300468472,
      "loss": 5.1901,
      "mean_token_accuracy": 0.2288077712059021,
      "num_tokens": 130354502.0,
      "step": 69915
    },
    {
      "entropy": 6.071783781051636,
      "epoch": 6.174496644295302,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001960720918883555,
      "loss": 5.2372,
      "mean_token_accuracy": 0.2238459274172783,
      "num_tokens": 130363978.0,
      "step": 69920
    },
    {
      "entropy": 6.04493932723999,
      "epoch": 6.174938184387142,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019604260231817906,
      "loss": 5.1898,
      "mean_token_accuracy": 0.2297036439180374,
      "num_tokens": 130374449.0,
      "step": 69925
    },
    {
      "entropy": 5.997508192062378,
      "epoch": 6.175379724478983,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00019601311429473306,
      "loss": 5.1274,
      "mean_token_accuracy": 0.2345612734556198,
      "num_tokens": 130383928.0,
      "step": 69930
    },
    {
      "entropy": 5.9959362030029295,
      "epoch": 6.175821264570823,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019598362781859537,
      "loss": 5.1173,
      "mean_token_accuracy": 0.2383663296699524,
      "num_tokens": 130393399.0,
      "step": 69935
    },
    {
      "entropy": 6.051262378692627,
      "epoch": 6.176262804662663,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001959541428903434,
      "loss": 5.2101,
      "mean_token_accuracy": 0.22398554533720016,
      "num_tokens": 130403698.0,
      "step": 69940
    },
    {
      "entropy": 6.100199556350708,
      "epoch": 6.176704344754504,
      "grad_norm": 1.5,
      "learning_rate": 0.00019592465951055483,
      "loss": 5.2337,
      "mean_token_accuracy": 0.22375778555870057,
      "num_tokens": 130413020.0,
      "step": 69945
    },
    {
      "entropy": 5.966310930252075,
      "epoch": 6.177145884846344,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019589517767980714,
      "loss": 5.153,
      "mean_token_accuracy": 0.23097554743289947,
      "num_tokens": 130422507.0,
      "step": 69950
    },
    {
      "entropy": 6.0385571956634525,
      "epoch": 6.177587424938184,
      "grad_norm": 1.4375,
      "learning_rate": 0.000195865697398678,
      "loss": 5.1965,
      "mean_token_accuracy": 0.22592678666114807,
      "num_tokens": 130431829.0,
      "step": 69955
    },
    {
      "entropy": 5.979645919799805,
      "epoch": 6.178028965030025,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001958362186677448,
      "loss": 5.0839,
      "mean_token_accuracy": 0.23814509063959122,
      "num_tokens": 130441394.0,
      "step": 69960
    },
    {
      "entropy": 5.9859854221344,
      "epoch": 6.178470505121865,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001958067414875851,
      "loss": 5.1622,
      "mean_token_accuracy": 0.2390788897871971,
      "num_tokens": 130450069.0,
      "step": 69965
    },
    {
      "entropy": 5.970899820327759,
      "epoch": 6.178912045213705,
      "grad_norm": 1.265625,
      "learning_rate": 0.00019577726585877624,
      "loss": 5.1699,
      "mean_token_accuracy": 0.23079435527324677,
      "num_tokens": 130460594.0,
      "step": 69970
    },
    {
      "entropy": 6.021809673309326,
      "epoch": 6.179353585305546,
      "grad_norm": 1.5,
      "learning_rate": 0.00019574779178189566,
      "loss": 5.2369,
      "mean_token_accuracy": 0.23041145652532577,
      "num_tokens": 130469768.0,
      "step": 69975
    },
    {
      "entropy": 6.0449583530426025,
      "epoch": 6.179795125397386,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00019571831925752086,
      "loss": 5.1425,
      "mean_token_accuracy": 0.2305113360285759,
      "num_tokens": 130479244.0,
      "step": 69980
    },
    {
      "entropy": 6.040430021286011,
      "epoch": 6.180236665489226,
      "grad_norm": 1.375,
      "learning_rate": 0.00019568884828622912,
      "loss": 5.1178,
      "mean_token_accuracy": 0.2350597381591797,
      "num_tokens": 130489156.0,
      "step": 69985
    },
    {
      "entropy": 6.056056833267212,
      "epoch": 6.180678205581067,
      "grad_norm": 1.5,
      "learning_rate": 0.0001956593788685977,
      "loss": 5.271,
      "mean_token_accuracy": 0.22522613555192947,
      "num_tokens": 130499312.0,
      "step": 69990
    },
    {
      "entropy": 5.934847211837768,
      "epoch": 6.181119745672907,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019562991100520384,
      "loss": 5.106,
      "mean_token_accuracy": 0.2375900775194168,
      "num_tokens": 130509393.0,
      "step": 69995
    },
    {
      "entropy": 6.063180923461914,
      "epoch": 6.181561285764747,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019560044469662497,
      "loss": 5.1975,
      "mean_token_accuracy": 0.2289879560470581,
      "num_tokens": 130519811.0,
      "step": 70000
    },
    {
      "entropy": 6.047967624664307,
      "epoch": 6.182002825856587,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019557097994343824,
      "loss": 5.2436,
      "mean_token_accuracy": 0.2235588788986206,
      "num_tokens": 130529609.0,
      "step": 70005
    },
    {
      "entropy": 6.050255250930786,
      "epoch": 6.182444365948428,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001955415167462208,
      "loss": 5.2064,
      "mean_token_accuracy": 0.22814861685037613,
      "num_tokens": 130539461.0,
      "step": 70010
    },
    {
      "entropy": 6.061982917785644,
      "epoch": 6.182885906040268,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00019551205510554982,
      "loss": 5.1869,
      "mean_token_accuracy": 0.23100291043519974,
      "num_tokens": 130548091.0,
      "step": 70015
    },
    {
      "entropy": 5.970192289352417,
      "epoch": 6.183327446132108,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019548259502200256,
      "loss": 5.164,
      "mean_token_accuracy": 0.23190092742443086,
      "num_tokens": 130558700.0,
      "step": 70020
    },
    {
      "entropy": 6.00423002243042,
      "epoch": 6.183768986223949,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019545313649615599,
      "loss": 5.1082,
      "mean_token_accuracy": 0.23298227936029434,
      "num_tokens": 130567083.0,
      "step": 70025
    },
    {
      "entropy": 6.083107948303223,
      "epoch": 6.184210526315789,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001954236795285872,
      "loss": 5.261,
      "mean_token_accuracy": 0.22380055189132692,
      "num_tokens": 130576091.0,
      "step": 70030
    },
    {
      "entropy": 6.03643102645874,
      "epoch": 6.184652066407629,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001953942241198733,
      "loss": 5.2183,
      "mean_token_accuracy": 0.22701927721500398,
      "num_tokens": 130585173.0,
      "step": 70035
    },
    {
      "entropy": 5.991338586807251,
      "epoch": 6.18509360649947,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019536477027059124,
      "loss": 5.1776,
      "mean_token_accuracy": 0.2331307426095009,
      "num_tokens": 130594875.0,
      "step": 70040
    },
    {
      "entropy": 6.075282096862793,
      "epoch": 6.18553514659131,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001953353179813181,
      "loss": 5.2321,
      "mean_token_accuracy": 0.22366651892662048,
      "num_tokens": 130604036.0,
      "step": 70045
    },
    {
      "entropy": 6.058257246017456,
      "epoch": 6.1859766866831505,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019530586725263082,
      "loss": 5.2355,
      "mean_token_accuracy": 0.2301165997982025,
      "num_tokens": 130613135.0,
      "step": 70050
    },
    {
      "entropy": 5.999743795394897,
      "epoch": 6.186418226774991,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001952764180851062,
      "loss": 5.1404,
      "mean_token_accuracy": 0.22902783006429672,
      "num_tokens": 130621715.0,
      "step": 70055
    },
    {
      "entropy": 6.003078508377075,
      "epoch": 6.1868597668668315,
      "grad_norm": 1.375,
      "learning_rate": 0.00019524697047932117,
      "loss": 5.1667,
      "mean_token_accuracy": 0.2364387646317482,
      "num_tokens": 130630561.0,
      "step": 70060
    },
    {
      "entropy": 6.022984266281128,
      "epoch": 6.1873013069586715,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00019521752443585273,
      "loss": 5.2237,
      "mean_token_accuracy": 0.22167448252439498,
      "num_tokens": 130639855.0,
      "step": 70065
    },
    {
      "entropy": 6.076980400085449,
      "epoch": 6.1877428470505125,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019518807995527754,
      "loss": 5.1951,
      "mean_token_accuracy": 0.2169481635093689,
      "num_tokens": 130648924.0,
      "step": 70070
    },
    {
      "entropy": 6.0188699722290036,
      "epoch": 6.1881843871423525,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019515863703817256,
      "loss": 5.2031,
      "mean_token_accuracy": 0.2309545874595642,
      "num_tokens": 130658706.0,
      "step": 70075
    },
    {
      "entropy": 5.992365550994873,
      "epoch": 6.188625927234193,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001951291956851143,
      "loss": 5.1826,
      "mean_token_accuracy": 0.2340908855199814,
      "num_tokens": 130668124.0,
      "step": 70080
    },
    {
      "entropy": 5.943693542480469,
      "epoch": 6.1890674673260335,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019509975589667984,
      "loss": 5.1479,
      "mean_token_accuracy": 0.24210933446884156,
      "num_tokens": 130676446.0,
      "step": 70085
    },
    {
      "entropy": 6.112762928009033,
      "epoch": 6.189509007417874,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019507031767344564,
      "loss": 5.2508,
      "mean_token_accuracy": 0.21969628036022187,
      "num_tokens": 130686198.0,
      "step": 70090
    },
    {
      "entropy": 6.1015944480896,
      "epoch": 6.189950547509714,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001950408810159885,
      "loss": 5.2919,
      "mean_token_accuracy": 0.2177841395139694,
      "num_tokens": 130696786.0,
      "step": 70095
    },
    {
      "entropy": 6.101739740371704,
      "epoch": 6.190392087601555,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00019501144592488497,
      "loss": 5.2246,
      "mean_token_accuracy": 0.22574732154607774,
      "num_tokens": 130705439.0,
      "step": 70100
    },
    {
      "entropy": 6.0868267059326175,
      "epoch": 6.190833627693395,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019498201240071173,
      "loss": 5.2786,
      "mean_token_accuracy": 0.22094228863716125,
      "num_tokens": 130714285.0,
      "step": 70105
    },
    {
      "entropy": 6.050561046600341,
      "epoch": 6.191275167785235,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001949525804440454,
      "loss": 5.2463,
      "mean_token_accuracy": 0.2315311923623085,
      "num_tokens": 130723054.0,
      "step": 70110
    },
    {
      "entropy": 6.034504032135009,
      "epoch": 6.191716707877076,
      "grad_norm": 1.53125,
      "learning_rate": 0.00019492315005546253,
      "loss": 5.175,
      "mean_token_accuracy": 0.23649818301200867,
      "num_tokens": 130730723.0,
      "step": 70115
    },
    {
      "entropy": 6.052843618392944,
      "epoch": 6.192158247968916,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019489372123553955,
      "loss": 5.1329,
      "mean_token_accuracy": 0.23088832795619965,
      "num_tokens": 130740442.0,
      "step": 70120
    },
    {
      "entropy": 6.018727254867554,
      "epoch": 6.192599788060756,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019486429398485295,
      "loss": 5.1687,
      "mean_token_accuracy": 0.23480471819639206,
      "num_tokens": 130749264.0,
      "step": 70125
    },
    {
      "entropy": 6.000924062728882,
      "epoch": 6.193041328152597,
      "grad_norm": 1.265625,
      "learning_rate": 0.00019483486830397935,
      "loss": 5.1615,
      "mean_token_accuracy": 0.23322925120592117,
      "num_tokens": 130758133.0,
      "step": 70130
    },
    {
      "entropy": 6.06435194015503,
      "epoch": 6.193482868244437,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019480544419349504,
      "loss": 5.2177,
      "mean_token_accuracy": 0.2271888241171837,
      "num_tokens": 130767789.0,
      "step": 70135
    },
    {
      "entropy": 6.005453443527221,
      "epoch": 6.193924408336277,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019477602165397657,
      "loss": 5.1417,
      "mean_token_accuracy": 0.2324712321162224,
      "num_tokens": 130776634.0,
      "step": 70140
    },
    {
      "entropy": 6.104427003860474,
      "epoch": 6.194365948428118,
      "grad_norm": 1.375,
      "learning_rate": 0.00019474660068600003,
      "loss": 5.3187,
      "mean_token_accuracy": 0.21400994956493377,
      "num_tokens": 130785991.0,
      "step": 70145
    },
    {
      "entropy": 6.006189441680908,
      "epoch": 6.194807488519958,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001947171812901421,
      "loss": 5.1334,
      "mean_token_accuracy": 0.23225252330303192,
      "num_tokens": 130795395.0,
      "step": 70150
    },
    {
      "entropy": 6.072198343276978,
      "epoch": 6.195249028611798,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019468776346697887,
      "loss": 5.2571,
      "mean_token_accuracy": 0.22375068366527556,
      "num_tokens": 130804472.0,
      "step": 70155
    },
    {
      "entropy": 6.098814916610718,
      "epoch": 6.195690568703638,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019465834721708663,
      "loss": 5.2429,
      "mean_token_accuracy": 0.22679976969957352,
      "num_tokens": 130813738.0,
      "step": 70160
    },
    {
      "entropy": 6.052027177810669,
      "epoch": 6.196132108795479,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001946289325410418,
      "loss": 5.1496,
      "mean_token_accuracy": 0.23175779730081558,
      "num_tokens": 130824097.0,
      "step": 70165
    },
    {
      "entropy": 6.0447381973266605,
      "epoch": 6.196573648887319,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019459951943942027,
      "loss": 5.2191,
      "mean_token_accuracy": 0.2303340584039688,
      "num_tokens": 130833381.0,
      "step": 70170
    },
    {
      "entropy": 6.036796808242798,
      "epoch": 6.197015188979159,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001945701079127986,
      "loss": 5.1687,
      "mean_token_accuracy": 0.22819215804338455,
      "num_tokens": 130841991.0,
      "step": 70175
    },
    {
      "entropy": 5.976150798797607,
      "epoch": 6.197456729071,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019454069796175278,
      "loss": 5.1411,
      "mean_token_accuracy": 0.23563386499881744,
      "num_tokens": 130850956.0,
      "step": 70180
    },
    {
      "entropy": 6.005601406097412,
      "epoch": 6.19789826916284,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019451128958685888,
      "loss": 5.2265,
      "mean_token_accuracy": 0.22484132051467895,
      "num_tokens": 130861415.0,
      "step": 70185
    },
    {
      "entropy": 6.030205392837525,
      "epoch": 6.19833980925468,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019448188278869293,
      "loss": 5.1613,
      "mean_token_accuracy": 0.2310629218816757,
      "num_tokens": 130869958.0,
      "step": 70190
    },
    {
      "entropy": 6.015576171875,
      "epoch": 6.198781349346521,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00019445247756783124,
      "loss": 5.1046,
      "mean_token_accuracy": 0.2334784671664238,
      "num_tokens": 130879045.0,
      "step": 70195
    },
    {
      "entropy": 6.036749982833863,
      "epoch": 6.199222889438361,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001944230739248497,
      "loss": 5.2337,
      "mean_token_accuracy": 0.2307177245616913,
      "num_tokens": 130887800.0,
      "step": 70200
    },
    {
      "entropy": 6.002271270751953,
      "epoch": 6.199664429530201,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019439367186032435,
      "loss": 5.1962,
      "mean_token_accuracy": 0.22624898403882981,
      "num_tokens": 130897296.0,
      "step": 70205
    },
    {
      "entropy": 6.053735971450806,
      "epoch": 6.200105969622042,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00019436427137483098,
      "loss": 5.1925,
      "mean_token_accuracy": 0.22484144419431687,
      "num_tokens": 130906659.0,
      "step": 70210
    },
    {
      "entropy": 6.124346590042114,
      "epoch": 6.200547509713882,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019433487246894578,
      "loss": 5.2464,
      "mean_token_accuracy": 0.22820061594247817,
      "num_tokens": 130916591.0,
      "step": 70215
    },
    {
      "entropy": 5.996373796463013,
      "epoch": 6.200989049805722,
      "grad_norm": 1.53125,
      "learning_rate": 0.00019430547514324452,
      "loss": 5.1404,
      "mean_token_accuracy": 0.23245691359043122,
      "num_tokens": 130925746.0,
      "step": 70220
    },
    {
      "entropy": 6.085538005828857,
      "epoch": 6.201430589897563,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001942760793983031,
      "loss": 5.2497,
      "mean_token_accuracy": 0.22478357702493668,
      "num_tokens": 130935122.0,
      "step": 70225
    },
    {
      "entropy": 6.055449819564819,
      "epoch": 6.201872129989403,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019424668523469746,
      "loss": 5.2802,
      "mean_token_accuracy": 0.22259458303451538,
      "num_tokens": 130944992.0,
      "step": 70230
    },
    {
      "entropy": 6.022356081008911,
      "epoch": 6.202313670081243,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019421729265300315,
      "loss": 5.1514,
      "mean_token_accuracy": 0.23503215461969376,
      "num_tokens": 130953650.0,
      "step": 70235
    },
    {
      "entropy": 6.051391696929931,
      "epoch": 6.202755210173084,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001941879016537963,
      "loss": 5.1729,
      "mean_token_accuracy": 0.2317590370774269,
      "num_tokens": 130963050.0,
      "step": 70240
    },
    {
      "entropy": 6.011316871643066,
      "epoch": 6.203196750264924,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001941585122376524,
      "loss": 5.2207,
      "mean_token_accuracy": 0.23915302008390427,
      "num_tokens": 130972477.0,
      "step": 70245
    },
    {
      "entropy": 5.984680891036987,
      "epoch": 6.203638290356764,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019412912440514734,
      "loss": 5.142,
      "mean_token_accuracy": 0.2341024622321129,
      "num_tokens": 130981258.0,
      "step": 70250
    },
    {
      "entropy": 6.030043125152588,
      "epoch": 6.204079830448605,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00019409973815685667,
      "loss": 5.0787,
      "mean_token_accuracy": 0.23605849742889404,
      "num_tokens": 130990019.0,
      "step": 70255
    },
    {
      "entropy": 5.993234252929687,
      "epoch": 6.204521370540445,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019407035349335616,
      "loss": 5.1801,
      "mean_token_accuracy": 0.23311799317598342,
      "num_tokens": 131000868.0,
      "step": 70260
    },
    {
      "entropy": 6.047357654571533,
      "epoch": 6.204962910632285,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019404097041522144,
      "loss": 5.2506,
      "mean_token_accuracy": 0.22435102462768555,
      "num_tokens": 131010203.0,
      "step": 70265
    },
    {
      "entropy": 6.017895650863648,
      "epoch": 6.205404450724126,
      "grad_norm": 1.34375,
      "learning_rate": 0.000194011588923028,
      "loss": 5.1498,
      "mean_token_accuracy": 0.23852481842041015,
      "num_tokens": 131020653.0,
      "step": 70270
    },
    {
      "entropy": 6.047212409973144,
      "epoch": 6.205845990815966,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019398220901735148,
      "loss": 5.1532,
      "mean_token_accuracy": 0.23624677062034607,
      "num_tokens": 131029918.0,
      "step": 70275
    },
    {
      "entropy": 5.989521217346192,
      "epoch": 6.206287530907806,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001939528306987674,
      "loss": 5.1532,
      "mean_token_accuracy": 0.22736977636814118,
      "num_tokens": 131037649.0,
      "step": 70280
    },
    {
      "entropy": 5.977216148376465,
      "epoch": 6.206729070999646,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019392345396785133,
      "loss": 5.1662,
      "mean_token_accuracy": 0.23617999106645585,
      "num_tokens": 131046499.0,
      "step": 70285
    },
    {
      "entropy": 6.053515529632568,
      "epoch": 6.207170611091487,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019389407882517863,
      "loss": 5.2517,
      "mean_token_accuracy": 0.23134464174509048,
      "num_tokens": 131056680.0,
      "step": 70290
    },
    {
      "entropy": 6.079757452011108,
      "epoch": 6.207612151183327,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00019386470527132482,
      "loss": 5.1481,
      "mean_token_accuracy": 0.2317191481590271,
      "num_tokens": 131066026.0,
      "step": 70295
    },
    {
      "entropy": 6.003868103027344,
      "epoch": 6.208053691275167,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019383533330686527,
      "loss": 5.1573,
      "mean_token_accuracy": 0.23286425471305847,
      "num_tokens": 131075210.0,
      "step": 70300
    },
    {
      "entropy": 5.857029008865356,
      "epoch": 6.208495231367008,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001938059629323754,
      "loss": 5.0363,
      "mean_token_accuracy": 0.2474694862961769,
      "num_tokens": 131084414.0,
      "step": 70305
    },
    {
      "entropy": 5.937809610366822,
      "epoch": 6.208936771458848,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019377659414843056,
      "loss": 5.1413,
      "mean_token_accuracy": 0.237186698615551,
      "num_tokens": 131094199.0,
      "step": 70310
    },
    {
      "entropy": 6.015278673171997,
      "epoch": 6.209378311550688,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00019374722695560603,
      "loss": 5.1188,
      "mean_token_accuracy": 0.23353148251771927,
      "num_tokens": 131102946.0,
      "step": 70315
    },
    {
      "entropy": 6.010131311416626,
      "epoch": 6.209819851642529,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019371786135447704,
      "loss": 5.1294,
      "mean_token_accuracy": 0.23459388166666031,
      "num_tokens": 131112055.0,
      "step": 70320
    },
    {
      "entropy": 5.990596008300781,
      "epoch": 6.210261391734369,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000193688497345619,
      "loss": 5.1037,
      "mean_token_accuracy": 0.2336116388440132,
      "num_tokens": 131121669.0,
      "step": 70325
    },
    {
      "entropy": 6.065993928909302,
      "epoch": 6.2107029318262095,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000193659134929607,
      "loss": 5.1744,
      "mean_token_accuracy": 0.22914063036441804,
      "num_tokens": 131131113.0,
      "step": 70330
    },
    {
      "entropy": 6.0794164657592775,
      "epoch": 6.21114447191805,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019362977410701637,
      "loss": 5.1709,
      "mean_token_accuracy": 0.2337048903107643,
      "num_tokens": 131140636.0,
      "step": 70335
    },
    {
      "entropy": 5.983165168762207,
      "epoch": 6.2115860120098905,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019360041487842205,
      "loss": 5.1239,
      "mean_token_accuracy": 0.2384308844804764,
      "num_tokens": 131150614.0,
      "step": 70340
    },
    {
      "entropy": 5.984439945220947,
      "epoch": 6.2120275521017305,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001935710572443995,
      "loss": 5.2095,
      "mean_token_accuracy": 0.22774600833654404,
      "num_tokens": 131159850.0,
      "step": 70345
    },
    {
      "entropy": 6.057158279418945,
      "epoch": 6.2124690921935715,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001935417012055235,
      "loss": 5.2515,
      "mean_token_accuracy": 0.22383858114480973,
      "num_tokens": 131169125.0,
      "step": 70350
    },
    {
      "entropy": 6.142516994476319,
      "epoch": 6.2129106322854115,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001935123467623693,
      "loss": 5.3193,
      "mean_token_accuracy": 0.22107348144054412,
      "num_tokens": 131179590.0,
      "step": 70355
    },
    {
      "entropy": 5.990843439102173,
      "epoch": 6.213352172377252,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019348299391551183,
      "loss": 5.1517,
      "mean_token_accuracy": 0.23903966844081878,
      "num_tokens": 131189212.0,
      "step": 70360
    },
    {
      "entropy": 6.024285745620728,
      "epoch": 6.2137937124690925,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001934536426655261,
      "loss": 5.1568,
      "mean_token_accuracy": 0.231465445458889,
      "num_tokens": 131197731.0,
      "step": 70365
    },
    {
      "entropy": 5.984620189666748,
      "epoch": 6.214235252560933,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019342429301298726,
      "loss": 5.1481,
      "mean_token_accuracy": 0.23504835963249207,
      "num_tokens": 131206417.0,
      "step": 70370
    },
    {
      "entropy": 5.9503191947937015,
      "epoch": 6.214676792652773,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019339494495847003,
      "loss": 5.1341,
      "mean_token_accuracy": 0.23686912804841995,
      "num_tokens": 131215750.0,
      "step": 70375
    },
    {
      "entropy": 6.059508657455444,
      "epoch": 6.215118332744614,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019336559850254953,
      "loss": 5.213,
      "mean_token_accuracy": 0.22855344265699387,
      "num_tokens": 131225572.0,
      "step": 70380
    },
    {
      "entropy": 6.037755918502808,
      "epoch": 6.215559872836454,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001933362536458004,
      "loss": 5.1915,
      "mean_token_accuracy": 0.22707168608903885,
      "num_tokens": 131235105.0,
      "step": 70385
    },
    {
      "entropy": 6.011550712585449,
      "epoch": 6.216001412928294,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019330691038879766,
      "loss": 5.1524,
      "mean_token_accuracy": 0.23052674531936646,
      "num_tokens": 131244482.0,
      "step": 70390
    },
    {
      "entropy": 5.988933563232422,
      "epoch": 6.216442953020135,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00019327756873211612,
      "loss": 5.1192,
      "mean_token_accuracy": 0.2314819872379303,
      "num_tokens": 131253836.0,
      "step": 70395
    },
    {
      "entropy": 5.99286847114563,
      "epoch": 6.216884493111975,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001932482286763305,
      "loss": 5.1139,
      "mean_token_accuracy": 0.23920626938343048,
      "num_tokens": 131263007.0,
      "step": 70400
    },
    {
      "entropy": 6.008177423477173,
      "epoch": 6.217326033203815,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019321889022201562,
      "loss": 5.1557,
      "mean_token_accuracy": 0.2331790506839752,
      "num_tokens": 131271890.0,
      "step": 70405
    },
    {
      "entropy": 6.064021444320678,
      "epoch": 6.217767573295656,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019318955336974607,
      "loss": 5.2091,
      "mean_token_accuracy": 0.22623585015535355,
      "num_tokens": 131280617.0,
      "step": 70410
    },
    {
      "entropy": 6.051123809814453,
      "epoch": 6.218209113387496,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019316021812009672,
      "loss": 5.164,
      "mean_token_accuracy": 0.23225396275520324,
      "num_tokens": 131290075.0,
      "step": 70415
    },
    {
      "entropy": 6.078009033203125,
      "epoch": 6.218650653479336,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019313088447364218,
      "loss": 5.1896,
      "mean_token_accuracy": 0.2298753246665001,
      "num_tokens": 131298705.0,
      "step": 70420
    },
    {
      "entropy": 5.9998174667358395,
      "epoch": 6.219092193571177,
      "grad_norm": 1.328125,
      "learning_rate": 0.000193101552430957,
      "loss": 5.1026,
      "mean_token_accuracy": 0.23370446711778642,
      "num_tokens": 131307912.0,
      "step": 70425
    },
    {
      "entropy": 6.050546169281006,
      "epoch": 6.219533733663017,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019307222199261575,
      "loss": 5.243,
      "mean_token_accuracy": 0.22615303993225097,
      "num_tokens": 131317380.0,
      "step": 70430
    },
    {
      "entropy": 6.128535556793213,
      "epoch": 6.219975273754857,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001930428931591932,
      "loss": 5.2649,
      "mean_token_accuracy": 0.22363306432962418,
      "num_tokens": 131327337.0,
      "step": 70435
    },
    {
      "entropy": 6.104124212265015,
      "epoch": 6.220416813846697,
      "grad_norm": 1.5,
      "learning_rate": 0.00019301356593126378,
      "loss": 5.3019,
      "mean_token_accuracy": 0.22052939236164093,
      "num_tokens": 131336400.0,
      "step": 70440
    },
    {
      "entropy": 6.119446706771851,
      "epoch": 6.220858353938538,
      "grad_norm": 1.53125,
      "learning_rate": 0.00019298424030940198,
      "loss": 5.208,
      "mean_token_accuracy": 0.22894802838563919,
      "num_tokens": 131345987.0,
      "step": 70445
    },
    {
      "entropy": 6.07658314704895,
      "epoch": 6.221299894030378,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001929549162941821,
      "loss": 5.2989,
      "mean_token_accuracy": 0.22618066668510436,
      "num_tokens": 131355852.0,
      "step": 70450
    },
    {
      "entropy": 6.083773040771485,
      "epoch": 6.221741434122218,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019292559388617886,
      "loss": 5.2653,
      "mean_token_accuracy": 0.2259996637701988,
      "num_tokens": 131365327.0,
      "step": 70455
    },
    {
      "entropy": 6.035950183868408,
      "epoch": 6.222182974214059,
      "grad_norm": 1.546875,
      "learning_rate": 0.00019289627308596665,
      "loss": 5.1971,
      "mean_token_accuracy": 0.22507049590349198,
      "num_tokens": 131374866.0,
      "step": 70460
    },
    {
      "entropy": 6.1159961223602295,
      "epoch": 6.222624514305899,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019286695389411962,
      "loss": 5.2389,
      "mean_token_accuracy": 0.2189537301659584,
      "num_tokens": 131384037.0,
      "step": 70465
    },
    {
      "entropy": 5.96142086982727,
      "epoch": 6.223066054397739,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019283763631121225,
      "loss": 5.093,
      "mean_token_accuracy": 0.2384645774960518,
      "num_tokens": 131392735.0,
      "step": 70470
    },
    {
      "entropy": 6.065029144287109,
      "epoch": 6.22350759448958,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019280832033781892,
      "loss": 5.1747,
      "mean_token_accuracy": 0.23368444442749023,
      "num_tokens": 131402408.0,
      "step": 70475
    },
    {
      "entropy": 6.080005931854248,
      "epoch": 6.22394913458142,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019277900597451382,
      "loss": 5.208,
      "mean_token_accuracy": 0.22671598345041274,
      "num_tokens": 131411137.0,
      "step": 70480
    },
    {
      "entropy": 5.954145240783691,
      "epoch": 6.22439067467326,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00019274969322187124,
      "loss": 5.1196,
      "mean_token_accuracy": 0.23631925731897355,
      "num_tokens": 131421361.0,
      "step": 70485
    },
    {
      "entropy": 6.0357630252838135,
      "epoch": 6.224832214765101,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019272038208046545,
      "loss": 5.2016,
      "mean_token_accuracy": 0.23249805569648743,
      "num_tokens": 131430829.0,
      "step": 70490
    },
    {
      "entropy": 6.045338344573975,
      "epoch": 6.225273754856941,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001926910725508705,
      "loss": 5.2693,
      "mean_token_accuracy": 0.22446349561214446,
      "num_tokens": 131439734.0,
      "step": 70495
    },
    {
      "entropy": 5.958064603805542,
      "epoch": 6.225715294948781,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019266176463366065,
      "loss": 5.1008,
      "mean_token_accuracy": 0.23591423630714417,
      "num_tokens": 131450120.0,
      "step": 70500
    },
    {
      "entropy": 6.060775947570801,
      "epoch": 6.226156835040622,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019263245832941,
      "loss": 5.2365,
      "mean_token_accuracy": 0.2192584455013275,
      "num_tokens": 131459254.0,
      "step": 70505
    },
    {
      "entropy": 6.075843858718872,
      "epoch": 6.226598375132462,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019260315363869268,
      "loss": 5.2162,
      "mean_token_accuracy": 0.2291179582476616,
      "num_tokens": 131468887.0,
      "step": 70510
    },
    {
      "entropy": 6.111383438110352,
      "epoch": 6.227039915224302,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001925738505620826,
      "loss": 5.3896,
      "mean_token_accuracy": 0.2181619018316269,
      "num_tokens": 131479820.0,
      "step": 70515
    },
    {
      "entropy": 6.128213691711426,
      "epoch": 6.227481455316143,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019254454910015408,
      "loss": 5.3027,
      "mean_token_accuracy": 0.21894320100545883,
      "num_tokens": 131488894.0,
      "step": 70520
    },
    {
      "entropy": 6.057976102828979,
      "epoch": 6.227922995407983,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019251524925348085,
      "loss": 5.2455,
      "mean_token_accuracy": 0.22374192476272584,
      "num_tokens": 131499768.0,
      "step": 70525
    },
    {
      "entropy": 6.078690147399902,
      "epoch": 6.228364535499823,
      "grad_norm": 1.578125,
      "learning_rate": 0.000192485951022637,
      "loss": 5.192,
      "mean_token_accuracy": 0.22602156549692154,
      "num_tokens": 131508122.0,
      "step": 70530
    },
    {
      "entropy": 6.055310010910034,
      "epoch": 6.228806075591664,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019245665440819642,
      "loss": 5.2253,
      "mean_token_accuracy": 0.2290438309311867,
      "num_tokens": 131517531.0,
      "step": 70535
    },
    {
      "entropy": 6.071304559707642,
      "epoch": 6.229247615683504,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019242735941073303,
      "loss": 5.2414,
      "mean_token_accuracy": 0.22640594094991684,
      "num_tokens": 131526125.0,
      "step": 70540
    },
    {
      "entropy": 6.034442710876465,
      "epoch": 6.229689155775344,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019239806603082071,
      "loss": 5.1285,
      "mean_token_accuracy": 0.2363371193408966,
      "num_tokens": 131535002.0,
      "step": 70545
    },
    {
      "entropy": 6.020877981185913,
      "epoch": 6.230130695867185,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019236877426903333,
      "loss": 5.1923,
      "mean_token_accuracy": 0.22898783832788466,
      "num_tokens": 131544897.0,
      "step": 70550
    },
    {
      "entropy": 6.0592142105102536,
      "epoch": 6.230572235959025,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001923394841259447,
