{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.4668352460610776,
  "eval_steps": 3000,
  "global_step": 6000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 7.525209426879883,
      "epoch": 0.00038902937171756465,
      "grad_norm": 1.1640625,
      "learning_rate": 2e-06,
      "loss": 7.1164,
      "mean_token_accuracy": 0.09126382395625114,
      "num_tokens": 8095.0,
      "step": 5
    },
    {
      "entropy": 7.524957656860352,
      "epoch": 0.0007780587434351293,
      "grad_norm": 1.046875,
      "learning_rate": 4.5e-06,
      "loss": 7.1341,
      "mean_token_accuracy": 0.09478832259774209,
      "num_tokens": 17329.0,
      "step": 10
    },
    {
      "entropy": 7.446990966796875,
      "epoch": 0.001167088115152694,
      "grad_norm": 1.1171875,
      "learning_rate": 7e-06,
      "loss": 6.9396,
      "mean_token_accuracy": 0.09858316257596016,
      "num_tokens": 25729.0,
      "step": 15
    },
    {
      "entropy": 7.479873752593994,
      "epoch": 0.0015561174868702586,
      "grad_norm": 1.1328125,
      "learning_rate": 9.5e-06,
      "loss": 7.0656,
      "mean_token_accuracy": 0.10022840201854706,
      "num_tokens": 33980.0,
      "step": 20
    },
    {
      "entropy": 7.4935675144195555,
      "epoch": 0.0019451468585878235,
      "grad_norm": 1.171875,
      "learning_rate": 1.2e-05,
      "loss": 6.9001,
      "mean_token_accuracy": 0.10436431169509888,
      "num_tokens": 41675.0,
      "step": 25
    },
    {
      "entropy": 7.4462706565856935,
      "epoch": 0.002334176230305388,
      "grad_norm": 1.0859375,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 6.9849,
      "mean_token_accuracy": 0.09949584901332856,
      "num_tokens": 50277.0,
      "step": 30
    },
    {
      "entropy": 7.490463447570801,
      "epoch": 0.0027232056020229526,
      "grad_norm": 1.1484375,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 7.0679,
      "mean_token_accuracy": 0.09498238340020179,
      "num_tokens": 58790.0,
      "step": 35
    },
    {
      "entropy": 7.469117593765259,
      "epoch": 0.0031122349737405172,
      "grad_norm": 1.078125,
      "learning_rate": 1.95e-05,
      "loss": 7.1151,
      "mean_token_accuracy": 0.09345170259475707,
      "num_tokens": 67247.0,
      "step": 40
    },
    {
      "entropy": 7.437713384628296,
      "epoch": 0.003501264345458082,
      "grad_norm": 1.1328125,
      "learning_rate": 2.2e-05,
      "loss": 6.9327,
      "mean_token_accuracy": 0.09667690098285675,
      "num_tokens": 75174.0,
      "step": 45
    },
    {
      "entropy": 7.389952373504639,
      "epoch": 0.003890293717175647,
      "grad_norm": 1.0625,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 6.9055,
      "mean_token_accuracy": 0.10531265288591385,
      "num_tokens": 83203.0,
      "step": 50
    },
    {
      "entropy": 7.395076322555542,
      "epoch": 0.004279323088893211,
      "grad_norm": 0.98046875,
      "learning_rate": 2.7e-05,
      "loss": 6.9499,
      "mean_token_accuracy": 0.10203081965446473,
      "num_tokens": 91798.0,
      "step": 55
    },
    {
      "entropy": 7.40048041343689,
      "epoch": 0.004668352460610776,
      "grad_norm": 0.96875,
      "learning_rate": 2.95e-05,
      "loss": 6.9907,
      "mean_token_accuracy": 0.09655389487743378,
      "num_tokens": 100869.0,
      "step": 60
    },
    {
      "entropy": 7.454555511474609,
      "epoch": 0.0050573818323283405,
      "grad_norm": 1.03125,
      "learning_rate": 3.2e-05,
      "loss": 6.9875,
      "mean_token_accuracy": 0.09880984500050545,
      "num_tokens": 109697.0,
      "step": 65
    },
    {
      "entropy": 7.498321151733398,
      "epoch": 0.005446411204045905,
      "grad_norm": 1.078125,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 7.1231,
      "mean_token_accuracy": 0.09559949561953544,
      "num_tokens": 118656.0,
      "step": 70
    },
    {
      "entropy": 7.496587228775025,
      "epoch": 0.00583544057576347,
      "grad_norm": 1.03125,
      "learning_rate": 3.7e-05,
      "loss": 7.1172,
      "mean_token_accuracy": 0.09125430211424827,
      "num_tokens": 127748.0,
      "step": 75
    },
    {
      "entropy": 7.504852151870727,
      "epoch": 0.0062244699474810344,
      "grad_norm": 1.1015625,
      "learning_rate": 3.95e-05,
      "loss": 7.0387,
      "mean_token_accuracy": 0.1014796257019043,
      "num_tokens": 136621.0,
      "step": 80
    },
    {
      "entropy": 7.48614993095398,
      "epoch": 0.006613499319198599,
      "grad_norm": 1.1171875,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 6.9412,
      "mean_token_accuracy": 0.102222141623497,
      "num_tokens": 145793.0,
      "step": 85
    },
    {
      "entropy": 7.3585365295410154,
      "epoch": 0.007002528690916164,
      "grad_norm": 1.0078125,
      "learning_rate": 4.45e-05,
      "loss": 6.8998,
      "mean_token_accuracy": 0.09987122416496277,
      "num_tokens": 154466.0,
      "step": 90
    },
    {
      "entropy": 7.381277418136596,
      "epoch": 0.007391558062633729,
      "grad_norm": 1.03125,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 7.0817,
      "mean_token_accuracy": 0.09664711579680443,
      "num_tokens": 163297.0,
      "step": 95
    },
    {
      "entropy": 7.4567954540252686,
      "epoch": 0.007780587434351294,
      "grad_norm": 1.0703125,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 6.9582,
      "mean_token_accuracy": 0.0995919868350029,
      "num_tokens": 171747.0,
      "step": 100
    },
    {
      "entropy": 7.461597013473511,
      "epoch": 0.008169616806068859,
      "grad_norm": 1.0390625,
      "learning_rate": 5.2e-05,
      "loss": 6.9737,
      "mean_token_accuracy": 0.09625533521175385,
      "num_tokens": 180499.0,
      "step": 105
    },
    {
      "entropy": 7.4339855194091795,
      "epoch": 0.008558646177786422,
      "grad_norm": 1.0703125,
      "learning_rate": 5.45e-05,
      "loss": 6.9069,
      "mean_token_accuracy": 0.10029028877615928,
      "num_tokens": 189227.0,
      "step": 110
    },
    {
      "entropy": 7.355867671966553,
      "epoch": 0.008947675549503988,
      "grad_norm": 0.98046875,
      "learning_rate": 5.7e-05,
      "loss": 6.986,
      "mean_token_accuracy": 0.10084915980696678,
      "num_tokens": 198388.0,
      "step": 115
    },
    {
      "entropy": 7.4226056098937985,
      "epoch": 0.009336704921221552,
      "grad_norm": 1.015625,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 6.9883,
      "mean_token_accuracy": 0.10193781480193138,
      "num_tokens": 206854.0,
      "step": 120
    },
    {
      "entropy": 7.44698486328125,
      "epoch": 0.009725734292939117,
      "grad_norm": 1.109375,
      "learning_rate": 6.2e-05,
      "loss": 6.9597,
      "mean_token_accuracy": 0.10039241313934326,
      "num_tokens": 215506.0,
      "step": 125
    },
    {
      "entropy": 7.457604932785034,
      "epoch": 0.010114763664656681,
      "grad_norm": 1.1015625,
      "learning_rate": 6.450000000000001e-05,
      "loss": 6.9474,
      "mean_token_accuracy": 0.09563101530075073,
      "num_tokens": 223717.0,
      "step": 130
    },
    {
      "entropy": 7.4058609962463375,
      "epoch": 0.010503793036374246,
      "grad_norm": 0.9921875,
      "learning_rate": 6.7e-05,
      "loss": 6.9154,
      "mean_token_accuracy": 0.10078920200467109,
      "num_tokens": 232480.0,
      "step": 135
    },
    {
      "entropy": 7.396328163146973,
      "epoch": 0.01089282240809181,
      "grad_norm": 1.125,
      "learning_rate": 6.950000000000001e-05,
      "loss": 6.9653,
      "mean_token_accuracy": 0.10731882452964783,
      "num_tokens": 241350.0,
      "step": 140
    },
    {
      "entropy": 7.449981260299682,
      "epoch": 0.011281851779809376,
      "grad_norm": 1.1796875,
      "learning_rate": 7.2e-05,
      "loss": 6.9697,
      "mean_token_accuracy": 0.10272065401077271,
      "num_tokens": 250610.0,
      "step": 145
    },
    {
      "entropy": 7.4221306324005125,
      "epoch": 0.01167088115152694,
      "grad_norm": 1.0390625,
      "learning_rate": 7.45e-05,
      "loss": 6.9118,
      "mean_token_accuracy": 0.10345016792416573,
      "num_tokens": 259672.0,
      "step": 150
    },
    {
      "entropy": 7.369914913177491,
      "epoch": 0.012059910523244505,
      "grad_norm": 1.140625,
      "learning_rate": 7.7e-05,
      "loss": 6.9904,
      "mean_token_accuracy": 0.09594599083065987,
      "num_tokens": 267826.0,
      "step": 155
    },
    {
      "entropy": 7.477785062789917,
      "epoch": 0.012448939894962069,
      "grad_norm": 1.1328125,
      "learning_rate": 7.950000000000001e-05,
      "loss": 6.995,
      "mean_token_accuracy": 0.09624672085046768,
      "num_tokens": 277324.0,
      "step": 160
    },
    {
      "entropy": 7.4215576171875,
      "epoch": 0.012837969266679634,
      "grad_norm": 1.171875,
      "learning_rate": 8.2e-05,
      "loss": 6.8296,
      "mean_token_accuracy": 0.10975822508335113,
      "num_tokens": 285482.0,
      "step": 165
    },
    {
      "entropy": 7.35403299331665,
      "epoch": 0.013226998638397198,
      "grad_norm": 1.234375,
      "learning_rate": 8.450000000000001e-05,
      "loss": 6.893,
      "mean_token_accuracy": 0.10332747399806977,
      "num_tokens": 293684.0,
      "step": 170
    },
    {
      "entropy": 7.344775819778443,
      "epoch": 0.013616028010114764,
      "grad_norm": 1.078125,
      "learning_rate": 8.7e-05,
      "loss": 6.8705,
      "mean_token_accuracy": 0.10466427579522133,
      "num_tokens": 302605.0,
      "step": 175
    },
    {
      "entropy": 7.469928503036499,
      "epoch": 0.014005057381832327,
      "grad_norm": 1.1015625,
      "learning_rate": 8.95e-05,
      "loss": 6.9479,
      "mean_token_accuracy": 0.10224307030439377,
      "num_tokens": 310849.0,
      "step": 180
    },
    {
      "entropy": 7.377802467346191,
      "epoch": 0.014394086753549893,
      "grad_norm": 1.1015625,
      "learning_rate": 9.2e-05,
      "loss": 6.9647,
      "mean_token_accuracy": 0.09813596233725548,
      "num_tokens": 319752.0,
      "step": 185
    },
    {
      "entropy": 7.5093926906585695,
      "epoch": 0.014783116125267459,
      "grad_norm": 1.296875,
      "learning_rate": 9.45e-05,
      "loss": 7.0233,
      "mean_token_accuracy": 0.09324661493301392,
      "num_tokens": 328009.0,
      "step": 190
    },
    {
      "entropy": 7.352363014221192,
      "epoch": 0.015172145496985022,
      "grad_norm": 1.109375,
      "learning_rate": 9.7e-05,
      "loss": 6.909,
      "mean_token_accuracy": 0.10263535231351853,
      "num_tokens": 337414.0,
      "step": 195
    },
    {
