eus-latn-10mb-after-ppt-Dp-100mb-ckpt500_seed3407/checkpoint-9000/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.7947721653126104,
  "eval_steps": 3000,
  "global_step": 9000,
  "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
    }
  ],
  "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": 3680638699438080.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}