      "loss": 5.252,
      "mean_token_accuracy": 0.23365134447813035,
      "num_tokens": 131555119.0,
      "step": 70555
    },
    {
      "entropy": 6.0171479225158695,
      "epoch": 6.231013776050865,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019231019560212841,
      "loss": 5.1549,
      "mean_token_accuracy": 0.23445098251104354,
      "num_tokens": 131563878.0,
      "step": 70560
    },
    {
      "entropy": 6.017384195327759,
      "epoch": 6.231455316142705,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019228090869815852,
      "loss": 5.1536,
      "mean_token_accuracy": 0.2250404551625252,
      "num_tokens": 131572734.0,
      "step": 70565
    },
    {
      "entropy": 6.06180329322815,
      "epoch": 6.231896856234546,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00019225162341460856,
      "loss": 5.2611,
      "mean_token_accuracy": 0.2241060256958008,
      "num_tokens": 131581953.0,
      "step": 70570
    },
    {
      "entropy": 6.098414278030395,
      "epoch": 6.232338396326386,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019222233975205223,
      "loss": 5.2934,
      "mean_token_accuracy": 0.21997176706790925,
      "num_tokens": 131591769.0,
      "step": 70575
    },
    {
      "entropy": 6.0406341552734375,
      "epoch": 6.232779936418226,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00019219305771106304,
      "loss": 5.13,
      "mean_token_accuracy": 0.2321540743112564,
      "num_tokens": 131601253.0,
      "step": 70580
    },
    {
      "entropy": 5.987278318405151,
      "epoch": 6.233221476510067,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00019216377729221496,
      "loss": 5.0985,
      "mean_token_accuracy": 0.2375786632299423,
      "num_tokens": 131610894.0,
      "step": 70585
    },
    {
      "entropy": 6.093285179138183,
      "epoch": 6.233663016601907,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019213449849608127,
      "loss": 5.2776,
      "mean_token_accuracy": 0.22606266289949417,
      "num_tokens": 131619893.0,
      "step": 70590
    },
    {
      "entropy": 5.986587238311768,
      "epoch": 6.234104556693747,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001921052213232357,
      "loss": 5.1599,
      "mean_token_accuracy": 0.22969234138727188,
      "num_tokens": 131629436.0,
      "step": 70595
    },
    {
      "entropy": 6.008390808105469,
      "epoch": 6.234546096785588,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00019207594577425158,
      "loss": 5.1689,
      "mean_token_accuracy": 0.2247485935688019,
      "num_tokens": 131638395.0,
      "step": 70600
    },
    {
      "entropy": 6.0978411674499515,
      "epoch": 6.234987636877428,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001920466718497026,
      "loss": 5.2212,
      "mean_token_accuracy": 0.2190760314464569,
      "num_tokens": 131648342.0,
      "step": 70605
    },
    {
      "entropy": 6.116161298751831,
      "epoch": 6.2354291769692685,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001920173995501621,
      "loss": 5.3091,
      "mean_token_accuracy": 0.21680993139743804,
      "num_tokens": 131657067.0,
      "step": 70610
    },
    {
      "entropy": 6.018145132064819,
      "epoch": 6.235870717061109,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001919881288762036,
      "loss": 5.1661,
      "mean_token_accuracy": 0.23714479058980942,
      "num_tokens": 131666892.0,
      "step": 70615
    },
    {
      "entropy": 6.007786226272583,
      "epoch": 6.2363122571529495,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019195885982840045,
      "loss": 5.1916,
      "mean_token_accuracy": 0.22640355229377745,
      "num_tokens": 131677213.0,
      "step": 70620
    },
    {
      "entropy": 6.0716921329498295,
      "epoch": 6.2367537972447895,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019192959240732588,
      "loss": 5.2387,
      "mean_token_accuracy": 0.22372835576534272,
      "num_tokens": 131686731.0,
      "step": 70625
    },
    {
      "entropy": 6.0401082038879395,
      "epoch": 6.2371953373366305,
      "grad_norm": 1.375,
      "learning_rate": 0.0001919003266135535,
      "loss": 5.2271,
      "mean_token_accuracy": 0.23630117624998093,
      "num_tokens": 131696463.0,
      "step": 70630
    },
    {
      "entropy": 6.000450277328492,
      "epoch": 6.2376368774284705,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001918710624476564,
      "loss": 5.1732,
      "mean_token_accuracy": 0.2284565731883049,
      "num_tokens": 131705884.0,
      "step": 70635
    },
    {
      "entropy": 6.074663305282593,
      "epoch": 6.238078417520311,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00019184179991020794,
      "loss": 5.2269,
      "mean_token_accuracy": 0.23034612238407134,
      "num_tokens": 131714310.0,
      "step": 70640
    },
    {
      "entropy": 6.049645471572876,
      "epoch": 6.2385199576121515,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001918125390017813,
      "loss": 5.1564,
      "mean_token_accuracy": 0.23177050203084945,
      "num_tokens": 131723267.0,
      "step": 70645
    },
    {
      "entropy": 6.07457799911499,
      "epoch": 6.2389614977039916,
      "grad_norm": 1.53125,
      "learning_rate": 0.00019178327972294967,
      "loss": 5.1751,
      "mean_token_accuracy": 0.23254122585058212,
      "num_tokens": 131732075.0,
      "step": 70650
    },
    {
      "entropy": 6.143970966339111,
      "epoch": 6.239403037795832,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019175402207428634,
      "loss": 5.3356,
      "mean_token_accuracy": 0.21504477560520172,
      "num_tokens": 131742370.0,
      "step": 70655
    },
    {
      "entropy": 6.052599239349365,
      "epoch": 6.2398445778876725,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019172476605636436,
      "loss": 5.1208,
      "mean_token_accuracy": 0.23536004722118378,
      "num_tokens": 131751296.0,
      "step": 70660
    },
    {
      "entropy": 6.045238494873047,
      "epoch": 6.240286117979513,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001916955116697568,
      "loss": 5.222,
      "mean_token_accuracy": 0.22910029590129852,
      "num_tokens": 131761256.0,
      "step": 70665
    },
    {
      "entropy": 5.955167293548584,
      "epoch": 6.240727658071353,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019166625891503674,
      "loss": 5.1837,
      "mean_token_accuracy": 0.23309992551803588,
      "num_tokens": 131771301.0,
      "step": 70670
    },
    {
      "entropy": 6.088653564453125,
      "epoch": 6.241169198163194,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00019163700779277736,
      "loss": 5.2644,
      "mean_token_accuracy": 0.2175549879670143,
      "num_tokens": 131780539.0,
      "step": 70675
    },
    {
      "entropy": 6.094666242599487,
      "epoch": 6.241610738255034,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001916077583035516,
      "loss": 5.2,
      "mean_token_accuracy": 0.22718369215726852,
      "num_tokens": 131790464.0,
      "step": 70680
    },
    {
      "entropy": 6.069143056869507,
      "epoch": 6.242052278346874,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019157851044793246,
      "loss": 5.1753,
      "mean_token_accuracy": 0.23328556269407272,
      "num_tokens": 131799553.0,
      "step": 70685
    },
    {
      "entropy": 6.038270330429077,
      "epoch": 6.242493818438715,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019154926422649271,
      "loss": 5.2321,
      "mean_token_accuracy": 0.21882692724466324,
      "num_tokens": 131808835.0,
      "step": 70690
    },
    {
      "entropy": 6.009792184829712,
      "epoch": 6.242935358530555,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019152001963980552,
      "loss": 5.2004,
      "mean_token_accuracy": 0.2274614691734314,
      "num_tokens": 131818395.0,
      "step": 70695
    },
    {
      "entropy": 6.028041744232178,
      "epoch": 6.243376898622395,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019149077668844367,
      "loss": 5.1423,
      "mean_token_accuracy": 0.2356012210249901,
      "num_tokens": 131827267.0,
      "step": 70700
    },
    {
      "entropy": 6.081906080245972,
      "epoch": 6.243818438714236,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019146153537298,
      "loss": 5.2135,
      "mean_token_accuracy": 0.2252044141292572,
      "num_tokens": 131836656.0,
      "step": 70705
    },
    {
      "entropy": 6.053481101989746,
      "epoch": 6.244259978806076,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019143229569398734,
      "loss": 5.1758,
      "mean_token_accuracy": 0.23584651499986647,
      "num_tokens": 131845267.0,
      "step": 70710
    },
    {
      "entropy": 5.994307136535644,
      "epoch": 6.244701518897916,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00019140305765203843,
      "loss": 5.0877,
      "mean_token_accuracy": 0.2464616999030113,
      "num_tokens": 131854499.0,
      "step": 70715
    },
    {
      "entropy": 6.032397890090943,
      "epoch": 6.245143058989756,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019137382124770607,
      "loss": 5.2222,
      "mean_token_accuracy": 0.2281668156385422,
      "num_tokens": 131863786.0,
      "step": 70720
    },
    {
      "entropy": 5.947571754455566,
      "epoch": 6.245584599081597,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001913445864815631,
      "loss": 5.1235,
      "mean_token_accuracy": 0.23750457912683487,
      "num_tokens": 131873893.0,
      "step": 70725
    },
    {
      "entropy": 6.055856227874756,
      "epoch": 6.246026139173437,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019131535335418205,
      "loss": 5.2681,
      "mean_token_accuracy": 0.22133603692054749,
      "num_tokens": 131883980.0,
      "step": 70730
    },
    {
      "entropy": 5.981865358352661,
      "epoch": 6.246467679265277,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019128612186613558,
      "loss": 5.1411,
      "mean_token_accuracy": 0.23258715718984604,
      "num_tokens": 131892710.0,
      "step": 70735
    },
    {
      "entropy": 5.971516275405884,
      "epoch": 6.246909219357118,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001912568920179964,
      "loss": 5.1713,
      "mean_token_accuracy": 0.2261808529496193,
      "num_tokens": 131901793.0,
      "step": 70740
    },
    {
      "entropy": 6.14372820854187,
      "epoch": 6.247350759448958,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019122766381033713,
      "loss": 5.3709,
      "mean_token_accuracy": 0.21067800521850585,
      "num_tokens": 131910986.0,
      "step": 70745
    },
    {
      "entropy": 6.130183267593384,
      "epoch": 6.247792299540798,
      "grad_norm": 1.609375,
      "learning_rate": 0.00019119843724373027,
      "loss": 5.2648,
      "mean_token_accuracy": 0.22420018911361694,
      "num_tokens": 131919182.0,
      "step": 70750
    },
    {
      "entropy": 6.016904497146607,
      "epoch": 6.248233839632639,
      "grad_norm": 1.375,
      "learning_rate": 0.0001911692123187483,
      "loss": 5.1697,
      "mean_token_accuracy": 0.22247554808855058,
      "num_tokens": 131928734.0,
      "step": 70755
    },
    {
      "entropy": 6.075521802902221,
      "epoch": 6.248675379724479,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001911399890359638,
      "loss": 5.1745,
      "mean_token_accuracy": 0.23544640392065047,
      "num_tokens": 131937394.0,
      "step": 70760
    },
    {
      "entropy": 6.036504220962525,
      "epoch": 6.249116919816319,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019111076739594928,
      "loss": 5.2942,
      "mean_token_accuracy": 0.2212664470076561,
      "num_tokens": 131946646.0,
      "step": 70765
    },
    {
      "entropy": 6.002288341522217,
      "epoch": 6.24955845990816,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019108154739927707,
      "loss": 5.1941,
      "mean_token_accuracy": 0.2242419645190239,
      "num_tokens": 131956726.0,
      "step": 70770
    },
    {
      "entropy": 6.063654518127441,
      "epoch": 6.25,
      "grad_norm": 1.5,
      "learning_rate": 0.00019105232904651965,
      "loss": 5.2343,
      "mean_token_accuracy": 0.21615184098482132,
      "num_tokens": 131966385.0,
      "step": 70775
    },
    {
      "entropy": 6.03478741645813,
      "epoch": 6.25044154009184,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00019102311233824932,
      "loss": 5.1707,
      "mean_token_accuracy": 0.2305392935872078,
      "num_tokens": 131975615.0,
      "step": 70780
    },
    {
      "entropy": 6.101922607421875,
      "epoch": 6.250883080183681,
      "grad_norm": 1.578125,
      "learning_rate": 0.0001909938972750385,
      "loss": 5.1835,
      "mean_token_accuracy": 0.23119794726371765,
      "num_tokens": 131984837.0,
      "step": 70785
    },
    {
      "entropy": 6.05284276008606,
      "epoch": 6.251324620275521,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001909646838574594,
      "loss": 5.2068,
      "mean_token_accuracy": 0.22236884087324144,
      "num_tokens": 131993789.0,
      "step": 70790
    },
    {
      "entropy": 5.971593427658081,
      "epoch": 6.251766160367361,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001909354720860844,
      "loss": 5.0961,
      "mean_token_accuracy": 0.2426919624209404,
      "num_tokens": 132002993.0,
      "step": 70795
    },
    {
      "entropy": 6.022984170913697,
      "epoch": 6.252207700459202,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019090626196148564,
      "loss": 5.1597,
      "mean_token_accuracy": 0.2266380786895752,
      "num_tokens": 132012518.0,
      "step": 70800
    },
    {
      "entropy": 5.918777322769165,
      "epoch": 6.252649240551042,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001908770534842354,
      "loss": 5.1258,
      "mean_token_accuracy": 0.2328443005681038,
      "num_tokens": 132020923.0,
      "step": 70805
    },
    {
      "entropy": 6.026339149475097,
      "epoch": 6.253090780642882,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001908478466549059,
      "loss": 5.2171,
      "mean_token_accuracy": 0.21738932877779008,
      "num_tokens": 132030097.0,
      "step": 70810
    },
    {
      "entropy": 6.004847192764283,
      "epoch": 6.253532320734723,
      "grad_norm": 1.59375,
      "learning_rate": 0.0001908186414740692,
      "loss": 5.1833,
      "mean_token_accuracy": 0.22758155614137648,
      "num_tokens": 132038454.0,
      "step": 70815
    },
    {
      "entropy": 6.126782464981079,
      "epoch": 6.253973860826563,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019078943794229735,
      "loss": 5.3521,
      "mean_token_accuracy": 0.22022780328989028,
      "num_tokens": 132048435.0,
      "step": 70820
    },
    {
      "entropy": 6.063246345520019,
      "epoch": 6.254415400918403,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019076023606016262,
      "loss": 5.2291,
      "mean_token_accuracy": 0.22427231967449188,
      "num_tokens": 132058096.0,
      "step": 70825
    },
    {
      "entropy": 6.043432712554932,
      "epoch": 6.254856941010244,
      "grad_norm": 1.265625,
      "learning_rate": 0.00019073103582823692,
      "loss": 5.2037,
      "mean_token_accuracy": 0.22874394357204436,
      "num_tokens": 132068325.0,
      "step": 70830
    },
    {
      "entropy": 6.033414554595947,
      "epoch": 6.255298481102084,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019070183724709234,
      "loss": 5.1636,
      "mean_token_accuracy": 0.23110356330871581,
      "num_tokens": 132078175.0,
      "step": 70835
    },
    {
      "entropy": 5.991284942626953,
      "epoch": 6.255740021193924,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001906726403173008,
      "loss": 5.196,
      "mean_token_accuracy": 0.2222386986017227,
      "num_tokens": 132087494.0,
      "step": 70840
    },
    {
      "entropy": 6.0166419506072994,
      "epoch": 6.256181561285764,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0001906434450394342,
      "loss": 5.173,
      "mean_token_accuracy": 0.22716087549924852,
      "num_tokens": 132096355.0,
      "step": 70845
    },
    {
      "entropy": 5.985293769836426,
      "epoch": 6.256623101377605,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019061425141406464,
      "loss": 5.1241,
      "mean_token_accuracy": 0.24272262901067734,
      "num_tokens": 132105386.0,
      "step": 70850
    },
    {
      "entropy": 5.984474563598633,
      "epoch": 6.257064641469445,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019058505944176385,
      "loss": 5.1407,
      "mean_token_accuracy": 0.23190411627292634,
      "num_tokens": 132114311.0,
      "step": 70855
    },
    {
      "entropy": 6.0589393138885494,
      "epoch": 6.257506181561285,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019055586912310375,
      "loss": 5.2297,
      "mean_token_accuracy": 0.2232778400182724,
      "num_tokens": 132123316.0,
      "step": 70860
    },
    {
      "entropy": 5.935713768005371,
      "epoch": 6.257947721653126,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001905266804586561,
      "loss": 5.0357,
      "mean_token_accuracy": 0.25028409957885744,
      "num_tokens": 132133033.0,
      "step": 70865
    },
    {
      "entropy": 5.954876708984375,
      "epoch": 6.258389261744966,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001904974934489928,
      "loss": 5.1265,
      "mean_token_accuracy": 0.23657174408435822,
      "num_tokens": 132143242.0,
      "step": 70870
    },
    {
      "entropy": 6.0727091312408445,
      "epoch": 6.258830801836806,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019046830809468557,
      "loss": 5.1503,
      "mean_token_accuracy": 0.231142121553421,
      "num_tokens": 132152411.0,
      "step": 70875
    },
    {
      "entropy": 6.08269510269165,
      "epoch": 6.259272341928647,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00019043912439630607,
      "loss": 5.2684,
      "mean_token_accuracy": 0.22747729420661927,
      "num_tokens": 132162231.0,
      "step": 70880
    },
    {
      "entropy": 5.932106924057007,
      "epoch": 6.259713882020487,
      "grad_norm": 1.5,
      "learning_rate": 0.00019040994235442605,
      "loss": 5.0577,
      "mean_token_accuracy": 0.24423237890005112,
      "num_tokens": 132171061.0,
      "step": 70885
    },
    {
      "entropy": 6.088262224197388,
      "epoch": 6.2601554221123275,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019038076196961707,
      "loss": 5.1971,
      "mean_token_accuracy": 0.22435904145240784,
      "num_tokens": 132180274.0,
      "step": 70890
    },
    {
      "entropy": 6.050924253463745,
      "epoch": 6.260596962204168,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001903515832424509,
      "loss": 5.252,
      "mean_token_accuracy": 0.22088946402072906,
      "num_tokens": 132189211.0,
      "step": 70895
    },
    {
      "entropy": 6.0969523906707765,
      "epoch": 6.2610385022960084,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019032240617349912,
      "loss": 5.2355,
      "mean_token_accuracy": 0.22361504137516022,
      "num_tokens": 132198624.0,
      "step": 70900
    },
    {
      "entropy": 6.036590576171875,
      "epoch": 6.2614800423878485,
      "grad_norm": 1.28125,
      "learning_rate": 0.00019029323076333315,
      "loss": 5.241,
      "mean_token_accuracy": 0.22776662111282348,
      "num_tokens": 132209001.0,
      "step": 70905
    },
    {
      "entropy": 6.044690179824829,
      "epoch": 6.261921582479689,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019026405701252463,
      "loss": 5.197,
      "mean_token_accuracy": 0.22982307076454161,
      "num_tokens": 132217626.0,
      "step": 70910
    },
    {
      "entropy": 6.033207750320434,
      "epoch": 6.2623631225715295,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00019023488492164508,
      "loss": 5.214,
      "mean_token_accuracy": 0.22582222819328307,
      "num_tokens": 132227038.0,
      "step": 70915
    },
    {
      "entropy": 5.975070858001709,
      "epoch": 6.2628046626633695,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001902057144912659,
      "loss": 5.1286,
      "mean_token_accuracy": 0.22748504877090453,
      "num_tokens": 132236611.0,
      "step": 70920
    },
    {
      "entropy": 5.998774671554566,
      "epoch": 6.2632462027552105,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019017654572195858,
      "loss": 5.1275,
      "mean_token_accuracy": 0.21925905495882034,
      "num_tokens": 132245998.0,
      "step": 70925
    },
    {
      "entropy": 5.973236417770385,
      "epoch": 6.2636877428470505,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019014737861429442,
      "loss": 5.1052,
      "mean_token_accuracy": 0.23318101912736894,
      "num_tokens": 132254277.0,
      "step": 70930
    },
    {
      "entropy": 6.063209867477417,
      "epoch": 6.264129282938891,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019011821316884493,
      "loss": 5.2263,
      "mean_token_accuracy": 0.22322594970464707,
      "num_tokens": 132263184.0,
      "step": 70935
    },
    {
      "entropy": 5.9720055103302006,
      "epoch": 6.2645708230307315,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00019008904938618132,
      "loss": 5.1019,
      "mean_token_accuracy": 0.24089468866586686,
      "num_tokens": 132272227.0,
      "step": 70940
    },
    {
      "entropy": 5.926523923873901,
      "epoch": 6.265012363122572,
      "grad_norm": 1.484375,
      "learning_rate": 0.000190059887266875,
      "loss": 5.0609,
      "mean_token_accuracy": 0.2455934464931488,
      "num_tokens": 132280924.0,
      "step": 70945
    },
    {
      "entropy": 5.940511035919189,
      "epoch": 6.265453903214412,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019003072681149719,
      "loss": 5.0896,
      "mean_token_accuracy": 0.2436904415488243,
      "num_tokens": 132291274.0,
      "step": 70950
    },
    {
      "entropy": 5.93324875831604,
      "epoch": 6.265895443306253,
      "grad_norm": 1.609375,
      "learning_rate": 0.000190001568020619,
      "loss": 5.0204,
      "mean_token_accuracy": 0.24417092204093932,
      "num_tokens": 132299754.0,
      "step": 70955
    },
    {
      "entropy": 5.998865079879761,
      "epoch": 6.266336983398093,
      "grad_norm": 1.453125,
      "learning_rate": 0.00018997241089481189,
      "loss": 5.1853,
      "mean_token_accuracy": 0.22911240458488463,
      "num_tokens": 132308514.0,
      "step": 70960
    },
    {
      "entropy": 6.014474821090698,
      "epoch": 6.266778523489933,
      "grad_norm": 1.5,
      "learning_rate": 0.0001899432554346468,
      "loss": 5.1045,
      "mean_token_accuracy": 0.24051001965999602,
      "num_tokens": 132317186.0,
      "step": 70965
    },
    {
      "entropy": 6.0258832454681395,
      "epoch": 6.267220063581774,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000189914101640695,
      "loss": 5.1929,
      "mean_token_accuracy": 0.22628147155046463,
      "num_tokens": 132326803.0,
      "step": 70970
    },
    {
      "entropy": 6.096849536895752,
      "epoch": 6.267661603673614,
      "grad_norm": 1.421875,
      "learning_rate": 0.00018988494951352754,
      "loss": 5.2128,
      "mean_token_accuracy": 0.2262249231338501,
      "num_tokens": 132335840.0,
      "step": 70975
    },
    {
      "entropy": 6.056491041183472,
      "epoch": 6.268103143765454,
      "grad_norm": 1.3125,
      "learning_rate": 0.00018985579905371553,
      "loss": 5.1836,
      "mean_token_accuracy": 0.2246082380414009,
      "num_tokens": 132345438.0,
      "step": 70980
    },
    {
      "entropy": 6.035224866867066,
      "epoch": 6.268544683857295,
      "grad_norm": 1.546875,
      "learning_rate": 0.00018982665026182993,
      "loss": 5.1772,
      "mean_token_accuracy": 0.22927243262529373,
      "num_tokens": 132354695.0,
      "step": 70985
    },
    {
      "entropy": 6.037731122970581,
      "epoch": 6.268986223949135,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00018979750313844192,
      "loss": 5.225,
      "mean_token_accuracy": 0.2180541068315506,
      "num_tokens": 132363477.0,
      "step": 70990
    },
    {
      "entropy": 6.023589134216309,
      "epoch": 6.269427764040975,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018976835768412227,
      "loss": 5.1648,
      "mean_token_accuracy": 0.2268005356192589,
      "num_tokens": 132372292.0,
      "step": 70995
    },
    {
      "entropy": 6.009967899322509,
      "epoch": 6.269869304132815,
      "grad_norm": 1.375,
      "learning_rate": 0.000189739213899442,
      "loss": 5.1207,
      "mean_token_accuracy": 0.23384899944067,
      "num_tokens": 132381168.0,
      "step": 71000
    },
    {
      "entropy": 5.950336027145386,
      "epoch": 6.270310844224656,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018971007178497212,
      "loss": 5.1743,
      "mean_token_accuracy": 0.23342535197734832,
      "num_tokens": 132389072.0,
      "step": 71005
    },
    {
      "entropy": 6.063155937194824,
      "epoch": 6.270752384316496,
      "grad_norm": 1.453125,
      "learning_rate": 0.00018968093134128346,
      "loss": 5.2172,
      "mean_token_accuracy": 0.22281744778156282,
      "num_tokens": 132398207.0,
      "step": 71010
    },
    {
      "entropy": 5.976085758209228,
      "epoch": 6.271193924408336,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018965179256894677,
      "loss": 5.1576,
      "mean_token_accuracy": 0.2386045426130295,
      "num_tokens": 132407759.0,
      "step": 71015
    },
    {
      "entropy": 5.894367790222168,
      "epoch": 6.271635464500177,
      "grad_norm": 1.515625,
      "learning_rate": 0.00018962265546853285,
      "loss": 5.0281,
      "mean_token_accuracy": 0.24198568165302276,
      "num_tokens": 132416560.0,
      "step": 71020
    },
    {
      "entropy": 6.024225473403931,
      "epoch": 6.272077004592017,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018959352004061265,
      "loss": 5.2228,
      "mean_token_accuracy": 0.2269394278526306,
      "num_tokens": 132425703.0,
      "step": 71025
    },
    {
      "entropy": 5.970879983901978,
      "epoch": 6.272518544683857,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00018956438628575683,
      "loss": 5.2061,
      "mean_token_accuracy": 0.22097702026367189,
      "num_tokens": 132435250.0,
      "step": 71030
    },
    {
      "entropy": 6.014341163635254,
      "epoch": 6.272960084775698,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001895352542045361,
      "loss": 5.1902,
      "mean_token_accuracy": 0.2289462313055992,
      "num_tokens": 132444073.0,
      "step": 71035
    },
    {
      "entropy": 6.04700083732605,
      "epoch": 6.273401624867538,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018950612379752102,
      "loss": 5.2554,
      "mean_token_accuracy": 0.22108971178531647,
      "num_tokens": 132453702.0,
      "step": 71040
    },
    {
      "entropy": 6.039056444168091,
      "epoch": 6.273843164959378,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00018947699506528242,
      "loss": 5.1903,
      "mean_token_accuracy": 0.2251739591360092,
      "num_tokens": 132462641.0,
      "step": 71045
    },
    {
      "entropy": 6.099861669540405,
      "epoch": 6.274284705051219,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018944786800839086,
      "loss": 5.1917,
      "mean_token_accuracy": 0.23312900215387344,
      "num_tokens": 132472003.0,
      "step": 71050
    },
    {
      "entropy": 6.048695087432861,
      "epoch": 6.274726245143059,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018941874262741693,
      "loss": 5.1696,
      "mean_token_accuracy": 0.23703863173723222,
      "num_tokens": 132480482.0,
      "step": 71055
    },
    {
      "entropy": 6.031598997116089,
      "epoch": 6.275167785234899,
      "grad_norm": 1.5,
      "learning_rate": 0.00018938961892293104,
      "loss": 5.1858,
      "mean_token_accuracy": 0.23450889438390732,
      "num_tokens": 132490313.0,
      "step": 71060
    },
    {
      "entropy": 6.033468580245971,
      "epoch": 6.27560932532674,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00018936049689550395,
      "loss": 5.1126,
      "mean_token_accuracy": 0.23400310724973677,
      "num_tokens": 132498822.0,
      "step": 71065
    },
    {
      "entropy": 6.0919413566589355,
      "epoch": 6.27605086541858,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018933137654570604,
      "loss": 5.3359,
      "mean_token_accuracy": 0.21729195415973662,
      "num_tokens": 132508390.0,
      "step": 71070
    },
    {
      "entropy": 6.055806016921997,
      "epoch": 6.27649240551042,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001893022578741077,
      "loss": 5.2249,
      "mean_token_accuracy": 0.21937346160411836,
      "num_tokens": 132517356.0,
      "step": 71075
    },
    {
      "entropy": 6.090053939819336,
      "epoch": 6.276933945602261,
      "grad_norm": 1.53125,
      "learning_rate": 0.00018927314088127938,
      "loss": 5.2619,
      "mean_token_accuracy": 0.22957577854394912,
      "num_tokens": 132526253.0,
      "step": 71080
    },
    {
      "entropy": 6.00076961517334,
      "epoch": 6.277375485694101,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001892440255677914,
      "loss": 5.1251,
      "mean_token_accuracy": 0.2334897294640541,
      "num_tokens": 132535118.0,
      "step": 71085
    },
    {
      "entropy": 6.038171911239624,
      "epoch": 6.277817025785941,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001892149119342143,
      "loss": 5.1654,
      "mean_token_accuracy": 0.22388256192207337,
      "num_tokens": 132544427.0,
      "step": 71090
    },
    {
      "entropy": 6.064270114898681,
      "epoch": 6.278258565877782,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001891857999811183,
      "loss": 5.2111,
      "mean_token_accuracy": 0.22366274893283844,
      "num_tokens": 132553306.0,
      "step": 71095
    },
    {
      "entropy": 6.026139831542968,
      "epoch": 6.278700105969622,
      "grad_norm": 1.421875,
      "learning_rate": 0.00018915668970907363,
      "loss": 5.1692,
      "mean_token_accuracy": 0.22980479747056962,
      "num_tokens": 132562386.0,
      "step": 71100
    },
    {
      "entropy": 6.135712289810181,
      "epoch": 6.279141646061462,
      "grad_norm": 1.5625,
      "learning_rate": 0.00018912758111865047,
      "loss": 5.3192,
      "mean_token_accuracy": 0.22199077606201173,
      "num_tokens": 132571570.0,
      "step": 71105
    },
    {
      "entropy": 6.023627758026123,
      "epoch": 6.279583186153303,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018909847421041932,
      "loss": 5.283,
      "mean_token_accuracy": 0.2249707445502281,
      "num_tokens": 132581144.0,
      "step": 71110
    },
    {
      "entropy": 6.067333221435547,
      "epoch": 6.280024726245143,
      "grad_norm": 1.453125,
      "learning_rate": 0.00018906936898495009,
      "loss": 5.2364,
      "mean_token_accuracy": 0.23021851927042009,
      "num_tokens": 132590259.0,
      "step": 71115
    },
    {
      "entropy": 6.0224652767181395,
      "epoch": 6.280466266336983,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018904026544281318,
      "loss": 5.1796,
      "mean_token_accuracy": 0.23491788357496263,
      "num_tokens": 132599353.0,
      "step": 71120
    },
    {
      "entropy": 6.054140233993531,
      "epoch": 6.280907806428823,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018901116358457855,
      "loss": 5.3135,
      "mean_token_accuracy": 0.21808030009269713,
      "num_tokens": 132610013.0,
      "step": 71125
    },
    {
      "entropy": 6.118954277038574,
      "epoch": 6.281349346520664,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001889820634108162,
      "loss": 5.2832,
      "mean_token_accuracy": 0.21780659705400468,
      "num_tokens": 132619612.0,
      "step": 71130
    },
    {
      "entropy": 6.079602766036987,
      "epoch": 6.281790886612504,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001889529649220964,
      "loss": 5.2032,
      "mean_token_accuracy": 0.22865204364061356,
      "num_tokens": 132627854.0,
      "step": 71135
    },
    {
      "entropy": 6.037886095046997,
      "epoch": 6.282232426704344,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001889238681189891,
      "loss": 5.193,
      "mean_token_accuracy": 0.22712101340293883,
      "num_tokens": 132637506.0,
      "step": 71140
    },
    {
      "entropy": 6.019790697097778,
      "epoch": 6.282673966796185,
      "grad_norm": 1.546875,
      "learning_rate": 0.00018889477300206427,
      "loss": 5.1862,
      "mean_token_accuracy": 0.22931152880191802,
      "num_tokens": 132646732.0,
      "step": 71145
    },
    {
      "entropy": 6.075607109069824,
      "epoch": 6.283115506888025,
      "grad_norm": 1.53125,
      "learning_rate": 0.00018886567957189176,
      "loss": 5.2324,
      "mean_token_accuracy": 0.223495951294899,
      "num_tokens": 132655828.0,
      "step": 71150
    },
    {
      "entropy": 6.043700647354126,
      "epoch": 6.283557046979865,
      "grad_norm": 1.515625,
      "learning_rate": 0.00018883658782904166,
      "loss": 5.2422,
      "mean_token_accuracy": 0.22816627621650695,
      "num_tokens": 132665688.0,
      "step": 71155
    },
    {
      "entropy": 6.0258887767791744,
      "epoch": 6.283998587071706,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018880749777408384,
      "loss": 5.299,
      "mean_token_accuracy": 0.22202372848987578,
      "num_tokens": 132675512.0,
      "step": 71160
    },
    {
      "entropy": 6.137737989425659,
      "epoch": 6.284440127163546,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001887784094075881,
      "loss": 5.2941,
      "mean_token_accuracy": 0.2201131835579872,
      "num_tokens": 132684777.0,
      "step": 71165
    },
    {
      "entropy": 6.112903499603272,
      "epoch": 6.284881667255386,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00018874932273012425,
      "loss": 5.21,
      "mean_token_accuracy": 0.22644573599100112,
      "num_tokens": 132694600.0,
      "step": 71170
    },
    {
      "entropy": 6.0706719875335695,
      "epoch": 6.285323207347227,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00018872023774226215,
      "loss": 5.2338,
      "mean_token_accuracy": 0.22611554712057114,
      "num_tokens": 132703340.0,
      "step": 71175
    },
    {
      "entropy": 6.003203821182251,
      "epoch": 6.285764747439067,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018869115444457153,
      "loss": 5.0697,
      "mean_token_accuracy": 0.2366401359438896,
      "num_tokens": 132712863.0,
      "step": 71180
    },
    {
      "entropy": 6.02967677116394,
      "epoch": 6.2862062875309075,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00018866207283762211,
      "loss": 5.1953,
      "mean_token_accuracy": 0.22641806453466415,
      "num_tokens": 132723056.0,
      "step": 71185
    },
    {
      "entropy": 6.050200510025024,
      "epoch": 6.286647827622748,
      "grad_norm": 1.5,
      "learning_rate": 0.0001886329929219836,
      "loss": 5.2276,
      "mean_token_accuracy": 0.22515298724174498,
      "num_tokens": 132732551.0,
      "step": 71190
    },
    {