      "entropy": 7.411336231231689,
      "epoch": 0.015561174868702588,
      "grad_norm": 1.15625,
      "learning_rate": 9.95e-05,
      "loss": 6.9275,
      "mean_token_accuracy": 0.10379069447517394,
      "num_tokens": 345419.0,
      "step": 200
    },
    {
      "entropy": 7.4287652492523195,
      "epoch": 0.01595020424042015,
      "grad_norm": 1.203125,
      "learning_rate": 0.000102,
      "loss": 6.9079,
      "mean_token_accuracy": 0.10015243589878083,
      "num_tokens": 353435.0,
      "step": 205
    },
    {
      "entropy": 7.346658992767334,
      "epoch": 0.016339233612137717,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00010449999999999999,
      "loss": 6.9221,
      "mean_token_accuracy": 0.1055522657930851,
      "num_tokens": 361840.0,
      "step": 210
    },
    {
      "entropy": 7.52528715133667,
      "epoch": 0.016728262983855283,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000107,
      "loss": 7.0276,
      "mean_token_accuracy": 0.09967184960842132,
      "num_tokens": 370873.0,
      "step": 215
    },
    {
      "entropy": 7.3679437160491945,
      "epoch": 0.017117292355572845,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001095,
      "loss": 6.9343,
      "mean_token_accuracy": 0.09932355582714081,
      "num_tokens": 379558.0,
      "step": 220
    },
    {
      "entropy": 7.469588994979858,
      "epoch": 0.01750632172729041,
      "grad_norm": 1.234375,
      "learning_rate": 0.000112,
      "loss": 6.9429,
      "mean_token_accuracy": 0.10046415254473687,
      "num_tokens": 387870.0,
      "step": 225
    },
    {
      "entropy": 7.386438465118408,
      "epoch": 0.017895351099007976,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001145,
      "loss": 7.0274,
      "mean_token_accuracy": 0.10568641722202302,
      "num_tokens": 396442.0,
      "step": 230
    },
    {
      "entropy": 7.447541284561157,
      "epoch": 0.01828438047072554,
      "grad_norm": 1.21875,
      "learning_rate": 0.00011700000000000001,
      "loss": 6.9675,
      "mean_token_accuracy": 0.10170508250594139,
      "num_tokens": 404764.0,
      "step": 235
    },
    {
      "entropy": 7.413917064666748,
      "epoch": 0.018673409842443103,
      "grad_norm": 1.203125,
      "learning_rate": 0.00011949999999999999,
      "loss": 6.9034,
      "mean_token_accuracy": 0.10267666056752205,
      "num_tokens": 413355.0,
      "step": 240
    },
    {
      "entropy": 7.331215286254883,
      "epoch": 0.01906243921416067,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000122,
      "loss": 6.8107,
      "mean_token_accuracy": 0.10804717987775803,
      "num_tokens": 421868.0,
      "step": 245
    },
    {
      "entropy": 7.342196226119995,
      "epoch": 0.019451468585878234,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0001245,
      "loss": 6.8674,
      "mean_token_accuracy": 0.10438908636569977,
      "num_tokens": 431106.0,
      "step": 250
    },
    {
      "entropy": 7.457882690429687,
      "epoch": 0.0198404979575958,
      "grad_norm": 1.28125,
      "learning_rate": 0.000127,
      "loss": 7.0068,
      "mean_token_accuracy": 0.09497510865330697,
      "num_tokens": 439459.0,
      "step": 255
    },
    {
      "entropy": 7.371812105178833,
      "epoch": 0.020229527329313362,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001295,
      "loss": 6.8334,
      "mean_token_accuracy": 0.10420059189200401,
      "num_tokens": 447943.0,
      "step": 260
    },
    {
      "entropy": 7.375357580184937,
      "epoch": 0.020618556701030927,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000132,
      "loss": 6.9826,
      "mean_token_accuracy": 0.09710169360041618,
      "num_tokens": 456691.0,
      "step": 265
    },
    {
      "entropy": 7.364774465560913,
      "epoch": 0.021007586072748493,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00013450000000000002,
      "loss": 6.8583,
      "mean_token_accuracy": 0.10362368077039719,
      "num_tokens": 466388.0,
      "step": 270
    },
    {
      "entropy": 7.392491149902344,
      "epoch": 0.02139661544446606,
      "grad_norm": 1.171875,
      "learning_rate": 0.00013700000000000002,
      "loss": 6.9669,
      "mean_token_accuracy": 0.10099864155054092,
      "num_tokens": 474544.0,
      "step": 275
    },
    {
      "entropy": 7.414118957519531,
      "epoch": 0.02178564481618362,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001395,
      "loss": 6.9797,
      "mean_token_accuracy": 0.10356927365064621,
      "num_tokens": 483561.0,
      "step": 280
    },
    {
      "entropy": 7.296314764022827,
      "epoch": 0.022174674187901186,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00014199999999999998,
      "loss": 6.8018,
      "mean_token_accuracy": 0.10049450471997261,
      "num_tokens": 492036.0,
      "step": 285
    },
    {
      "entropy": 7.479441213607788,
      "epoch": 0.02256370355961875,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001445,
      "loss": 6.9413,
      "mean_token_accuracy": 0.10272752419114113,
      "num_tokens": 500978.0,
      "step": 290
    },
    {
      "entropy": 7.315587186813355,
      "epoch": 0.022952732931336317,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000147,
      "loss": 6.9848,
      "mean_token_accuracy": 0.10035317316651345,
      "num_tokens": 509719.0,
      "step": 295
    },
    {
      "entropy": 7.50186858177185,
      "epoch": 0.02334176230305388,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001495,
      "loss": 6.8866,
      "mean_token_accuracy": 0.1016937792301178,
      "num_tokens": 518422.0,
      "step": 300
    },
    {
      "entropy": 7.289468431472779,
      "epoch": 0.023730791674771445,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000152,
      "loss": 6.9335,
      "mean_token_accuracy": 0.09595205262303352,
      "num_tokens": 527172.0,
      "step": 305
    },
    {
      "entropy": 7.423691844940185,
      "epoch": 0.02411982104648901,
      "grad_norm": 1.171875,
      "learning_rate": 0.00015450000000000001,
      "loss": 6.8618,
      "mean_token_accuracy": 0.10495315194129944,
      "num_tokens": 536070.0,
      "step": 310
    },
    {
      "entropy": 7.324627017974853,
      "epoch": 0.024508850418206576,
      "grad_norm": 1.265625,
      "learning_rate": 0.000157,
      "loss": 6.8709,
      "mean_token_accuracy": 0.10348411202430725,
      "num_tokens": 544556.0,
      "step": 315
    },
    {
      "entropy": 7.411664628982544,
      "epoch": 0.024897879789924138,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001595,
      "loss": 6.8337,
      "mean_token_accuracy": 0.10683412104845047,
      "num_tokens": 553122.0,
      "step": 320
    },
    {
      "entropy": 7.352542924880981,
      "epoch": 0.025286909161641703,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000162,
      "loss": 6.9213,
      "mean_token_accuracy": 0.09781907498836517,
      "num_tokens": 562265.0,
      "step": 325
    },
    {
      "entropy": 7.3105535984039305,
      "epoch": 0.02567593853335927,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00016450000000000001,
      "loss": 6.8794,
      "mean_token_accuracy": 0.1058124616742134,
      "num_tokens": 570938.0,
      "step": 330
    },
    {
      "entropy": 7.363333702087402,
      "epoch": 0.026064967905076834,
      "grad_norm": 1.203125,
      "learning_rate": 0.00016700000000000002,
      "loss": 6.8159,
      "mean_token_accuracy": 0.10493286699056625,
      "num_tokens": 578568.0,
      "step": 335
    },
    {
      "entropy": 7.4085417747497555,
      "epoch": 0.026453997276794396,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016950000000000003,
      "loss": 6.8823,
      "mean_token_accuracy": 0.10770140811800957,
      "num_tokens": 587013.0,
      "step": 340
    },
    {
      "entropy": 7.380347776412964,
      "epoch": 0.026843026648511962,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00017199999999999998,
      "loss": 6.9395,
      "mean_token_accuracy": 0.10302414670586586,
      "num_tokens": 595354.0,
      "step": 345
    },
    {
      "entropy": 7.401273250579834,
      "epoch": 0.027232056020229527,
      "grad_norm": 1.1875,
      "learning_rate": 0.00017449999999999999,
      "loss": 6.9143,
      "mean_token_accuracy": 0.09726170152425766,
      "num_tokens": 603971.0,
      "step": 350
    },
    {
      "entropy": 7.264584875106811,
      "epoch": 0.027621085391947093,
      "grad_norm": 1.109375,
      "learning_rate": 0.000177,
      "loss": 6.8246,
      "mean_token_accuracy": 0.10611001029610634,
      "num_tokens": 612097.0,
      "step": 355
    },
    {
      "entropy": 7.374355459213257,
      "epoch": 0.028010114763664655,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0001795,
      "loss": 6.8914,
      "mean_token_accuracy": 0.1076267570257187,
      "num_tokens": 621445.0,
      "step": 360
    },
    {
      "entropy": 7.367084646224976,
      "epoch": 0.02839914413538222,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000182,
      "loss": 6.9642,
      "mean_token_accuracy": 0.10026197507977486,
      "num_tokens": 630041.0,
      "step": 365
    },
    {
      "entropy": 7.286877632141113,
      "epoch": 0.028788173507099786,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001845,
      "loss": 6.793,
      "mean_token_accuracy": 0.10361501201987267,
      "num_tokens": 638317.0,
      "step": 370
    },
    {
      "entropy": 7.358971881866455,
      "epoch": 0.02917720287881735,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000187,
      "loss": 6.9153,
      "mean_token_accuracy": 0.10330267027020454,
      "num_tokens": 647297.0,
      "step": 375
    },
    {
      "entropy": 7.467724800109863,
      "epoch": 0.029566232250534917,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001895,
      "loss": 6.9596,
      "mean_token_accuracy": 0.10418846607208251,
      "num_tokens": 655309.0,
      "step": 380
    },
    {
      "entropy": 7.318552494049072,
      "epoch": 0.02995526162225248,
      "grad_norm": 1.203125,
      "learning_rate": 0.000192,
      "loss": 6.8423,
      "mean_token_accuracy": 0.09740504622459412,
      "num_tokens": 663879.0,
      "step": 385
    },
    {
      "entropy": 7.267914342880249,
      "epoch": 0.030344290993970045,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001945,
      "loss": 6.7288,
      "mean_token_accuracy": 0.10926953107118606,
      "num_tokens": 672913.0,
      "step": 390
    },
    {
      "entropy": 7.249853038787842,
      "epoch": 0.03073332036568761,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019700000000000002,