      "entropy": 6.091233015060425,
      "epoch": 6.2870893677145885,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018860391469822553,
      "loss": 5.2456,
      "mean_token_accuracy": 0.22089869678020477,
      "num_tokens": 132742562.0,
      "step": 71195
    },
    {
      "entropy": 6.041773128509521,
      "epoch": 6.2875309078064285,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018857483816691774,
      "loss": 5.1125,
      "mean_token_accuracy": 0.23475242853164674,
      "num_tokens": 132753097.0,
      "step": 71200
    },
    {
      "entropy": 6.14368085861206,
      "epoch": 6.2879724478982695,
      "grad_norm": 1.421875,
      "learning_rate": 0.00018854576332862967,
      "loss": 5.3328,
      "mean_token_accuracy": 0.21837098300457,
      "num_tokens": 132762089.0,
      "step": 71205
    },
    {
      "entropy": 6.019692468643188,
      "epoch": 6.2884139879901095,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018851669018393099,
      "loss": 5.1846,
      "mean_token_accuracy": 0.22994695454835892,
      "num_tokens": 132770883.0,
      "step": 71210
    },
    {
      "entropy": 6.04076771736145,
      "epoch": 6.28885552808195,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018848761873339103,
      "loss": 5.1315,
      "mean_token_accuracy": 0.23729992061853408,
      "num_tokens": 132779715.0,
      "step": 71215
    },
    {
      "entropy": 6.008563375473022,
      "epoch": 6.2892970681737905,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018845854897757953,
      "loss": 5.242,
      "mean_token_accuracy": 0.2289489597082138,
      "num_tokens": 132788297.0,
      "step": 71220
    },
    {
      "entropy": 6.023489761352539,
      "epoch": 6.289738608265631,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00018842948091706585,
      "loss": 5.1827,
      "mean_token_accuracy": 0.23306150138378143,
      "num_tokens": 132797995.0,
      "step": 71225
    },
    {
      "entropy": 6.1361957550048825,
      "epoch": 6.290180148357471,
      "grad_norm": 1.640625,
      "learning_rate": 0.0001884004145524193,
      "loss": 5.2679,
      "mean_token_accuracy": 0.22515598237514495,
      "num_tokens": 132806567.0,
      "step": 71230
    },
    {
      "entropy": 6.068236494064331,
      "epoch": 6.290621688449312,
      "grad_norm": 1.53125,
      "learning_rate": 0.00018837134988420944,
      "loss": 5.2882,
      "mean_token_accuracy": 0.22127320915460585,
      "num_tokens": 132815912.0,
      "step": 71235
    },
    {
      "entropy": 6.120700407028198,
      "epoch": 6.291063228541152,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001883422869130056,
      "loss": 5.2689,
      "mean_token_accuracy": 0.2187100663781166,
      "num_tokens": 132827436.0,
      "step": 71240
    },
    {
      "entropy": 6.051955223083496,
      "epoch": 6.291504768632992,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00018831322563937704,
      "loss": 5.2069,
      "mean_token_accuracy": 0.24068057388067246,
      "num_tokens": 132836774.0,
      "step": 71245
    },
    {
      "entropy": 6.024342203140259,
      "epoch": 6.291946308724833,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001882841660638931,
      "loss": 5.1784,
      "mean_token_accuracy": 0.2343178942799568,
      "num_tokens": 132845062.0,
      "step": 71250
    },
    {
      "entropy": 6.037579011917114,
      "epoch": 6.292387848816673,
      "grad_norm": 1.4375,
      "learning_rate": 0.00018825510818712304,
      "loss": 5.2372,
      "mean_token_accuracy": 0.22808992713689805,
      "num_tokens": 132854095.0,
      "step": 71255
    },
    {
      "entropy": 6.051355695724487,
      "epoch": 6.292829388908513,
      "grad_norm": 1.375,
      "learning_rate": 0.000188226052009636,
      "loss": 5.1802,
      "mean_token_accuracy": 0.22557108849287033,
      "num_tokens": 132863419.0,
      "step": 71260
    },
    {
      "entropy": 6.035331726074219,
      "epoch": 6.293270929000354,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018819699753200132,
      "loss": 5.1889,
      "mean_token_accuracy": 0.22888516783714294,
      "num_tokens": 132873219.0,
      "step": 71265
    },
    {
      "entropy": 6.0694104671478275,
      "epoch": 6.293712469092194,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018816794475478804,
      "loss": 5.1691,
      "mean_token_accuracy": 0.22815874069929123,
      "num_tokens": 132882846.0,
      "step": 71270
    },
    {
      "entropy": 5.96825385093689,
      "epoch": 6.294154009184034,
      "grad_norm": 1.5,
      "learning_rate": 0.00018813889367856542,
      "loss": 5.1082,
      "mean_token_accuracy": 0.2367638349533081,
      "num_tokens": 132892418.0,
      "step": 71275
    },
    {
      "entropy": 6.049703359603882,
      "epoch": 6.294595549275874,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018810984430390237,
      "loss": 5.2049,
      "mean_token_accuracy": 0.22379643172025682,
      "num_tokens": 132901914.0,
      "step": 71280
    },
    {
      "entropy": 6.01566891670227,
      "epoch": 6.295037089367715,
      "grad_norm": 1.578125,
      "learning_rate": 0.00018808079663136813,
      "loss": 5.1638,
      "mean_token_accuracy": 0.2341097190976143,
      "num_tokens": 132911103.0,
      "step": 71285
    },
    {
      "entropy": 6.012403678894043,
      "epoch": 6.295478629459555,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001880517506615317,
      "loss": 5.1952,
      "mean_token_accuracy": 0.2274521693587303,
      "num_tokens": 132920370.0,
      "step": 71290
    },
    {
      "entropy": 6.072741508483887,
      "epoch": 6.295920169551395,
      "grad_norm": 1.515625,
      "learning_rate": 0.00018802270639496188,
      "loss": 5.2062,
      "mean_token_accuracy": 0.2335180312395096,
      "num_tokens": 132929194.0,
      "step": 71295
    },
    {
      "entropy": 6.044211196899414,
      "epoch": 6.296361709643236,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00018799366383222784,
      "loss": 5.1101,
      "mean_token_accuracy": 0.2275906577706337,
      "num_tokens": 132937786.0,
      "step": 71300
    },
    {
      "entropy": 6.059840965270996,
      "epoch": 6.296803249735076,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00018796462297389848,
      "loss": 5.2352,
      "mean_token_accuracy": 0.23221810162067413,
      "num_tokens": 132946894.0,
      "step": 71305
    },
    {
      "entropy": 6.145320558547974,
      "epoch": 6.297244789826916,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001879355838205426,
      "loss": 5.3223,
      "mean_token_accuracy": 0.2225992426276207,
      "num_tokens": 132956068.0,
      "step": 71310
    },
    {
      "entropy": 5.9840576171875,
      "epoch": 6.297686329918757,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001879065463727292,
      "loss": 5.1368,
      "mean_token_accuracy": 0.23306874781847,
      "num_tokens": 132965340.0,
      "step": 71315
    },
    {
      "entropy": 5.980219078063965,
      "epoch": 6.298127870010597,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000187877510631027,
      "loss": 5.1384,
      "mean_token_accuracy": 0.2357727661728859,
      "num_tokens": 132975050.0,
      "step": 71320
    },
    {
      "entropy": 5.974828815460205,
      "epoch": 6.298569410102437,
      "grad_norm": 1.453125,
      "learning_rate": 0.00018784847659600467,
      "loss": 5.2284,
      "mean_token_accuracy": 0.23097517043352128,
      "num_tokens": 132985148.0,
      "step": 71325
    },
    {
      "entropy": 6.107257080078125,
      "epoch": 6.299010950194278,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00018781944426823126,
      "loss": 5.3014,
      "mean_token_accuracy": 0.22037504017353057,
      "num_tokens": 132994090.0,
      "step": 71330
    },
    {
      "entropy": 6.060730409622193,
      "epoch": 6.299452490286118,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00018779041364827537,
      "loss": 5.1374,
      "mean_token_accuracy": 0.22420258522033693,
      "num_tokens": 133002541.0,
      "step": 71335
    },
    {
      "entropy": 6.024450826644897,
      "epoch": 6.299894030377958,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018776138473670562,
      "loss": 5.1469,
      "mean_token_accuracy": 0.23155563473701476,
      "num_tokens": 133012055.0,
      "step": 71340
    },
    {
      "entropy": 6.006728601455689,
      "epoch": 6.300335570469799,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00018773235753409068,
      "loss": 5.1408,
      "mean_token_accuracy": 0.23532328754663467,
      "num_tokens": 133021333.0,
      "step": 71345
    },
    {
      "entropy": 5.994809484481811,
      "epoch": 6.300777110561639,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018770333204099933,
      "loss": 5.1966,
      "mean_token_accuracy": 0.22944044470787048,
      "num_tokens": 133031881.0,
      "step": 71350
    },
    {
      "entropy": 6.039773607254029,
      "epoch": 6.301218650653479,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00018767430825800002,
      "loss": 5.1847,
      "mean_token_accuracy": 0.23751015067100525,
      "num_tokens": 133041591.0,
      "step": 71355
    },
    {
      "entropy": 6.0708150386810305,
      "epoch": 6.30166019074532,
      "grad_norm": 1.265625,
      "learning_rate": 0.00018764528618566133,
      "loss": 5.2945,
      "mean_token_accuracy": 0.22454102635383605,
      "num_tokens": 133051017.0,
      "step": 71360
    },
    {
      "entropy": 6.047233724594117,
      "epoch": 6.30210173083716,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001876162658245519,
      "loss": 5.1584,
      "mean_token_accuracy": 0.2366548404097557,
      "num_tokens": 133059064.0,
      "step": 71365
    },
    {
      "entropy": 6.032393741607666,
      "epoch": 6.302543270929,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00018758724717523996,
      "loss": 5.2211,
      "mean_token_accuracy": 0.2275093674659729,
      "num_tokens": 133067836.0,
      "step": 71370
    },
    {
      "entropy": 6.011809062957764,
      "epoch": 6.302984811020841,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018755823023829416,
      "loss": 5.1393,
      "mean_token_accuracy": 0.2328682065010071,
      "num_tokens": 133076368.0,
      "step": 71375
    },
    {
      "entropy": 5.998836946487427,
      "epoch": 6.303426351112681,
      "grad_norm": 1.546875,
      "learning_rate": 0.000187529215014283,
      "loss": 5.1615,
      "mean_token_accuracy": 0.23052206933498381,
      "num_tokens": 133084518.0,
      "step": 71380
    },
    {
      "entropy": 6.034792518615722,
      "epoch": 6.303867891204521,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00018750020150377472,
      "loss": 5.166,
      "mean_token_accuracy": 0.22623866498470308,
      "num_tokens": 133094193.0,
      "step": 71385
    },
    {
      "entropy": 5.9654604434967045,
      "epoch": 6.304309431296362,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018747118970733768,
      "loss": 5.117,
      "mean_token_accuracy": 0.23181407451629638,
      "num_tokens": 133103538.0,
      "step": 71390
    },
    {
      "entropy": 6.075441360473633,
      "epoch": 6.304750971388202,
      "grad_norm": 1.203125,
      "learning_rate": 0.00018744217962554034,
      "loss": 5.2921,
      "mean_token_accuracy": 0.21690315306186675,
      "num_tokens": 133114723.0,
      "step": 71395
    },
    {
      "entropy": 6.053264713287353,
      "epoch": 6.305192511480042,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00018741317125895086,
      "loss": 5.2296,
      "mean_token_accuracy": 0.22875017076730728,
      "num_tokens": 133123624.0,
      "step": 71400
    },
    {
      "entropy": 6.036983633041382,
      "epoch": 6.305634051571882,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00018738416460813755,
      "loss": 5.1495,
      "mean_token_accuracy": 0.22869692742824554,
      "num_tokens": 133132711.0,
      "step": 71405
    },
    {
      "entropy": 5.998521947860718,
      "epoch": 6.306075591663723,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00018735515967366863,
      "loss": 5.2013,
      "mean_token_accuracy": 0.22521072775125503,
      "num_tokens": 133142115.0,
      "step": 71410
    },
    {
      "entropy": 6.076993751525879,
      "epoch": 6.306517131755563,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00018732615645611228,
      "loss": 5.2598,
      "mean_token_accuracy": 0.22425293773412705,
      "num_tokens": 133150932.0,
      "step": 71415
    },
    {
      "entropy": 6.118753433227539,
      "epoch": 6.306958671847403,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018729715495603672,
      "loss": 5.2435,
      "mean_token_accuracy": 0.22116129249334335,
      "num_tokens": 133160390.0,
      "step": 71420
    },
    {
      "entropy": 6.091535329818726,
      "epoch": 6.307400211939244,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001872681551740099,
      "loss": 5.1216,
      "mean_token_accuracy": 0.2318122312426567,
      "num_tokens": 133169845.0,
      "step": 71425
    },
    {
      "entropy": 5.961934804916382,
      "epoch": 6.307841752031084,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018723915711060007,
      "loss": 5.0817,
      "mean_token_accuracy": 0.23918834179639817,
      "num_tokens": 133179607.0,
      "step": 71430
    },
    {
      "entropy": 6.02396068572998,
      "epoch": 6.308283292122924,
      "grad_norm": 1.375,
      "learning_rate": 0.00018721016076637525,
      "loss": 5.2923,
      "mean_token_accuracy": 0.21974751502275466,
      "num_tokens": 133190120.0,
      "step": 71435
    },
    {
      "entropy": 6.051429176330567,
      "epoch": 6.308724832214765,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018718116614190345,
      "loss": 5.2448,
      "mean_token_accuracy": 0.22522326558828354,
      "num_tokens": 133199439.0,
      "step": 71440
    },
    {
      "entropy": 6.027507495880127,
      "epoch": 6.309166372306605,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018715217323775267,
      "loss": 5.1248,
      "mean_token_accuracy": 0.2325780138373375,
      "num_tokens": 133207908.0,
      "step": 71445
    },
    {
      "entropy": 6.094397640228271,
      "epoch": 6.309607912398445,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018712318205449087,
      "loss": 5.3059,
      "mean_token_accuracy": 0.22188162207603454,
      "num_tokens": 133217490.0,
      "step": 71450
    },
    {
      "entropy": 5.933057260513306,
      "epoch": 6.310049452490286,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001870941925926859,
      "loss": 5.0589,
      "mean_token_accuracy": 0.24564255475997926,
      "num_tokens": 133226323.0,
      "step": 71455
    },
    {
      "entropy": 6.0316415309906,
      "epoch": 6.310490992582126,
      "grad_norm": 1.453125,
      "learning_rate": 0.00018706520485290573,
      "loss": 5.2185,
      "mean_token_accuracy": 0.22687285095453263,
      "num_tokens": 133236282.0,
      "step": 71460
    },
    {
      "entropy": 5.98782753944397,
      "epoch": 6.3109325326739665,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00018703621883571825,
      "loss": 5.1547,
      "mean_token_accuracy": 0.22766625881195068,
      "num_tokens": 133245645.0,
      "step": 71465
    },
    {
      "entropy": 5.986543989181518,
      "epoch": 6.311374072765807,
      "grad_norm": 1.453125,
      "learning_rate": 0.00018700723454169116,
      "loss": 5.2075,
      "mean_token_accuracy": 0.22145345509052278,
      "num_tokens": 133255231.0,
      "step": 71470
    },
    {
      "entropy": 6.026941394805908,
      "epoch": 6.3118156128576475,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018697825197139224,
      "loss": 5.1932,
      "mean_token_accuracy": 0.22441125363111497,
      "num_tokens": 133264328.0,
      "step": 71475
    },
    {
      "entropy": 6.121951007843018,
      "epoch": 6.3122571529494875,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001869492711253894,
      "loss": 5.2797,
      "mean_token_accuracy": 0.22173542082309722,
      "num_tokens": 133272541.0,
      "step": 71480
    },
    {
      "entropy": 6.030793142318726,
      "epoch": 6.3126986930413285,
      "grad_norm": 1.625,
      "learning_rate": 0.0001869202920042502,
      "loss": 5.2057,
      "mean_token_accuracy": 0.23179744333028793,
      "num_tokens": 133282097.0,
      "step": 71485
    },
    {
      "entropy": 6.028719568252564,
      "epoch": 6.3131402331331685,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001868913146085424,
      "loss": 5.1302,
      "mean_token_accuracy": 0.23742564469575883,
      "num_tokens": 133290999.0,
      "step": 71490
    },
    {
      "entropy": 5.935369682312012,
      "epoch": 6.313581773225009,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001868623389388337,
      "loss": 5.1108,
      "mean_token_accuracy": 0.23748937994241714,
      "num_tokens": 133299077.0,
      "step": 71495
    },
    {
      "entropy": 5.932492637634278,
      "epoch": 6.3140233133168495,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018683336499569154,
      "loss": 5.0869,
      "mean_token_accuracy": 0.24132556319236756,
      "num_tokens": 133308390.0,
      "step": 71500
    },
    {
      "entropy": 5.965198755264282,
      "epoch": 6.31446485340869,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001868043927796837,
      "loss": 5.1251,
      "mean_token_accuracy": 0.23844507038593293,
      "num_tokens": 133317282.0,
      "step": 71505
    },
    {
      "entropy": 6.058956813812256,
      "epoch": 6.31490639350053,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018677542229137762,
      "loss": 5.2767,
      "mean_token_accuracy": 0.2207501083612442,
      "num_tokens": 133327290.0,
      "step": 71510
    },
    {
      "entropy": 6.043591403961182,
      "epoch": 6.315347933592371,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018674645353134085,
      "loss": 5.2369,
      "mean_token_accuracy": 0.2255664974451065,
      "num_tokens": 133337527.0,
      "step": 71515
    },
    {
      "entropy": 6.077365684509277,
      "epoch": 6.315789473684211,
      "grad_norm": 1.625,
      "learning_rate": 0.0001867174865001408,
      "loss": 5.2308,
      "mean_token_accuracy": 0.22708680331707,
      "num_tokens": 133347329.0,
      "step": 71520
    },
    {
      "entropy": 5.999406480789185,
      "epoch": 6.316231013776051,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018668852119834507,
      "loss": 5.1576,
      "mean_token_accuracy": 0.23275349885225297,
      "num_tokens": 133356444.0,
      "step": 71525
    },
    {
      "entropy": 5.961482095718384,
      "epoch": 6.316672553867892,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018665955762652098,
      "loss": 5.1373,
      "mean_token_accuracy": 0.23734770715236664,
      "num_tokens": 133365436.0,
      "step": 71530
    },
    {
      "entropy": 6.012049722671509,
      "epoch": 6.317114093959732,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001866305957852359,
      "loss": 5.1568,
      "mean_token_accuracy": 0.23278596997261047,
      "num_tokens": 133374486.0,
      "step": 71535
    },
    {
      "entropy": 6.087688207626343,
      "epoch": 6.317555634051572,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001866016356750571,
      "loss": 5.1792,
      "mean_token_accuracy": 0.2286951422691345,
      "num_tokens": 133383821.0,
      "step": 71540
    },
    {
      "entropy": 6.130600023269653,
      "epoch": 6.317997174143413,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001865726772965521,
      "loss": 5.3095,
      "mean_token_accuracy": 0.2156551018357277,
      "num_tokens": 133393069.0,
      "step": 71545
    },
    {
      "entropy": 5.978632831573487,
      "epoch": 6.318438714235253,
      "grad_norm": 1.5,
      "learning_rate": 0.000186543720650288,
      "loss": 5.1091,
      "mean_token_accuracy": 0.2278567373752594,
      "num_tokens": 133402449.0,
      "step": 71550
    },
    {
      "entropy": 6.032812452316284,
      "epoch": 6.318880254327093,
      "grad_norm": 1.53125,
      "learning_rate": 0.00018651476573683213,
      "loss": 5.2403,
      "mean_token_accuracy": 0.22400777339935302,
      "num_tokens": 133412671.0,
      "step": 71555
    },
    {
      "entropy": 6.105469226837158,
      "epoch": 6.319321794418933,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0001864858125567517,
      "loss": 5.2459,
      "mean_token_accuracy": 0.2273216500878334,
      "num_tokens": 133422439.0,
      "step": 71560
    },
    {
      "entropy": 6.0240377426147464,
      "epoch": 6.319763334510774,
      "grad_norm": 1.578125,
      "learning_rate": 0.00018645686111061371,
      "loss": 5.1492,
      "mean_token_accuracy": 0.23596205264329911,
      "num_tokens": 133431579.0,
      "step": 71565
    },
    {
      "entropy": 6.029260206222534,
      "epoch": 6.320204874602614,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018642791139898563,
      "loss": 5.2058,
      "mean_token_accuracy": 0.23633532971143723,
      "num_tokens": 133440939.0,
      "step": 71570
    },
    {
      "entropy": 6.028343534469604,
      "epoch": 6.320646414694454,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001863989634224344,
      "loss": 5.3104,
      "mean_token_accuracy": 0.21829820722341536,
      "num_tokens": 133450377.0,
      "step": 71575
    },
    {
      "entropy": 6.0321245193481445,
      "epoch": 6.321087954786295,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018637001718152703,
      "loss": 5.1933,
      "mean_token_accuracy": 0.22346394807100295,
      "num_tokens": 133459577.0,
      "step": 71580
    },
    {
      "entropy": 6.15658655166626,
      "epoch": 6.321529494878135,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018634107267683053,
      "loss": 5.2813,
      "mean_token_accuracy": 0.21453115791082383,
      "num_tokens": 133469184.0,
      "step": 71585
    },
    {
      "entropy": 6.028821802139282,
      "epoch": 6.321971034969975,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001863121299089121,
      "loss": 5.1946,
      "mean_token_accuracy": 0.23323245346546173,
      "num_tokens": 133478630.0,
      "step": 71590
    },
    {
      "entropy": 6.114125061035156,
      "epoch": 6.322412575061816,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018628318887833857,
      "loss": 5.2812,
      "mean_token_accuracy": 0.21933623999357224,
      "num_tokens": 133488014.0,
      "step": 71595
    },
    {
      "entropy": 5.955132579803466,
      "epoch": 6.322854115153656,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018625424958567695,
      "loss": 5.0556,
      "mean_token_accuracy": 0.247459214925766,
      "num_tokens": 133496378.0,
      "step": 71600
    },
    {
      "entropy": 6.080935144424439,
      "epoch": 6.323295655245496,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00018622531203149405,
      "loss": 5.1843,
      "mean_token_accuracy": 0.2284678488969803,
      "num_tokens": 133504387.0,
      "step": 71605
    },
    {
      "entropy": 6.055029249191284,
      "epoch": 6.323737195337337,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018619637621635683,
      "loss": 5.2509,
      "mean_token_accuracy": 0.22656511813402175,
      "num_tokens": 133514155.0,
      "step": 71610
    },
    {
      "entropy": 5.977121734619141,
      "epoch": 6.324178735429177,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00018616744214083208,
      "loss": 5.1436,
      "mean_token_accuracy": 0.23608821928501128,
      "num_tokens": 133523305.0,
      "step": 71615
    },
    {
      "entropy": 6.081372022628784,
      "epoch": 6.324620275521017,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00018613850980548664,
      "loss": 5.2956,
      "mean_token_accuracy": 0.21727113872766496,
      "num_tokens": 133532795.0,
      "step": 71620
    },
    {
      "entropy": 6.080343198776245,
      "epoch": 6.325061815612858,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001861095792108873,
      "loss": 5.258,
      "mean_token_accuracy": 0.22226392179727555,
      "num_tokens": 133542599.0,
      "step": 71625
    },
    {
      "entropy": 6.0903167724609375,
      "epoch": 6.325503355704698,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018608065035760057,
      "loss": 5.229,
      "mean_token_accuracy": 0.22431954592466355,
      "num_tokens": 133552114.0,
      "step": 71630
    },
    {
      "entropy": 6.10559983253479,
      "epoch": 6.325944895796538,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00018605172324619347,
      "loss": 5.2957,
      "mean_token_accuracy": 0.2196951240301132,
      "num_tokens": 133561579.0,
      "step": 71635
    },
    {
      "entropy": 6.0299169540405275,
      "epoch": 6.326386435888379,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00018602279787723253,
      "loss": 5.186,
      "mean_token_accuracy": 0.23303710520267487,
      "num_tokens": 133570475.0,
      "step": 71640
    },
    {
      "entropy": 6.009503555297852,
      "epoch": 6.326827975980219,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001859938742512844,
      "loss": 5.1379,
      "mean_token_accuracy": 0.23376000076532363,
      "num_tokens": 133579314.0,
      "step": 71645
    },
    {
      "entropy": 6.100499153137207,
      "epoch": 6.327269516072059,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00018596495236891552,
      "loss": 5.2461,
      "mean_token_accuracy": 0.22469794005155563,
      "num_tokens": 133589403.0,
      "step": 71650
    },
    {
      "entropy": 6.045414304733276,
      "epoch": 6.3277110561639,
      "grad_norm": 1.53125,
      "learning_rate": 0.00018593603223069266,
      "loss": 5.1798,
      "mean_token_accuracy": 0.2299298107624054,
      "num_tokens": 133598396.0,
      "step": 71655
    },
    {
      "entropy": 6.086488723754883,
      "epoch": 6.32815259625574,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00018590711383718235,
      "loss": 5.2326,
      "mean_token_accuracy": 0.22215885072946548,
      "num_tokens": 133608110.0,
      "step": 71660
    },
    {
      "entropy": 6.085900592803955,
      "epoch": 6.32859413634758,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018587819718895093,
      "loss": 5.1965,
      "mean_token_accuracy": 0.22366163432598113,
      "num_tokens": 133617687.0,
      "step": 71665
    },
    {
      "entropy": 6.004135608673096,
      "epoch": 6.329035676439421,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00018584928228656495,
      "loss": 5.1849,
      "mean_token_accuracy": 0.22883089482784272,
      "num_tokens": 133626705.0,
      "step": 71670
    },
    {
      "entropy": 6.030672407150268,
      "epoch": 6.329477216531261,
      "grad_norm": 1.59375,
      "learning_rate": 0.0001858203691305908,
      "loss": 5.1056,
      "mean_token_accuracy": 0.23990027755498886,
      "num_tokens": 133635377.0,
      "step": 71675
    },
    {
      "entropy": 6.074524736404419,
      "epoch": 6.329918756623101,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000185791457721595,
      "loss": 5.19,
      "mean_token_accuracy": 0.22013155668973922,
      "num_tokens": 133644311.0,
      "step": 71680
    },
    {
      "entropy": 6.032068347930908,
      "epoch": 6.330360296714941,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001857625480601437,
      "loss": 5.3215,
      "mean_token_accuracy": 0.21975128799676896,
      "num_tokens": 133653751.0,
      "step": 71685
    },
    {
      "entropy": 6.001954889297485,
      "epoch": 6.330801836806782,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001857336401468034,
      "loss": 5.1354,
      "mean_token_accuracy": 0.236459381878376,
      "num_tokens": 133663493.0,
      "step": 71690
    },
    {
      "entropy": 6.05225625038147,
      "epoch": 6.331243376898622,
      "grad_norm": 1.421875,
      "learning_rate": 0.00018570473398214023,
      "loss": 5.2313,
      "mean_token_accuracy": 0.22821792215108871,
      "num_tokens": 133674210.0,
      "step": 71695
    },
    {
      "entropy": 6.013869428634644,
      "epoch": 6.331684916990462,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018567582956672063,
      "loss": 5.1445,
      "mean_token_accuracy": 0.22940249592065812,
      "num_tokens": 133683803.0,
      "step": 71700
    },
    {
      "entropy": 6.024902772903443,
      "epoch": 6.332126457082303,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001856469269011107,
      "loss": 5.1934,
      "mean_token_accuracy": 0.22560469359159468,
      "num_tokens": 133692662.0,
      "step": 71705
    },
    {
      "entropy": 5.914103174209595,
      "epoch": 6.332567997174143,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018561802598587667,
      "loss": 5.1514,
      "mean_token_accuracy": 0.24423212856054305,
      "num_tokens": 133701836.0,
      "step": 71710
    },
    {
      "entropy": 6.014426040649414,
      "epoch": 6.333009537265983,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018558912682158457,
      "loss": 5.163,
      "mean_token_accuracy": 0.22621484994888305,
      "num_tokens": 133711316.0,
      "step": 71715
    },
    {
      "entropy": 6.0334240913391115,
      "epoch": 6.333451077357824,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00018556022940880075,
      "loss": 5.1726,
      "mean_token_accuracy": 0.23157422095537186,
      "num_tokens": 133721693.0,
      "step": 71720
    },
    {
      "entropy": 6.0639080047607425,
      "epoch": 6.333892617449664,
      "grad_norm": 1.453125,
      "learning_rate": 0.00018553133374809116,
      "loss": 5.1995,
      "mean_token_accuracy": 0.22890183627605437,
      "num_tokens": 133731263.0,
      "step": 71725
    },
    {
      "entropy": 6.0829095363616945,
      "epoch": 6.334334157541504,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018550243984002184,
      "loss": 5.2329,
      "mean_token_accuracy": 0.22802646607160568,
      "num_tokens": 133741030.0,
      "step": 71730
    },
    {
      "entropy": 6.037301349639892,
      "epoch": 6.334775697633345,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018547354768515877,
      "loss": 5.146,
      "mean_token_accuracy": 0.2318353310227394,
      "num_tokens": 133750801.0,
      "step": 71735
    },
    {
      "entropy": 5.99811692237854,
      "epoch": 6.335217237725185,
      "grad_norm": 1.53125,
      "learning_rate": 0.00018544465728406802,
      "loss": 5.089,
      "mean_token_accuracy": 0.2287026673555374,
      "num_tokens": 133760296.0,
      "step": 71740
    },
    {
      "entropy": 6.054973506927491,
      "epoch": 6.3356587778170255,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018541576863731558,
      "loss": 5.1469,
      "mean_token_accuracy": 0.23923446238040924,
      "num_tokens": 133771695.0,
      "step": 71745
    },
    {
      "entropy": 6.0633398532867435,
      "epoch": 6.336100317908866,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018538688174546714,
      "loss": 5.2218,
      "mean_token_accuracy": 0.22583234161138535,
      "num_tokens": 133781441.0,
      "step": 71750
    },
    {
      "entropy": 6.009413337707519,
      "epoch": 6.3365418580007065,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001853579966090888,
      "loss": 5.1452,
      "mean_token_accuracy": 0.2303926542401314,
      "num_tokens": 133790668.0,
      "step": 71755
    },
    {
      "entropy": 6.040688323974609,
      "epoch": 6.3369833980925465,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00018532911322874629,
      "loss": 5.2009,
      "mean_token_accuracy": 0.2281615436077118,
      "num_tokens": 133800404.0,
      "step": 71760
    },
    {
      "entropy": 6.04460916519165,
      "epoch": 6.3374249381843875,
      "grad_norm": 1.515625,
      "learning_rate": 0.00018530023160500547,
      "loss": 5.2794,
      "mean_token_accuracy": 0.22168779671192168,
      "num_tokens": 133809895.0,
      "step": 71765
    },
    {
      "entropy": 6.055094003677368,
      "epoch": 6.3378664782762275,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018527135173843212,
      "loss": 5.2673,
      "mean_token_accuracy": 0.2180749386548996,
      "num_tokens": 133820260.0,
      "step": 71770
    },
    {
      "entropy": 6.075963878631592,
      "epoch": 6.338308018368068,
      "grad_norm": 1.515625,
      "learning_rate": 0.00018524247362959202,
      "loss": 5.186,
      "mean_token_accuracy": 0.22615518122911454,
      "num_tokens": 133829963.0,
      "step": 71775
    },
    {
      "entropy": 6.020014667510987,
      "epoch": 6.3387495584599085,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018521359727905062,
      "loss": 5.1598,
      "mean_token_accuracy": 0.2334679514169693,
      "num_tokens": 133839326.0,
      "step": 71780
    },
    {
      "entropy": 6.018413162231445,
      "epoch": 6.339191098551749,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018518472268737397,
      "loss": 5.1002,
      "mean_token_accuracy": 0.2362337201833725,
      "num_tokens": 133848432.0,
      "step": 71785
    },
    {
      "entropy": 6.108249855041504,
      "epoch": 6.339632638643589,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018515584985512746,
      "loss": 5.2818,
      "mean_token_accuracy": 0.21993620842695236,
      "num_tokens": 133858869.0,
      "step": 71790
    },
    {
      "entropy": 6.059883069992066,
      "epoch": 6.3400741787354296,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018512697878287677,
      "loss": 5.2405,
      "mean_token_accuracy": 0.22426854968070983,
      "num_tokens": 133868846.0,
      "step": 71795
    },
    {
      "entropy": 6.0413844108581545,
      "epoch": 6.34051571882727,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018509810947118743,
      "loss": 5.2259,
      "mean_token_accuracy": 0.22219707071781158,
      "num_tokens": 133878128.0,
      "step": 71800
    },
    {
      "entropy": 5.966335964202881,
      "epoch": 6.34095725891911,
      "grad_norm": 1.609375,
      "learning_rate": 0.00018506924192062497,
      "loss": 5.1226,
      "mean_token_accuracy": 0.23044585436582565,
      "num_tokens": 133887796.0,
      "step": 71805
    },
    {
      "entropy": 6.09179458618164,
      "epoch": 6.341398799010951,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018504037613175495,
      "loss": 5.1994,
      "mean_token_accuracy": 0.22945682257413863,
      "num_tokens": 133896559.0,
      "step": 71810
    },
    {
      "entropy": 6.092132329940796,
      "epoch": 6.341840339102791,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018501151210514278,
      "loss": 5.2341,
      "mean_token_accuracy": 0.2173902601003647,
      "num_tokens": 133906430.0,
      "step": 71815
    },
    {
      "entropy": 5.993502283096314,
      "epoch": 6.342281879194631,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00018498264984135394,
      "loss": 5.1142,
      "mean_token_accuracy": 0.23119350224733354,
      "num_tokens": 133915735.0,
      "step": 71820
    },
    {
      "entropy": 6.135320043563842,
      "epoch": 6.342723419286472,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001849537893409537,
      "loss": 5.3405,
      "mean_token_accuracy": 0.21593308299779893,
      "num_tokens": 133925198.0,
      "step": 71825
    },
    {
      "entropy": 6.101360321044922,
      "epoch": 6.343164959378312,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001849249306045076,
      "loss": 5.321,
      "mean_token_accuracy": 0.2162655010819435,
      "num_tokens": 133935007.0,
      "step": 71830
    },
    {
      "entropy": 6.088851737976074,
      "epoch": 6.343606499470152,
      "grad_norm": 1.375,
      "learning_rate": 0.00018489607363258092,
      "loss": 5.322,
      "mean_token_accuracy": 0.22384214401245117,
      "num_tokens": 133944627.0,
      "step": 71835
    },
    {
      "entropy": 6.097633361816406,
      "epoch": 6.344048039561992,
      "grad_norm": 1.515625,
      "learning_rate": 0.00018486721842573883,
      "loss": 5.2736,
      "mean_token_accuracy": 0.22668585628271104,
      "num_tokens": 133953618.0,
      "step": 71840
    },
    {
      "entropy": 6.186729478836059,
      "epoch": 6.344489579653833,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018483836498454676,
      "loss": 5.3474,
      "mean_token_accuracy": 0.21739975064992906,
      "num_tokens": 133964213.0,
      "step": 71845
    },
    {
      "entropy": 6.064111948013306,
      "epoch": 6.344931119745673,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018480951330956973,
      "loss": 5.2146,
      "mean_token_accuracy": 0.23092542439699174,
      "num_tokens": 133974027.0,
      "step": 71850
    },
    {
      "entropy": 6.190155267715454,
      "epoch": 6.345372659837513,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018478066340137312,
      "loss": 5.3251,
      "mean_token_accuracy": 0.21825735867023469,
      "num_tokens": 133983749.0,
      "step": 71855
    },
    {
      "entropy": 6.060257148742676,
      "epoch": 6.345814199929354,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000184751815260522,
      "loss": 5.2133,
      "mean_token_accuracy": 0.22293200790882112,
      "num_tokens": 133993438.0,
      "step": 71860
    },
    {
      "entropy": 5.934889221191407,
      "epoch": 6.346255740021194,
      "grad_norm": 1.4375,
      "learning_rate": 0.00018472296888758145,
      "loss": 5.1345,
      "mean_token_accuracy": 0.23871717602014542,
      "num_tokens": 134003585.0,
      "step": 71865
    },
    {
      "entropy": 6.054064798355102,
      "epoch": 6.346697280113034,
      "grad_norm": 1.53125,
      "learning_rate": 0.00018469412428311654,
      "loss": 5.1972,
      "mean_token_accuracy": 0.2269358143210411,
      "num_tokens": 134011894.0,
      "step": 71870
    },
    {
      "entropy": 6.0308677673339846,
      "epoch": 6.347138820204875,
      "grad_norm": 1.5,
      "learning_rate": 0.00018466528144769253,
      "loss": 5.2382,
      "mean_token_accuracy": 0.22313284426927565,
      "num_tokens": 134020978.0,
      "step": 71875
    },
    {
      "entropy": 6.136392641067505,
      "epoch": 6.347580360296715,
      "grad_norm": 1.53125,
      "learning_rate": 0.00018463644038187415,
      "loss": 5.314,
      "mean_token_accuracy": 0.21653055697679519,
      "num_tokens": 134030087.0,
      "step": 71880
    },
    {
      "entropy": 6.036927032470703,
      "epoch": 6.348021900388555,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001846076010862266,
      "loss": 5.1232,
      "mean_token_accuracy": 0.2333551436662674,
      "num_tokens": 134039134.0,
      "step": 71885
    },
    {
      "entropy": 6.020253467559814,
      "epoch": 6.348463440480396,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018457876356131465,
      "loss": 5.1391,
      "mean_token_accuracy": 0.23424115926027297,
      "num_tokens": 134048519.0,
      "step": 71890
    },
    {
      "entropy": 6.108287334442139,
      "epoch": 6.348904980572236,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018454992780770336,
      "loss": 5.3258,
      "mean_token_accuracy": 0.21349294930696489,
      "num_tokens": 134058634.0,
      "step": 71895
    },
    {
      "entropy": 6.013665962219238,
      "epoch": 6.349346520664076,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018452109382595755,
      "loss": 5.1862,
      "mean_token_accuracy": 0.23081749826669692,
      "num_tokens": 134068194.0,
      "step": 71900
    },
    {
      "entropy": 6.063305234909057,
      "epoch": 6.349788060755917,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018449226161664205,
      "loss": 5.2354,
      "mean_token_accuracy": 0.22473645806312562,
      "num_tokens": 134077161.0,
      "step": 71905
    },
    {
      "entropy": 6.0514991760253904,
      "epoch": 6.350229600847757,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018446343118032167,
      "loss": 5.1442,
      "mean_token_accuracy": 0.23411550372838974,
      "num_tokens": 134086126.0,
      "step": 71910
    },
    {
      "entropy": 6.034863233566284,
      "epoch": 6.350671140939597,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001844346025175611,
      "loss": 5.1216,
      "mean_token_accuracy": 0.23330852538347244,
      "num_tokens": 134095935.0,
      "step": 71915
    },
    {
      "entropy": 6.131738805770874,
      "epoch": 6.351112681031438,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018440577562892524,
      "loss": 5.245,
      "mean_token_accuracy": 0.22364227026700972,
      "num_tokens": 134105337.0,
      "step": 71920
    },
    {
      "entropy": 6.0607603073120115,
      "epoch": 6.351554221123278,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018437695051497872,
      "loss": 5.2137,
      "mean_token_accuracy": 0.22617647051811218,
      "num_tokens": 134114397.0,
      "step": 71925
    },
    {
      "entropy": 5.989152145385742,
      "epoch": 6.351995761215118,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001843481271762862,
      "loss": 5.1128,
      "mean_token_accuracy": 0.2359983056783676,
      "num_tokens": 134123092.0,
      "step": 71930
    },
    {
      "entropy": 6.019545698165894,
      "epoch": 6.352437301306959,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0001843193056134122,
      "loss": 5.1534,
      "mean_token_accuracy": 0.228353750705719,
      "num_tokens": 134132622.0,
      "step": 71935
    },
    {
      "entropy": 6.012885570526123,
      "epoch": 6.352878841398799,
      "grad_norm": 1.609375,
      "learning_rate": 0.00018429048582692154,
      "loss": 5.206,
      "mean_token_accuracy": 0.22618315517902374,
      "num_tokens": 134140803.0,
      "step": 71940
    },
    {
      "entropy": 6.034342861175537,
      "epoch": 6.353320381490639,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018426166781737864,
      "loss": 5.1799,
      "mean_token_accuracy": 0.22982617616653442,
      "num_tokens": 134150084.0,
      "step": 71945
    },
    {
      "entropy": 6.074199438095093,
      "epoch": 6.35376192158248,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018423285158534808,
      "loss": 5.1798,
      "mean_token_accuracy": 0.22783196419477464,
      "num_tokens": 134159168.0,
      "step": 71950
    },
    {
      "entropy": 6.10211501121521,
      "epoch": 6.35420346167432,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018420403713139427,
      "loss": 5.3299,
      "mean_token_accuracy": 0.21808500438928605,
      "num_tokens": 134168339.0,
      "step": 71955
    },
    {
      "entropy": 6.049385213851929,
      "epoch": 6.35464500176616,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001841752244560818,
      "loss": 5.2346,
      "mean_token_accuracy": 0.22871629148721695,
      "num_tokens": 134177548.0,
      "step": 71960
    },
    {
      "entropy": 5.974206352233887,
      "epoch": 6.355086541858,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018414641355997502,
      "loss": 5.1002,
      "mean_token_accuracy": 0.23921826034784316,
      "num_tokens": 134187997.0,
      "step": 71965
    },
    {
      "entropy": 6.078206014633179,
      "epoch": 6.355528081949841,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001841176044436384,
      "loss": 5.2294,
      "mean_token_accuracy": 0.22074040323495864,
      "num_tokens": 134197248.0,
      "step": 71970
    },
    {
      "entropy": 5.973174715042115,
      "epoch": 6.355969622041681,
      "grad_norm": 1.4375,
      "learning_rate": 0.00018408879710763614,
      "loss": 5.0842,
      "mean_token_accuracy": 0.2338806301355362,
      "num_tokens": 134206613.0,
      "step": 71975
    },
    {
      "entropy": 6.04902606010437,
      "epoch": 6.356411162133521,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001840599915525326,
      "loss": 5.1515,
      "mean_token_accuracy": 0.23446519672870636,
      "num_tokens": 134216558.0,
      "step": 71980
    },
    {
      "entropy": 6.037983703613281,
      "epoch": 6.356852702225362,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00018403118777889215,
      "loss": 5.2055,
      "mean_token_accuracy": 0.22480901181697846,
      "num_tokens": 134225873.0,
      "step": 71985
    },
    {
      "entropy": 6.036578702926636,
      "epoch": 6.357294242317202,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018400238578727908,
      "loss": 5.2298,
      "mean_token_accuracy": 0.22949354648590087,
      "num_tokens": 134235341.0,
      "step": 71990
    },
    {
      "entropy": 6.15181245803833,
      "epoch": 6.357735782409042,
      "grad_norm": 1.5,
      "learning_rate": 0.0001839735855782574,
      "loss": 5.2877,
      "mean_token_accuracy": 0.2200804516673088,
      "num_tokens": 134244425.0,
      "step": 71995
    },
    {
      "entropy": 6.016364288330078,
      "epoch": 6.358177322500883,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018394478715239143,
      "loss": 5.2101,
      "mean_token_accuracy": 0.22035905718803406,
      "num_tokens": 134253615.0,
      "step": 72000
    },
    {
      "epoch": 6.358177322500883,
      "eval_entropy": 5.728417662288972,
      "eval_loss": 5.284831523895264,
      "eval_mean_token_accuracy": 0.23087794704959724,
      "eval_num_tokens": 134253615.0,
      "eval_runtime": 20.3542,
      "eval_samples_per_second": 1735.019,
      "eval_steps_per_second": 216.908,
      "step": 72000
    },
    {
      "entropy": 6.085148572921753,
      "epoch": 6.358618862592723,
      "grad_norm": 1.375,
      "learning_rate": 0.00018391599051024527,
      "loss": 5.2201,
      "mean_token_accuracy": 0.22828047275543212,
      "num_tokens": 134262965.0,
      "step": 72005
    },
    {
      "entropy": 6.0027632236480715,
      "epoch": 6.359060402684563,
      "grad_norm": 1.4375,
      "learning_rate": 0.00018388719565238317,
      "loss": 5.0668,
      "mean_token_accuracy": 0.23916891664266587,
      "num_tokens": 134271799.0,
      "step": 72010
    },
    {
      "entropy": 6.006814241409302,
      "epoch": 6.359501942776404,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018385840257936897,
      "loss": 5.1441,
      "mean_token_accuracy": 0.23960642367601395,
      "num_tokens": 134282299.0,
      "step": 72015
    },
    {
      "entropy": 6.030652475357056,
      "epoch": 6.359943482868244,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001838296112917668,
      "loss": 5.2309,
      "mean_token_accuracy": 0.21852647960186006,
      "num_tokens": 134290997.0,
      "step": 72020
    },
    {
      "entropy": 6.04334979057312,
      "epoch": 6.3603850229600845,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001838008217901408,
      "loss": 5.1681,
      "mean_token_accuracy": 0.22406495958566666,
      "num_tokens": 134300573.0,
      "step": 72025
    },
    {
      "entropy": 5.927307224273681,
      "epoch": 6.360826563051925,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00018377203407505483,
      "loss": 5.1428,
      "mean_token_accuracy": 0.23906634747982025,
      "num_tokens": 134310059.0,
      "step": 72030
    },
    {
      "entropy": 6.039495420455933,
      "epoch": 6.3612681031437655,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00018374324814707281,
      "loss": 5.1316,
      "mean_token_accuracy": 0.23014816492795945,
      "num_tokens": 134318544.0,
      "step": 72035
    },
    {
      "entropy": 6.0379705905914305,
      "epoch": 6.3617096432356055,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018371446400675869,
      "loss": 5.1935,
      "mean_token_accuracy": 0.22897524982690812,
      "num_tokens": 134327600.0,
      "step": 72040
    },
    {
      "entropy": 6.0315008640289305,
      "epoch": 6.3621511833274464,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018368568165467615,
      "loss": 5.1243,
      "mean_token_accuracy": 0.23515188097953796,
      "num_tokens": 134336822.0,
      "step": 72045
    },
    {
      "entropy": 6.091099262237549,
      "epoch": 6.3625927234192865,
      "grad_norm": 1.4375,
      "learning_rate": 0.00018365690109138934,
      "loss": 5.249,
      "mean_token_accuracy": 0.2201238214969635,
      "num_tokens": 134346095.0,
      "step": 72050
    },
    {
      "entropy": 6.065782260894776,
      "epoch": 6.3630342635111266,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00018362812231746185,
      "loss": 5.1878,
      "mean_token_accuracy": 0.22660498470067977,
      "num_tokens": 134354679.0,
      "step": 72055
    },
    {
      "entropy": 6.070665121078491,
      "epoch": 6.3634758036029675,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001835993453334574,
      "loss": 5.3062,
      "mean_token_accuracy": 0.21757742762565613,
      "num_tokens": 134364035.0,
      "step": 72060
    },
    {
      "entropy": 6.0852437019348145,
      "epoch": 6.3639173436948075,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00018357057013993988,
      "loss": 5.2453,
      "mean_token_accuracy": 0.21555272936820985,
      "num_tokens": 134373460.0,
      "step": 72065
    },
    {
      "entropy": 6.082564926147461,
      "epoch": 6.364358883786648,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001835417967374728,
      "loss": 5.2198,
      "mean_token_accuracy": 0.23350777179002763,
      "num_tokens": 134383667.0,
      "step": 72070
    },
    {
      "entropy": 6.121385002136231,
      "epoch": 6.3648004238784885,
      "grad_norm": 1.546875,
      "learning_rate": 0.00018351302512662,
      "loss": 5.2371,
      "mean_token_accuracy": 0.225783634185791,
      "num_tokens": 134391855.0,
      "step": 72075
    },
    {
      "entropy": 6.061810684204102,
      "epoch": 6.365241963970329,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018348425530794499,
      "loss": 5.2215,
      "mean_token_accuracy": 0.22842950671911239,
      "num_tokens": 134402180.0,
      "step": 72080
    },
    {
      "entropy": 6.050688171386719,
      "epoch": 6.365683504062169,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00018345548728201138,
      "loss": 5.251,
      "mean_token_accuracy": 0.23183012902736663,
      "num_tokens": 134411583.0,
      "step": 72085
    },
    {
      "entropy": 6.002784490585327,
      "epoch": 6.36612504415401,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018342672104938262,
      "loss": 5.1456,
      "mean_token_accuracy": 0.23523531407117843,
      "num_tokens": 134421557.0,
      "step": 72090
    },
    {
      "entropy": 6.047684097290039,
      "epoch": 6.36656658424585,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018339795661062236,
      "loss": 5.3041,
      "mean_token_accuracy": 0.21969198882579805,
      "num_tokens": 134430832.0,
      "step": 72095
    },
    {
      "entropy": 6.069177150726318,
      "epoch": 6.36700812433769,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001833691939662941,
      "loss": 5.2034,
      "mean_token_accuracy": 0.22383444905281066,
      "num_tokens": 134440038.0,
      "step": 72100
    },
    {
      "entropy": 6.111990833282471,
      "epoch": 6.367449664429531,
      "grad_norm": 1.578125,
      "learning_rate": 0.0001833404331169612,
      "loss": 5.2769,
      "mean_token_accuracy": 0.22681757509708406,
      "num_tokens": 134448375.0,
      "step": 72105
    },
    {
      "entropy": 6.0318958282470705,
      "epoch": 6.367891204521371,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000183311674063187,
      "loss": 5.1515,
      "mean_token_accuracy": 0.23162134736776352,
      "num_tokens": 134456625.0,
      "step": 72110
    },
    {
      "entropy": 6.066334295272827,
      "epoch": 6.368332744613211,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018328291680553505,
      "loss": 5.2881,
      "mean_token_accuracy": 0.22342354357242583,
      "num_tokens": 134466885.0,
      "step": 72115
    },
    {
      "entropy": 6.067664909362793,
      "epoch": 6.368774284705051,
      "grad_norm": 1.4375,
      "learning_rate": 0.00018325416134456858,
      "loss": 5.1775,
      "mean_token_accuracy": 0.23923462331295015,
      "num_tokens": 134475712.0,
      "step": 72120
    },
    {
      "entropy": 6.050408124923706,
      "epoch": 6.369215824796892,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018322540768085088,
      "loss": 5.2117,
      "mean_token_accuracy": 0.2232234612107277,
      "num_tokens": 134484745.0,
      "step": 72125
    },
    {
      "entropy": 6.0352463722229,
      "epoch": 6.369657364888732,
      "grad_norm": 1.5,
      "learning_rate": 0.0001831966558149452,
      "loss": 5.1518,
      "mean_token_accuracy": 0.2291746973991394,
      "num_tokens": 134493781.0,
      "step": 72130
    },
    {
      "entropy": 6.0243922710418705,
      "epoch": 6.370098904980572,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00018316790574741487,
      "loss": 5.1436,
      "mean_token_accuracy": 0.23145317286252975,
      "num_tokens": 134502075.0,
      "step": 72135
    },
    {
      "entropy": 6.059496545791626,
      "epoch": 6.370540445072413,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00018313915747882308,
      "loss": 5.1927,
      "mean_token_accuracy": 0.21959083080291747,
      "num_tokens": 134509955.0,
      "step": 72140
    },
    {
      "entropy": 6.068021583557129,
      "epoch": 6.370981985164253,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00018311041100973292,
      "loss": 5.2556,
      "mean_token_accuracy": 0.2245980218052864,
      "num_tokens": 134518584.0,
      "step": 72145
    },
    {
      "entropy": 6.05255160331726,
      "epoch": 6.371423525256093,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001830816663407076,
      "loss": 5.1368,
      "mean_token_accuracy": 0.23568432927131652,
      "num_tokens": 134527162.0,
      "step": 72150
    },
    {
      "entropy": 5.905821514129639,
      "epoch": 6.371865065347934,
      "grad_norm": 1.578125,
      "learning_rate": 0.00018305292347231,
      "loss": 5.0507,
      "mean_token_accuracy": 0.2331707939505577,
      "num_tokens": 134537091.0,
      "step": 72155
    },
    {
      "entropy": 6.061464643478393,
      "epoch": 6.372306605439774,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001830241824051035,
      "loss": 5.2448,
      "mean_token_accuracy": 0.23269792050123214,
      "num_tokens": 134546401.0,
      "step": 72160
    },
    {
      "entropy": 6.08590726852417,
      "epoch": 6.372748145531614,
      "grad_norm": 1.546875,
      "learning_rate": 0.00018299544313965093,
      "loss": 5.2509,
      "mean_token_accuracy": 0.22458924502134323,
      "num_tokens": 134555625.0,
      "step": 72165
    },
    {
      "entropy": 6.093982219696045,
      "epoch": 6.373189685623455,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001829667056765153,
      "loss": 5.3071,
      "mean_token_accuracy": 0.21975326389074326,
      "num_tokens": 134564823.0,
      "step": 72170
    },
    {
      "entropy": 5.981622648239136,
      "epoch": 6.373631225715295,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018293797001625944,
      "loss": 5.1535,
      "mean_token_accuracy": 0.24049674421548844,
      "num_tokens": 134574050.0,
      "step": 72175
    },
    {
      "entropy": 6.006125640869141,
      "epoch": 6.374072765807135,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018290923615944655,
      "loss": 5.1168,
      "mean_token_accuracy": 0.22852312326431273,
      "num_tokens": 134583051.0,
      "step": 72180
    },
    {
      "entropy": 6.037536382675171,
      "epoch": 6.374514305898976,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001828805041066393,
      "loss": 5.2512,
      "mean_token_accuracy": 0.22087409794330598,
      "num_tokens": 134593232.0,
      "step": 72185
    },
    {
      "entropy": 6.119208240509034,
      "epoch": 6.374955845990816,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018285177385840058,
      "loss": 5.2992,
      "mean_token_accuracy": 0.21303593665361403,
      "num_tokens": 134602595.0,
      "step": 72190
    },
    {
      "entropy": 6.007679653167725,
      "epoch": 6.375397386082656,
      "grad_norm": 1.546875,
      "learning_rate": 0.00018282304541529314,
      "loss": 5.1361,
      "mean_token_accuracy": 0.2313630238175392,
      "num_tokens": 134613285.0,
      "step": 72195
    },
    {
      "entropy": 6.001399374008178,
      "epoch": 6.375838926174497,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001827943187778799,
      "loss": 5.152,
      "mean_token_accuracy": 0.23259779065847397,
      "num_tokens": 134622769.0,
      "step": 72200
    },
    {
      "entropy": 6.0975995540618895,
      "epoch": 6.376280466266337,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001827655939467235,
      "loss": 5.259,
      "mean_token_accuracy": 0.21885378211736678,
      "num_tokens": 134632971.0,
      "step": 72205
    },
    {
      "entropy": 6.021077394485474,
      "epoch": 6.376722006358177,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018273687092238667,
      "loss": 5.1037,
      "mean_token_accuracy": 0.236461079120636,
      "num_tokens": 134642468.0,
      "step": 72210
    },
    {
      "entropy": 5.993970727920532,
      "epoch": 6.377163546450018,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018270814970543204,
      "loss": 5.1182,
      "mean_token_accuracy": 0.2346041277050972,
      "num_tokens": 134652213.0,
      "step": 72215
    },
    {
      "entropy": 5.980411243438721,
      "epoch": 6.377605086541858,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001826794302964222,
      "loss": 5.1247,
      "mean_token_accuracy": 0.23456307798624038,
      "num_tokens": 134661893.0,
      "step": 72220
    },
    {
      "entropy": 6.121136474609375,
      "epoch": 6.378046626633698,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018265071269591993,
      "loss": 5.2474,
      "mean_token_accuracy": 0.22336681932210922,
      "num_tokens": 134671356.0,
      "step": 72225
    },
    {
      "entropy": 5.959933710098267,
      "epoch": 6.378488166725539,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018262199690448762,
      "loss": 5.045,
      "mean_token_accuracy": 0.24430338740348817,
      "num_tokens": 134680781.0,
      "step": 72230
    },
    {
      "entropy": 6.027146720886231,
      "epoch": 6.378929706817379,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001825932829226879,
      "loss": 5.1437,
      "mean_token_accuracy": 0.23318926841020585,
      "num_tokens": 134689695.0,
      "step": 72235
    },
    {
      "entropy": 6.021663188934326,
      "epoch": 6.379371246909219,
      "grad_norm": 1.4375,
      "learning_rate": 0.00018256457075108308,
      "loss": 5.1806,
      "mean_token_accuracy": 0.2295595273375511,
      "num_tokens": 134698787.0,
      "step": 72240
    },
    {
      "entropy": 6.031795930862427,
      "epoch": 6.379812787001059,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001825358603902359,
      "loss": 5.156,
      "mean_token_accuracy": 0.22870527505874633,
      "num_tokens": 134707180.0,
      "step": 72245
    },
    {
      "entropy": 5.996192932128906,
      "epoch": 6.3802543270929,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018250715184070848,
      "loss": 5.1731,
      "mean_token_accuracy": 0.22901031672954558,
      "num_tokens": 134717296.0,
      "step": 72250
    },
    {
      "entropy": 6.0560140132904055,
      "epoch": 6.38069586718474,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001824784451030635,
      "loss": 5.1412,
      "mean_token_accuracy": 0.22493072748184204,
      "num_tokens": 134727097.0,
      "step": 72255
    },
    {
      "entropy": 5.990032482147217,
      "epoch": 6.38113740727658,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018244974017786304,
      "loss": 5.1807,
      "mean_token_accuracy": 0.23333459496498107,
      "num_tokens": 134736043.0,
      "step": 72260
    },
    {
      "entropy": 5.999984455108643,
      "epoch": 6.381578947368421,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001824210370656696,
      "loss": 5.2109,
      "mean_token_accuracy": 0.23424049019813536,
      "num_tokens": 134745468.0,
      "step": 72265
    },
    {
      "entropy": 6.10510287284851,
      "epoch": 6.382020487460261,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001823923357670454,
      "loss": 5.315,
      "mean_token_accuracy": 0.2134024292230606,
      "num_tokens": 134754557.0,
      "step": 72270
    },
    {
      "entropy": 6.032690191268921,
      "epoch": 6.382462027552101,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0001823636362825527,
      "loss": 5.1078,
      "mean_token_accuracy": 0.24045825898647308,
      "num_tokens": 134762999.0,
      "step": 72275
    },
    {
      "entropy": 6.0439623355865475,
      "epoch": 6.382903567643942,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0001823349386127537,
      "loss": 5.1994,
      "mean_token_accuracy": 0.2229466915130615,
      "num_tokens": 134772175.0,
      "step": 72280
    },
    {
      "entropy": 6.089414024353028,
      "epoch": 6.383345107735782,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001823062427582104,
      "loss": 5.176,
      "mean_token_accuracy": 0.230042864382267,
      "num_tokens": 134781343.0,
      "step": 72285
    },
    {
      "entropy": 6.0786964893341064,
      "epoch": 6.383786647827622,
      "grad_norm": 1.4375,
      "learning_rate": 0.00018227754871948516,
      "loss": 5.2228,
      "mean_token_accuracy": 0.22355578988790512,
      "num_tokens": 134790214.0,
      "step": 72290
    },
    {
      "entropy": 6.0471819877624515,
      "epoch": 6.384228187919463,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018224885649714006,
      "loss": 5.1682,
      "mean_token_accuracy": 0.23038221299648284,
      "num_tokens": 134799638.0,
      "step": 72295
    },
    {
      "entropy": 5.9862799644470215,
      "epoch": 6.384669728011303,
      "grad_norm": 1.453125,
      "learning_rate": 0.00018222016609173713,
      "loss": 5.1102,
      "mean_token_accuracy": 0.24133894890546798,
      "num_tokens": 134809195.0,
      "step": 72300
    },
    {
      "entropy": 6.0599006652832035,
      "epoch": 6.3851112681031434,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00018219147750383823,
      "loss": 5.191,
      "mean_token_accuracy": 0.22356169372797013,
      "num_tokens": 134818086.0,
      "step": 72305
    },
    {
      "entropy": 6.096798658370972,
      "epoch": 6.385552808194984,
      "grad_norm": 1.5,
      "learning_rate": 0.00018216279073400565,
      "loss": 5.2368,
      "mean_token_accuracy": 0.22317418456077576,
      "num_tokens": 134826606.0,
      "step": 72310
    },
    {
      "entropy": 5.986972379684448,
      "epoch": 6.385994348286824,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018213410578280109,
      "loss": 5.1133,
      "mean_token_accuracy": 0.23316074013710023,
      "num_tokens": 134835057.0,
      "step": 72315
    },
    {
      "entropy": 6.007153701782227,
      "epoch": 6.3864358883786645,
      "grad_norm": 1.421875,
      "learning_rate": 0.00018210542265078666,
      "loss": 5.1352,
      "mean_token_accuracy": 0.23170446753501892,
      "num_tokens": 134844242.0,
      "step": 72320
    },
    {
      "entropy": 5.949256849288941,
      "epoch": 6.386877428470505,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001820767413385242,
      "loss": 5.0831,
      "mean_token_accuracy": 0.23868893682956696,
      "num_tokens": 134854197.0,
      "step": 72325
    },
    {
      "entropy": 6.017849111557007,
      "epoch": 6.3873189685623455,
      "grad_norm": 1.515625,
      "learning_rate": 0.00018204806184657542,
      "loss": 5.1637,
      "mean_token_accuracy": 0.22911248058080674,
      "num_tokens": 134863509.0,
      "step": 72330
    },
    {
      "entropy": 6.1228721141815186,
      "epoch": 6.3877605086541855,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0001820193841755023,
      "loss": 5.2883,
      "mean_token_accuracy": 0.2229285329580307,
      "num_tokens": 134872316.0,
      "step": 72335
    },
    {
      "entropy": 6.101109313964844,
      "epoch": 6.3882020487460265,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001819907083258666,
      "loss": 5.2688,
      "mean_token_accuracy": 0.2279080256819725,
      "num_tokens": 134881956.0,
      "step": 72340
    },
    {
      "entropy": 5.971071434020996,
      "epoch": 6.3886435888378665,
      "grad_norm": 1.375,
      "learning_rate": 0.00018196203429823,
      "loss": 5.085,
      "mean_token_accuracy": 0.24474502056837083,
      "num_tokens": 134891121.0,
      "step": 72345
    },
    {
      "entropy": 6.082256078720093,
      "epoch": 6.389085128929707,
      "grad_norm": 1.609375,
      "learning_rate": 0.0001819333620931542,
      "loss": 5.25,
      "mean_token_accuracy": 0.22520736902952193,
      "num_tokens": 134900074.0,
      "step": 72350
    },
    {
      "entropy": 6.113578414916992,
      "epoch": 6.3895266690215475,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018190469171120094,
      "loss": 5.1955,
      "mean_token_accuracy": 0.2271719053387642,
      "num_tokens": 134908945.0,
      "step": 72355
    },
    {
      "entropy": 6.104056167602539,
      "epoch": 6.389968209113388,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018187602315293183,
      "loss": 5.2939,
      "mean_token_accuracy": 0.21680342257022858,
      "num_tokens": 134918374.0,
      "step": 72360
    },
    {
      "entropy": 6.071239995956421,
      "epoch": 6.390409749205228,
      "grad_norm": 1.59375,
      "learning_rate": 0.00018184735641890842,
      "loss": 5.2916,
      "mean_token_accuracy": 0.213956056535244,
      "num_tokens": 134927043.0,
      "step": 72365
    },
    {
      "entropy": 6.056768417358398,
      "epoch": 6.390851289297069,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018181869150969232,
      "loss": 5.213,
      "mean_token_accuracy": 0.23092517852783204,
      "num_tokens": 134936898.0,
      "step": 72370
    },
    {
      "entropy": 6.00008807182312,
      "epoch": 6.391292829388909,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018179002842584513,
      "loss": 5.1759,
      "mean_token_accuracy": 0.23216013461351395,
      "num_tokens": 134946466.0,
      "step": 72375
    },
    {
      "entropy": 6.020346832275391,
      "epoch": 6.391734369480749,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00018176136716792818,
      "loss": 5.2128,
      "mean_token_accuracy": 0.22320578396320342,
      "num_tokens": 134955944.0,
      "step": 72380
    },
    {
      "entropy": 5.950846910476685,
      "epoch": 6.39217590957259,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018173270773650312,
      "loss": 5.0887,
      "mean_token_accuracy": 0.23422413617372512,
      "num_tokens": 134965246.0,
      "step": 72385
    },
    {
      "entropy": 6.076325750350952,
      "epoch": 6.39261744966443,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001817040501321312,
      "loss": 5.2398,
      "mean_token_accuracy": 0.21459944397211075,
      "num_tokens": 134974780.0,
      "step": 72390
    },
    {
      "entropy": 5.96962251663208,
      "epoch": 6.39305898975627,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018167539435537383,
      "loss": 5.1021,
      "mean_token_accuracy": 0.23563334941864014,
      "num_tokens": 134983811.0,
      "step": 72395
    },
    {
      "entropy": 6.109715795516967,
      "epoch": 6.39350052984811,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018164674040679253,
      "loss": 5.2033,
      "mean_token_accuracy": 0.23287330120801925,
      "num_tokens": 134993058.0,
      "step": 72400
    },
    {
      "entropy": 6.0496265411376955,
      "epoch": 6.393942069939951,
      "grad_norm": 1.453125,
      "learning_rate": 0.00018161808828694848,
      "loss": 5.2296,
      "mean_token_accuracy": 0.2276323288679123,
      "num_tokens": 135002479.0,
      "step": 72405
    },
    {
      "entropy": 6.068730974197388,
      "epoch": 6.394383610031791,
      "grad_norm": 1.5,
      "learning_rate": 0.00018158943799640294,
      "loss": 5.2456,
      "mean_token_accuracy": 0.22675777077674866,
      "num_tokens": 135011471.0,
      "step": 72410
    },
    {
      "entropy": 6.043718004226685,
      "epoch": 6.394825150123631,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001815607895357171,
      "loss": 5.1453,
      "mean_token_accuracy": 0.241004142165184,
      "num_tokens": 135019672.0,
      "step": 72415
    },
    {
      "entropy": 6.10523042678833,
      "epoch": 6.395266690215472,
      "grad_norm": 1.421875,
      "learning_rate": 0.00018153214290545235,
      "loss": 5.1863,
      "mean_token_accuracy": 0.2269708663225174,
      "num_tokens": 135028714.0,
      "step": 72420
    },
    {
      "entropy": 6.059910011291504,
      "epoch": 6.395708230307312,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001815034981061698,
      "loss": 5.2093,
      "mean_token_accuracy": 0.22729528099298477,
      "num_tokens": 135037599.0,
      "step": 72425
    },
    {
      "entropy": 6.034963369369507,
      "epoch": 6.396149770399152,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00018147485513843044,
      "loss": 5.1831,
      "mean_token_accuracy": 0.22807004749774934,
      "num_tokens": 135046261.0,
      "step": 72430
    },
    {
      "entropy": 5.93724799156189,
      "epoch": 6.396591310490993,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018144621400279548,
      "loss": 5.131,
      "mean_token_accuracy": 0.23599084466695786,
      "num_tokens": 135056046.0,
      "step": 72435
    },
    {
      "entropy": 6.027021360397339,
      "epoch": 6.397032850582833,
      "grad_norm": 1.53125,
      "learning_rate": 0.000181417574699826,
      "loss": 5.0964,
      "mean_token_accuracy": 0.23317359387874603,
      "num_tokens": 135065094.0,
      "step": 72440
    },
    {
      "entropy": 6.050240850448608,
      "epoch": 6.397474390674673,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018138893723008303,
      "loss": 5.1267,
      "mean_token_accuracy": 0.23690550029277802,
      "num_tokens": 135073502.0,
      "step": 72445
    },
    {
      "entropy": 6.032764625549317,
      "epoch": 6.397915930766514,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018136030159412764,
      "loss": 5.1638,
      "mean_token_accuracy": 0.2343120038509369,
      "num_tokens": 135082739.0,
      "step": 72450
    },
    {
      "entropy": 5.951064777374268,
      "epoch": 6.398357470858354,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00018133166779252057,
      "loss": 5.1627,
      "mean_token_accuracy": 0.2333879292011261,
      "num_tokens": 135091673.0,
      "step": 72455
    },
    {
      "entropy": 5.980304431915283,
      "epoch": 6.398799010950194,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001813030358258228,
      "loss": 5.1996,
      "mean_token_accuracy": 0.2327078253030777,
      "num_tokens": 135100849.0,
      "step": 72460
    },
    {