      "loss": 6.8682,
      "mean_token_accuracy": 0.10577411428093911,
      "num_tokens": 682282.0,
      "step": 395
    },
    {
      "entropy": 7.347216081619263,
      "epoch": 0.031122349737405176,
      "grad_norm": 1.125,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.8777,
      "mean_token_accuracy": 0.10061730667948723,
      "num_tokens": 691432.0,
      "step": 400
    },
    {
      "entropy": 7.2628637790679935,
      "epoch": 0.03151137910912274,
      "grad_norm": 1.28125,
      "learning_rate": 0.000202,
      "loss": 6.8792,
      "mean_token_accuracy": 0.10118450745940208,
      "num_tokens": 700639.0,
      "step": 405
    },
    {
      "entropy": 7.387156391143799,
      "epoch": 0.0319004084808403,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.946,
      "mean_token_accuracy": 0.0988673098385334,
      "num_tokens": 709200.0,
      "step": 410
    },
    {
      "entropy": 7.180784082412719,
      "epoch": 0.03228943785255787,
      "grad_norm": 1.171875,
      "learning_rate": 0.000207,
      "loss": 6.7769,
      "mean_token_accuracy": 0.10579730719327926,
      "num_tokens": 717266.0,
      "step": 415
    },
    {
      "entropy": 7.3383214473724365,
      "epoch": 0.032678467224275434,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002095,
      "loss": 6.7761,
      "mean_token_accuracy": 0.10209827125072479,
      "num_tokens": 725799.0,
      "step": 420
    },
    {
      "entropy": 7.2920979976654055,
      "epoch": 0.033067496595993,
      "grad_norm": 1.328125,
      "learning_rate": 0.000212,
      "loss": 6.9062,
      "mean_token_accuracy": 0.10427533909678459,
      "num_tokens": 734375.0,
      "step": 425
    },
    {
      "entropy": 7.3845654964447025,
      "epoch": 0.033456525967710565,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002145,
      "loss": 6.9307,
      "mean_token_accuracy": 0.10116986557841301,
      "num_tokens": 743453.0,
      "step": 430
    },
    {
      "entropy": 7.256317663192749,
      "epoch": 0.033845555339428124,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.7835,
      "mean_token_accuracy": 0.1099538803100586,
      "num_tokens": 752443.0,
      "step": 435
    },
    {
      "entropy": 7.239885091781616,
      "epoch": 0.03423458471114569,
      "grad_norm": 1.25,
      "learning_rate": 0.0002195,
      "loss": 6.7755,
      "mean_token_accuracy": 0.103090950101614,
      "num_tokens": 761294.0,
      "step": 440
    },
    {
      "entropy": 7.315788888931275,
      "epoch": 0.034623614082863255,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000222,
      "loss": 6.7633,
      "mean_token_accuracy": 0.10775340422987938,
      "num_tokens": 769489.0,
      "step": 445
    },
    {
      "entropy": 7.27491250038147,
      "epoch": 0.03501264345458082,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002245,
      "loss": 6.8251,
      "mean_token_accuracy": 0.1056965485215187,
      "num_tokens": 777836.0,
      "step": 450
    },
    {
      "entropy": 7.317339181900024,
      "epoch": 0.035401672826298386,
      "grad_norm": 1.25,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.8512,
      "mean_token_accuracy": 0.09845920503139496,
      "num_tokens": 786543.0,
      "step": 455
    },
    {
      "entropy": 7.298014497756958,
      "epoch": 0.03579070219801595,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.9104,
      "mean_token_accuracy": 0.10431987568736076,
      "num_tokens": 794525.0,
      "step": 460
    },
    {
      "entropy": 7.248893308639526,
      "epoch": 0.03617973156973352,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.8222,
      "mean_token_accuracy": 0.1065070852637291,
      "num_tokens": 803537.0,
      "step": 465
    },
    {
      "entropy": 7.270473957061768,
      "epoch": 0.03656876094145108,
      "grad_norm": 1.15625,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.7685,
      "mean_token_accuracy": 0.10664488673210144,
      "num_tokens": 811706.0,
      "step": 470
    },
    {
      "entropy": 7.247415494918823,
      "epoch": 0.03695779031316864,
      "grad_norm": 1.15625,
      "learning_rate": 0.000237,
      "loss": 6.8359,
      "mean_token_accuracy": 0.10338398665189744,
      "num_tokens": 820182.0,
      "step": 475
    },
    {
      "entropy": 7.244891786575318,
      "epoch": 0.03734681968488621,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002395,
      "loss": 6.7711,
      "mean_token_accuracy": 0.10853901281952857,
      "num_tokens": 829808.0,
      "step": 480
    },
    {
      "entropy": 7.287813949584961,
      "epoch": 0.03773584905660377,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000242,
      "loss": 6.721,
      "mean_token_accuracy": 0.10383491292595863,
      "num_tokens": 838838.0,
      "step": 485
    },
    {
      "entropy": 7.207957983016968,
      "epoch": 0.03812487842832134,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002445,
      "loss": 6.8069,
      "mean_token_accuracy": 0.1081025905907154,
      "num_tokens": 847800.0,
      "step": 490
    },
    {
      "entropy": 7.361988687515259,
      "epoch": 0.0385139078000389,
      "grad_norm": 1.21875,
      "learning_rate": 0.000247,
      "loss": 6.9005,
      "mean_token_accuracy": 0.10296034291386605,
      "num_tokens": 856503.0,
      "step": 495
    },
    {
      "entropy": 7.193755149841309,
      "epoch": 0.03890293717175647,
      "grad_norm": 1.375,
      "learning_rate": 0.0002495,
      "loss": 6.7864,
      "mean_token_accuracy": 0.10174561962485314,
      "num_tokens": 865245.0,
      "step": 500
    },
    {
      "entropy": 7.283534049987793,
      "epoch": 0.039291966543474034,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000252,
      "loss": 6.8082,
      "mean_token_accuracy": 0.10216192156076431,
      "num_tokens": 874485.0,
      "step": 505
    },
    {
      "entropy": 7.117496156692505,
      "epoch": 0.0396809959151916,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002545,
      "loss": 6.6771,
      "mean_token_accuracy": 0.11026940196752548,
      "num_tokens": 883301.0,
      "step": 510
    },
    {
      "entropy": 7.131668567657471,
      "epoch": 0.04007002528690916,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000257,
      "loss": 6.7448,
      "mean_token_accuracy": 0.10374188125133514,
      "num_tokens": 892387.0,
      "step": 515
    },
    {
      "entropy": 7.24387755393982,
      "epoch": 0.040459054658626724,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002595,
      "loss": 6.732,
      "mean_token_accuracy": 0.11151306107640266,
      "num_tokens": 901522.0,
      "step": 520
    },
    {
      "entropy": 7.128845930099487,
      "epoch": 0.04084808403034429,
      "grad_norm": 1.1875,
      "learning_rate": 0.000262,
      "loss": 6.6974,
      "mean_token_accuracy": 0.10741285160183907,
      "num_tokens": 909848.0,
      "step": 525
    },
    {
      "entropy": 7.312799263000488,
      "epoch": 0.041237113402061855,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.9168,
      "mean_token_accuracy": 0.1004093937575817,
      "num_tokens": 919079.0,
      "step": 530
    },
    {
      "entropy": 7.166108179092407,
      "epoch": 0.04162614277377942,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.6571,
      "mean_token_accuracy": 0.1108947217464447,
      "num_tokens": 927807.0,
      "step": 535
    },
    {
      "entropy": 7.130585718154907,
      "epoch": 0.042015172145496986,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.7716,
      "mean_token_accuracy": 0.10743746683001518,
      "num_tokens": 936498.0,
      "step": 540
    },
    {
      "entropy": 7.2852715969085695,
      "epoch": 0.04240420151721455,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.8486,
      "mean_token_accuracy": 0.1115134410560131,
      "num_tokens": 945436.0,
      "step": 545
    },
    {
      "entropy": 7.238163709640503,
      "epoch": 0.04279323088893212,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002745,
      "loss": 6.76,
      "mean_token_accuracy": 0.10598877593874931,
      "num_tokens": 953990.0,
      "step": 550
    },
    {
      "entropy": 7.096475458145141,
      "epoch": 0.04318226026064968,
      "grad_norm": 1.296875,
      "learning_rate": 0.000277,
      "loss": 6.6979,
      "mean_token_accuracy": 0.10942429900169373,
      "num_tokens": 962008.0,
      "step": 555
    },
    {
      "entropy": 7.238119173049927,
      "epoch": 0.04357128963236724,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002795,
      "loss": 6.8697,
      "mean_token_accuracy": 0.10453333854675292,
      "num_tokens": 970935.0,
      "step": 560
    },
    {
      "entropy": 7.282825183868408,
      "epoch": 0.04396031900408481,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.8089,
      "mean_token_accuracy": 0.09897146746516228,
      "num_tokens": 979492.0,
      "step": 565
    },
    {
      "entropy": 7.173346710205078,
      "epoch": 0.04434934837580237,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002845,
      "loss": 6.725,
      "mean_token_accuracy": 0.1077153742313385,
      "num_tokens": 987866.0,
      "step": 570
    },
    {
      "entropy": 7.165633726119995,
      "epoch": 0.04473837774751994,
      "grad_norm": 1.0625,
      "learning_rate": 0.000287,
      "loss": 6.7609,
      "mean_token_accuracy": 0.11032681092619896,
      "num_tokens": 996760.0,
      "step": 575
    },
    {
      "entropy": 7.229097557067871,
      "epoch": 0.0451274071192375,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002895,
      "loss": 6.6913,
      "mean_token_accuracy": 0.10864157006144523,
      "num_tokens": 1005432.0,
      "step": 580
    },
    {
      "entropy": 7.137337875366211,
      "epoch": 0.04551643649095507,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000292,
      "loss": 6.7743,
      "mean_token_accuracy": 0.10533035397529603,
      "num_tokens": 1013729.0,
      "step": 585
    },
    {
      "entropy": 7.049478149414062,
      "epoch": 0.045905465862672634,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002945,
      "loss": 6.6116,
      "mean_token_accuracy": 0.11751440018415452,
      "num_tokens": 1022485.0,
      "step": 590
    },
    {
      "entropy": 7.1416089057922365,
      "epoch": 0.0462944952343902,
      "grad_norm": 1.328125,
      "learning_rate": 0.000297,
      "loss": 6.6885,
      "mean_token_accuracy": 0.10569961667060852,
      "num_tokens": 1030523.0,
      "step": 595
    },
    {
      "entropy": 7.179096317291259,
      "epoch": 0.04668352460610776,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002995,