      "entropy": 5.9653637409210205,
      "epoch": 6.399240551042035,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00018127440569459536,
      "loss": 5.18,
      "mean_token_accuracy": 0.2282908722758293,
      "num_tokens": 135110661.0,
      "step": 72465
    },
    {
      "entropy": 5.974596548080444,
      "epoch": 6.399682091133875,
      "grad_norm": 1.625,
      "learning_rate": 0.00018124577739939897,
      "loss": 5.0982,
      "mean_token_accuracy": 0.2339981898665428,
      "num_tokens": 135119021.0,
      "step": 72470
    },
    {
      "entropy": 6.077662420272827,
      "epoch": 6.400123631225715,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018121715094079454,
      "loss": 5.2248,
      "mean_token_accuracy": 0.23031542897224427,
      "num_tokens": 135128066.0,
      "step": 72475
    },
    {
      "entropy": 6.064073896408081,
      "epoch": 6.400565171317556,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018118852631934275,
      "loss": 5.2314,
      "mean_token_accuracy": 0.22573940008878707,
      "num_tokens": 135137885.0,
      "step": 72480
    },
    {
      "entropy": 6.011835098266602,
      "epoch": 6.401006711409396,
      "grad_norm": 1.4375,
      "learning_rate": 0.00018115990353560435,
      "loss": 5.1308,
      "mean_token_accuracy": 0.23218374997377395,
      "num_tokens": 135147222.0,
      "step": 72485
    },
    {
      "entropy": 6.034543561935425,
      "epoch": 6.401448251501236,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018113128259014005,
      "loss": 5.2155,
      "mean_token_accuracy": 0.22804097086191177,
      "num_tokens": 135156483.0,
      "step": 72490
    },
    {
      "entropy": 5.984066486358643,
      "epoch": 6.401889791593077,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001811026634835107,
      "loss": 5.164,
      "mean_token_accuracy": 0.2347170516848564,
      "num_tokens": 135166367.0,
      "step": 72495
    },
    {
      "entropy": 5.981552076339722,
      "epoch": 6.402331331684917,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001810740462162765,
      "loss": 5.102,
      "mean_token_accuracy": 0.2323054626584053,
      "num_tokens": 135175228.0,
      "step": 72500
    },
    {
      "entropy": 6.010305309295655,
      "epoch": 6.402772871776757,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018104543078899853,
      "loss": 5.2162,
      "mean_token_accuracy": 0.22650054097175598,
      "num_tokens": 135184291.0,
      "step": 72505
    },
    {
      "entropy": 6.007086610794067,
      "epoch": 6.403214411868598,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018101681720223705,
      "loss": 5.1258,
      "mean_token_accuracy": 0.23331202268600465,
      "num_tokens": 135192932.0,
      "step": 72510
    },
    {
      "entropy": 6.038142395019531,
      "epoch": 6.403655951960438,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018098820545655277,
      "loss": 5.2478,
      "mean_token_accuracy": 0.2224357768893242,
      "num_tokens": 135202248.0,
      "step": 72515
    },
    {
      "entropy": 6.048619508743286,
      "epoch": 6.404097492052278,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018095959555250595,
      "loss": 5.2614,
      "mean_token_accuracy": 0.21857171803712844,
      "num_tokens": 135211701.0,
      "step": 72520
    },
    {
      "entropy": 5.9716565132141115,
      "epoch": 6.404539032144118,
      "grad_norm": 1.4375,
      "learning_rate": 0.00018093098749065722,
      "loss": 5.1213,
      "mean_token_accuracy": 0.24042104184627533,
      "num_tokens": 135220545.0,
      "step": 72525
    },
    {
      "entropy": 6.05140872001648,
      "epoch": 6.404980572235959,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018090238127156693,
      "loss": 5.1593,
      "mean_token_accuracy": 0.22807050347328187,
      "num_tokens": 135230550.0,
      "step": 72530
    },
    {
      "entropy": 6.077653503417968,
      "epoch": 6.405422112327799,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00018087377689579549,
      "loss": 5.1579,
      "mean_token_accuracy": 0.2310902327299118,
      "num_tokens": 135239032.0,
      "step": 72535
    },
    {
      "entropy": 5.991582584381104,
      "epoch": 6.405863652419639,
      "grad_norm": 1.265625,
      "learning_rate": 0.00018084517436390323,
      "loss": 5.1151,
      "mean_token_accuracy": 0.2414459466934204,
      "num_tokens": 135249478.0,
      "step": 72540
    },
    {
      "entropy": 5.981379413604737,
      "epoch": 6.40630519251148,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00018081657367645049,
      "loss": 5.1205,
      "mean_token_accuracy": 0.237137271463871,
      "num_tokens": 135258428.0,
      "step": 72545
    },
    {
      "entropy": 6.058144807815552,
      "epoch": 6.40674673260332,
      "grad_norm": 1.5,
      "learning_rate": 0.00018078797483399745,
      "loss": 5.2651,
      "mean_token_accuracy": 0.2225346863269806,
      "num_tokens": 135267613.0,
      "step": 72550
    },
    {
      "entropy": 6.012741184234619,
      "epoch": 6.40718827269516,
      "grad_norm": 1.421875,
      "learning_rate": 0.00018075937783710439,
      "loss": 5.1645,
      "mean_token_accuracy": 0.23498202115297318,
      "num_tokens": 135277648.0,
      "step": 72555
    },
    {
      "entropy": 6.030074596405029,
      "epoch": 6.407629812787001,
      "grad_norm": 1.4375,
      "learning_rate": 0.00018073078268633164,
      "loss": 5.1445,
      "mean_token_accuracy": 0.235527166724205,
      "num_tokens": 135286864.0,
      "step": 72560
    },
    {
      "entropy": 6.054480266571045,
      "epoch": 6.408071352878841,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001807021893822391,
      "loss": 5.1855,
      "mean_token_accuracy": 0.22996231615543367,
      "num_tokens": 135295864.0,
      "step": 72565
    },
    {
      "entropy": 5.989073896408081,
      "epoch": 6.408512892970681,
      "grad_norm": 1.375,
      "learning_rate": 0.00018067359792538706,
      "loss": 5.1551,
      "mean_token_accuracy": 0.236356483399868,
      "num_tokens": 135305331.0,
      "step": 72570
    },
    {
      "entropy": 6.03484992980957,
      "epoch": 6.408954433062522,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018064500831633563,
      "loss": 5.2017,
      "mean_token_accuracy": 0.2279114693403244,
      "num_tokens": 135315792.0,
      "step": 72575
    },
    {
      "entropy": 5.924392557144165,
      "epoch": 6.409395973154362,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018061642055564492,
      "loss": 5.0267,
      "mean_token_accuracy": 0.2485029488801956,
      "num_tokens": 135324456.0,
      "step": 72580
    },
    {
      "entropy": 6.021338939666748,
      "epoch": 6.409837513246202,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018058783464387475,
      "loss": 5.1637,
      "mean_token_accuracy": 0.2295669436454773,
      "num_tokens": 135334352.0,
      "step": 72585
    },
    {
      "entropy": 6.0577000141143795,
      "epoch": 6.410279053338043,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018055925058158513,
      "loss": 5.2825,
      "mean_token_accuracy": 0.22493149638175963,
      "num_tokens": 135344157.0,
      "step": 72590
    },
    {
      "entropy": 5.953558540344238,
      "epoch": 6.410720593429883,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018053066836933612,
      "loss": 5.1267,
      "mean_token_accuracy": 0.2365507423877716,
      "num_tokens": 135352950.0,
      "step": 72595
    },
    {
      "entropy": 6.107327222824097,
      "epoch": 6.4111621335217235,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018050208800768765,
      "loss": 5.3816,
      "mean_token_accuracy": 0.2107352524995804,
      "num_tokens": 135361802.0,
      "step": 72600
    },
    {
      "entropy": 6.049883508682251,
      "epoch": 6.411603673613564,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018047350949719953,
      "loss": 5.2858,
      "mean_token_accuracy": 0.22516858130693435,
      "num_tokens": 135371574.0,
      "step": 72605
    },
    {
      "entropy": 6.067880725860595,
      "epoch": 6.4120452137054045,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018044493283843156,
      "loss": 5.2224,
      "mean_token_accuracy": 0.21864799857139589,
      "num_tokens": 135381229.0,
      "step": 72610
    },
    {
      "entropy": 6.075825786590576,
      "epoch": 6.4124867537972445,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018041635803194356,
      "loss": 5.2519,
      "mean_token_accuracy": 0.22892017811536788,
      "num_tokens": 135390818.0,
      "step": 72615
    },
    {
      "entropy": 6.025710153579712,
      "epoch": 6.4129282938890855,
      "grad_norm": 1.5625,
      "learning_rate": 0.00018038778507829535,
      "loss": 5.1947,
      "mean_token_accuracy": 0.2292865812778473,
      "num_tokens": 135399801.0,
      "step": 72620
    },
    {
      "entropy": 6.006979942321777,
      "epoch": 6.4133698339809255,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018035921397804668,
      "loss": 5.2049,
      "mean_token_accuracy": 0.23453116714954375,
      "num_tokens": 135409248.0,
      "step": 72625
    },
    {
      "entropy": 6.038823080062866,
      "epoch": 6.413811374072766,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018033064473175713,
      "loss": 5.2181,
      "mean_token_accuracy": 0.223002989590168,
      "num_tokens": 135417495.0,
      "step": 72630
    },
    {
      "entropy": 5.995086145401001,
      "epoch": 6.4142529141646065,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018030207733998638,
      "loss": 5.2099,
      "mean_token_accuracy": 0.22526072859764099,
      "num_tokens": 135426774.0,
      "step": 72635
    },
    {
      "entropy": 6.004425525665283,
      "epoch": 6.414694454256447,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001802735118032941,
      "loss": 5.2009,
      "mean_token_accuracy": 0.22071910798549652,
      "num_tokens": 135436368.0,
      "step": 72640
    },
    {
      "entropy": 6.032795095443726,
      "epoch": 6.415135994348287,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00018024494812223985,
      "loss": 5.1304,
      "mean_token_accuracy": 0.22970362156629562,
      "num_tokens": 135445401.0,
      "step": 72645
    },
    {
      "entropy": 5.988029193878174,
      "epoch": 6.415577534440128,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00018021638629738318,
      "loss": 5.1606,
      "mean_token_accuracy": 0.2334589660167694,
      "num_tokens": 135454285.0,
      "step": 72650
    },
    {
      "entropy": 6.034882593154907,
      "epoch": 6.416019074531968,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001801878263292836,
      "loss": 5.2058,
      "mean_token_accuracy": 0.23045666217803956,
      "num_tokens": 135463639.0,
      "step": 72655
    },
    {
      "entropy": 6.069443798065185,
      "epoch": 6.416460614623808,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00018015926821850055,
      "loss": 5.191,
      "mean_token_accuracy": 0.2280579164624214,
      "num_tokens": 135472842.0,
      "step": 72660
    },
    {
      "entropy": 6.115406703948975,
      "epoch": 6.416902154715649,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001801307119655935,
      "loss": 5.2547,
      "mean_token_accuracy": 0.2236936569213867,
      "num_tokens": 135482326.0,
      "step": 72665
    },
    {
      "entropy": 6.064954328536987,
      "epoch": 6.417343694807489,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001801021575711219,
      "loss": 5.2543,
      "mean_token_accuracy": 0.22488884180784224,
      "num_tokens": 135491373.0,
      "step": 72670
    },
    {
      "entropy": 6.093793344497681,
      "epoch": 6.417785234899329,
      "grad_norm": 1.53125,
      "learning_rate": 0.000180073605035645,
      "loss": 5.2063,
      "mean_token_accuracy": 0.22879040539264678,
      "num_tokens": 135500690.0,
      "step": 72675
    },
    {
      "entropy": 6.0260755062103275,
      "epoch": 6.418226774991169,
      "grad_norm": 1.4375,
      "learning_rate": 0.00018004505435972225,
      "loss": 5.1969,
      "mean_token_accuracy": 0.22394685447216034,
      "num_tokens": 135510085.0,
      "step": 72680
    },
    {
      "entropy": 6.078405952453613,
      "epoch": 6.41866831508301,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001800165055439128,
      "loss": 5.2794,
      "mean_token_accuracy": 0.21833257228136063,
      "num_tokens": 135520197.0,
      "step": 72685
    },
    {
      "entropy": 6.01177978515625,
      "epoch": 6.41910985517485,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001799879585887761,
      "loss": 5.131,
      "mean_token_accuracy": 0.22887562215328217,
      "num_tokens": 135530540.0,
      "step": 72690
    },
    {
      "entropy": 6.081964445114136,
      "epoch": 6.41955139526669,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017995941349487117,
      "loss": 5.2264,
      "mean_token_accuracy": 0.2286106362938881,
      "num_tokens": 135540285.0,
      "step": 72695
    },
    {
      "entropy": 6.100079727172852,
      "epoch": 6.419992935358531,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017993087026275728,
      "loss": 5.287,
      "mean_token_accuracy": 0.22621574997901917,
      "num_tokens": 135549227.0,
      "step": 72700
    },
    {
      "entropy": 6.0997720718383786,
      "epoch": 6.420434475450371,
      "grad_norm": 1.453125,
      "learning_rate": 0.00017990232889299358,
      "loss": 5.2506,
      "mean_token_accuracy": 0.2247163638472557,
      "num_tokens": 135558621.0,
      "step": 72705
    },
    {
      "entropy": 6.003881883621216,
      "epoch": 6.420876015542211,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00017987378938613924,
      "loss": 5.0886,
      "mean_token_accuracy": 0.240548238158226,
      "num_tokens": 135567372.0,
      "step": 72710
    },
    {
      "entropy": 6.036970758438111,
      "epoch": 6.421317555634052,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017984525174275317,
      "loss": 5.1749,
      "mean_token_accuracy": 0.23597031831741333,
      "num_tokens": 135577380.0,
      "step": 72715
    },
    {
      "entropy": 6.0232501983642575,
      "epoch": 6.421759095725892,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001798167159633945,
      "loss": 5.113,
      "mean_token_accuracy": 0.23664651513099672,
      "num_tokens": 135586678.0,
      "step": 72720
    },
    {
      "entropy": 6.044790935516358,
      "epoch": 6.422200635817732,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001797881820486223,
      "loss": 5.147,
      "mean_token_accuracy": 0.23220470398664475,
      "num_tokens": 135594682.0,
      "step": 72725
    },
    {
      "entropy": 6.031757402420044,
      "epoch": 6.422642175909573,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017975964999899542,
      "loss": 5.2371,
      "mean_token_accuracy": 0.21936984062194825,
      "num_tokens": 135605644.0,
      "step": 72730
    },
    {
      "entropy": 5.949765157699585,
      "epoch": 6.423083716001413,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001797311198150729,
      "loss": 5.0997,
      "mean_token_accuracy": 0.23229114562273026,
      "num_tokens": 135614012.0,
      "step": 72735
    },
    {
      "entropy": 6.073131561279297,
      "epoch": 6.423525256093253,
      "grad_norm": 1.578125,
      "learning_rate": 0.00017970259149741337,
      "loss": 5.2405,
      "mean_token_accuracy": 0.22619881331920624,
      "num_tokens": 135623122.0,
      "step": 72740
    },
    {
      "entropy": 6.063552284240723,
      "epoch": 6.423966796185094,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00017967406504657606,
      "loss": 5.2756,
      "mean_token_accuracy": 0.22300105392932892,
      "num_tokens": 135632703.0,
      "step": 72745
    },
    {
      "entropy": 6.047269487380982,
      "epoch": 6.424408336276934,
      "grad_norm": 1.546875,
      "learning_rate": 0.00017964554046311955,
      "loss": 5.1842,
      "mean_token_accuracy": 0.22982044667005538,
      "num_tokens": 135642613.0,
      "step": 72750
    },
    {
      "entropy": 6.035674810409546,
      "epoch": 6.424849876368774,
      "grad_norm": 1.6875,
      "learning_rate": 0.00017961701774760276,
      "loss": 5.1862,
      "mean_token_accuracy": 0.23057285100221633,
      "num_tokens": 135651229.0,
      "step": 72755
    },
    {
      "entropy": 6.066130971908569,
      "epoch": 6.425291416460615,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001795884969005842,
      "loss": 5.2518,
      "mean_token_accuracy": 0.21908183991909028,
      "num_tokens": 135661725.0,
      "step": 72760
    },
    {
      "entropy": 6.04557466506958,
      "epoch": 6.425732956552455,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00017955997792262274,
      "loss": 5.2349,
      "mean_token_accuracy": 0.2217671424150467,
      "num_tokens": 135670622.0,
      "step": 72765
    },
    {
      "entropy": 6.093409490585327,
      "epoch": 6.426174496644295,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017953146081427706,
      "loss": 5.2393,
      "mean_token_accuracy": 0.2271408572793007,
      "num_tokens": 135679604.0,
      "step": 72770
    },
    {
      "entropy": 6.078974819183349,
      "epoch": 6.426616036736136,
      "grad_norm": 1.375,
      "learning_rate": 0.00017950294557610586,
      "loss": 5.1826,
      "mean_token_accuracy": 0.2272156298160553,
      "num_tokens": 135688346.0,
      "step": 72775
    },
    {
      "entropy": 6.105705690383911,
      "epoch": 6.427057576827976,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017947443220866756,
      "loss": 5.2825,
      "mean_token_accuracy": 0.22246816158294677,
      "num_tokens": 135698113.0,
      "step": 72780
    },
    {
      "entropy": 6.06584997177124,
      "epoch": 6.427499116919816,
      "grad_norm": 1.484375,
      "learning_rate": 0.00017944592071252074,
      "loss": 5.2473,
      "mean_token_accuracy": 0.22436997145414353,
      "num_tokens": 135707839.0,
      "step": 72785
    },
    {
      "entropy": 6.088120460510254,
      "epoch": 6.427940657011657,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000179417411088224,
      "loss": 5.2006,
      "mean_token_accuracy": 0.2280411198735237,
      "num_tokens": 135717619.0,
      "step": 72790
    },
    {
      "entropy": 6.053931140899659,
      "epoch": 6.428382197103497,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001793889033363359,
      "loss": 5.2255,
      "mean_token_accuracy": 0.22513751685619354,
      "num_tokens": 135727683.0,
      "step": 72795
    },
    {
      "entropy": 6.079124784469604,
      "epoch": 6.428823737195337,
      "grad_norm": 1.546875,
      "learning_rate": 0.00017936039745741484,
      "loss": 5.1675,
      "mean_token_accuracy": 0.2302924320101738,
      "num_tokens": 135736801.0,
      "step": 72800
    },
    {
      "entropy": 6.03086404800415,
      "epoch": 6.429265277287177,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017933189345201912,
      "loss": 5.2738,
      "mean_token_accuracy": 0.22466396987438203,
      "num_tokens": 135746759.0,
      "step": 72805
    },
    {
      "entropy": 6.109205532073974,
      "epoch": 6.429706817379018,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017930339132070716,
      "loss": 5.2677,
      "mean_token_accuracy": 0.21552886813879013,
      "num_tokens": 135757233.0,
      "step": 72810
    },
    {
      "entropy": 6.011569309234619,
      "epoch": 6.430148357470858,
      "grad_norm": 1.53125,
      "learning_rate": 0.00017927489106403738,
      "loss": 5.1611,
      "mean_token_accuracy": 0.23705787509679793,
      "num_tokens": 135766826.0,
      "step": 72815
    },
    {
      "entropy": 5.966382789611816,
      "epoch": 6.430589897562699,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017924639268256814,
      "loss": 5.1089,
      "mean_token_accuracy": 0.24546022713184357,
      "num_tokens": 135775501.0,
      "step": 72820
    },
    {
      "entropy": 6.05725998878479,
      "epoch": 6.431031437654539,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001792178961768575,
      "loss": 5.1984,
      "mean_token_accuracy": 0.22697965204715728,
      "num_tokens": 135785054.0,
      "step": 72825
    },
    {
      "entropy": 6.073242330551148,
      "epoch": 6.431472977746379,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00017918940154746383,
      "loss": 5.175,
      "mean_token_accuracy": 0.23305580914020538,
      "num_tokens": 135793689.0,
      "step": 72830
    },
    {
      "entropy": 5.989125728607178,
      "epoch": 6.431914517838219,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017916090879494524,
      "loss": 5.1765,
      "mean_token_accuracy": 0.23383914232254027,
      "num_tokens": 135802542.0,
      "step": 72835
    },
    {
      "entropy": 6.024219036102295,
      "epoch": 6.43235605793006,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017913241791986007,
      "loss": 5.1914,
      "mean_token_accuracy": 0.22826112806797028,
      "num_tokens": 135811990.0,
      "step": 72840
    },
    {
      "entropy": 6.025516223907471,
      "epoch": 6.4327975980219,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017910392892276622,
      "loss": 5.1944,
      "mean_token_accuracy": 0.233308807015419,
      "num_tokens": 135821191.0,
      "step": 72845
    },
    {
      "entropy": 6.127788400650024,
      "epoch": 6.43323913811374,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017907544180422187,
      "loss": 5.3002,
      "mean_token_accuracy": 0.21696486026048661,
      "num_tokens": 135830320.0,
      "step": 72850
    },
    {
      "entropy": 5.986446809768677,
      "epoch": 6.433680678205581,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017904695656478504,
      "loss": 5.1947,
      "mean_token_accuracy": 0.23067243546247482,
      "num_tokens": 135839765.0,
      "step": 72855
    },
    {
      "entropy": 5.97621169090271,
      "epoch": 6.434122218297421,
      "grad_norm": 1.453125,
      "learning_rate": 0.00017901847320501375,
      "loss": 5.0602,
      "mean_token_accuracy": 0.2403232216835022,
      "num_tokens": 135848795.0,
      "step": 72860
    },
    {
      "entropy": 5.993353033065796,
      "epoch": 6.434563758389261,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001789899917254661,
      "loss": 5.1029,
      "mean_token_accuracy": 0.23297332972288132,
      "num_tokens": 135856943.0,
      "step": 72865
    },
    {
      "entropy": 6.040097904205322,
      "epoch": 6.435005298481102,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017896151212669982,
      "loss": 5.187,
      "mean_token_accuracy": 0.232087205350399,
      "num_tokens": 135867252.0,
      "step": 72870
    },
    {
      "entropy": 6.058801794052124,
      "epoch": 6.435446838572942,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00017893303440927284,
      "loss": 5.1756,
      "mean_token_accuracy": 0.2308831438422203,
      "num_tokens": 135875978.0,
      "step": 72875
    },
    {
      "entropy": 6.068047189712525,
      "epoch": 6.4358883786647825,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001789045585737431,
      "loss": 5.207,
      "mean_token_accuracy": 0.2146911710500717,
      "num_tokens": 135884373.0,
      "step": 72880
    },
    {
      "entropy": 6.089624547958374,
      "epoch": 6.436329918756623,
      "grad_norm": 1.375,
      "learning_rate": 0.0001788760846206685,
      "loss": 5.2364,
      "mean_token_accuracy": 0.22204395830631257,
      "num_tokens": 135893820.0,
      "step": 72885
    },
    {
      "entropy": 5.964035940170288,
      "epoch": 6.4367714588484635,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017884761255060666,
      "loss": 5.1147,
      "mean_token_accuracy": 0.2320214942097664,
      "num_tokens": 135903069.0,
      "step": 72890
    },
    {
      "entropy": 6.0412843227386475,
      "epoch": 6.4372129989403035,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001788191423641154,
      "loss": 5.1345,
      "mean_token_accuracy": 0.23536669015884398,
      "num_tokens": 135913372.0,
      "step": 72895
    },
    {
      "entropy": 6.066806745529175,
      "epoch": 6.4376545390321445,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00017879067406175247,
      "loss": 5.2195,
      "mean_token_accuracy": 0.2276594579219818,
      "num_tokens": 135923524.0,
      "step": 72900
    },
    {
      "entropy": 6.120207977294922,
      "epoch": 6.4380960791239845,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00017876220764407558,
      "loss": 5.2786,
      "mean_token_accuracy": 0.2202603444457054,
      "num_tokens": 135933607.0,
      "step": 72905
    },
    {
      "entropy": 6.032462072372437,
      "epoch": 6.438537619215825,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001787337431116422,
      "loss": 5.241,
      "mean_token_accuracy": 0.22507500797510147,
      "num_tokens": 135943892.0,
      "step": 72910
    },
    {
      "entropy": 6.096452617645264,
      "epoch": 6.4389791593076655,
      "grad_norm": 1.5,
      "learning_rate": 0.00017870528046501005,
      "loss": 5.2757,
      "mean_token_accuracy": 0.22944428771734238,
      "num_tokens": 135953194.0,
      "step": 72915
    },
    {
      "entropy": 6.057935428619385,
      "epoch": 6.439420699399506,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001786768197047367,
      "loss": 5.1942,
      "mean_token_accuracy": 0.23242301493883133,
      "num_tokens": 135962211.0,
      "step": 72920
    },
    {
      "entropy": 6.085458660125733,
      "epoch": 6.439862239491346,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001786483608313797,
      "loss": 5.2436,
      "mean_token_accuracy": 0.22516833543777465,
      "num_tokens": 135972071.0,
      "step": 72925
    },
    {
      "entropy": 6.072771644592285,
      "epoch": 6.440303779583187,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00017861990384549658,
      "loss": 5.1919,
      "mean_token_accuracy": 0.22807253003120423,
      "num_tokens": 135980756.0,
      "step": 72930
    },
    {
      "entropy": 6.074890184402466,
      "epoch": 6.440745319675027,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001785914487476446,
      "loss": 5.246,
      "mean_token_accuracy": 0.2220063701272011,
      "num_tokens": 135989618.0,
      "step": 72935
    },
    {
      "entropy": 5.993072843551635,
      "epoch": 6.441186859766867,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017856299553838134,
      "loss": 5.1339,
      "mean_token_accuracy": 0.2292139232158661,
      "num_tokens": 135999447.0,
      "step": 72940
    },
    {
      "entropy": 6.034934091567993,
      "epoch": 6.441628399858708,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001785345442182642,
      "loss": 5.1812,
      "mean_token_accuracy": 0.23508769720792771,
      "num_tokens": 136007252.0,
      "step": 72945
    },
    {
      "entropy": 6.0317892074584964,
      "epoch": 6.442069939950548,
      "grad_norm": 1.484375,
      "learning_rate": 0.00017850609478785052,
      "loss": 5.2495,
      "mean_token_accuracy": 0.22953770756721498,
      "num_tokens": 136017034.0,
      "step": 72950
    },
    {
      "entropy": 6.01522536277771,
      "epoch": 6.442511480042388,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017847764724769743,
      "loss": 5.154,
      "mean_token_accuracy": 0.2354934573173523,
      "num_tokens": 136026417.0,
      "step": 72955
    },
    {
      "entropy": 6.022701072692871,
      "epoch": 6.442953020134228,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017844920159836236,
      "loss": 5.1163,
      "mean_token_accuracy": 0.23261548280715943,
      "num_tokens": 136035396.0,
      "step": 72960
    },
    {
      "entropy": 6.069689798355102,
      "epoch": 6.443394560226069,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00017842075784040256,
      "loss": 5.3057,
      "mean_token_accuracy": 0.22041568607091905,
      "num_tokens": 136043531.0,
      "step": 72965
    },
    {
      "entropy": 6.079697465896606,
      "epoch": 6.443836100317909,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017839231597437528,
      "loss": 5.2573,
      "mean_token_accuracy": 0.22357288151979446,
      "num_tokens": 136053411.0,
      "step": 72970
    },
    {
      "entropy": 6.10012412071228,
      "epoch": 6.444277640409749,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017836387600083747,
      "loss": 5.2969,
      "mean_token_accuracy": 0.2212923750281334,
      "num_tokens": 136063525.0,
      "step": 72975
    },
    {
      "entropy": 6.07824854850769,
      "epoch": 6.44471918050159,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017833543792034634,
      "loss": 5.2515,
      "mean_token_accuracy": 0.2282644659280777,
      "num_tokens": 136074586.0,
      "step": 72980
    },
    {
      "entropy": 6.0378098487854,
      "epoch": 6.44516072059343,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001783070017334591,
      "loss": 5.1247,
      "mean_token_accuracy": 0.23260941058397294,
      "num_tokens": 136084229.0,
      "step": 72985
    },
    {
      "entropy": 6.08452091217041,
      "epoch": 6.44560226068527,
      "grad_norm": 1.375,
      "learning_rate": 0.00017827856744073262,
      "loss": 5.2493,
      "mean_token_accuracy": 0.22269176691770554,
      "num_tokens": 136093873.0,
      "step": 72990
    },
    {
      "entropy": 6.047620058059692,
      "epoch": 6.446043800777111,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00017825013504272407,
      "loss": 5.1511,
      "mean_token_accuracy": 0.233565354347229,
      "num_tokens": 136104046.0,
      "step": 72995
    },
    {
      "entropy": 6.002977705001831,
      "epoch": 6.446485340868951,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001782217045399903,
      "loss": 5.1785,
      "mean_token_accuracy": 0.23265644609928132,
      "num_tokens": 136113189.0,
      "step": 73000
    },
    {
      "entropy": 5.979156732559204,
      "epoch": 6.446926880960791,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017819327593308827,
      "loss": 5.117,
      "mean_token_accuracy": 0.23390013575553895,
      "num_tokens": 136122528.0,
      "step": 73005
    },
    {
      "entropy": 6.066040754318237,
      "epoch": 6.447368421052632,
      "grad_norm": 1.3125,
      "learning_rate": 0.000178164849222575,
      "loss": 5.2606,
      "mean_token_accuracy": 0.22955990135669707,
      "num_tokens": 136132547.0,
      "step": 73010
    },
    {
      "entropy": 6.106465482711792,
      "epoch": 6.447809961144472,
      "grad_norm": 1.515625,
      "learning_rate": 0.00017813642440900722,
      "loss": 5.212,
      "mean_token_accuracy": 0.22762126922607423,
      "num_tokens": 136141008.0,
      "step": 73015
    },
    {
      "entropy": 6.006130838394165,
      "epoch": 6.448251501236312,
      "grad_norm": 1.546875,
      "learning_rate": 0.00017810800149294177,
      "loss": 5.1556,
      "mean_token_accuracy": 0.2369928315281868,
      "num_tokens": 136150335.0,
      "step": 73020
    },
    {
      "entropy": 6.028720903396606,
      "epoch": 6.448693041328153,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001780795804749355,
      "loss": 5.1979,
      "mean_token_accuracy": 0.23157595247030258,
      "num_tokens": 136159094.0,
      "step": 73025
    },
    {
      "entropy": 6.075287246704102,
      "epoch": 6.449134581419993,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017805116135554506,
      "loss": 5.2444,
      "mean_token_accuracy": 0.22427046447992324,
      "num_tokens": 136168856.0,
      "step": 73030
    },
    {
      "entropy": 6.033271026611328,
      "epoch": 6.449576121511833,
      "grad_norm": 1.515625,
      "learning_rate": 0.00017802274413532732,
      "loss": 5.2259,
      "mean_token_accuracy": 0.23029589354991914,
      "num_tokens": 136178625.0,
      "step": 73035
    },
    {
      "entropy": 6.02096700668335,
      "epoch": 6.450017661603674,
      "grad_norm": 1.515625,
      "learning_rate": 0.00017799432881483886,
      "loss": 5.1042,
      "mean_token_accuracy": 0.23523277789354324,
      "num_tokens": 136187194.0,
      "step": 73040
    },
    {
      "entropy": 6.094587087631226,
      "epoch": 6.450459201695514,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00017796591539463624,
      "loss": 5.281,
      "mean_token_accuracy": 0.2245142713189125,
      "num_tokens": 136196186.0,
      "step": 73045
    },
    {
      "entropy": 5.977595233917237,
      "epoch": 6.450900741787354,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001779375038752762,
      "loss": 5.12,
      "mean_token_accuracy": 0.24131109863519667,
      "num_tokens": 136205923.0,
      "step": 73050
    },
    {
      "entropy": 5.9579860210418705,
      "epoch": 6.451342281879195,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00017790909425731522,
      "loss": 5.1168,
      "mean_token_accuracy": 0.23213835656642914,
      "num_tokens": 136216030.0,
      "step": 73055
    },
    {
      "entropy": 6.003146171569824,
      "epoch": 6.451783821971035,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00017788068654131002,
      "loss": 5.1574,
      "mean_token_accuracy": 0.23309746384620667,
      "num_tokens": 136225474.0,
      "step": 73060
    },
    {
      "entropy": 6.074593162536621,
      "epoch": 6.452225362062875,
      "grad_norm": 1.625,
      "learning_rate": 0.0001778522807278168,
      "loss": 5.2748,
      "mean_token_accuracy": 0.22115447968244553,
      "num_tokens": 136234645.0,
      "step": 73065
    },
    {
      "entropy": 6.035504293441773,
      "epoch": 6.452666902154716,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00017782387681739215,
      "loss": 5.1881,
      "mean_token_accuracy": 0.23201334178447724,
      "num_tokens": 136243374.0,
      "step": 73070
    },
    {
      "entropy": 6.035064840316773,
      "epoch": 6.453108442246556,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001777954748105925,
      "loss": 5.1463,
      "mean_token_accuracy": 0.2391817182302475,
      "num_tokens": 136253414.0,
      "step": 73075
    },
    {
      "entropy": 5.992511224746704,
      "epoch": 6.453549982338396,
      "grad_norm": 1.5,
      "learning_rate": 0.00017776707470797427,
      "loss": 5.1678,
      "mean_token_accuracy": 0.23089695274829863,
      "num_tokens": 136262409.0,
      "step": 73080
    },
    {
      "entropy": 6.065598630905152,
      "epoch": 6.453991522430236,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017773867651009372,
      "loss": 5.2583,
      "mean_token_accuracy": 0.2170376420021057,
      "num_tokens": 136271501.0,
      "step": 73085
    },
    {
      "entropy": 6.023087501525879,
      "epoch": 6.454433062522077,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00017771028021750713,
      "loss": 5.195,
      "mean_token_accuracy": 0.23250950872898102,
      "num_tokens": 136280901.0,
      "step": 73090
    },
    {
      "entropy": 6.096102571487426,
      "epoch": 6.454874602613917,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017768188583077083,
      "loss": 5.2615,
      "mean_token_accuracy": 0.21971475034952165,
      "num_tokens": 136290089.0,
      "step": 73095
    },
    {
      "entropy": 6.051176691055298,
      "epoch": 6.455316142705758,
      "grad_norm": 1.5,
      "learning_rate": 0.00017765349335044112,
      "loss": 5.1587,
      "mean_token_accuracy": 0.23420050889253616,