      "loss": 6.7869,
      "mean_token_accuracy": 0.10787507370114327,
      "num_tokens": 1039268.0,
      "step": 600
    },
    {
      "entropy": 7.120957326889038,
      "epoch": 0.047072553977825324,
      "grad_norm": 1.15625,
      "learning_rate": 0.000302,
      "loss": 6.684,
      "mean_token_accuracy": 0.10937019288539887,
      "num_tokens": 1048590.0,
      "step": 605
    },
    {
      "entropy": 7.0393298149108885,
      "epoch": 0.04746158334954289,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003045,
      "loss": 6.617,
      "mean_token_accuracy": 0.11176003962755203,
      "num_tokens": 1057232.0,
      "step": 610
    },
    {
      "entropy": 7.098176956176758,
      "epoch": 0.047850612721260455,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000307,
      "loss": 6.6998,
      "mean_token_accuracy": 0.1081328421831131,
      "num_tokens": 1066404.0,
      "step": 615
    },
    {
      "entropy": 7.031232643127441,
      "epoch": 0.04823964209297802,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003095,
      "loss": 6.7638,
      "mean_token_accuracy": 0.10122817903757095,
      "num_tokens": 1075093.0,
      "step": 620
    },
    {
      "entropy": 7.155955123901367,
      "epoch": 0.048628671464695586,
      "grad_norm": 1.1875,
      "learning_rate": 0.000312,
      "loss": 6.667,
      "mean_token_accuracy": 0.10735566318035125,
      "num_tokens": 1084454.0,
      "step": 625
    },
    {
      "entropy": 7.108900642395019,
      "epoch": 0.04901770083641315,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003145,
      "loss": 6.5969,
      "mean_token_accuracy": 0.10924701541662216,
      "num_tokens": 1093452.0,
      "step": 630
    },
    {
      "entropy": 7.034401082992554,
      "epoch": 0.04940673020813072,
      "grad_norm": 1.328125,
      "learning_rate": 0.000317,
      "loss": 6.6196,
      "mean_token_accuracy": 0.11470122635364532,
      "num_tokens": 1101814.0,
      "step": 635
    },
    {
      "entropy": 7.051713323593139,
      "epoch": 0.049795759579848276,
      "grad_norm": 1.25,
      "learning_rate": 0.0003195,
      "loss": 6.7,
      "mean_token_accuracy": 0.11058250665664673,
      "num_tokens": 1110386.0,
      "step": 640
    },
    {
      "entropy": 7.14307336807251,
      "epoch": 0.05018478895156584,
      "grad_norm": 1.265625,
      "learning_rate": 0.000322,
      "loss": 6.8031,
      "mean_token_accuracy": 0.10201985463500023,
      "num_tokens": 1120141.0,
      "step": 645
    },
    {
      "entropy": 7.1510227680206295,
      "epoch": 0.05057381832328341,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.7396,
      "mean_token_accuracy": 0.1102594219148159,
      "num_tokens": 1129920.0,
      "step": 650
    },
    {
      "entropy": 6.983091688156128,
      "epoch": 0.05096284769500097,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.6196,
      "mean_token_accuracy": 0.11163505613803863,
      "num_tokens": 1138518.0,
      "step": 655
    },
    {
      "entropy": 7.117662715911865,
      "epoch": 0.05135187706671854,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.7209,
      "mean_token_accuracy": 0.10830123797059059,
      "num_tokens": 1147113.0,
      "step": 660
    },
    {
      "entropy": 7.052978134155273,
      "epoch": 0.0517409064384361,
      "grad_norm": 1.390625,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.5612,
      "mean_token_accuracy": 0.11946347504854202,
      "num_tokens": 1155886.0,
      "step": 665
    },
    {
      "entropy": 7.017060613632202,
      "epoch": 0.05212993581015367,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.725,
      "mean_token_accuracy": 0.10746086090803146,
      "num_tokens": 1164200.0,
      "step": 670
    },
    {
      "entropy": 7.051968240737915,
      "epoch": 0.052518965181871234,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000337,
      "loss": 6.5857,
      "mean_token_accuracy": 0.1125464141368866,
      "num_tokens": 1172720.0,
      "step": 675
    },
    {
      "entropy": 6.996215152740478,
      "epoch": 0.05290799455358879,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003395,
      "loss": 6.5468,
      "mean_token_accuracy": 0.1077719546854496,
      "num_tokens": 1181225.0,
      "step": 680
    },
    {
      "entropy": 7.120228815078735,
      "epoch": 0.05329702392530636,
      "grad_norm": 1.203125,
      "learning_rate": 0.000342,
      "loss": 6.7546,
      "mean_token_accuracy": 0.10120292380452156,
      "num_tokens": 1190070.0,
      "step": 685
    },
    {
      "entropy": 7.013375902175904,
      "epoch": 0.053686053297023924,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.6267,
      "mean_token_accuracy": 0.11402207687497139,
      "num_tokens": 1198995.0,
      "step": 690
    },
    {
      "entropy": 7.120525217056274,
      "epoch": 0.05407508266874149,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000347,
      "loss": 6.6744,
      "mean_token_accuracy": 0.11539597958326339,
      "num_tokens": 1207664.0,
      "step": 695
    },
    {
      "entropy": 7.102228593826294,
      "epoch": 0.054464112040459055,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003495,
      "loss": 6.5789,
      "mean_token_accuracy": 0.12240418270230294,
      "num_tokens": 1216232.0,
      "step": 700
    },
    {
      "entropy": 6.950713348388672,
      "epoch": 0.05485314141217662,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000352,
      "loss": 6.6748,
      "mean_token_accuracy": 0.11159612014889717,
      "num_tokens": 1224106.0,
      "step": 705
    },
    {
      "entropy": 6.960762357711792,
      "epoch": 0.055242170783894186,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003545,
      "loss": 6.5619,
      "mean_token_accuracy": 0.11645937711000443,
      "num_tokens": 1232514.0,
      "step": 710
    },
    {
      "entropy": 7.127431774139405,
      "epoch": 0.05563120015561175,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000357,
      "loss": 6.6152,
      "mean_token_accuracy": 0.11127564162015915,
      "num_tokens": 1240178.0,
      "step": 715
    },
    {
      "entropy": 6.829823637008667,
      "epoch": 0.05602022952732931,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003595,
      "loss": 6.5236,
      "mean_token_accuracy": 0.11545735970139503,
      "num_tokens": 1248635.0,
      "step": 720
    },
    {
      "entropy": 7.094886350631714,
      "epoch": 0.056409258899046875,
      "grad_norm": 1.265625,
      "learning_rate": 0.000362,
      "loss": 6.6258,
      "mean_token_accuracy": 0.11595786064863205,
      "num_tokens": 1256389.0,
      "step": 725
    },
    {
      "entropy": 6.9064311504364015,
      "epoch": 0.05679828827076444,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003645,
      "loss": 6.5896,
      "mean_token_accuracy": 0.11666487753391266,
      "num_tokens": 1264035.0,
      "step": 730
    },
    {
      "entropy": 7.034950304031372,
      "epoch": 0.057187317642482006,
      "grad_norm": 1.25,
      "learning_rate": 0.000367,
      "loss": 6.5838,
      "mean_token_accuracy": 0.11167205199599266,
      "num_tokens": 1272941.0,
      "step": 735
    },
    {
      "entropy": 6.862753534317017,
      "epoch": 0.05757634701419957,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003695,
      "loss": 6.5714,
      "mean_token_accuracy": 0.11405754834413528,
      "num_tokens": 1281767.0,
      "step": 740
    },
    {
      "entropy": 7.123349332809449,
      "epoch": 0.05796537638591714,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000372,
      "loss": 6.6469,
      "mean_token_accuracy": 0.11400221288204193,
      "num_tokens": 1290049.0,
      "step": 745
    },
    {
      "entropy": 6.815326452255249,
      "epoch": 0.0583544057576347,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003745,
      "loss": 6.4981,
      "mean_token_accuracy": 0.12375147640705109,
      "num_tokens": 1299032.0,
      "step": 750
    },
    {
      "entropy": 6.98442530632019,
      "epoch": 0.05874343512935227,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000377,
      "loss": 6.7388,
      "mean_token_accuracy": 0.10899372920393943,
      "num_tokens": 1308508.0,
      "step": 755
    },
    {
      "entropy": 7.064965343475341,
      "epoch": 0.059132464501069834,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003795,
      "loss": 6.5673,
      "mean_token_accuracy": 0.1138399638235569,
      "num_tokens": 1317044.0,
      "step": 760
    },
    {
      "entropy": 7.0013751029968265,
      "epoch": 0.05952149387278739,
      "grad_norm": 1.171875,
      "learning_rate": 0.000382,
      "loss": 6.5988,
      "mean_token_accuracy": 0.11330810338258743,
      "num_tokens": 1325554.0,
      "step": 765
    },
    {
      "entropy": 6.8747217655181885,
      "epoch": 0.05991052324450496,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003845,
      "loss": 6.4944,
      "mean_token_accuracy": 0.11092947795987129,
      "num_tokens": 1334201.0,
      "step": 770
    },
    {
      "entropy": 6.932094430923462,
      "epoch": 0.060299552616222524,
      "grad_norm": 1.25,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.5156,
      "mean_token_accuracy": 0.11926758959889412,
      "num_tokens": 1343061.0,
      "step": 775
    },
    {
      "entropy": 6.882827472686768,
      "epoch": 0.06068858198794009,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.4792,
      "mean_token_accuracy": 0.12102379724383354,
      "num_tokens": 1352262.0,
      "step": 780
    },
    {
      "entropy": 6.8652026653289795,
      "epoch": 0.061077611359657655,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.5417,
      "mean_token_accuracy": 0.11120004132390023,
      "num_tokens": 1361400.0,
      "step": 785
    },
    {
      "entropy": 6.9827672958374025,
      "epoch": 0.06146664073137522,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.6545,
      "mean_token_accuracy": 0.1084800124168396,
      "num_tokens": 1369850.0,
      "step": 790
    },
    {
      "entropy": 6.896434259414673,
      "epoch": 0.061855670103092786,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.5041,
      "mean_token_accuracy": 0.11742325201630592,
      "num_tokens": 1378342.0,
      "step": 795
    },
    {
      "entropy": 6.971995687484741,
      "epoch": 0.06224469947481035,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003995,
      "loss": 6.5897,
      "mean_token_accuracy": 0.109882403165102,
      "num_tokens": 1386981.0,
      "step": 800
    },
    {
      "entropy": 6.938254642486572,