      "num_tokens": 136299653.0,
      "step": 73100
    },
    {
      "entropy": 6.068878126144409,
      "epoch": 6.455757682797598,
      "grad_norm": 1.46875,
      "learning_rate": 0.00017762510277707406,
      "loss": 5.201,
      "mean_token_accuracy": 0.22303453087806702,
      "num_tokens": 136308414.0,
      "step": 73105
    },
    {
      "entropy": 5.934784555435181,
      "epoch": 6.456199222889438,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001775967141112258,
      "loss": 5.1873,
      "mean_token_accuracy": 0.23411233723163605,
      "num_tokens": 136319418.0,
      "step": 73110
    },
    {
      "entropy": 6.067537260055542,
      "epoch": 6.456640762981278,
      "grad_norm": 1.390625,
      "learning_rate": 0.00017756832735345258,
      "loss": 5.2271,
      "mean_token_accuracy": 0.2290120989084244,
      "num_tokens": 136329074.0,
      "step": 73115
    },
    {
      "entropy": 5.9754656791687015,
      "epoch": 6.457082303073119,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00017753994250431038,
      "loss": 5.0972,
      "mean_token_accuracy": 0.23549572080373765,
      "num_tokens": 136338334.0,
      "step": 73120
    },
    {
      "entropy": 6.088043069839477,
      "epoch": 6.457523843164959,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00017751155956435532,
      "loss": 5.3113,
      "mean_token_accuracy": 0.21903034001588823,
      "num_tokens": 136348090.0,
      "step": 73125
    },
    {
      "entropy": 6.043156671524048,
      "epoch": 6.457965383256799,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00017748317853414337,
      "loss": 5.1556,
      "mean_token_accuracy": 0.23175428956747054,
      "num_tokens": 136357696.0,
      "step": 73130
    },
    {
      "entropy": 6.037970542907715,
      "epoch": 6.45840692334864,
      "grad_norm": 1.515625,
      "learning_rate": 0.00017745479941423044,
      "loss": 5.1857,
      "mean_token_accuracy": 0.22642245143651962,
      "num_tokens": 136366408.0,
      "step": 73135
    },
    {
      "entropy": 6.015361309051514,
      "epoch": 6.45884846344048,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00017742642220517252,
      "loss": 5.1431,
      "mean_token_accuracy": 0.22613377273082733,
      "num_tokens": 136375128.0,
      "step": 73140
    },
    {
      "entropy": 6.063375806808471,
      "epoch": 6.45929000353232,
      "grad_norm": 1.296875,
      "learning_rate": 0.00017739804690752552,
      "loss": 5.2237,
      "mean_token_accuracy": 0.22316358387470245,
      "num_tokens": 136384703.0,
      "step": 73145
    },
    {
      "entropy": 6.029549789428711,
      "epoch": 6.459731543624161,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017736967352184529,
      "loss": 5.1864,
      "mean_token_accuracy": 0.22994934469461442,
      "num_tokens": 136394126.0,
      "step": 73150
    },
    {
      "entropy": 6.139075660705567,
      "epoch": 6.460173083716001,
      "grad_norm": 1.578125,
      "learning_rate": 0.0001773413020486876,
      "loss": 5.2417,
      "mean_token_accuracy": 0.22606227397918702,
      "num_tokens": 136402742.0,
      "step": 73155
    },
    {
      "entropy": 6.1108485698699955,
      "epoch": 6.4606146238078415,
      "grad_norm": 1.390625,
      "learning_rate": 0.00017731293248860825,
      "loss": 5.2325,
      "mean_token_accuracy": 0.22725572139024736,
      "num_tokens": 136412500.0,
      "step": 73160
    },
    {
      "entropy": 6.075052404403687,
      "epoch": 6.461056163899682,
      "grad_norm": 1.5625,
      "learning_rate": 0.00017728456484216305,
      "loss": 5.235,
      "mean_token_accuracy": 0.23014767915010453,
      "num_tokens": 136421449.0,
      "step": 73165
    },
    {
      "entropy": 6.056708192825317,
      "epoch": 6.4614977039915225,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00017725619910990758,
      "loss": 5.2376,
      "mean_token_accuracy": 0.22303118854761123,
      "num_tokens": 136429809.0,
      "step": 73170
    },
    {
      "entropy": 6.0315131664276125,
      "epoch": 6.4619392440833625,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001772278352923976,
      "loss": 5.1859,
      "mean_token_accuracy": 0.23100538551807404,
      "num_tokens": 136438894.0,
      "step": 73175
    },
    {
      "entropy": 5.987519598007202,
      "epoch": 6.4623807841752035,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001771994733901887,
      "loss": 5.155,
      "mean_token_accuracy": 0.23271118551492692,
      "num_tokens": 136448060.0,
      "step": 73180
    },
    {
      "entropy": 6.053967237472534,
      "epoch": 6.4628223242670435,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001771711134038365,
      "loss": 5.2027,
      "mean_token_accuracy": 0.2204044744372368,
      "num_tokens": 136457741.0,
      "step": 73185
    },
    {
      "entropy": 6.044374084472656,
      "epoch": 6.463263864358884,
      "grad_norm": 1.53125,
      "learning_rate": 0.00017714275533389667,
      "loss": 5.2351,
      "mean_token_accuracy": 0.23008078336715698,
      "num_tokens": 136467776.0,
      "step": 73190
    },
    {
      "entropy": 6.029800701141357,
      "epoch": 6.4637054044507245,
      "grad_norm": 1.484375,
      "learning_rate": 0.00017711439918092448,
      "loss": 5.1935,
      "mean_token_accuracy": 0.22310391813516617,
      "num_tokens": 136476864.0,
      "step": 73195
    },
    {
      "entropy": 6.01916470527649,
      "epoch": 6.4641469445425646,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017708604494547558,
      "loss": 5.142,
      "mean_token_accuracy": 0.2275850296020508,
      "num_tokens": 136486323.0,
      "step": 73200
    },
    {
      "entropy": 6.079987573623657,
      "epoch": 6.464588484634405,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017705769262810534,
      "loss": 5.2153,
      "mean_token_accuracy": 0.21771495938301086,
      "num_tokens": 136495887.0,
      "step": 73205
    },
    {
      "entropy": 5.992080163955689,
      "epoch": 6.4650300247262455,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001770293422293693,
      "loss": 5.0724,
      "mean_token_accuracy": 0.2418474793434143,
      "num_tokens": 136505044.0,
      "step": 73210
    },
    {
      "entropy": 6.025130271911621,
      "epoch": 6.465471564818086,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001770009937498226,
      "loss": 5.1621,
      "mean_token_accuracy": 0.23315207809209823,
      "num_tokens": 136515248.0,
      "step": 73215
    },
    {
      "entropy": 6.032058477401733,
      "epoch": 6.465913104909926,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00017697264719002077,
      "loss": 5.1537,
      "mean_token_accuracy": 0.23747752159833907,
      "num_tokens": 136525032.0,
      "step": 73220
    },
    {
      "entropy": 6.0087646484375,
      "epoch": 6.466354645001767,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00017694430255051896,
      "loss": 5.2468,
      "mean_token_accuracy": 0.2264498472213745,
      "num_tokens": 136534718.0,
      "step": 73225
    },
    {
      "entropy": 5.975346660614013,
      "epoch": 6.466796185093607,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017691595983187263,
      "loss": 5.126,
      "mean_token_accuracy": 0.23357695937156678,
      "num_tokens": 136544946.0,
      "step": 73230
    },
    {
      "entropy": 6.095744276046753,
      "epoch": 6.467237725185447,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00017688761903463673,
      "loss": 5.2772,
      "mean_token_accuracy": 0.22278822511434554,
      "num_tokens": 136553645.0,
      "step": 73235
    },
    {
      "entropy": 6.086370992660522,
      "epoch": 6.467679265277287,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001768592801593666,
      "loss": 5.2577,
      "mean_token_accuracy": 0.229888354241848,
      "num_tokens": 136562930.0,
      "step": 73240
    },
    {
      "entropy": 5.963274574279785,
      "epoch": 6.468120805369128,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00017683094320661736,
      "loss": 5.114,
      "mean_token_accuracy": 0.23181760013103486,
      "num_tokens": 136572330.0,
      "step": 73245
    },
    {
      "entropy": 6.018288040161133,
      "epoch": 6.468562345460968,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00017680260817694415,
      "loss": 5.1605,
      "mean_token_accuracy": 0.22931186109781265,
      "num_tokens": 136581796.0,
      "step": 73250
    },
    {
      "entropy": 6.076360988616943,
      "epoch": 6.469003885552808,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017677427507090205,
      "loss": 5.2183,
      "mean_token_accuracy": 0.2252792552113533,
      "num_tokens": 136591349.0,
      "step": 73255
    },
    {
      "entropy": 6.084792852401733,
      "epoch": 6.469445425644649,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001767459438890459,
      "loss": 5.1772,
      "mean_token_accuracy": 0.22431631982326508,
      "num_tokens": 136601290.0,
      "step": 73260
    },
    {
      "entropy": 6.072015476226807,
      "epoch": 6.469886965736489,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017671761463193092,
      "loss": 5.2112,
      "mean_token_accuracy": 0.21841478496789932,
      "num_tokens": 136610769.0,
      "step": 73265
    },
    {
      "entropy": 6.074857473373413,
      "epoch": 6.470328505828329,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0001766892873001119,
      "loss": 5.2472,
      "mean_token_accuracy": 0.22558358907699586,
      "num_tokens": 136620536.0,
      "step": 73270
    },
    {
      "entropy": 6.03356671333313,
      "epoch": 6.47077004592017,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017666096189414393,
      "loss": 5.222,
      "mean_token_accuracy": 0.22517517656087876,
      "num_tokens": 136630536.0,
      "step": 73275
    },
    {
      "entropy": 5.998409080505371,
      "epoch": 6.47121158601201,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00017663263841458172,
      "loss": 5.0852,
      "mean_token_accuracy": 0.23846740573644637,
      "num_tokens": 136639445.0,
      "step": 73280
    },
    {
      "entropy": 6.101090288162231,
      "epoch": 6.47165312610385,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001766043168619802,
      "loss": 5.2749,
      "mean_token_accuracy": 0.22272457629442216,
      "num_tokens": 136648920.0,
      "step": 73285
    },
    {
      "entropy": 6.0838501930236815,
      "epoch": 6.472094666195691,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001765759972368941,
      "loss": 5.2297,
      "mean_token_accuracy": 0.22453545033931732,
      "num_tokens": 136657395.0,
      "step": 73290
    },
    {
      "entropy": 5.990595960617066,
      "epoch": 6.472536206287531,
      "grad_norm": 1.515625,
      "learning_rate": 0.00017654767953987838,
      "loss": 5.1822,
      "mean_token_accuracy": 0.227402026951313,
      "num_tokens": 136665987.0,
      "step": 73295
    },
    {
      "entropy": 6.048631811141968,
      "epoch": 6.472977746379371,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00017651936377148752,
      "loss": 5.2153,
      "mean_token_accuracy": 0.2250726193189621,
      "num_tokens": 136675047.0,
      "step": 73300
    },
    {
      "entropy": 6.084477090835572,
      "epoch": 6.473419286471212,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00017649104993227628,
      "loss": 5.2426,
      "mean_token_accuracy": 0.2270253822207451,
      "num_tokens": 136684955.0,
      "step": 73305
    },
    {
      "entropy": 6.114155387878418,
      "epoch": 6.473860826563052,
      "grad_norm": 1.59375,
      "learning_rate": 0.00017646273802279937,
      "loss": 5.269,
      "mean_token_accuracy": 0.22370671182870866,
      "num_tokens": 136694542.0,
      "step": 73310
    },
    {
      "entropy": 6.058563566207885,
      "epoch": 6.474302366654892,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017643442804361138,
      "loss": 5.2203,
      "mean_token_accuracy": 0.23274377584457398,
      "num_tokens": 136704505.0,
      "step": 73315
    },
    {
      "entropy": 6.017111444473267,
      "epoch": 6.474743906746733,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017640611999526697,
      "loss": 5.2144,
      "mean_token_accuracy": 0.2344801113009453,
      "num_tokens": 136715075.0,
      "step": 73320
    },
    {
      "entropy": 6.0997484683990475,
      "epoch": 6.475185446838573,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017637781387832053,
      "loss": 5.3048,
      "mean_token_accuracy": 0.22146795988082885,
      "num_tokens": 136724884.0,
      "step": 73325
    },
    {
      "entropy": 6.048447132110596,
      "epoch": 6.475626986930413,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001763495096933266,
      "loss": 5.1761,
      "mean_token_accuracy": 0.2327228680253029,
      "num_tokens": 136734677.0,
      "step": 73330
    },
    {
      "entropy": 6.0585675716400145,
      "epoch": 6.476068527022254,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001763212074408397,
      "loss": 5.2056,
      "mean_token_accuracy": 0.23067445904016495,
      "num_tokens": 136744184.0,
      "step": 73335
    },
    {
      "entropy": 5.985987424850464,
      "epoch": 6.476510067114094,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017629290712141426,
      "loss": 5.1442,
      "mean_token_accuracy": 0.23965974301099777,
      "num_tokens": 136753930.0,
      "step": 73340
    },
    {
      "entropy": 5.953957986831665,
      "epoch": 6.476951607205934,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017626460873560463,
      "loss": 5.0796,
      "mean_token_accuracy": 0.2368362307548523,
      "num_tokens": 136763236.0,
      "step": 73345
    },
    {
      "entropy": 6.001727104187012,
      "epoch": 6.477393147297775,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001762363122839651,
      "loss": 5.1499,
      "mean_token_accuracy": 0.22609837353229523,
      "num_tokens": 136772993.0,
      "step": 73350
    },
    {
      "entropy": 6.067659854888916,
      "epoch": 6.477834687389615,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017620801776705003,
      "loss": 5.2512,
      "mean_token_accuracy": 0.21412401646375656,
      "num_tokens": 136783172.0,
      "step": 73355
    },
    {
      "entropy": 5.968433904647827,
      "epoch": 6.478276227481455,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001761797251854138,
      "loss": 5.1156,
      "mean_token_accuracy": 0.23860300928354264,
      "num_tokens": 136792754.0,
      "step": 73360
    },
    {
      "entropy": 6.041513586044312,
      "epoch": 6.478717767573295,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017615143453961053,
      "loss": 5.1886,
      "mean_token_accuracy": 0.2306237354874611,
      "num_tokens": 136802834.0,
      "step": 73365
    },
    {
      "entropy": 6.069870376586914,
      "epoch": 6.479159307665136,
      "grad_norm": 1.453125,
      "learning_rate": 0.00017612314583019434,
      "loss": 5.1608,
      "mean_token_accuracy": 0.23091342002153398,
      "num_tokens": 136811449.0,
      "step": 73370
    },
    {
      "entropy": 6.02090048789978,
      "epoch": 6.479600847756976,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017609485905771956,
      "loss": 5.1539,
      "mean_token_accuracy": 0.22992697060108186,
      "num_tokens": 136820317.0,
      "step": 73375
    },
    {
      "entropy": 6.060851526260376,
      "epoch": 6.480042387848817,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001760665742227402,
      "loss": 5.2195,
      "mean_token_accuracy": 0.22020160108804704,
      "num_tokens": 136830022.0,
      "step": 73380
    },
    {
      "entropy": 6.035616111755371,
      "epoch": 6.480483927940657,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001760382913258105,
      "loss": 5.1778,
      "mean_token_accuracy": 0.22761378884315492,
      "num_tokens": 136839871.0,
      "step": 73385
    },
    {
      "entropy": 6.006584930419922,
      "epoch": 6.480925468032497,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00017601001036748426,
      "loss": 5.1659,
      "mean_token_accuracy": 0.2302108660340309,
      "num_tokens": 136848988.0,
      "step": 73390
    },
    {
      "entropy": 5.965969896316528,
      "epoch": 6.481367008124337,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00017598173134831564,
      "loss": 5.1634,
      "mean_token_accuracy": 0.23370348960161208,
      "num_tokens": 136858430.0,
      "step": 73395
    },
    {
      "entropy": 5.988442039489746,
      "epoch": 6.481808548216178,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017595345426885868,
      "loss": 5.18,
      "mean_token_accuracy": 0.22939983904361724,
      "num_tokens": 136867776.0,
      "step": 73400
    },
    {
      "entropy": 6.033003902435302,
      "epoch": 6.482250088308018,
      "grad_norm": 1.53125,
      "learning_rate": 0.00017592517912966716,
      "loss": 5.1864,
      "mean_token_accuracy": 0.22534752488136292,
      "num_tokens": 136876815.0,
      "step": 73405
    },
    {
      "entropy": 5.991541242599487,
      "epoch": 6.482691628399858,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017589690593129503,
      "loss": 5.1091,
      "mean_token_accuracy": 0.2340376541018486,
      "num_tokens": 136884984.0,
      "step": 73410
    },
    {
      "entropy": 6.0422357559204105,
      "epoch": 6.483133168491699,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017586863467429614,
      "loss": 5.2068,
      "mean_token_accuracy": 0.2335924595594406,
      "num_tokens": 136893910.0,
      "step": 73415
    },
    {
      "entropy": 6.043364191055298,
      "epoch": 6.483574708583539,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017584036535922432,
      "loss": 5.1739,
      "mean_token_accuracy": 0.23066525906324387,
      "num_tokens": 136903205.0,
      "step": 73420
    },
    {
      "entropy": 6.136893701553345,
      "epoch": 6.484016248675379,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001758120979866334,
      "loss": 5.3048,
      "mean_token_accuracy": 0.21892004311084748,
      "num_tokens": 136912250.0,
      "step": 73425
    },
    {
      "entropy": 6.006074094772339,
      "epoch": 6.48445778876722,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00017578383255707704,
      "loss": 5.1411,
      "mean_token_accuracy": 0.22984661012887955,
      "num_tokens": 136920713.0,
      "step": 73430
    },
    {
      "entropy": 6.0203711032867435,
      "epoch": 6.48489932885906,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001757555690711089,
      "loss": 5.2018,
      "mean_token_accuracy": 0.2271927461028099,
      "num_tokens": 136930435.0,
      "step": 73435
    },
    {
      "entropy": 6.086066102981567,
      "epoch": 6.4853408689509005,
      "grad_norm": 1.515625,
      "learning_rate": 0.00017572730752928278,
      "loss": 5.2881,
      "mean_token_accuracy": 0.2231282413005829,
      "num_tokens": 136939937.0,
      "step": 73440
    },
    {
      "entropy": 6.046640872955322,
      "epoch": 6.485782409042741,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001756990479321522,
      "loss": 5.1931,
      "mean_token_accuracy": 0.23101030588150023,
      "num_tokens": 136949549.0,
      "step": 73445
    },
    {
      "entropy": 6.040156745910645,
      "epoch": 6.4862239491345814,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00017567079028027094,
      "loss": 5.264,
      "mean_token_accuracy": 0.22213930636644363,
      "num_tokens": 136959481.0,
      "step": 73450
    },
    {
      "entropy": 6.009092330932617,
      "epoch": 6.4866654892264215,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017564253457419228,
      "loss": 5.1827,
      "mean_token_accuracy": 0.22827948331832887,
      "num_tokens": 136969437.0,
      "step": 73455
    },
    {
      "entropy": 6.0895552158355715,
      "epoch": 6.487107029318262,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001756142808144699,
      "loss": 5.263,
      "mean_token_accuracy": 0.2235394909977913,
      "num_tokens": 136979546.0,
      "step": 73460
    },
    {
      "entropy": 6.074250316619873,
      "epoch": 6.4875485694101025,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017558602900165715,
      "loss": 5.2253,
      "mean_token_accuracy": 0.22974590212106705,
      "num_tokens": 136989408.0,
      "step": 73465
    },
    {
      "entropy": 6.078907585144043,
      "epoch": 6.4879901095019425,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001755577791363077,
      "loss": 5.2187,
      "mean_token_accuracy": 0.2296836256980896,
      "num_tokens": 136999735.0,
      "step": 73470
    },
    {
      "entropy": 6.019113779067993,
      "epoch": 6.4884316495937835,
      "grad_norm": 1.28125,
      "learning_rate": 0.00017552953121897469,
      "loss": 5.1619,
      "mean_token_accuracy": 0.2250885099172592,
      "num_tokens": 137009429.0,
      "step": 73475
    },
    {
      "entropy": 6.062746000289917,
      "epoch": 6.4888731896856235,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00017550128525021158,
      "loss": 5.2404,
      "mean_token_accuracy": 0.22507989704608916,
      "num_tokens": 137018667.0,
      "step": 73480
    },
    {
      "entropy": 6.095451402664184,
      "epoch": 6.489314729777464,
      "grad_norm": 1.375,
      "learning_rate": 0.0001754730412305718,
      "loss": 5.2353,
      "mean_token_accuracy": 0.22623298913240433,
      "num_tokens": 137028410.0,
      "step": 73485
    },
    {
      "entropy": 6.060564184188843,
      "epoch": 6.4897562698693045,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001754447991606085,
      "loss": 5.1962,
      "mean_token_accuracy": 0.221963731944561,
      "num_tokens": 137037819.0,
      "step": 73490
    },
    {
      "entropy": 6.054397201538086,
      "epoch": 6.490197809961145,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017541655904087494,
      "loss": 5.2365,
      "mean_token_accuracy": 0.22797078341245652,
      "num_tokens": 137047697.0,
      "step": 73495
    },
    {
      "entropy": 6.006433343887329,
      "epoch": 6.490639350052985,
      "grad_norm": 1.625,
      "learning_rate": 0.00017538832087192436,
      "loss": 5.1706,
      "mean_token_accuracy": 0.22834858894348145,
      "num_tokens": 137056365.0,
      "step": 73500
    },
    {
      "entropy": 6.060852909088135,
      "epoch": 6.491080890144826,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001753600846543099,
      "loss": 5.2464,
      "mean_token_accuracy": 0.22611636966466903,
      "num_tokens": 137065740.0,
      "step": 73505
    },
    {
      "entropy": 5.967998743057251,
      "epoch": 6.491522430236666,
      "grad_norm": 1.453125,
      "learning_rate": 0.00017533185038858472,
      "loss": 5.12,
      "mean_token_accuracy": 0.24312337338924409,
      "num_tokens": 137074737.0,
      "step": 73510
    },
    {
      "entropy": 6.086010837554932,
      "epoch": 6.491963970328506,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017530361807530198,
      "loss": 5.2706,
      "mean_token_accuracy": 0.21696659922599792,
      "num_tokens": 137085265.0,
      "step": 73515
    },
    {
      "entropy": 6.00913438796997,
      "epoch": 6.492405510420346,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001752753877150146,
      "loss": 5.1206,
      "mean_token_accuracy": 0.23486064225435258,
      "num_tokens": 137094649.0,
      "step": 73520
    },
    {
      "entropy": 6.08743839263916,
      "epoch": 6.492847050512187,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017524715930827578,
      "loss": 5.3032,
      "mean_token_accuracy": 0.21491850316524505,
      "num_tokens": 137104789.0,
      "step": 73525
    },
    {
      "entropy": 6.054756450653076,
      "epoch": 6.493288590604027,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001752189328556381,
      "loss": 5.1909,
      "mean_token_accuracy": 0.2293651208281517,
      "num_tokens": 137113502.0,
      "step": 73530
    },
    {
      "entropy": 6.066201829910279,
      "epoch": 6.493730130695867,
      "grad_norm": 1.375,
      "learning_rate": 0.00017519070835765504,
      "loss": 5.2242,
      "mean_token_accuracy": 0.23163980096578599,
      "num_tokens": 137124545.0,
      "step": 73535
    },
    {
      "entropy": 6.089739847183227,
      "epoch": 6.494171670787708,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0001751624858148791,
      "loss": 5.1962,
      "mean_token_accuracy": 0.2311623439192772,
      "num_tokens": 137133106.0,
      "step": 73540
    },
    {
      "entropy": 6.035658931732177,
      "epoch": 6.494613210879548,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0001751342652278633,
      "loss": 5.1277,
      "mean_token_accuracy": 0.23613449335098266,
      "num_tokens": 137141433.0,
      "step": 73545
    },
    {
      "entropy": 6.105919265747071,
      "epoch": 6.495054750971388,
      "grad_norm": 1.515625,
      "learning_rate": 0.00017510604659716044,
      "loss": 5.2828,
      "mean_token_accuracy": 0.21105647832155228,
      "num_tokens": 137150480.0,
      "step": 73550
    },
    {
      "entropy": 6.02418360710144,
      "epoch": 6.495496291063229,
      "grad_norm": 1.296875,
      "learning_rate": 0.00017507782992332333,
      "loss": 5.2259,
      "mean_token_accuracy": 0.2254432260990143,
      "num_tokens": 137159980.0,
      "step": 73555
    },
    {
      "entropy": 6.0170066356658936,
      "epoch": 6.495937831155069,
      "grad_norm": 1.46875,
      "learning_rate": 0.00017504961520690483,
      "loss": 5.1666,
      "mean_token_accuracy": 0.2301329866051674,
      "num_tokens": 137169005.0,
      "step": 73560
    },
    {
      "entropy": 6.015260028839111,
      "epoch": 6.496379371246909,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017502140244845738,
      "loss": 5.19,
      "mean_token_accuracy": 0.22830753177404403,
      "num_tokens": 137178684.0,
      "step": 73565
    },
    {
      "entropy": 6.095593595504761,
      "epoch": 6.49682091133875,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001749931916485338,
      "loss": 5.2512,
      "mean_token_accuracy": 0.22537850141525267,
      "num_tokens": 137187608.0,
      "step": 73570
    },
    {
      "entropy": 6.1165729522705075,
      "epoch": 6.49726245143059,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017496498280768672,
      "loss": 5.2519,
      "mean_token_accuracy": 0.22198622524738312,
      "num_tokens": 137196235.0,
      "step": 73575
    },
    {
      "entropy": 6.0338586330413815,
      "epoch": 6.49770399152243,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017493677592646885,
      "loss": 5.2046,
      "mean_token_accuracy": 0.225533065199852,
      "num_tokens": 137204793.0,
      "step": 73580
    },
    {
      "entropy": 5.964408588409424,
      "epoch": 6.498145531614271,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017490857100543248,
      "loss": 5.1241,
      "mean_token_accuracy": 0.2288973033428192,
      "num_tokens": 137215496.0,
      "step": 73585
    },
    {
      "entropy": 6.008149194717407,
      "epoch": 6.498587071706111,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001748803680451304,
      "loss": 5.172,
      "mean_token_accuracy": 0.23424824476242065,
      "num_tokens": 137224313.0,
      "step": 73590
    },
    {
      "entropy": 6.081565713882446,
      "epoch": 6.499028611797951,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001748521670461148,
      "loss": 5.2278,
      "mean_token_accuracy": 0.22606250941753386,
      "num_tokens": 137232666.0,
      "step": 73595
    },
    {
      "entropy": 6.056109952926636,
      "epoch": 6.499470151889792,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017482396800893847,
      "loss": 5.1942,
      "mean_token_accuracy": 0.2244621619582176,
      "num_tokens": 137242219.0,
      "step": 73600
    },
    {
      "entropy": 6.1284407615661625,
      "epoch": 6.499911691981632,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017479577093415356,
      "loss": 5.2711,
      "mean_token_accuracy": 0.21427861899137496,
      "num_tokens": 137250804.0,
      "step": 73605
    },
    {
      "entropy": 6.109929513931275,
      "epoch": 6.500353232073472,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017476757582231246,
      "loss": 5.3211,
      "mean_token_accuracy": 0.22161277532577514,
      "num_tokens": 137261023.0,
      "step": 73610
    },
    {
      "entropy": 6.143712759017944,
      "epoch": 6.500794772165313,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017473938267396756,
      "loss": 5.2592,
      "mean_token_accuracy": 0.23034058809280394,
      "num_tokens": 137270041.0,
      "step": 73615
    },
    {
      "entropy": 6.050892782211304,
      "epoch": 6.501236312257153,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017471119148967113,
      "loss": 5.2155,
      "mean_token_accuracy": 0.2227567493915558,
      "num_tokens": 137279522.0,
      "step": 73620
    },
    {
      "entropy": 6.0276933193206785,
      "epoch": 6.501677852348993,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017468300226997554,
      "loss": 5.1677,
      "mean_token_accuracy": 0.22665615677833556,
      "num_tokens": 137289677.0,
      "step": 73625
    },
    {
      "entropy": 6.046933650970459,
      "epoch": 6.502119392440834,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001746548150154328,
      "loss": 5.1147,
      "mean_token_accuracy": 0.23763570934534073,
      "num_tokens": 137299436.0,
      "step": 73630
    },
    {
      "entropy": 6.051575660705566,
      "epoch": 6.502560932532674,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017462662972659513,
      "loss": 5.2473,
      "mean_token_accuracy": 0.2247756913304329,
      "num_tokens": 137308900.0,
      "step": 73635
    },
    {
      "entropy": 6.023081922531128,
      "epoch": 6.503002472624514,
      "grad_norm": 1.375,
      "learning_rate": 0.0001745984464040147,
      "loss": 5.1884,
      "mean_token_accuracy": 0.226188588142395,
      "num_tokens": 137318579.0,
      "step": 73640
    },
    {
      "entropy": 6.112138032913208,
      "epoch": 6.503444012716354,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001745702650482437,
      "loss": 5.2465,
      "mean_token_accuracy": 0.2196335643529892,
      "num_tokens": 137328122.0,
      "step": 73645
    },
    {
      "entropy": 6.0794390678405765,
      "epoch": 6.503885552808195,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000174542085659834,
      "loss": 5.2339,
      "mean_token_accuracy": 0.22607847154140473,
      "num_tokens": 137337187.0,
      "step": 73650
    },
    {
      "entropy": 6.013528060913086,
      "epoch": 6.504327092900035,
      "grad_norm": 1.59375,
      "learning_rate": 0.00017451390823933782,
      "loss": 5.1799,
      "mean_token_accuracy": 0.23196058571338654,
      "num_tokens": 137345697.0,
      "step": 73655
    },
    {
      "entropy": 6.000797843933105,
      "epoch": 6.504768632991876,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017448573278730685,
      "loss": 5.1912,
      "mean_token_accuracy": 0.2271404355764389,
      "num_tokens": 137355507.0,
      "step": 73660
    },
    {
      "entropy": 6.013786697387696,
      "epoch": 6.505210173083716,
      "grad_norm": 1.5625,
      "learning_rate": 0.00017445755930429332,
      "loss": 5.1851,
      "mean_token_accuracy": 0.23644232451915742,
      "num_tokens": 137365492.0,
      "step": 73665
    },
    {
      "entropy": 6.002637195587158,
      "epoch": 6.505651713175556,
      "grad_norm": 1.40625,
      "learning_rate": 0.000174429387790849,
      "loss": 5.1098,
      "mean_token_accuracy": 0.2346872568130493,
      "num_tokens": 137375399.0,
      "step": 73670
    },
    {
      "entropy": 5.894897174835205,
      "epoch": 6.506093253267396,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017440121824752574,
      "loss": 5.0661,
      "mean_token_accuracy": 0.2398289993405342,
      "num_tokens": 137384590.0,
      "step": 73675
    },
    {
      "entropy": 6.007190561294555,
      "epoch": 6.506534793359237,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017437305067487548,
      "loss": 5.1759,
      "mean_token_accuracy": 0.23135334253311157,
      "num_tokens": 137394372.0,
      "step": 73680
    },
    {
      "entropy": 6.11349515914917,
      "epoch": 6.506976333451077,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00017434488507344983,
      "loss": 5.2509,
      "mean_token_accuracy": 0.2220567673444748,
      "num_tokens": 137403821.0,
      "step": 73685
    },
    {
      "entropy": 5.983781242370606,
      "epoch": 6.507417873542917,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00017431672144380077,
      "loss": 5.0859,
      "mean_token_accuracy": 0.2361291155219078,
      "num_tokens": 137412626.0,
      "step": 73690
    },
    {
      "entropy": 6.083795547485352,
      "epoch": 6.507859413634758,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001742885597864798,
      "loss": 5.2848,
      "mean_token_accuracy": 0.22352520674467086,
      "num_tokens": 137422714.0,
      "step": 73695
    },
    {
      "entropy": 6.063307857513427,
      "epoch": 6.508300953726598,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001742604001020386,
      "loss": 5.2523,
      "mean_token_accuracy": 0.2265275999903679,
      "num_tokens": 137432439.0,
      "step": 73700
    },
    {
      "entropy": 6.119938468933105,
      "epoch": 6.508742493818438,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00017423224239102893,
      "loss": 5.2844,
      "mean_token_accuracy": 0.21455201953649522,
      "num_tokens": 137441635.0,
      "step": 73705
    },
    {
      "entropy": 6.0603433609008786,
      "epoch": 6.509184033910279,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017420408665400237,
      "loss": 5.1938,
      "mean_token_accuracy": 0.22681653499603271,
      "num_tokens": 137450348.0,
      "step": 73710
    },
    {
      "entropy": 6.082636070251465,
      "epoch": 6.509625574002119,
      "grad_norm": 1.390625,
      "learning_rate": 0.00017417593289151037,
      "loss": 5.2054,
      "mean_token_accuracy": 0.220847187936306,
      "num_tokens": 137460037.0,
      "step": 73715
    },
    {
      "entropy": 6.122673654556275,
      "epoch": 6.510067114093959,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017414778110410456,
      "loss": 5.327,
      "mean_token_accuracy": 0.21617611050605773,
      "num_tokens": 137469828.0,
      "step": 73720
    },
    {
      "entropy": 6.023561286926269,
      "epoch": 6.5105086541858,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00017411963129233623,
      "loss": 5.1238,
      "mean_token_accuracy": 0.23821661472320557,
      "num_tokens": 137478716.0,
      "step": 73725
    },
    {
      "entropy": 6.078034830093384,
      "epoch": 6.51095019427764,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00017409148345675707,
      "loss": 5.2293,
      "mean_token_accuracy": 0.22603893876075745,
      "num_tokens": 137488377.0,
      "step": 73730
    },
    {
      "entropy": 6.074365425109863,
      "epoch": 6.5113917343694805,
      "grad_norm": 1.5,
      "learning_rate": 0.00017406333759791833,
      "loss": 5.1836,
      "mean_token_accuracy": 0.23222060054540633,
      "num_tokens": 137497081.0,
      "step": 73735
    },
    {
      "entropy": 5.969721174240112,
      "epoch": 6.511833274461321,