      "epoch": 0.06263372884652792,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000402,
      "loss": 6.6818,
      "mean_token_accuracy": 0.10729405432939529,
      "num_tokens": 1395801.0,
      "step": 805
    },
    {
      "entropy": 7.06087851524353,
      "epoch": 0.06302275821824548,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004045,
      "loss": 6.5776,
      "mean_token_accuracy": 0.11447910293936729,
      "num_tokens": 1403715.0,
      "step": 810
    },
    {
      "entropy": 6.9617109298706055,
      "epoch": 0.06341178758996305,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.6583,
      "mean_token_accuracy": 0.11031345576047898,
      "num_tokens": 1413199.0,
      "step": 815
    },
    {
      "entropy": 6.929412412643432,
      "epoch": 0.0638008169616806,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004095,
      "loss": 6.632,
      "mean_token_accuracy": 0.11478308364748954,
      "num_tokens": 1421918.0,
      "step": 820
    },
    {
      "entropy": 6.956843757629395,
      "epoch": 0.06418984633339817,
      "grad_norm": 1.15625,
      "learning_rate": 0.000412,
      "loss": 6.5245,
      "mean_token_accuracy": 0.11161991730332374,
      "num_tokens": 1430625.0,
      "step": 825
    },
    {
      "entropy": 6.826650476455688,
      "epoch": 0.06457887570511574,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004145,
      "loss": 6.562,
      "mean_token_accuracy": 0.11291277259588242,
      "num_tokens": 1439158.0,
      "step": 830
    },
    {
      "entropy": 6.870626020431518,
      "epoch": 0.0649679050768333,
      "grad_norm": 1.109375,
      "learning_rate": 0.000417,
      "loss": 6.527,
      "mean_token_accuracy": 0.11909559741616249,
      "num_tokens": 1448266.0,
      "step": 835
    },
    {
      "entropy": 6.925184917449951,
      "epoch": 0.06535693444855087,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004195,
      "loss": 6.6033,
      "mean_token_accuracy": 0.11738861948251725,
      "num_tokens": 1457322.0,
      "step": 840
    },
    {
      "entropy": 7.063237190246582,
      "epoch": 0.06574596382026843,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000422,
      "loss": 6.7224,
      "mean_token_accuracy": 0.10825124606490136,
      "num_tokens": 1466120.0,
      "step": 845
    },
    {
      "entropy": 6.864966869354248,
      "epoch": 0.066134993191986,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004245,
      "loss": 6.4761,
      "mean_token_accuracy": 0.11677760556340218,
      "num_tokens": 1474965.0,
      "step": 850
    },
    {
      "entropy": 6.859512901306152,
      "epoch": 0.06652402256370356,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000427,
      "loss": 6.5646,
      "mean_token_accuracy": 0.11498897522687912,
      "num_tokens": 1483483.0,
      "step": 855
    },
    {
      "entropy": 6.937557744979858,
      "epoch": 0.06691305193542113,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004295,
      "loss": 6.5105,
      "mean_token_accuracy": 0.1108026348054409,
      "num_tokens": 1492313.0,
      "step": 860
    },
    {
      "entropy": 6.8773322105407715,
      "epoch": 0.06730208130713869,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000432,
      "loss": 6.4427,
      "mean_token_accuracy": 0.11592724919319153,
      "num_tokens": 1501081.0,
      "step": 865
    },
    {
      "entropy": 6.809585762023926,
      "epoch": 0.06769111067885625,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004345,
      "loss": 6.4583,
      "mean_token_accuracy": 0.11120192930102349,
      "num_tokens": 1510169.0,
      "step": 870
    },
    {
      "entropy": 6.838544702529907,
      "epoch": 0.06808014005057382,
      "grad_norm": 1.21875,
      "learning_rate": 0.000437,
      "loss": 6.4244,
      "mean_token_accuracy": 0.11338047757744789,
      "num_tokens": 1519546.0,
      "step": 875
    },
    {
      "entropy": 6.820107889175415,
      "epoch": 0.06846916942229138,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004395,
      "loss": 6.5435,
      "mean_token_accuracy": 0.1123778223991394,
      "num_tokens": 1529104.0,
      "step": 880
    },
    {
      "entropy": 6.7314704895019535,
      "epoch": 0.06885819879400895,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000442,
      "loss": 6.4938,
      "mean_token_accuracy": 0.12263969853520393,
      "num_tokens": 1537033.0,
      "step": 885
    },
    {
      "entropy": 6.9164563655853275,
      "epoch": 0.06924722816572651,
      "grad_norm": 1.25,
      "learning_rate": 0.0004445,
      "loss": 6.5018,
      "mean_token_accuracy": 0.11451332420110702,
      "num_tokens": 1545059.0,
      "step": 890
    },
    {
      "entropy": 6.853432083129883,
      "epoch": 0.06963625753744408,
      "grad_norm": 1.09375,
      "learning_rate": 0.000447,
      "loss": 6.5062,
      "mean_token_accuracy": 0.11352940872311593,
      "num_tokens": 1553784.0,
      "step": 895
    },
    {
      "entropy": 6.785615253448486,
      "epoch": 0.07002528690916164,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.4413,
      "mean_token_accuracy": 0.11957875341176986,
      "num_tokens": 1562059.0,
      "step": 900
    },
    {
      "entropy": 6.781708908081055,
      "epoch": 0.0704143162808792,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.5083,
      "mean_token_accuracy": 0.1147128850221634,
      "num_tokens": 1570061.0,
      "step": 905
    },
    {
      "entropy": 6.792767333984375,
      "epoch": 0.07080334565259677,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.4642,
      "mean_token_accuracy": 0.11603316441178321,
      "num_tokens": 1578261.0,
      "step": 910
    },
    {
      "entropy": 6.81901068687439,
      "epoch": 0.07119237502431433,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.3991,
      "mean_token_accuracy": 0.11860067769885063,
      "num_tokens": 1586260.0,
      "step": 915
    },
    {
      "entropy": 6.7977457523345945,
      "epoch": 0.0715814043960319,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.5922,
      "mean_token_accuracy": 0.11235073879361153,
      "num_tokens": 1595234.0,
      "step": 920
    },
    {
      "entropy": 6.873261642456055,
      "epoch": 0.07197043376774946,
      "grad_norm": 1.265625,
      "learning_rate": 0.000462,
      "loss": 6.4801,
      "mean_token_accuracy": 0.11916077733039857,
      "num_tokens": 1604212.0,
      "step": 925
    },
    {
      "entropy": 6.792505550384521,
      "epoch": 0.07235946313946703,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004645,
      "loss": 6.5283,
      "mean_token_accuracy": 0.11446316465735436,
      "num_tokens": 1612925.0,
      "step": 930
    },
    {
      "entropy": 6.767776250839233,
      "epoch": 0.07274849251118459,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000467,
      "loss": 6.4187,
      "mean_token_accuracy": 0.12004713863134384,
      "num_tokens": 1621804.0,
      "step": 935
    },
    {
      "entropy": 6.757959938049316,
      "epoch": 0.07313752188290217,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004695,
      "loss": 6.3969,
      "mean_token_accuracy": 0.11997036337852478,
      "num_tokens": 1630943.0,
      "step": 940
    },
    {
      "entropy": 6.713178873062134,
      "epoch": 0.07352655125461972,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000472,
      "loss": 6.3106,
      "mean_token_accuracy": 0.12384014949202538,
      "num_tokens": 1638899.0,
      "step": 945
    },
    {
      "entropy": 6.813256025314331,
      "epoch": 0.07391558062633728,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004745,
      "loss": 6.5183,
      "mean_token_accuracy": 0.11929329931735992,
      "num_tokens": 1647485.0,
      "step": 950
    },
    {
      "entropy": 6.792909765243531,
      "epoch": 0.07430460999805485,
      "grad_norm": 1.28125,
      "learning_rate": 0.000477,
      "loss": 6.485,
      "mean_token_accuracy": 0.11955598518252372,
      "num_tokens": 1656290.0,
      "step": 955
    },
    {
      "entropy": 6.814577341079712,
      "epoch": 0.07469363936977241,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004795,
      "loss": 6.5025,
      "mean_token_accuracy": 0.11491848975419998,
      "num_tokens": 1665474.0,
      "step": 960
    },
    {
      "entropy": 6.6340234756469725,
      "epoch": 0.07508266874148999,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000482,
      "loss": 6.3785,
      "mean_token_accuracy": 0.12735505029559135,
      "num_tokens": 1673599.0,
      "step": 965
    },
    {
      "entropy": 6.7855254173278805,
      "epoch": 0.07547169811320754,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004845,
      "loss": 6.3513,
      "mean_token_accuracy": 0.11603628024458885,
      "num_tokens": 1683282.0,
      "step": 970
    },
    {
      "entropy": 6.784381628036499,
      "epoch": 0.07586072748492512,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000487,
      "loss": 6.4793,
      "mean_token_accuracy": 0.12123103961348533,
      "num_tokens": 1691094.0,
      "step": 975
    },
    {
      "entropy": 6.6339301586151125,
      "epoch": 0.07624975685664268,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004895,
      "loss": 6.356,
      "mean_token_accuracy": 0.1204170361161232,
      "num_tokens": 1700053.0,
      "step": 980
    },
    {
      "entropy": 6.736316251754761,
      "epoch": 0.07663878622836025,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000492,
      "loss": 6.2828,
      "mean_token_accuracy": 0.12172069922089576,
      "num_tokens": 1707732.0,
      "step": 985
    },
    {
      "entropy": 6.719516134262085,
      "epoch": 0.0770278156000778,
      "grad_norm": 1.25,
      "learning_rate": 0.0004945,
      "loss": 6.5251,
      "mean_token_accuracy": 0.11388466954231262,
      "num_tokens": 1716406.0,
      "step": 990
    },
    {
      "entropy": 6.887849283218384,
      "epoch": 0.07741684497179536,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000497,
      "loss": 6.5579,
      "mean_token_accuracy": 0.11262270659208298,
      "num_tokens": 1725832.0,
      "step": 995
    },
    {
      "entropy": 6.818440103530884,
      "epoch": 0.07780587434351294,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004995,
      "loss": 6.4678,
      "mean_token_accuracy": 0.11311930790543556,
      "num_tokens": 1734853.0,
      "step": 1000
    },
    {
      "entropy": 6.694705009460449,
      "epoch": 0.0781949037152305,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004999999989075146,
      "loss": 6.4807,
      "mean_token_accuracy": 0.12094770967960358,
      "num_tokens": 1744240.0,