      "grad_norm": 1.296875,
      "learning_rate": 0.00017403519371637138,
      "loss": 5.185,
      "mean_token_accuracy": 0.2302725210785866,
      "num_tokens": 137507125.0,
      "step": 73740
    },
    {
      "entropy": 6.043236351013183,
      "epoch": 6.5122748145531615,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017400705181266758,
      "loss": 5.1926,
      "mean_token_accuracy": 0.23356905579566956,
      "num_tokens": 137516498.0,
      "step": 73745
    },
    {
      "entropy": 6.118697738647461,
      "epoch": 6.5127163546450015,
      "grad_norm": 1.546875,
      "learning_rate": 0.00017397891188735814,
      "loss": 5.241,
      "mean_token_accuracy": 0.22681759893894196,
      "num_tokens": 137525653.0,
      "step": 73750
    },
    {
      "entropy": 6.070454025268555,
      "epoch": 6.5131578947368425,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017395077394099446,
      "loss": 5.2195,
      "mean_token_accuracy": 0.22353679239749907,
      "num_tokens": 137535006.0,
      "step": 73755
    },
    {
      "entropy": 6.031857967376709,
      "epoch": 6.5135994348286825,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001739226379741276,
      "loss": 5.1852,
      "mean_token_accuracy": 0.23372479677200317,
      "num_tokens": 137543941.0,
      "step": 73760
    },
    {
      "entropy": 6.102281904220581,
      "epoch": 6.514040974920523,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017389450398730885,
      "loss": 5.2939,
      "mean_token_accuracy": 0.21943538784980773,
      "num_tokens": 137553498.0,
      "step": 73765
    },
    {
      "entropy": 6.038562917709351,
      "epoch": 6.514482515012363,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001738663719810891,
      "loss": 5.2103,
      "mean_token_accuracy": 0.22244756817817687,
      "num_tokens": 137563312.0,
      "step": 73770
    },
    {
      "entropy": 6.01675763130188,
      "epoch": 6.514924055104204,
      "grad_norm": 1.515625,
      "learning_rate": 0.00017383824195601978,
      "loss": 5.1346,
      "mean_token_accuracy": 0.2326972007751465,
      "num_tokens": 137572109.0,
      "step": 73775
    },
    {
      "entropy": 6.073830413818359,
      "epoch": 6.515365595196044,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017381011391265168,
      "loss": 5.2013,
      "mean_token_accuracy": 0.23145166635513306,
      "num_tokens": 137581585.0,
      "step": 73780
    },
    {
      "entropy": 6.029830741882324,
      "epoch": 6.515807135287885,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00017378198785153592,
      "loss": 5.1497,
      "mean_token_accuracy": 0.23201336413621904,
      "num_tokens": 137589942.0,
      "step": 73785
    },
    {
      "entropy": 6.087540340423584,
      "epoch": 6.516248675379725,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00017375386377322332,
      "loss": 5.2772,
      "mean_token_accuracy": 0.2186218500137329,
      "num_tokens": 137598883.0,
      "step": 73790
    },
    {
      "entropy": 5.997995090484619,
      "epoch": 6.516690215471565,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017372574167826517,
      "loss": 5.1714,
      "mean_token_accuracy": 0.23207465559244156,
      "num_tokens": 137608275.0,
      "step": 73795
    },
    {
      "entropy": 6.061982774734497,
      "epoch": 6.517131755563405,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00017369762156721203,
      "loss": 5.2417,
      "mean_token_accuracy": 0.21731424629688262,
      "num_tokens": 137616869.0,
      "step": 73800
    },
    {
      "entropy": 6.079756212234497,
      "epoch": 6.517573295655246,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00017366950344061482,
      "loss": 5.2119,
      "mean_token_accuracy": 0.22609217315912247,
      "num_tokens": 137625778.0,
      "step": 73805
    },
    {
      "entropy": 6.089027786254883,
      "epoch": 6.518014835747086,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017364138729902446,
      "loss": 5.2202,
      "mean_token_accuracy": 0.2258329674601555,
      "num_tokens": 137636163.0,
      "step": 73810
    },
    {
      "entropy": 5.930283069610596,
      "epoch": 6.518456375838926,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017361327314299168,
      "loss": 5.0097,
      "mean_token_accuracy": 0.2418830245733261,
      "num_tokens": 137645556.0,
      "step": 73815
    },
    {
      "entropy": 6.086055612564087,
      "epoch": 6.518897915930767,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00017358516097306726,
      "loss": 5.2225,
      "mean_token_accuracy": 0.2264561265707016,
      "num_tokens": 137655133.0,
      "step": 73820
    },
    {
      "entropy": 6.0369726657867435,
      "epoch": 6.519339456022607,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00017355705078980183,
      "loss": 5.2515,
      "mean_token_accuracy": 0.2262669175863266,
      "num_tokens": 137664112.0,
      "step": 73825
    },
    {
      "entropy": 5.964975166320801,
      "epoch": 6.519780996114447,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017352894259374613,
      "loss": 5.1858,
      "mean_token_accuracy": 0.22578066736459732,
      "num_tokens": 137674829.0,
      "step": 73830
    },
    {
      "entropy": 6.093663215637207,
      "epoch": 6.520222536206288,
      "grad_norm": 1.625,
      "learning_rate": 0.00017350083638545058,
      "loss": 5.322,
      "mean_token_accuracy": 0.21495845466852187,
      "num_tokens": 137684759.0,
      "step": 73835
    },
    {
      "entropy": 6.010812902450562,
      "epoch": 6.520664076298128,
      "grad_norm": 1.515625,
      "learning_rate": 0.00017347273216546604,
      "loss": 5.1713,
      "mean_token_accuracy": 0.23656477481126786,
      "num_tokens": 137693828.0,
      "step": 73840
    },
    {
      "entropy": 6.090016508102417,
      "epoch": 6.521105616389968,
      "grad_norm": 1.5,
      "learning_rate": 0.0001734446299343429,
      "loss": 5.2616,
      "mean_token_accuracy": 0.21784615218639375,
      "num_tokens": 137703592.0,
      "step": 73845
    },
    {
      "entropy": 6.07504677772522,
      "epoch": 6.521547156481809,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00017341652969263176,
      "loss": 5.254,
      "mean_token_accuracy": 0.21922514140605925,
      "num_tokens": 137713229.0,
      "step": 73850
    },
    {
      "entropy": 5.992056131362915,
      "epoch": 6.521988696573649,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001733884314408829,
      "loss": 5.1715,
      "mean_token_accuracy": 0.2275712639093399,
      "num_tokens": 137723393.0,
      "step": 73855
    },
    {
      "entropy": 6.101328611373901,
      "epoch": 6.522430236665489,
      "grad_norm": 1.375,
      "learning_rate": 0.00017336033517964707,
      "loss": 5.2124,
      "mean_token_accuracy": 0.22652495056390762,
      "num_tokens": 137732784.0,
      "step": 73860
    },
    {
      "entropy": 6.161338996887207,
      "epoch": 6.52287177675733,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001733322409094743,
      "loss": 5.2879,
      "mean_token_accuracy": 0.21493514776229858,
      "num_tokens": 137741649.0,
      "step": 73865
    },
    {
      "entropy": 6.039220237731934,
      "epoch": 6.52331331684917,
      "grad_norm": 1.46875,
      "learning_rate": 0.00017330414863091516,
      "loss": 5.2216,
      "mean_token_accuracy": 0.2275017663836479,
      "num_tokens": 137750948.0,
      "step": 73870
    },
    {
      "entropy": 5.962779855728149,
      "epoch": 6.52375485694101,
      "grad_norm": 1.484375,
      "learning_rate": 0.00017327605834451987,
      "loss": 5.0928,
      "mean_token_accuracy": 0.23918237537145615,
      "num_tokens": 137759812.0,
      "step": 73875
    },
    {
      "entropy": 6.018794918060303,
      "epoch": 6.524196397032851,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00017324797005083877,
      "loss": 5.2097,
      "mean_token_accuracy": 0.22474794387817382,
      "num_tokens": 137768367.0,
      "step": 73880
    },
    {
      "entropy": 5.977225351333618,
      "epoch": 6.524637937124691,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001732198837504222,
      "loss": 5.134,
      "mean_token_accuracy": 0.23218592256307602,
      "num_tokens": 137777043.0,
      "step": 73885
    },
    {
      "entropy": 6.045046615600586,
      "epoch": 6.525079477216531,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00017319179944382004,
      "loss": 5.1972,
      "mean_token_accuracy": 0.22951098680496215,
      "num_tokens": 137785596.0,
      "step": 73890
    },
    {
      "entropy": 6.114206123352051,
      "epoch": 6.525521017308372,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0001731637171315828,
      "loss": 5.2502,
      "mean_token_accuracy": 0.21752891838550567,
      "num_tokens": 137794246.0,
      "step": 73895
    },
    {
      "entropy": 5.989264106750488,
      "epoch": 6.525962557400212,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001731356368142602,
      "loss": 5.0943,
      "mean_token_accuracy": 0.2394038572907448,
      "num_tokens": 137804721.0,
      "step": 73900
    },
    {
      "entropy": 6.108598518371582,
      "epoch": 6.526404097492052,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00017310755849240272,
      "loss": 5.2314,
      "mean_token_accuracy": 0.22770176082849503,
      "num_tokens": 137813664.0,
      "step": 73905
    },
    {
      "entropy": 6.071500635147094,
      "epoch": 6.526845637583893,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00017307948216656015,
      "loss": 5.296,
      "mean_token_accuracy": 0.22154698073863982,
      "num_tokens": 137824442.0,
      "step": 73910
    },
    {
      "entropy": 5.9570770263671875,
      "epoch": 6.527287177675733,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00017305140783728263,
      "loss": 5.1444,
      "mean_token_accuracy": 0.22693825513124466,
      "num_tokens": 137833901.0,
      "step": 73915
    },
    {
      "entropy": 5.978899812698364,
      "epoch": 6.527728717767573,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001730233355051199,
      "loss": 5.1353,
      "mean_token_accuracy": 0.23146208226680756,
      "num_tokens": 137843361.0,
      "step": 73920
    },
    {
      "entropy": 6.11105170249939,
      "epoch": 6.528170257859413,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001729952651706222,
      "loss": 5.2574,
      "mean_token_accuracy": 0.2186020642518997,
      "num_tokens": 137852447.0,
      "step": 73925
    },
    {
      "entropy": 6.033225440979004,
      "epoch": 6.528611797951254,
      "grad_norm": 1.484375,
      "learning_rate": 0.00017296719683433914,
      "loss": 5.1989,
      "mean_token_accuracy": 0.22645100355148315,
      "num_tokens": 137862283.0,
      "step": 73930
    },
    {
      "entropy": 6.083249473571778,
      "epoch": 6.529053338043094,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00017293913049682064,
      "loss": 5.1668,
      "mean_token_accuracy": 0.22774318158626555,
      "num_tokens": 137870636.0,
      "step": 73935
    },
    {
      "entropy": 6.016557788848877,
      "epoch": 6.529494878134935,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00017291106615861657,
      "loss": 5.1165,
      "mean_token_accuracy": 0.23751264214515685,
      "num_tokens": 137879109.0,
      "step": 73940
    },
    {
      "entropy": 5.972432613372803,
      "epoch": 6.529936418226775,
      "grad_norm": 1.546875,
      "learning_rate": 0.00017288300382027665,
      "loss": 5.1,
      "mean_token_accuracy": 0.2363693669438362,
      "num_tokens": 137887958.0,
      "step": 73945
    },
    {
      "entropy": 5.955268430709839,
      "epoch": 6.530377958318615,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0001728549434823507,
      "loss": 5.1135,
      "mean_token_accuracy": 0.23242974579334258,
      "num_tokens": 137896387.0,
      "step": 73950
    },
    {
      "entropy": 6.051337671279907,
      "epoch": 6.530819498410455,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001728268851453882,
      "loss": 5.1934,
      "mean_token_accuracy": 0.22939758151769638,
      "num_tokens": 137906650.0,
      "step": 73955
    },
    {
      "entropy": 6.0836316585540775,
      "epoch": 6.531261038502296,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00017279882880993902,
      "loss": 5.2706,
      "mean_token_accuracy": 0.22202375084161757,
      "num_tokens": 137915961.0,
      "step": 73960
    },
    {
      "entropy": 6.098011875152588,
      "epoch": 6.531702578594136,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017277077447655246,
      "loss": 5.2196,
      "mean_token_accuracy": 0.2276412159204483,
      "num_tokens": 137925965.0,
      "step": 73965
    },
    {
      "entropy": 5.934897422790527,
      "epoch": 6.532144118685976,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017274272214577849,
      "loss": 5.0548,
      "mean_token_accuracy": 0.24164528399705887,
      "num_tokens": 137935588.0,
      "step": 73970
    },
    {
      "entropy": 6.083372497558594,
      "epoch": 6.532585658777817,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017271467181816634,
      "loss": 5.2655,
      "mean_token_accuracy": 0.23169106096029282,
      "num_tokens": 137945242.0,
      "step": 73975
    },
    {
      "entropy": 6.003134441375733,
      "epoch": 6.533027198869657,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00017268662349426568,
      "loss": 5.1385,
      "mean_token_accuracy": 0.23746754974126816,
      "num_tokens": 137953991.0,
      "step": 73980
    },
    {
      "entropy": 6.071116495132446,
      "epoch": 6.533468738961497,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00017265857717462568,
      "loss": 5.2001,
      "mean_token_accuracy": 0.2342442974448204,
      "num_tokens": 137962439.0,
      "step": 73985
    },
    {
      "entropy": 6.054548072814941,
      "epoch": 6.533910279053338,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017263053285979617,
      "loss": 5.2762,
      "mean_token_accuracy": 0.22260052859783172,
      "num_tokens": 137972270.0,
      "step": 73990
    },
    {
      "entropy": 6.012239170074463,
      "epoch": 6.534351819145178,
      "grad_norm": 1.5,
      "learning_rate": 0.00017260249055032613,
      "loss": 5.1832,
      "mean_token_accuracy": 0.23281716108322142,
      "num_tokens": 137981441.0,
      "step": 73995
    },
    {
      "entropy": 6.023110008239746,
      "epoch": 6.534793359237018,
      "grad_norm": 1.5,
      "learning_rate": 0.00017257445024676516,
      "loss": 5.2252,
      "mean_token_accuracy": 0.23397188186645507,
      "num_tokens": 137990587.0,
      "step": 74000
    },
    {
      "entropy": 6.004404497146607,
      "epoch": 6.535234899328859,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001725464119496625,
      "loss": 5.138,
      "mean_token_accuracy": 0.23345817029476165,
      "num_tokens": 137999415.0,
      "step": 74005
    },
    {
      "entropy": 6.042609453201294,
      "epoch": 6.535676439420699,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017251837565956718,
      "loss": 5.2609,
      "mean_token_accuracy": 0.21765935570001602,
      "num_tokens": 138009184.0,
      "step": 74010
    },
    {
      "entropy": 5.992401170730591,
      "epoch": 6.5361179795125395,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001724903413770288,
      "loss": 5.0431,
      "mean_token_accuracy": 0.23801366686820985,
      "num_tokens": 138018497.0,
      "step": 74015
    },
    {
      "entropy": 6.02591643333435,
      "epoch": 6.53655951960438,
      "grad_norm": 1.46875,
      "learning_rate": 0.00017246230910259624,
      "loss": 5.158,
      "mean_token_accuracy": 0.22930193543434144,
      "num_tokens": 138027508.0,
      "step": 74020
    },
    {
      "entropy": 6.023580074310303,
      "epoch": 6.5370010596962205,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00017243427883681877,
      "loss": 5.2322,
      "mean_token_accuracy": 0.22306273728609086,
      "num_tokens": 138037413.0,
      "step": 74025
    },
    {
      "entropy": 6.04895544052124,
      "epoch": 6.5374425997880605,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001724062505802453,
      "loss": 5.2583,
      "mean_token_accuracy": 0.22140326499938964,
      "num_tokens": 138047118.0,
      "step": 74030
    },
    {
      "entropy": 6.0805192470550535,
      "epoch": 6.5378841398799015,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017237822433342523,
      "loss": 5.3159,
      "mean_token_accuracy": 0.2195608749985695,
      "num_tokens": 138055504.0,
      "step": 74035
    },
    {
      "entropy": 6.085501337051392,
      "epoch": 6.5383256799717415,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00017235020009690727,
      "loss": 5.2264,
      "mean_token_accuracy": 0.23139699697494506,
      "num_tokens": 138064219.0,
      "step": 74040
    },
    {
      "entropy": 6.0079936504364015,
      "epoch": 6.538767220063582,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001723221778712406,
      "loss": 5.1173,
      "mean_token_accuracy": 0.23448744118213655,
      "num_tokens": 138073238.0,
      "step": 74045
    },
    {
      "entropy": 6.069379806518555,
      "epoch": 6.539208760155422,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001722941576569739,
      "loss": 5.2848,
      "mean_token_accuracy": 0.22076168805360794,
      "num_tokens": 138082431.0,
      "step": 74050
    },
    {
      "entropy": 6.051317405700684,
      "epoch": 6.539650300247263,
      "grad_norm": 1.484375,
      "learning_rate": 0.00017226613945465642,
      "loss": 5.2199,
      "mean_token_accuracy": 0.2330279529094696,
      "num_tokens": 138091816.0,
      "step": 74055
    },
    {
      "entropy": 6.066849184036255,
      "epoch": 6.540091840339103,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017223812326483672,
      "loss": 5.2822,
      "mean_token_accuracy": 0.2211850568652153,
      "num_tokens": 138101110.0,
      "step": 74060
    },
    {
      "entropy": 6.092052745819092,
      "epoch": 6.540533380430944,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001722101090880638,
      "loss": 5.2627,
      "mean_token_accuracy": 0.22978471219539642,
      "num_tokens": 138110541.0,
      "step": 74065
    },
    {
      "entropy": 6.060283946990967,
      "epoch": 6.540974920522784,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017218209692488646,
      "loss": 5.2264,
      "mean_token_accuracy": 0.22198595255613326,
      "num_tokens": 138120865.0,
      "step": 74070
    },
    {
      "entropy": 6.097635269165039,
      "epoch": 6.541416460614624,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017215408677585315,
      "loss": 5.2236,
      "mean_token_accuracy": 0.22873733043670655,
      "num_tokens": 138129935.0,
      "step": 74075
    },
    {
      "entropy": 5.9842548847198485,
      "epoch": 6.541858000706464,
      "grad_norm": 1.515625,
      "learning_rate": 0.00017212607864151296,
      "loss": 5.1111,
      "mean_token_accuracy": 0.22962874174118042,
      "num_tokens": 138138991.0,
      "step": 74080
    },
    {
      "entropy": 6.017950201034546,
      "epoch": 6.542299540798305,
      "grad_norm": 1.515625,
      "learning_rate": 0.00017209807252241432,
      "loss": 5.205,
      "mean_token_accuracy": 0.23118755519390105,
      "num_tokens": 138148633.0,
      "step": 74085
    },
    {
      "entropy": 6.052536106109619,
      "epoch": 6.542741080890145,
      "grad_norm": 1.359375,
      "learning_rate": 0.000172070068419106,
      "loss": 5.2954,
      "mean_token_accuracy": 0.22076648324728013,
      "num_tokens": 138158352.0,
      "step": 74090
    },
    {
      "entropy": 6.057839393615723,
      "epoch": 6.543182620981985,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00017204206633213636,
      "loss": 5.2609,
      "mean_token_accuracy": 0.2229365810751915,
      "num_tokens": 138167008.0,
      "step": 74095
    },
    {
      "entropy": 6.0101570129394535,
      "epoch": 6.543624161073826,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001720140662620542,
      "loss": 5.1591,
      "mean_token_accuracy": 0.2264588311314583,
      "num_tokens": 138176684.0,
      "step": 74100
    },
    {
      "entropy": 6.001376628875732,
      "epoch": 6.544065701165666,
      "grad_norm": 1.5,
      "learning_rate": 0.00017198606820940785,
      "loss": 5.1511,
      "mean_token_accuracy": 0.23879822939634324,
      "num_tokens": 138184806.0,
      "step": 74105
    },
    {
      "entropy": 6.068196725845337,
      "epoch": 6.544507241257506,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001719580721747459,
      "loss": 5.1527,
      "mean_token_accuracy": 0.225324909389019,
      "num_tokens": 138193542.0,
      "step": 74110
    },
    {
      "entropy": 5.947190809249878,
      "epoch": 6.544948781349347,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00017193007815861656,
      "loss": 5.0925,
      "mean_token_accuracy": 0.23874665647745133,
      "num_tokens": 138201976.0,
      "step": 74115
    },
    {
      "entropy": 5.942461919784546,
      "epoch": 6.545390321441187,
      "grad_norm": 1.484375,
      "learning_rate": 0.00017190208616156858,
      "loss": 5.1324,
      "mean_token_accuracy": 0.2419552505016327,
      "num_tokens": 138211139.0,
      "step": 74120
    },
    {
      "entropy": 6.052747917175293,
      "epoch": 6.545831861533027,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00017187409618414995,
      "loss": 5.2483,
      "mean_token_accuracy": 0.22053516060113906,
      "num_tokens": 138219688.0,
      "step": 74125
    },
    {
      "entropy": 6.014463758468628,
      "epoch": 6.546273401624868,
      "grad_norm": 1.375,
      "learning_rate": 0.00017184610822690917,
      "loss": 5.1625,
      "mean_token_accuracy": 0.22655456215143205,
      "num_tokens": 138229111.0,
      "step": 74130
    },
    {
      "entropy": 5.976575469970703,
      "epoch": 6.546714941716708,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001718181222903945,
      "loss": 5.1482,
      "mean_token_accuracy": 0.2287372663617134,
      "num_tokens": 138238702.0,
      "step": 74135
    },
    {
      "entropy": 6.135239028930664,
      "epoch": 6.547156481808548,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000171790138375154,
      "loss": 5.2124,
      "mean_token_accuracy": 0.23122920840978622,
      "num_tokens": 138248091.0,
      "step": 74140
    },
    {
      "entropy": 6.017029190063477,
      "epoch": 6.547598021900389,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017176215648173615,
      "loss": 5.1924,
      "mean_token_accuracy": 0.22782187759876252,
      "num_tokens": 138257760.0,
      "step": 74145
    },
    {
      "entropy": 6.017115116119385,
      "epoch": 6.548039561992229,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017173417661068888,
      "loss": 5.1951,
      "mean_token_accuracy": 0.2349841445684433,
      "num_tokens": 138267293.0,
      "step": 74150
    },
    {
      "entropy": 6.026567268371582,
      "epoch": 6.548481102084069,
      "grad_norm": 1.546875,
      "learning_rate": 0.00017170619876256038,
      "loss": 5.1485,
      "mean_token_accuracy": 0.23303699791431426,
      "num_tokens": 138276724.0,
      "step": 74155
    },
    {
      "entropy": 6.029583883285523,
      "epoch": 6.54892264217591,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001716782229378986,
      "loss": 5.1628,
      "mean_token_accuracy": 0.2312607780098915,
      "num_tokens": 138285449.0,
      "step": 74160
    },
    {
      "entropy": 6.030110454559326,
      "epoch": 6.54936418226775,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001716502491372518,
      "loss": 5.1659,
      "mean_token_accuracy": 0.23134157359600066,
      "num_tokens": 138293573.0,
      "step": 74165
    },
    {
      "entropy": 5.980860233306885,
      "epoch": 6.54980572235959,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017162227736116774,
      "loss": 5.2041,
      "mean_token_accuracy": 0.23125953674316407,
      "num_tokens": 138303081.0,
      "step": 74170
    },
    {
      "entropy": 5.988902139663696,
      "epoch": 6.550247262451431,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001715943076101946,
      "loss": 5.1609,
      "mean_token_accuracy": 0.2344343438744545,
      "num_tokens": 138312858.0,
      "step": 74175
    },
    {
      "entropy": 6.003191566467285,
      "epoch": 6.550688802543271,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017156633988487998,
      "loss": 5.157,
      "mean_token_accuracy": 0.23876679092645645,
      "num_tokens": 138321532.0,
      "step": 74180
    },
    {
      "entropy": 6.092740821838379,
      "epoch": 6.551130342635111,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017153837418577213,
      "loss": 5.3048,
      "mean_token_accuracy": 0.2174499362707138,
      "num_tokens": 138330477.0,
      "step": 74185
    },
    {
      "entropy": 6.047281789779663,
      "epoch": 6.551571882726952,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00017151041051341858,
      "loss": 5.1607,
      "mean_token_accuracy": 0.23113223612308503,
      "num_tokens": 138339376.0,
      "step": 74190
    },
    {
      "entropy": 6.045635652542114,
      "epoch": 6.552013422818792,
      "grad_norm": 1.578125,
      "learning_rate": 0.00017148244886836724,
      "loss": 5.1111,
      "mean_token_accuracy": 0.2309350699186325,
      "num_tokens": 138348650.0,
      "step": 74195
    },
    {
      "entropy": 6.044003582000732,
      "epoch": 6.552454962910632,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017145448925116592,
      "loss": 5.2009,
      "mean_token_accuracy": 0.2221650078892708,
      "num_tokens": 138359659.0,
      "step": 74200
    },
    {
      "entropy": 6.029045152664184,
      "epoch": 6.552896503002472,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00017142653166236205,
      "loss": 5.1612,
      "mean_token_accuracy": 0.22871242612600326,
      "num_tokens": 138368440.0,
      "step": 74205
    },
    {
      "entropy": 6.049146032333374,
      "epoch": 6.553338043094313,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017139857610250374,
      "loss": 5.2039,
      "mean_token_accuracy": 0.2261536255478859,
      "num_tokens": 138378000.0,
      "step": 74210
    },
    {
      "entropy": 6.052721309661865,
      "epoch": 6.553779583186153,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017137062257213838,
      "loss": 5.2835,
      "mean_token_accuracy": 0.22128895819187164,
      "num_tokens": 138386923.0,
      "step": 74215
    },
    {
      "entropy": 6.029571771621704,
      "epoch": 6.554221123277994,
      "grad_norm": 1.28125,
      "learning_rate": 0.00017134267107181357,
      "loss": 5.1739,
      "mean_token_accuracy": 0.23286404013633727,
      "num_tokens": 138396683.0,
      "step": 74220
    },
    {
      "entropy": 6.083280944824219,
      "epoch": 6.554662663369834,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017131472160207672,
      "loss": 5.1963,
      "mean_token_accuracy": 0.22865627706050873,
      "num_tokens": 138406381.0,
      "step": 74225
    },
    {
      "entropy": 6.051854419708252,
      "epoch": 6.555104203461674,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017128677416347574,
      "loss": 5.1836,
      "mean_token_accuracy": 0.22351969331502913,
      "num_tokens": 138414877.0,
      "step": 74230
    },
    {
      "entropy": 6.024654912948608,
      "epoch": 6.555545743553514,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00017125882875655777,
      "loss": 5.1788,
      "mean_token_accuracy": 0.2229279786348343,
      "num_tokens": 138425283.0,
      "step": 74235
    },
    {
      "entropy": 6.051630115509033,
      "epoch": 6.555987283645355,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017123088538187038,
      "loss": 5.2348,
      "mean_token_accuracy": 0.22085834592580794,
      "num_tokens": 138434120.0,
      "step": 74240
    },
    {
      "entropy": 5.942744016647339,
      "epoch": 6.556428823737195,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017120294403996086,
      "loss": 5.0117,
      "mean_token_accuracy": 0.24946481585502625,
      "num_tokens": 138443612.0,
      "step": 74245
    },
    {
      "entropy": 6.011153650283814,
      "epoch": 6.556870363829035,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017117500473137667,
      "loss": 5.1669,
      "mean_token_accuracy": 0.227872571349144,
      "num_tokens": 138452730.0,
      "step": 74250
    },
    {
      "entropy": 5.986340427398682,
      "epoch": 6.557311903920876,
      "grad_norm": 1.484375,
      "learning_rate": 0.00017114706745666506,
      "loss": 5.1348,
      "mean_token_accuracy": 0.24012638628482819,
      "num_tokens": 138461187.0,
      "step": 74255
    },
    {
      "entropy": 6.083223056793213,
      "epoch": 6.557753444012716,
      "grad_norm": 1.484375,
      "learning_rate": 0.00017111913221637333,
      "loss": 5.2835,
      "mean_token_accuracy": 0.2229391634464264,
      "num_tokens": 138469900.0,
      "step": 74260
    },
    {
      "entropy": 6.023241806030273,
      "epoch": 6.558194984104556,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017109119901104883,
      "loss": 5.1401,
      "mean_token_accuracy": 0.23360887318849563,
      "num_tokens": 138479146.0,
      "step": 74265
    },
    {
      "entropy": 6.066507816314697,
      "epoch": 6.558636524196397,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00017106326784123844,
      "loss": 5.2242,
      "mean_token_accuracy": 0.22432800382375717,
      "num_tokens": 138487868.0,
      "step": 74270
    },
    {
      "entropy": 6.007065773010254,
      "epoch": 6.559078064288237,
      "grad_norm": 1.578125,
      "learning_rate": 0.00017103533870748973,
      "loss": 5.1253,
      "mean_token_accuracy": 0.23317156583070756,
      "num_tokens": 138496563.0,
      "step": 74275
    },
    {
      "entropy": 6.083238887786865,
      "epoch": 6.559519604380077,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001710074116103495,
      "loss": 5.2221,
      "mean_token_accuracy": 0.22069152146577836,
      "num_tokens": 138505610.0,
      "step": 74280
    },
    {
      "entropy": 6.0675498962402346,
      "epoch": 6.559961144471918,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017097948655036502,
      "loss": 5.1899,
      "mean_token_accuracy": 0.23124610781669616,
      "num_tokens": 138515519.0,
      "step": 74285
    },
    {
      "entropy": 5.990978574752807,
      "epoch": 6.560402684563758,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017095156352808305,
      "loss": 5.0716,
      "mean_token_accuracy": 0.2377513751387596,
      "num_tokens": 138524272.0,
      "step": 74290
    },
    {
      "entropy": 5.998846817016601,
      "epoch": 6.5608442246555985,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00017092364254405102,
      "loss": 5.1422,
      "mean_token_accuracy": 0.23827216774225235,
      "num_tokens": 138533167.0,
      "step": 74295
    },
    {
      "entropy": 6.11570405960083,
      "epoch": 6.561285764747439,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017089572359881548,
      "loss": 5.3088,
      "mean_token_accuracy": 0.2205163061618805,
      "num_tokens": 138543747.0,
      "step": 74300
    },
    {
      "entropy": 6.122352457046508,
      "epoch": 6.5617273048392795,
      "grad_norm": 1.46875,
      "learning_rate": 0.00017086780669292356,
      "loss": 5.2285,
      "mean_token_accuracy": 0.2339420959353447,
      "num_tokens": 138552762.0,
      "step": 74305
    },
    {
      "entropy": 5.976398849487305,
      "epoch": 6.5621688449311195,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00017083989182692212,
      "loss": 5.0721,
      "mean_token_accuracy": 0.23383170664310454,
      "num_tokens": 138561160.0,
      "step": 74310
    },
    {
      "entropy": 6.0500936031341555,
      "epoch": 6.5626103850229605,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017081197900135786,
      "loss": 5.1943,
      "mean_token_accuracy": 0.22736619114875795,
      "num_tokens": 138570463.0,
      "step": 74315
    },
    {
      "entropy": 6.040579986572266,
      "epoch": 6.5630519251148005,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017078406821677768,
      "loss": 5.2253,
      "mean_token_accuracy": 0.21926801800727844,
      "num_tokens": 138579824.0,
      "step": 74320
    },
    {
      "entropy": 5.973381042480469,
      "epoch": 6.563493465206641,
      "grad_norm": 1.5,
      "learning_rate": 0.00017075615947372837,
      "loss": 5.0908,
      "mean_token_accuracy": 0.2446690693497658,
      "num_tokens": 138588686.0,
      "step": 74325
    },
    {
      "entropy": 5.9881579875946045,
      "epoch": 6.563935005298481,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00017072825277275667,
      "loss": 5.2023,
      "mean_token_accuracy": 0.22648349404335022,
      "num_tokens": 138597240.0,
      "step": 74330
    },
    {
      "entropy": 5.976555442810058,
      "epoch": 6.564376545390322,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017070034811440907,
      "loss": 5.1337,
      "mean_token_accuracy": 0.22910628467798233,
      "num_tokens": 138607335.0,
      "step": 74335
    },
    {
      "entropy": 6.0792069911956785,
      "epoch": 6.564818085482162,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00017067244549923247,
      "loss": 5.2284,
      "mean_token_accuracy": 0.22080413699150087,
      "num_tokens": 138616503.0,
      "step": 74340
    },
    {
      "entropy": 6.054936122894287,
      "epoch": 6.5652596255740026,
      "grad_norm": 1.296875,
      "learning_rate": 0.00017064454492777325,
      "loss": 5.1782,
      "mean_token_accuracy": 0.23712183088064193,
      "num_tokens": 138626078.0,
      "step": 74345
    },
    {
      "entropy": 6.085271978378296,
      "epoch": 6.565701165665843,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017061664640057815,
      "loss": 5.169,
      "mean_token_accuracy": 0.22437054812908172,
      "num_tokens": 138635196.0,
      "step": 74350
    },
    {
      "entropy": 6.101207828521728,
      "epoch": 6.566142705757683,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017058874991819335,
      "loss": 5.3427,
      "mean_token_accuracy": 0.21542446315288544,
      "num_tokens": 138644896.0,
      "step": 74355
    },
    {
      "entropy": 6.044817066192627,
      "epoch": 6.566584245849523,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017056085548116577,
      "loss": 5.1803,
      "mean_token_accuracy": 0.22502127140760422,
      "num_tokens": 138653967.0,
      "step": 74360
    },
    {
      "entropy": 6.012141513824463,
      "epoch": 6.567025785941364,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00017053296309004154,
      "loss": 5.1668,
      "mean_token_accuracy": 0.23302168399095535,
      "num_tokens": 138662535.0,
      "step": 74365
    },
    {
      "entropy": 6.071602153778076,
      "epoch": 6.567467326033204,
      "grad_norm": 1.484375,
      "learning_rate": 0.00017050507274536715,
      "loss": 5.1778,
      "mean_token_accuracy": 0.22176117748022078,
      "num_tokens": 138671257.0,
      "step": 74370
    },
    {
      "entropy": 6.041357183456421,
      "epoch": 6.567908866125045,
      "grad_norm": 1.59375,
      "learning_rate": 0.00017047718444768904,
      "loss": 5.1842,
      "mean_token_accuracy": 0.23626243621110915,