      "step": 1005
    },
    {
      "entropy": 6.865083646774292,
      "epoch": 0.07858393308694807,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999999944692927,
      "loss": 6.4969,
      "mean_token_accuracy": 0.11460169404745102,
      "num_tokens": 1753334.0,
      "step": 1010
    },
    {
      "entropy": 6.603975772857666,
      "epoch": 0.07897296245866563,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499999986617054,
      "loss": 6.3948,
      "mean_token_accuracy": 0.11559040769934655,
      "num_tokens": 1762032.0,
      "step": 1015
    },
    {
      "entropy": 6.726358270645141,
      "epoch": 0.0793619918303832,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999999753507986,
      "loss": 6.3834,
      "mean_token_accuracy": 0.11660856455564499,
      "num_tokens": 1770508.0,
      "step": 1020
    },
    {
      "entropy": 6.621751356124878,
      "epoch": 0.07975102120210076,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999999606705267,
      "loss": 6.2867,
      "mean_token_accuracy": 0.12645310387015343,
      "num_tokens": 1780167.0,
      "step": 1025
    },
    {
      "entropy": 6.760381269454956,
      "epoch": 0.08014005057381832,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999999425762385,
      "loss": 6.4923,
      "mean_token_accuracy": 0.11203559264540672,
      "num_tokens": 1788809.0,
      "step": 1030
    },
    {
      "entropy": 6.707613611221314,
      "epoch": 0.08052907994553589,
      "grad_norm": 1.375,
      "learning_rate": 0.0004999999210679344,
      "loss": 6.4082,
      "mean_token_accuracy": 0.11932154521346092,
      "num_tokens": 1797445.0,
      "step": 1035
    },
    {
      "entropy": 6.708571434020996,
      "epoch": 0.08091810931725345,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999998961456145,
      "loss": 6.4093,
      "mean_token_accuracy": 0.11497489586472512,
      "num_tokens": 1806488.0,
      "step": 1040
    },
    {
      "entropy": 6.7836400985717775,
      "epoch": 0.08130713868897102,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999998678092794,
      "loss": 6.4865,
      "mean_token_accuracy": 0.12190944403409958,
      "num_tokens": 1815692.0,
      "step": 1045
    },
    {
      "entropy": 6.733627986907959,
      "epoch": 0.08169616806068858,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004999998360589293,
      "loss": 6.3935,
      "mean_token_accuracy": 0.12152979299426078,
      "num_tokens": 1824599.0,
      "step": 1050
    },
    {
      "entropy": 6.814380025863647,
      "epoch": 0.08208519743240615,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000499999800894565,
      "loss": 6.5522,
      "mean_token_accuracy": 0.11713027730584144,
      "num_tokens": 1833411.0,
      "step": 1055
    },
    {
      "entropy": 6.664803600311279,
      "epoch": 0.08247422680412371,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004999997623161866,
      "loss": 6.3602,
      "mean_token_accuracy": 0.11816268861293792,
      "num_tokens": 1842177.0,
      "step": 1060
    },
    {
      "entropy": 6.589741563796997,
      "epoch": 0.08286325617584128,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999997203237951,
      "loss": 6.3501,
      "mean_token_accuracy": 0.12157049998641015,
      "num_tokens": 1850176.0,
      "step": 1065
    },
    {
      "entropy": 6.725516271591187,
      "epoch": 0.08325228554755884,
      "grad_norm": 1.125,
      "learning_rate": 0.000499999674917391,
      "loss": 6.3956,
      "mean_token_accuracy": 0.12002233043313026,
      "num_tokens": 1859464.0,
      "step": 1070
    },
    {
      "entropy": 6.589758443832397,
      "epoch": 0.0836413149192764,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004999996260969748,
      "loss": 6.3058,
      "mean_token_accuracy": 0.11860100701451301,
      "num_tokens": 1867733.0,
      "step": 1075
    },
    {
      "entropy": 6.651362705230713,
      "epoch": 0.08403034429099397,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999995738625475,
      "loss": 6.2897,
      "mean_token_accuracy": 0.1241245798766613,
      "num_tokens": 1876010.0,
      "step": 1080
    },
    {
      "entropy": 6.645105648040771,
      "epoch": 0.08441937366271153,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999995182141099,
      "loss": 6.4517,
      "mean_token_accuracy": 0.11694926843047142,
      "num_tokens": 1884906.0,
      "step": 1085
    },
    {
      "entropy": 6.668650484085083,
      "epoch": 0.0848084030344291,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999994591516626,
      "loss": 6.36,
      "mean_token_accuracy": 0.1204823799431324,
      "num_tokens": 1893464.0,
      "step": 1090
    },
    {
      "entropy": 6.652740049362182,
      "epoch": 0.08519743240614666,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999993966752066,
      "loss": 6.4075,
      "mean_token_accuracy": 0.11751677766442299,
      "num_tokens": 1902769.0,
      "step": 1095
    },
    {
      "entropy": 6.724245500564575,
      "epoch": 0.08558646177786423,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499999330784743,
      "loss": 6.3414,
      "mean_token_accuracy": 0.11880838871002197,
      "num_tokens": 1911968.0,
      "step": 1100
    },
    {
      "entropy": 6.623199605941773,
      "epoch": 0.08597549114958179,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004999992614802725,
      "loss": 6.3306,
      "mean_token_accuracy": 0.12165946811437607,
      "num_tokens": 1920063.0,
      "step": 1105
    },
    {
      "entropy": 6.5714949607849125,
      "epoch": 0.08636452052129936,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004999991887617966,
      "loss": 6.2931,
      "mean_token_accuracy": 0.12126443311572074,
      "num_tokens": 1929192.0,
      "step": 1110
    },
    {
      "entropy": 6.723468399047851,
      "epoch": 0.08675354989301692,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999991126293158,
      "loss": 6.5039,
      "mean_token_accuracy": 0.11924195289611816,
      "num_tokens": 1939082.0,
      "step": 1115
    },
    {
      "entropy": 6.509707975387573,
      "epoch": 0.08714257926473448,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999990330828318,
      "loss": 6.3622,
      "mean_token_accuracy": 0.11828808337450028,
      "num_tokens": 1948313.0,
      "step": 1120
    },
    {
      "entropy": 6.764839315414429,
      "epoch": 0.08753160863645205,
      "grad_norm": 1.25,
      "learning_rate": 0.0004999989501223456,
      "loss": 6.3535,
      "mean_token_accuracy": 0.11773044615983963,
      "num_tokens": 1957035.0,
      "step": 1125
    },
    {
      "entropy": 6.571807384490967,
      "epoch": 0.08792063800816961,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999988637478584,
      "loss": 6.356,
      "mean_token_accuracy": 0.11375595852732659,
      "num_tokens": 1966041.0,
      "step": 1130
    },
    {
      "entropy": 6.682859468460083,
      "epoch": 0.08830966737988719,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004999987739593716,
      "loss": 6.3945,
      "mean_token_accuracy": 0.11131789907813072,
      "num_tokens": 1975019.0,
      "step": 1135
    },
    {
      "entropy": 6.595472717285157,
      "epoch": 0.08869869675160474,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999986807568865,
      "loss": 6.3177,
      "mean_token_accuracy": 0.12635206803679466,
      "num_tokens": 1982930.0,
      "step": 1140
    },
    {
      "entropy": 6.661126136779785,
      "epoch": 0.08908772612332232,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999985841404045,
      "loss": 6.4155,
      "mean_token_accuracy": 0.11563977152109146,
      "num_tokens": 1991888.0,
      "step": 1145
    },
    {
      "entropy": 6.647232627868652,
      "epoch": 0.08947675549503988,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999984841099271,
      "loss": 6.2743,
      "mean_token_accuracy": 0.11983647644519806,
      "num_tokens": 2000202.0,
      "step": 1150
    },
    {
      "entropy": 6.5404551982879635,
      "epoch": 0.08986578486675743,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499998380665456,
      "loss": 6.3057,
      "mean_token_accuracy": 0.12251394167542458,
      "num_tokens": 2009281.0,
      "step": 1155
    },
    {
      "entropy": 6.654289579391479,
      "epoch": 0.090254814238475,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999982738069924,
      "loss": 6.3102,
      "mean_token_accuracy": 0.12231542989611625,
      "num_tokens": 2017806.0,
      "step": 1160
    },
    {
      "entropy": 6.654610538482666,
      "epoch": 0.09064384361019256,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999981635345382,
      "loss": 6.3964,
      "mean_token_accuracy": 0.11891458630561828,
      "num_tokens": 2027582.0,
      "step": 1165
    },
    {
      "entropy": 6.530627679824829,
      "epoch": 0.09103287298191014,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999980498480949,
      "loss": 6.2699,
      "mean_token_accuracy": 0.12494958117604256,
      "num_tokens": 2035427.0,
      "step": 1170
    },
    {
      "entropy": 6.712484693527221,
      "epoch": 0.0914219023536277,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999979327476644,
      "loss": 6.4533,
      "mean_token_accuracy": 0.1177413173019886,
      "num_tokens": 2044603.0,
      "step": 1175
    },
    {
      "entropy": 6.584618806838989,
      "epoch": 0.09181093172534527,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999978122332485,
      "loss": 6.2931,
      "mean_token_accuracy": 0.12760861366987228,
      "num_tokens": 2053272.0,
      "step": 1180
    },
    {
      "entropy": 6.628946113586426,
      "epoch": 0.09219996109706283,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999976883048487,
      "loss": 6.3113,
      "mean_token_accuracy": 0.12247533276677132,
      "num_tokens": 2061677.0,
      "step": 1185
    },
    {
      "entropy": 6.565109968185425,
      "epoch": 0.0925889904687804,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999975609624673,
      "loss": 6.3337,
      "mean_token_accuracy": 0.12145892754197121,
      "num_tokens": 2070282.0,
      "step": 1190
    },
    {
      "entropy": 6.7010410785675045,
      "epoch": 0.09297801984049796,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999974302061059,
      "loss": 6.4161,
      "mean_token_accuracy": 0.11821920722723007,
      "num_tokens": 2078812.0,
      "step": 1195
    },
    {
      "entropy": 6.600937271118164,
      "epoch": 0.09336704921221552,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999972960357666,