      "num_tokens": 138680158.0,
      "step": 74375
    },
    {
      "entropy": 6.0610371112823485,
      "epoch": 6.568350406216885,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017044929819755338,
      "loss": 5.2235,
      "mean_token_accuracy": 0.22534480839967727,
      "num_tokens": 138689235.0,
      "step": 74380
    },
    {
      "entropy": 5.98968391418457,
      "epoch": 6.568791946308725,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00017042141399550647,
      "loss": 5.2095,
      "mean_token_accuracy": 0.22643092423677444,
      "num_tokens": 138698831.0,
      "step": 74385
    },
    {
      "entropy": 5.970379829406738,
      "epoch": 6.569233486400565,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001703935318420946,
      "loss": 5.1179,
      "mean_token_accuracy": 0.24340287446975709,
      "num_tokens": 138708448.0,
      "step": 74390
    },
    {
      "entropy": 5.98236575126648,
      "epoch": 6.569675026492406,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017036565173786402,
      "loss": 5.17,
      "mean_token_accuracy": 0.23099371343851088,
      "num_tokens": 138718148.0,
      "step": 74395
    },
    {
      "entropy": 6.040058422088623,
      "epoch": 6.570116566584246,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017033777368336068,
      "loss": 5.2376,
      "mean_token_accuracy": 0.22808988839387895,
      "num_tokens": 138727447.0,
      "step": 74400
    },
    {
      "entropy": 6.081261396408081,
      "epoch": 6.570558106676086,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017030989767913095,
      "loss": 5.2631,
      "mean_token_accuracy": 0.22298014014959336,
      "num_tokens": 138738232.0,
      "step": 74405
    },
    {
      "entropy": 6.048779487609863,
      "epoch": 6.570999646767927,
      "grad_norm": 1.390625,
      "learning_rate": 0.00017028202372572072,
      "loss": 5.1286,
      "mean_token_accuracy": 0.23315920531749726,
      "num_tokens": 138746994.0,
      "step": 74410
    },
    {
      "entropy": 6.003534460067749,
      "epoch": 6.571441186859767,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001702541518236762,
      "loss": 5.2061,
      "mean_token_accuracy": 0.2287626713514328,
      "num_tokens": 138757631.0,
      "step": 74415
    },
    {
      "entropy": 6.15169939994812,
      "epoch": 6.571882726951607,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017022628197354306,
      "loss": 5.3312,
      "mean_token_accuracy": 0.2111918792128563,
      "num_tokens": 138766511.0,
      "step": 74420
    },
    {
      "entropy": 6.080064964294434,
      "epoch": 6.572324267043448,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001701984141758677,
      "loss": 5.2015,
      "mean_token_accuracy": 0.22552327960729598,
      "num_tokens": 138774509.0,
      "step": 74425
    },
    {
      "entropy": 6.100792741775512,
      "epoch": 6.572765807135288,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017017054843119567,
      "loss": 5.2421,
      "mean_token_accuracy": 0.22868618071079255,
      "num_tokens": 138784773.0,
      "step": 74430
    },
    {
      "entropy": 6.057469606399536,
      "epoch": 6.573207347227128,
      "grad_norm": 1.53125,
      "learning_rate": 0.00017014268474007294,
      "loss": 5.249,
      "mean_token_accuracy": 0.22564428597688674,
      "num_tokens": 138794161.0,
      "step": 74435
    },
    {
      "entropy": 6.039336204528809,
      "epoch": 6.573648887318969,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017011482310304554,
      "loss": 5.1819,
      "mean_token_accuracy": 0.22625374794006348,
      "num_tokens": 138803969.0,
      "step": 74440
    },
    {
      "entropy": 6.048614120483398,
      "epoch": 6.574090427410809,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000170086963520659,
      "loss": 5.2063,
      "mean_token_accuracy": 0.22077168226242067,
      "num_tokens": 138813785.0,
      "step": 74445
    },
    {
      "entropy": 6.0774791717529295,
      "epoch": 6.574531967502649,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00017005910599345915,
      "loss": 5.2006,
      "mean_token_accuracy": 0.2230744704604149,
      "num_tokens": 138823643.0,
      "step": 74450
    },
    {
      "entropy": 5.979939317703247,
      "epoch": 6.57497350759449,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00017003125052199175,
      "loss": 5.1351,
      "mean_token_accuracy": 0.2412787050008774,
      "num_tokens": 138833266.0,
      "step": 74455
    },
    {
      "entropy": 5.992308139801025,
      "epoch": 6.57541504768633,
      "grad_norm": 1.53125,
      "learning_rate": 0.00017000339710680252,
      "loss": 5.139,
      "mean_token_accuracy": 0.22995719611644744,
      "num_tokens": 138842013.0,
      "step": 74460
    },
    {
      "entropy": 6.012792491912842,
      "epoch": 6.57585658777817,
      "grad_norm": 1.515625,
      "learning_rate": 0.00016997554574843686,
      "loss": 5.2409,
      "mean_token_accuracy": 0.2230142816901207,
      "num_tokens": 138850186.0,
      "step": 74465
    },
    {
      "entropy": 6.095190858840942,
      "epoch": 6.576298127870011,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001699476964474407,
      "loss": 5.2666,
      "mean_token_accuracy": 0.2244493305683136,
      "num_tokens": 138859816.0,
      "step": 74470
    },
    {
      "entropy": 6.1311369895935055,
      "epoch": 6.576739667961851,
      "grad_norm": 1.671875,
      "learning_rate": 0.00016991984920435938,
      "loss": 5.2871,
      "mean_token_accuracy": 0.22676447480916978,
      "num_tokens": 138868454.0,
      "step": 74475
    },
    {
      "entropy": 6.025729560852051,
      "epoch": 6.577181208053691,
      "grad_norm": 1.453125,
      "learning_rate": 0.00016989200401973847,
      "loss": 5.1848,
      "mean_token_accuracy": 0.22656766772270204,
      "num_tokens": 138876902.0,
      "step": 74480
    },
    {
      "entropy": 5.985172176361084,
      "epoch": 6.577622748145531,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00016986416089412337,
      "loss": 5.0949,
      "mean_token_accuracy": 0.2289160594344139,
      "num_tokens": 138885749.0,
      "step": 74485
    },
    {
      "entropy": 6.0734350204467775,
      "epoch": 6.578064288237372,
      "grad_norm": 1.640625,
      "learning_rate": 0.00016983631982805958,
      "loss": 5.1796,
      "mean_token_accuracy": 0.23504878878593444,
      "num_tokens": 138894462.0,
      "step": 74490
    },
    {
      "entropy": 6.109299993515014,
      "epoch": 6.578505828329212,
      "grad_norm": 1.484375,
      "learning_rate": 0.00016980848082209243,
      "loss": 5.282,
      "mean_token_accuracy": 0.22023896723985673,
      "num_tokens": 138903368.0,
      "step": 74495
    },
    {
      "entropy": 6.017741918563843,
      "epoch": 6.578947368421053,
      "grad_norm": 1.453125,
      "learning_rate": 0.00016978064387676735,
      "loss": 5.2266,
      "mean_token_accuracy": 0.23236165344715118,
      "num_tokens": 138913197.0,
      "step": 74500
    },
    {
      "entropy": 6.119561004638672,
      "epoch": 6.579388908512893,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016975280899262968,
      "loss": 5.251,
      "mean_token_accuracy": 0.2217475101351738,
      "num_tokens": 138923230.0,
      "step": 74505
    },
    {
      "entropy": 6.06779990196228,
      "epoch": 6.579830448604733,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001697249761702246,
      "loss": 5.1758,
      "mean_token_accuracy": 0.2357835665345192,
      "num_tokens": 138932041.0,
      "step": 74510
    },
    {
      "entropy": 6.039390182495117,
      "epoch": 6.580271988696573,
      "grad_norm": 1.375,
      "learning_rate": 0.00016969714541009728,
      "loss": 5.1881,
      "mean_token_accuracy": 0.22894273251295089,
      "num_tokens": 138941735.0,
      "step": 74515
    },
    {
      "entropy": 6.16040186882019,
      "epoch": 6.580713528788414,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000169669316712793,
      "loss": 5.3832,
      "mean_token_accuracy": 0.2079812094569206,
      "num_tokens": 138951130.0,
      "step": 74520
    },
    {
      "entropy": 6.05761194229126,
      "epoch": 6.581155068880254,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016964149007885705,
      "loss": 5.2586,
      "mean_token_accuracy": 0.22746537625789642,
      "num_tokens": 138961412.0,
      "step": 74525
    },
    {
      "entropy": 6.10659351348877,
      "epoch": 6.581596608972094,
      "grad_norm": 1.515625,
      "learning_rate": 0.00016961366550883418,
      "loss": 5.3202,
      "mean_token_accuracy": 0.2174544855952263,
      "num_tokens": 138970699.0,
      "step": 74530
    },
    {
      "entropy": 6.086166095733643,
      "epoch": 6.582038149063935,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00016958584300326985,
      "loss": 5.239,
      "mean_token_accuracy": 0.22677557468414306,
      "num_tokens": 138980736.0,
      "step": 74535
    },
    {
      "entropy": 6.045394992828369,
      "epoch": 6.582479689155775,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001695580225627088,
      "loss": 5.2136,
      "mean_token_accuracy": 0.23439743965864182,
      "num_tokens": 138990710.0,
      "step": 74540
    },
    {
      "entropy": 6.080216121673584,
      "epoch": 6.582921229247615,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00016953020418769622,
      "loss": 5.1801,
      "mean_token_accuracy": 0.22827853858470917,
      "num_tokens": 138999553.0,
      "step": 74545
    },
    {
      "entropy": 6.076859045028686,
      "epoch": 6.583362769339456,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016950238787877684,
      "loss": 5.2861,
      "mean_token_accuracy": 0.2204687014222145,
      "num_tokens": 139009501.0,
      "step": 74550
    },
    {
      "entropy": 6.057710027694702,
      "epoch": 6.583804309431296,
      "grad_norm": 1.28125,
      "learning_rate": 0.00016947457363649572,
      "loss": 5.1773,
      "mean_token_accuracy": 0.22887781709432603,
      "num_tokens": 139019407.0,
      "step": 74555
    },
    {
      "entropy": 6.020586252212524,
      "epoch": 6.584245849523136,
      "grad_norm": 1.453125,
      "learning_rate": 0.00016944676146139764,
      "loss": 5.1274,
      "mean_token_accuracy": 0.2357385575771332,
      "num_tokens": 139029453.0,
      "step": 74560
    },
    {
      "entropy": 6.013595819473267,
      "epoch": 6.584687389614977,
      "grad_norm": 1.609375,
      "learning_rate": 0.00016941895135402752,
      "loss": 5.144,
      "mean_token_accuracy": 0.23198023587465286,
      "num_tokens": 139038533.0,
      "step": 74565
    },
    {
      "entropy": 6.058824062347412,
      "epoch": 6.585128929706817,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00016939114331493017,
      "loss": 5.2051,
      "mean_token_accuracy": 0.22040344923734664,
      "num_tokens": 139048121.0,
      "step": 74570
    },
    {
      "entropy": 6.0199713706970215,
      "epoch": 6.5855704697986575,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00016936333734465016,
      "loss": 5.1133,
      "mean_token_accuracy": 0.2380879431962967,
      "num_tokens": 139057330.0,
      "step": 74575
    },
    {
      "entropy": 6.005889940261841,
      "epoch": 6.586012009890498,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001693355334437323,
      "loss": 5.2317,
      "mean_token_accuracy": 0.22247401177883147,
      "num_tokens": 139066553.0,
      "step": 74580
    },
    {
      "entropy": 6.109932327270508,
      "epoch": 6.5864535499823385,
      "grad_norm": 1.375,
      "learning_rate": 0.00016930773161272122,
      "loss": 5.2309,
      "mean_token_accuracy": 0.22528009861707687,
      "num_tokens": 139076201.0,
      "step": 74585
    },
    {
      "entropy": 5.97654242515564,
      "epoch": 6.5868950900741785,
      "grad_norm": 1.453125,
      "learning_rate": 0.00016927993185216172,
      "loss": 5.0896,
      "mean_token_accuracy": 0.23353525400161743,
      "num_tokens": 139085166.0,
      "step": 74590
    },
    {
      "entropy": 6.038797283172608,
      "epoch": 6.5873366301660194,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000169252134162598,
      "loss": 5.1527,
      "mean_token_accuracy": 0.227364644408226,
      "num_tokens": 139094431.0,
      "step": 74595
    },
    {
      "entropy": 6.079681062698365,
      "epoch": 6.5877781702578595,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00016922433854457504,
      "loss": 5.1876,
      "mean_token_accuracy": 0.2269825741648674,
      "num_tokens": 139103703.0,
      "step": 74600
    },
    {
      "entropy": 6.043679904937744,
      "epoch": 6.5882197103496996,
      "grad_norm": 1.4375,
      "learning_rate": 0.00016919654499863702,
      "loss": 5.2168,
      "mean_token_accuracy": 0.22823536098003389,
      "num_tokens": 139111291.0,
      "step": 74605
    },
    {
      "entropy": 5.998942565917969,
      "epoch": 6.58866125044154,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00016916875352532856,
      "loss": 5.1756,
      "mean_token_accuracy": 0.2322400376200676,
      "num_tokens": 139120864.0,
      "step": 74610
    },
    {
      "entropy": 5.930403995513916,
      "epoch": 6.5891027905333805,
      "grad_norm": 1.5,
      "learning_rate": 0.000169140964125194,
      "loss": 5.1018,
      "mean_token_accuracy": 0.23261443376541138,
      "num_tokens": 139129765.0,
      "step": 74615
    },
    {
      "entropy": 6.027949047088623,
      "epoch": 6.589544330625221,
      "grad_norm": 1.375,
      "learning_rate": 0.00016911317679877774,
      "loss": 5.182,
      "mean_token_accuracy": 0.23263518512248993,
      "num_tokens": 139140178.0,
      "step": 74620
    },
    {
      "entropy": 5.984851217269897,
      "epoch": 6.5899858707170615,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00016908539154662411,
      "loss": 5.0795,
      "mean_token_accuracy": 0.2324855789542198,
      "num_tokens": 139149699.0,
      "step": 74625
    },
    {
      "entropy": 6.057438516616822,
      "epoch": 6.590427410808902,
      "grad_norm": 1.421875,
      "learning_rate": 0.00016905760836927742,
      "loss": 5.2185,
      "mean_token_accuracy": 0.22894678115844727,
      "num_tokens": 139160129.0,
      "step": 74630
    },
    {
      "entropy": 6.092076063156128,
      "epoch": 6.590868950900742,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000169029827267282,
      "loss": 5.1637,
      "mean_token_accuracy": 0.23005165308713912,
      "num_tokens": 139169060.0,
      "step": 74635
    },
    {
      "entropy": 6.078687477111816,
      "epoch": 6.591310490992582,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00016900204824118187,
      "loss": 5.1845,
      "mean_token_accuracy": 0.2197719097137451,
      "num_tokens": 139178615.0,
      "step": 74640
    },
    {
      "entropy": 5.943011140823364,
      "epoch": 6.591752031084423,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016897427129152146,
      "loss": 5.0609,
      "mean_token_accuracy": 0.24624402672052384,
      "num_tokens": 139187689.0,
      "step": 74645
    },
    {
      "entropy": 5.98151969909668,
      "epoch": 6.592193571176263,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00016894649641884468,
      "loss": 5.1137,
      "mean_token_accuracy": 0.23581619709730148,
      "num_tokens": 139197184.0,
      "step": 74650
    },
    {
      "entropy": 6.07154598236084,
      "epoch": 6.592635111268104,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00016891872362369587,
      "loss": 5.2078,
      "mean_token_accuracy": 0.22565475404262542,
      "num_tokens": 139206171.0,
      "step": 74655
    },
    {
      "entropy": 6.069673871994018,
      "epoch": 6.593076651359944,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001688909529066187,
      "loss": 5.2153,
      "mean_token_accuracy": 0.22223198115825654,
      "num_tokens": 139215112.0,
      "step": 74660
    },
    {
      "entropy": 6.111437702178955,
      "epoch": 6.593518191451784,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016886318426815764,
      "loss": 5.2701,
      "mean_token_accuracy": 0.22066789716482163,
      "num_tokens": 139224305.0,
      "step": 74665
    },
    {
      "entropy": 6.094088315963745,
      "epoch": 6.593959731543624,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001688354177088563,
      "loss": 5.2085,
      "mean_token_accuracy": 0.22419090420007706,
      "num_tokens": 139234344.0,
      "step": 74670
    },
    {
      "entropy": 6.085861492156982,
      "epoch": 6.594401271635465,
      "grad_norm": 1.53125,
      "learning_rate": 0.00016880765322925886,
      "loss": 5.2477,
      "mean_token_accuracy": 0.22118126451969147,
      "num_tokens": 139244156.0,
      "step": 74675
    },
    {
      "entropy": 6.095066165924072,
      "epoch": 6.594842811727305,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016877989082990903,
      "loss": 5.2115,
      "mean_token_accuracy": 0.2276400402188301,
      "num_tokens": 139253265.0,
      "step": 74680
    },
    {
      "entropy": 6.139769554138184,
      "epoch": 6.595284351819145,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016875213051135068,
      "loss": 5.3143,
      "mean_token_accuracy": 0.21486877501010895,
      "num_tokens": 139262511.0,
      "step": 74685
    },
    {
      "entropy": 6.062497472763061,
      "epoch": 6.595725891910986,
      "grad_norm": 1.5,
      "learning_rate": 0.00016872437227412768,
      "loss": 5.2354,
      "mean_token_accuracy": 0.22296709567308426,
      "num_tokens": 139271716.0,
      "step": 74690
    },
    {
      "entropy": 6.0076268196105955,
      "epoch": 6.596167432002826,
      "grad_norm": 1.515625,
      "learning_rate": 0.00016869661611878384,
      "loss": 5.1215,
      "mean_token_accuracy": 0.23458350002765654,
      "num_tokens": 139281222.0,
      "step": 74695
    },
    {
      "entropy": 6.125835037231445,
      "epoch": 6.596608972094666,
      "grad_norm": 1.5,
      "learning_rate": 0.00016866886204586285,
      "loss": 5.2707,
      "mean_token_accuracy": 0.216384419798851,
      "num_tokens": 139291208.0,
      "step": 74700
    },
    {
      "entropy": 6.136457347869873,
      "epoch": 6.597050512186507,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016864111005590823,
      "loss": 5.2952,
      "mean_token_accuracy": 0.21848654001951218,
      "num_tokens": 139300492.0,
      "step": 74705
    },
    {
      "entropy": 6.031192922592163,
      "epoch": 6.597492052278347,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016861336014946393,
      "loss": 5.1584,
      "mean_token_accuracy": 0.2262047991156578,
      "num_tokens": 139310080.0,
      "step": 74710
    },
    {
      "entropy": 6.174390363693237,
      "epoch": 6.597933592370187,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00016858561232707336,
      "loss": 5.3715,
      "mean_token_accuracy": 0.21091823577880858,
      "num_tokens": 139319846.0,
      "step": 74715
    },
    {
      "entropy": 5.985805368423462,
      "epoch": 6.598375132462028,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016855786658928019,
      "loss": 5.1026,
      "mean_token_accuracy": 0.23371736705303192,
      "num_tokens": 139329276.0,
      "step": 74720
    },
    {
      "entropy": 6.054020786285401,
      "epoch": 6.598816672553868,
      "grad_norm": 1.484375,
      "learning_rate": 0.00016853012293662777,
      "loss": 5.2894,
      "mean_token_accuracy": 0.22424309849739074,
      "num_tokens": 139339004.0,
      "step": 74725
    },
    {
      "entropy": 6.101826858520508,
      "epoch": 6.599258212645708,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00016850238136965974,
      "loss": 5.2406,
      "mean_token_accuracy": 0.22040267884731293,
      "num_tokens": 139347133.0,
      "step": 74730
    },
    {
      "entropy": 6.129229593276977,
      "epoch": 6.599699752737549,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001684746418889194,
      "loss": 5.299,
      "mean_token_accuracy": 0.2195468634366989,
      "num_tokens": 139356546.0,
      "step": 74735
    },
    {
      "entropy": 6.060072994232177,
      "epoch": 6.600141292829389,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00016844690449495037,
      "loss": 5.2059,
      "mean_token_accuracy": 0.2291766032576561,
      "num_tokens": 139365728.0,
      "step": 74740
    },
    {
      "entropy": 6.05091814994812,
      "epoch": 6.600582832921229,
      "grad_norm": 1.46875,
      "learning_rate": 0.00016841916918829583,
      "loss": 5.1853,
      "mean_token_accuracy": 0.23330098688602446,
      "num_tokens": 139374784.0,
      "step": 74745
    },
    {
      "entropy": 6.025155019760132,
      "epoch": 6.60102437301307,
      "grad_norm": 1.375,
      "learning_rate": 0.00016839143596949913,
      "loss": 5.2156,
      "mean_token_accuracy": 0.22146908044815064,
      "num_tokens": 139383727.0,
      "step": 74750
    },
    {
      "entropy": 6.060332918167115,
      "epoch": 6.60146591310491,
      "grad_norm": 1.546875,
      "learning_rate": 0.00016836370483910352,
      "loss": 5.2049,
      "mean_token_accuracy": 0.23088509738445281,
      "num_tokens": 139392631.0,
      "step": 74755
    },
    {
      "entropy": 6.1489983081817625,
      "epoch": 6.60190745319675,
      "grad_norm": 1.484375,
      "learning_rate": 0.00016833597579765235,
      "loss": 5.262,
      "mean_token_accuracy": 0.22706206738948823,
      "num_tokens": 139401260.0,
      "step": 74760
    },
    {
      "entropy": 5.995348262786865,
      "epoch": 6.60234899328859,
      "grad_norm": 1.5,
      "learning_rate": 0.00016830824884568884,
      "loss": 5.1215,
      "mean_token_accuracy": 0.22873588353395463,
      "num_tokens": 139410217.0,
      "step": 74765
    },
    {
      "entropy": 6.065141391754151,
      "epoch": 6.602790533380431,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00016828052398375583,
      "loss": 5.2424,
      "mean_token_accuracy": 0.2250101625919342,
      "num_tokens": 139419339.0,
      "step": 74770
    },
    {
      "entropy": 6.007054281234741,
      "epoch": 6.603232073472271,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016825280121239683,
      "loss": 5.1019,
      "mean_token_accuracy": 0.23457452207803725,
      "num_tokens": 139428239.0,
      "step": 74775
    },
    {
      "entropy": 6.039247274398804,
      "epoch": 6.603673613564112,
      "grad_norm": 1.5,
      "learning_rate": 0.00016822508053215474,
      "loss": 5.167,
      "mean_token_accuracy": 0.23077090531587602,
      "num_tokens": 139437667.0,
      "step": 74780
    },
    {
      "entropy": 6.006346273422241,
      "epoch": 6.604115153655952,
      "grad_norm": 1.53125,
      "learning_rate": 0.00016819736194357264,
      "loss": 5.137,
      "mean_token_accuracy": 0.22831135541200637,
      "num_tokens": 139446839.0,
      "step": 74785
    },
    {
      "entropy": 5.988711786270142,
      "epoch": 6.604556693747792,
      "grad_norm": 1.453125,
      "learning_rate": 0.00016816964544719338,
      "loss": 5.1503,
      "mean_token_accuracy": 0.24010764360427855,
      "num_tokens": 139456611.0,
      "step": 74790
    },
    {
      "entropy": 6.046332788467407,
      "epoch": 6.604998233839632,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016814193104356,
      "loss": 5.2115,
      "mean_token_accuracy": 0.2244840830564499,
      "num_tokens": 139466929.0,
      "step": 74795
    },
    {
      "entropy": 6.062432336807251,
      "epoch": 6.605439773931473,
      "grad_norm": 1.703125,
      "learning_rate": 0.00016811421873321543,
      "loss": 5.2181,
      "mean_token_accuracy": 0.23138492852449416,
      "num_tokens": 139475748.0,
      "step": 74800
    },
    {
      "entropy": 6.0797301769256595,
      "epoch": 6.605881314023313,
      "grad_norm": 1.53125,
      "learning_rate": 0.00016808650851670259,
      "loss": 5.2453,
      "mean_token_accuracy": 0.2225626826286316,
      "num_tokens": 139484644.0,
      "step": 74805
    },
    {
      "entropy": 6.04109673500061,
      "epoch": 6.606322854115153,
      "grad_norm": 1.4375,
      "learning_rate": 0.00016805880039456417,
      "loss": 5.187,
      "mean_token_accuracy": 0.23169294148683547,
      "num_tokens": 139494102.0,
      "step": 74810
    },
    {
      "entropy": 5.922340393066406,
      "epoch": 6.606764394206994,
      "grad_norm": 1.453125,
      "learning_rate": 0.00016803109436734303,
      "loss": 5.1571,
      "mean_token_accuracy": 0.22576664090156556,
      "num_tokens": 139503837.0,
      "step": 74815
    },
    {
      "entropy": 6.001028156280517,
      "epoch": 6.607205934298834,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00016800339043558193,
      "loss": 5.0937,
      "mean_token_accuracy": 0.23912446796894074,
      "num_tokens": 139512948.0,
      "step": 74820
    },
    {
      "entropy": 6.054156827926636,
      "epoch": 6.607647474390674,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016797568859982357,
      "loss": 5.2667,
      "mean_token_accuracy": 0.22326294779777528,
      "num_tokens": 139522146.0,
      "step": 74825
    },
    {
      "entropy": 6.063079071044922,
      "epoch": 6.608089014482515,
      "grad_norm": 1.4375,
      "learning_rate": 0.00016794798886061066,
      "loss": 5.2008,
      "mean_token_accuracy": 0.22666812241077422,
      "num_tokens": 139531641.0,
      "step": 74830
    },
    {
      "entropy": 6.010265445709228,
      "epoch": 6.608530554574355,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00016792029121848556,
      "loss": 5.1811,
      "mean_token_accuracy": 0.22642983496189117,
      "num_tokens": 139540398.0,
      "step": 74835
    },
    {
      "entropy": 6.018852186203003,
      "epoch": 6.608972094666195,
      "grad_norm": 1.5625,
      "learning_rate": 0.00016789259567399128,
      "loss": 5.092,
      "mean_token_accuracy": 0.2411144718527794,
      "num_tokens": 139549530.0,
      "step": 74840
    },
    {
      "entropy": 6.025407600402832,
      "epoch": 6.609413634758036,
      "grad_norm": 1.484375,
      "learning_rate": 0.00016786490222767003,
      "loss": 5.1767,
      "mean_token_accuracy": 0.23308070302009581,
      "num_tokens": 139558552.0,
      "step": 74845
    },
    {
      "entropy": 6.1277998924255375,
      "epoch": 6.609855174849876,
      "grad_norm": 1.4375,
      "learning_rate": 0.00016783721088006455,
      "loss": 5.3182,
      "mean_token_accuracy": 0.21165740340948105,
      "num_tokens": 139568261.0,
      "step": 74850
    },
    {
      "entropy": 5.959741640090942,
      "epoch": 6.6102967149417164,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00016780952163171703,
      "loss": 5.0618,
      "mean_token_accuracy": 0.24421593695878982,
      "num_tokens": 139577518.0,
      "step": 74855
    },
    {
      "entropy": 6.049164724349976,
      "epoch": 6.610738255033557,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00016778183448317,
      "loss": 5.2071,
      "mean_token_accuracy": 0.23030039370059968,
      "num_tokens": 139586614.0,
      "step": 74860
    },
    {
      "entropy": 6.06880555152893,
      "epoch": 6.611179795125397,
      "grad_norm": 1.5,
      "learning_rate": 0.00016775414943496586,
      "loss": 5.1989,
      "mean_token_accuracy": 0.23213960081338883,
      "num_tokens": 139596155.0,
      "step": 74865
    },
    {
      "entropy": 6.042067956924439,
      "epoch": 6.6116213352172375,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016772646648764708,
      "loss": 5.2344,
      "mean_token_accuracy": 0.22153654545545579,
      "num_tokens": 139605020.0,
      "step": 74870
    },
    {
      "entropy": 6.097408580780029,
      "epoch": 6.612062875309078,
      "grad_norm": 1.578125,
      "learning_rate": 0.0001676987856417557,
      "loss": 5.2624,
      "mean_token_accuracy": 0.22370718866586686,
      "num_tokens": 139614263.0,
      "step": 74875
    },
    {
      "entropy": 6.05946044921875,
      "epoch": 6.6125044154009185,
      "grad_norm": 1.46875,
      "learning_rate": 0.00016767110689783405,
      "loss": 5.2255,
      "mean_token_accuracy": 0.22298363596200943,
      "num_tokens": 139623369.0,
      "step": 74880
    },
    {
      "entropy": 6.05640139579773,
      "epoch": 6.6129459554927585,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016764343025642438,
      "loss": 5.2187,
      "mean_token_accuracy": 0.22516158372163772,
      "num_tokens": 139633115.0,
      "step": 74885
    },
    {
      "entropy": 5.992241954803466,
      "epoch": 6.613387495584599,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016761575571806885,
      "loss": 5.1219,
      "mean_token_accuracy": 0.2333174929022789,
      "num_tokens": 139642940.0,
      "step": 74890
    },
    {
      "entropy": 6.097334861755371,
      "epoch": 6.6138290356764395,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001675880832833097,
      "loss": 5.2892,
      "mean_token_accuracy": 0.22668772637844087,
      "num_tokens": 139652831.0,
      "step": 74895
    },
    {
      "entropy": 6.100159645080566,
      "epoch": 6.61427057576828,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016756041295268878,
      "loss": 5.2583,
      "mean_token_accuracy": 0.210552479326725,
      "num_tokens": 139662391.0,
      "step": 74900
    },
    {
      "entropy": 6.072394943237304,
      "epoch": 6.6147121158601205,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016753274472674836,
      "loss": 5.2197,
      "mean_token_accuracy": 0.22707052677869796,
      "num_tokens": 139671616.0,
      "step": 74905
    },
    {
      "entropy": 6.055575752258301,
      "epoch": 6.615153655951961,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00016750507860603033,
      "loss": 5.2093,
      "mean_token_accuracy": 0.22054322957992553,
      "num_tokens": 139681508.0,
      "step": 74910
    },
    {
      "entropy": 6.042250490188598,
      "epoch": 6.615595196043801,
      "grad_norm": 1.484375,
      "learning_rate": 0.00016747741459107672,
      "loss": 5.2574,
      "mean_token_accuracy": 0.22483226358890535,
      "num_tokens": 139691221.0,
      "step": 74915
    },
    {
      "entropy": 6.0815284729003904,
      "epoch": 6.616036736135641,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00016744975268242934,
      "loss": 5.2444,
      "mean_token_accuracy": 0.2235145553946495,
      "num_tokens": 139701129.0,
      "step": 74920
    },
    {
      "entropy": 6.142758846282959,
      "epoch": 6.616478276227482,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016742209288063013,
      "loss": 5.3283,
      "mean_token_accuracy": 0.21779675632715226,
      "num_tokens": 139711099.0,
      "step": 74925
    },
    {
      "entropy": 6.019256210327148,
      "epoch": 6.616919816319322,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00016739443518622095,
      "loss": 5.0948,
      "mean_token_accuracy": 0.2428218111395836,
      "num_tokens": 139720001.0,
      "step": 74930
    },
    {
      "entropy": 6.004851865768432,
      "epoch": 6.617361356411163,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001673667795997436,
      "loss": 5.1172,
      "mean_token_accuracy": 0.23968183994293213,
      "num_tokens": 139729031.0,
      "step": 74935
    },
    {
      "entropy": 6.008338689804077,
      "epoch": 6.617802896503003,
      "grad_norm": 1.46875,
      "learning_rate": 0.00016733912612173986,
      "loss": 5.2447,
      "mean_token_accuracy": 0.22549937218427657,
      "num_tokens": 139738371.0,
      "step": 74940
    },
    {
      "entropy": 6.046562337875367,
      "epoch": 6.618244436594843,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016731147475275134,
      "loss": 5.229,
      "mean_token_accuracy": 0.2289349153637886,
      "num_tokens": 139747924.0,
      "step": 74945
    },
    {
      "entropy": 6.092154884338379,
      "epoch": 6.618685976686683,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00016728382549331978,
      "loss": 5.1984,
      "mean_token_accuracy": 0.22881451547145842,
      "num_tokens": 139757131.0,
      "step": 74950
    },
    {
      "entropy": 6.095964050292968,
      "epoch": 6.619127516778524,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001672561783439868,
      "loss": 5.2516,
      "mean_token_accuracy": 0.22571742236614228,
      "num_tokens": 139765932.0,
      "step": 74955
    },
    {
      "entropy": 6.042350625991821,
      "epoch": 6.619569056870364,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00016722853330529414,
      "loss": 5.199,
      "mean_token_accuracy": 0.23045468032360078,
      "num_tokens": 139776083.0,
      "step": 74960
    },
    {
      "entropy": 6.054132032394409,
      "epoch": 6.620010596962204,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016720089037778314,
      "loss": 5.193,
      "mean_token_accuracy": 0.23000480830669404,
      "num_tokens": 139784684.0,
      "step": 74965
    },
    {
      "entropy": 6.076971960067749,
      "epoch": 6.620452137054045,
      "grad_norm": 1.46875,
      "learning_rate": 0.00016717324956199532,
      "loss": 5.2051,
      "mean_token_accuracy": 0.23014216423034667,
      "num_tokens": 139793812.0,
      "step": 74970
    },
    {
      "entropy": 6.002180051803589,
      "epoch": 6.620893677145885,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00016714561085847226,
      "loss": 5.188,
      "mean_token_accuracy": 0.22946558594703675,
      "num_tokens": 139803308.0,
      "step": 74975
    },
    {
      "entropy": 6.061569929122925,
      "epoch": 6.621335217237725,
      "grad_norm": 1.46875,
      "learning_rate": 0.00016711797426775544,
      "loss": 5.2338,
      "mean_token_accuracy": 0.2252446487545967,
      "num_tokens": 139811745.0,
      "step": 74980
    },
    {
      "entropy": 6.096099328994751,
      "epoch": 6.621776757329566,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016709033979038607,
      "loss": 5.2845,
      "mean_token_accuracy": 0.21353610157966613,
      "num_tokens": 139821117.0,
      "step": 74985
    },
    {
      "entropy": 6.070365333557129,
      "epoch": 6.622218297421406,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00016706270742690551,
      "loss": 5.1996,
      "mean_token_accuracy": 0.22596663534641265,
      "num_tokens": 139830131.0,
      "step": 74990
    },
    {
      "entropy": 6.0857686519622805,
      "epoch": 6.622659837513246,
      "grad_norm": 1.375,
      "learning_rate": 0.0001670350771778552,
      "loss": 5.2425,
      "mean_token_accuracy": 0.22756047993898393,
      "num_tokens": 139839748.0,
      "step": 74995
    },
    {
      "entropy": 6.055009555816651,
      "epoch": 6.623101377605087,
      "grad_norm": 1.5,
      "learning_rate": 0.0001670074490437764,
      "loss": 5.1696,
      "mean_token_accuracy": 0.23866534978151321,
      "num_tokens": 139847842.0,
      "step": 75000
    },
    {
      "epoch": 6.623101377605087,
      "eval_entropy": 5.733027477091598,
      "eval_loss": 5.280413627624512,
      "eval_mean_token_accuracy": 0.23188823806470005,
      "eval_num_tokens": 139847842.0,
      "eval_runtime": 20.3992,
      "eval_samples_per_second": 1731.192,
      "eval_steps_per_second": 216.43,
      "step": 75000
    }
  ],
  "logging_steps": 5,
  "max_steps": 113230,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 3000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.073327557562368e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}