      "loss": 6.3138,
      "mean_token_accuracy": 0.11917631253600121,
      "num_tokens": 2087020.0,
      "step": 1200
    },
    {
      "entropy": 6.548227739334107,
      "epoch": 0.09375607858393309,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999971584514516,
      "loss": 6.2607,
      "mean_token_accuracy": 0.1289692461490631,
      "num_tokens": 2095933.0,
      "step": 1205
    },
    {
      "entropy": 6.49370927810669,
      "epoch": 0.09414510795565065,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999970174531627,
      "loss": 6.2709,
      "mean_token_accuracy": 0.1252399981021881,
      "num_tokens": 2105218.0,
      "step": 1210
    },
    {
      "entropy": 6.519537162780762,
      "epoch": 0.09453413732736822,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004999968730409022,
      "loss": 6.1868,
      "mean_token_accuracy": 0.12397775202989578,
      "num_tokens": 2113808.0,
      "step": 1215
    },
    {
      "entropy": 6.508232164382934,
      "epoch": 0.09492316669908578,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999967252146723,
      "loss": 6.3589,
      "mean_token_accuracy": 0.11807975843548775,
      "num_tokens": 2123014.0,
      "step": 1220
    },
    {
      "entropy": 6.678585195541382,
      "epoch": 0.09531219607080335,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999965739744752,
      "loss": 6.2823,
      "mean_token_accuracy": 0.12472278475761414,
      "num_tokens": 2131840.0,
      "step": 1225
    },
    {
      "entropy": 6.571667098999024,
      "epoch": 0.09570122544252091,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999964193203132,
      "loss": 6.3797,
      "mean_token_accuracy": 0.1198837198317051,
      "num_tokens": 2140467.0,
      "step": 1230
    },
    {
      "entropy": 6.62928991317749,
      "epoch": 0.09609025481423847,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999962612521888,
      "loss": 6.3252,
      "mean_token_accuracy": 0.12536743581295012,
      "num_tokens": 2149282.0,
      "step": 1235
    },
    {
      "entropy": 6.576823568344116,
      "epoch": 0.09647928418595604,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999960997701041,
      "loss": 6.3244,
      "mean_token_accuracy": 0.12110247910022735,
      "num_tokens": 2157781.0,
      "step": 1240
    },
    {
      "entropy": 6.597092962265014,
      "epoch": 0.0968683135576736,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999959348740618,
      "loss": 6.2967,
      "mean_token_accuracy": 0.12580249533057214,
      "num_tokens": 2166482.0,
      "step": 1245
    },
    {
      "entropy": 6.499213790893554,
      "epoch": 0.09725734292939117,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999957665640641,
      "loss": 6.3132,
      "mean_token_accuracy": 0.12070827260613441,
      "num_tokens": 2174559.0,
      "step": 1250
    },
    {
      "entropy": 6.5587629795074465,
      "epoch": 0.09764637230110873,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499995594840114,
      "loss": 6.2106,
      "mean_token_accuracy": 0.12732202261686326,
      "num_tokens": 2182374.0,
      "step": 1255
    },
    {
      "entropy": 6.497962141036988,
      "epoch": 0.0980354016728263,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999954197022138,
      "loss": 6.2986,
      "mean_token_accuracy": 0.1267983116209507,
      "num_tokens": 2191427.0,
      "step": 1260
    },
    {
      "entropy": 6.650804328918457,
      "epoch": 0.09842443104454386,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999952411503661,
      "loss": 6.3879,
      "mean_token_accuracy": 0.11929038390517235,
      "num_tokens": 2199698.0,
      "step": 1265
    },
    {
      "entropy": 6.565453624725341,
      "epoch": 0.09881346041626143,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999950591845738,
      "loss": 6.2348,
      "mean_token_accuracy": 0.12733448669314384,
      "num_tokens": 2208519.0,
      "step": 1270
    },
    {
      "entropy": 6.464875268936157,
      "epoch": 0.09920248978797899,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999948738048397,
      "loss": 6.2428,
      "mean_token_accuracy": 0.1302006796002388,
      "num_tokens": 2217453.0,
      "step": 1275
    },
    {
      "entropy": 6.554579877853394,
      "epoch": 0.09959151915969655,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999946850111663,
      "loss": 6.3577,
      "mean_token_accuracy": 0.12167153358459473,
      "num_tokens": 2225623.0,
      "step": 1280
    },
    {
      "entropy": 6.576758623123169,
      "epoch": 0.09998054853141412,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999944928035569,
      "loss": 6.2919,
      "mean_token_accuracy": 0.12363119274377823,
      "num_tokens": 2234519.0,
      "step": 1285
    },
    {
      "entropy": 6.5044145584106445,
      "epoch": 0.10036957790313168,
      "grad_norm": 1.421875,
      "learning_rate": 0.000499994297182014,
      "loss": 6.3061,
      "mean_token_accuracy": 0.12101315706968307,
      "num_tokens": 2242994.0,
      "step": 1290
    },
    {
      "entropy": 6.651056385040283,
      "epoch": 0.10075860727484925,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999940981465408,
      "loss": 6.3138,
      "mean_token_accuracy": 0.12628812938928605,
      "num_tokens": 2251687.0,
      "step": 1295
    },
    {
      "entropy": 6.394090366363526,
      "epoch": 0.10114763664656681,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999938956971404,
      "loss": 6.2409,
      "mean_token_accuracy": 0.12676058262586593,
      "num_tokens": 2260751.0,
      "step": 1300
    },
    {
      "entropy": 6.595192193984985,
      "epoch": 0.10153666601828439,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999936898338156,
      "loss": 6.1791,
      "mean_token_accuracy": 0.13165878504514694,
      "num_tokens": 2269125.0,
      "step": 1305
    },
    {
      "entropy": 6.401913452148437,
      "epoch": 0.10192569539000194,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999934805565698,
      "loss": 6.1097,
      "mean_token_accuracy": 0.12795391529798508,
      "num_tokens": 2277098.0,
      "step": 1310
    },
    {
      "entropy": 6.470233392715454,
      "epoch": 0.10231472476171952,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499993267865406,
      "loss": 6.3887,
      "mean_token_accuracy": 0.1236030898988247,
      "num_tokens": 2285953.0,
      "step": 1315
    },
    {
      "entropy": 6.676652717590332,
      "epoch": 0.10270375413343708,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999930517603275,
      "loss": 6.3036,
      "mean_token_accuracy": 0.12659136727452278,
      "num_tokens": 2294487.0,
      "step": 1320
    },
    {
      "entropy": 6.545140743255615,
      "epoch": 0.10309278350515463,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999928322413374,
      "loss": 6.3383,
      "mean_token_accuracy": 0.1220410332083702,
      "num_tokens": 2302915.0,
      "step": 1325
    },
    {
      "entropy": 6.493171358108521,
      "epoch": 0.1034818128768722,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999926093084394,
      "loss": 6.2879,
      "mean_token_accuracy": 0.11915498897433281,
      "num_tokens": 2310769.0,
      "step": 1330
    },
    {
      "entropy": 6.463140153884888,
      "epoch": 0.10387084224858976,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004999923829616365,
      "loss": 6.1782,
      "mean_token_accuracy": 0.12339231967926026,
      "num_tokens": 2319987.0,
      "step": 1335
    },
    {
      "entropy": 6.515197849273681,
      "epoch": 0.10425987162030734,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999921532009323,
      "loss": 6.2932,
      "mean_token_accuracy": 0.1183166541159153,
      "num_tokens": 2328964.0,
      "step": 1340
    },
    {
      "entropy": 6.563854217529297,
      "epoch": 0.1046489009920249,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999919200263304,
      "loss": 6.2556,
      "mean_token_accuracy": 0.11943673864006996,
      "num_tokens": 2337945.0,
      "step": 1345
    },
    {
      "entropy": 6.530140399932861,
      "epoch": 0.10503793036374247,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004999916834378342,
      "loss": 6.1306,
      "mean_token_accuracy": 0.12886034995317458,
      "num_tokens": 2345989.0,
      "step": 1350
    },
    {
      "entropy": 6.404300594329834,
      "epoch": 0.10542695973546003,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999914434354472,
      "loss": 6.2652,
      "mean_token_accuracy": 0.12330945879220963,
      "num_tokens": 2353752.0,
      "step": 1355
    },
    {
      "entropy": 6.504211378097534,
      "epoch": 0.10581598910717759,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999912000191733,
      "loss": 6.2419,
      "mean_token_accuracy": 0.12025356367230415,
      "num_tokens": 2361961.0,
      "step": 1360
    },
    {
      "entropy": 6.423410511016845,
      "epoch": 0.10620501847889516,
      "grad_norm": 1.140625,
      "learning_rate": 0.000499990953189016,
      "loss": 6.1589,
      "mean_token_accuracy": 0.12905707210302353,
      "num_tokens": 2370148.0,
      "step": 1365
    },
    {
      "entropy": 6.537519788742065,
      "epoch": 0.10659404785061272,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999907029449791,
      "loss": 6.186,
      "mean_token_accuracy": 0.131666212528944,
      "num_tokens": 2379105.0,
      "step": 1370
    },
    {
      "entropy": 6.4747106552124025,
      "epoch": 0.10698307722233029,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999904492870663,
      "loss": 6.2698,
      "mean_token_accuracy": 0.12513106688857079,
      "num_tokens": 2387417.0,
      "step": 1375
    },
    {
      "entropy": 6.521770191192627,
      "epoch": 0.10737210659404785,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999901922152816,
      "loss": 6.2904,
      "mean_token_accuracy": 0.13334674686193465,
      "num_tokens": 2396163.0,
      "step": 1380
    },
    {
      "entropy": 6.347804832458496,
      "epoch": 0.10776113596576542,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999899317296288,
      "loss": 6.1475,
      "mean_token_accuracy": 0.12342875301837922,
      "num_tokens": 2404661.0,
      "step": 1385
    },
    {
      "entropy": 6.447468996047974,
      "epoch": 0.10815016533748298,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004999896678301121,
      "loss": 6.2422,
      "mean_token_accuracy": 0.12361390069127083,
      "num_tokens": 2413459.0,
      "step": 1390
    },
    {
      "entropy": 6.573637056350708,
      "epoch": 0.10853919470920055,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999894005167352,
      "loss": 6.1903,
      "mean_token_accuracy": 0.1322077475488186,
      "num_tokens": 2421401.0,
      "step": 1395
    },
    {
      "entropy": 6.409449672698974,
      "epoch": 0.10892822408091811,
      "grad_norm": 1.15625,
      "learning_rate": 0.000499989129789502,
      "loss": 6.2017,
      "mean_token_accuracy": 0.1325265347957611,
      "num_tokens": 2429983.0,
      "step": 1400
    },
    {
      "entropy": 6.544249486923218,
      "epoch": 0.10931725345263567,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999888556484172,
      "loss": 6.3394,
      "mean_token_accuracy": 0.1189575508236885,
      "num_tokens": 2439206.0,
      "step": 1405
    },
    {
      "entropy": 6.494714403152466,
      "epoch": 0.10970628282435324,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999885780934845,
      "loss": 6.252,
      "mean_token_accuracy": 0.1334316536784172,
      "num_tokens": 2448801.0,
      "step": 1410
    },
    {
      "entropy": 6.461113309860229,
      "epoch": 0.1100953121960708,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004999882971247081,
      "loss": 6.2983,
      "mean_token_accuracy": 0.12159521356225014,
      "num_tokens": 2458007.0,
      "step": 1415
    },
    {
      "entropy": 6.482313394546509,
      "epoch": 0.11048434156778837,
      "grad_norm": 1.25,
      "learning_rate": 0.0004999880127420926,
      "loss": 6.1931,
      "mean_token_accuracy": 0.13716387897729873,
      "num_tokens": 2466574.0,
      "step": 1420
    },
    {
      "entropy": 6.530479192733765,
      "epoch": 0.11087337093950593,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999877249456421,
      "loss": 6.2612,
      "mean_token_accuracy": 0.12397516667842864,
      "num_tokens": 2475263.0,
      "step": 1425
    },
    {
      "entropy": 6.218584966659546,
      "epoch": 0.1112624003112235,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999874337353609,
      "loss": 6.0187,
      "mean_token_accuracy": 0.13271408900618553,
      "num_tokens": 2483738.0,
      "step": 1430
    },
    {
      "entropy": 6.438686561584473,
      "epoch": 0.11165142968294106,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999871391112535,
      "loss": 6.1274,
      "mean_token_accuracy": 0.1266042783856392,
      "num_tokens": 2492937.0,
      "step": 1435
    },
    {
      "entropy": 6.360862302780151,
      "epoch": 0.11204045905465862,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999868410733244,
      "loss": 6.2129,
      "mean_token_accuracy": 0.12471674755215645,
      "num_tokens": 2500716.0,
      "step": 1440
    },
    {
      "entropy": 6.442197370529175,
      "epoch": 0.11242948842637619,
      "grad_norm": 1.1875,
      "learning_rate": 0.000499986539621578,
      "loss": 6.1963,
      "mean_token_accuracy": 0.1276423953473568,
      "num_tokens": 2509385.0,
      "step": 1445
    },
    {
      "entropy": 6.529846048355102,
      "epoch": 0.11281851779809375,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999862347560191,
      "loss": 6.2579,
      "mean_token_accuracy": 0.12066849693655968,
      "num_tokens": 2518179.0,
      "step": 1450
    },
    {
      "entropy": 6.494664096832276,
      "epoch": 0.11320754716981132,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999859264766521,
      "loss": 6.2111,
      "mean_token_accuracy": 0.12627880051732063,
      "num_tokens": 2526644.0,
      "step": 1455
    },
    {
      "entropy": 6.364485549926758,
      "epoch": 0.11359657654152888,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999856147834817,
      "loss": 6.1292,
      "mean_token_accuracy": 0.1283652275800705,
      "num_tokens": 2534998.0,
      "step": 1460
    },
    {
      "entropy": 6.485772514343262,
      "epoch": 0.11398560591324645,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999852996765129,
      "loss": 6.2034,
      "mean_token_accuracy": 0.12745219245553016,
      "num_tokens": 2543562.0,
      "step": 1465
    },
    {
      "entropy": 6.43467493057251,
      "epoch": 0.11437463528496401,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999849811557502,
      "loss": 6.2066,
      "mean_token_accuracy": 0.1214485488831997,
      "num_tokens": 2551688.0,
      "step": 1470
    },
    {
      "entropy": 6.4273097038269045,
      "epoch": 0.11476366465668159,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999846592211984,
      "loss": 6.265,
      "mean_token_accuracy": 0.12918860763311385,
      "num_tokens": 2560227.0,
      "step": 1475
    },
    {
      "entropy": 6.4545982837677,
      "epoch": 0.11515269402839914,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999843338728626,
      "loss": 6.1509,
      "mean_token_accuracy": 0.13062360733747483,
      "num_tokens": 2569245.0,
      "step": 1480
    },
    {
      "entropy": 6.416768312454224,
      "epoch": 0.1155417234001167,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999840051107476,
      "loss": 6.1753,
      "mean_token_accuracy": 0.12632442712783815,
      "num_tokens": 2577706.0,
      "step": 1485
    },
    {
      "entropy": 6.441485500335693,
      "epoch": 0.11593075277183428,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999836729348584,
      "loss": 6.2026,
      "mean_token_accuracy": 0.1314474403858185,
      "num_tokens": 2586075.0,
      "step": 1490
    },
    {
      "entropy": 6.376056575775147,
      "epoch": 0.11631978214355183,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999833373452,
      "loss": 6.2116,
      "mean_token_accuracy": 0.12364233657717705,
      "num_tokens": 2594965.0,
      "step": 1495
    },
    {
      "entropy": 6.444352674484253,
      "epoch": 0.1167088115152694,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004999829983417778,
      "loss": 6.1165,
      "mean_token_accuracy": 0.13220934048295022,
      "num_tokens": 2603355.0,
      "step": 1500
    },
    {
      "entropy": 6.441127586364746,
      "epoch": 0.11709784088698696,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999826559245965,
      "loss": 6.2668,
      "mean_token_accuracy": 0.12569701820611953,
      "num_tokens": 2611697.0,
      "step": 1505
    },
    {
      "entropy": 6.447865724563599,
      "epoch": 0.11748687025870454,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999823100936615,
      "loss": 6.145,
      "mean_token_accuracy": 0.13168494999408722,
      "num_tokens": 2620821.0,
      "step": 1510
    },
    {
      "entropy": 6.413182401657105,
      "epoch": 0.1178758996304221,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999819608489781,
      "loss": 6.139,
      "mean_token_accuracy": 0.12816296517848969,
      "num_tokens": 2628891.0,
      "step": 1515
    },
    {
      "entropy": 6.4053560256958,
      "epoch": 0.11826492900213967,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999816081905515,
      "loss": 6.2701,
      "mean_token_accuracy": 0.12410432398319245,
      "num_tokens": 2638641.0,
      "step": 1520
    },
    {
      "entropy": 6.446357250213623,
      "epoch": 0.11865395837385723,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999812521183872,
      "loss": 6.1571,
      "mean_token_accuracy": 0.12258520424365997,
      "num_tokens": 2646898.0,
      "step": 1525
    },
    {
      "entropy": 6.494231557846069,
      "epoch": 0.11904298774557479,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004999808926324903,
      "loss": 6.2181,
      "mean_token_accuracy": 0.11994621977210045,
      "num_tokens": 2655404.0,
      "step": 1530
    },
    {
      "entropy": 6.3410632610321045,
      "epoch": 0.11943201711729236,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999805297328665,
      "loss": 6.0821,
      "mean_token_accuracy": 0.13641492202877997,
      "num_tokens": 2663851.0,
      "step": 1535
    },
    {
      "entropy": 6.395636558532715,
      "epoch": 0.11982104648900992,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999801634195214,
      "loss": 6.1436,
      "mean_token_accuracy": 0.12901071980595588,
      "num_tokens": 2672231.0,
      "step": 1540
    },
    {
      "entropy": 6.345880556106567,
      "epoch": 0.12021007586072749,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999797936924603,
      "loss": 6.1482,
      "mean_token_accuracy": 0.13240571096539497,
      "num_tokens": 2680198.0,
      "step": 1545
    },
    {
      "entropy": 6.534345436096191,
      "epoch": 0.12059910523244505,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499979420551689,
      "loss": 6.18,
      "mean_token_accuracy": 0.12448744922876358,
      "num_tokens": 2689208.0,
      "step": 1550
    },
    {
      "entropy": 6.408984041213989,
      "epoch": 0.12098813460416262,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499979043997213,
      "loss": 6.2361,
      "mean_token_accuracy": 0.1251956507563591,
      "num_tokens": 2698116.0,
      "step": 1555
    },
    {
      "entropy": 6.381083917617798,
      "epoch": 0.12137716397588018,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999786640290381,
      "loss": 6.1402,
      "mean_token_accuracy": 0.12055975571274757,
      "num_tokens": 2707046.0,
      "step": 1560
    },
    {
      "entropy": 6.352626132965088,
      "epoch": 0.12176619334759774,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999782806471701,
      "loss": 6.1304,
      "mean_token_accuracy": 0.12507664933800697,
      "num_tokens": 2714929.0,
      "step": 1565
    },
    {
      "entropy": 6.367618370056152,
      "epoch": 0.12215522271931531,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999778938516147,
      "loss": 6.1508,
      "mean_token_accuracy": 0.1280180834233761,
      "num_tokens": 2723436.0,
      "step": 1570
    },
    {
      "entropy": 6.3616029739379885,
      "epoch": 0.12254425209103287,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000499977503642378,
      "loss": 6.1914,
      "mean_token_accuracy": 0.12718005925416948,
      "num_tokens": 2731935.0,
      "step": 1575
    },
    {
      "entropy": 6.357917261123657,
      "epoch": 0.12293328146275044,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999771100194656,
      "loss": 6.03,
      "mean_token_accuracy": 0.13522375822067262,
      "num_tokens": 2740307.0,
      "step": 1580
    },
    {
      "entropy": 6.297295475006104,
      "epoch": 0.123322310834468,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999767129828837,
      "loss": 6.0432,
      "mean_token_accuracy": 0.13734385818243028,
      "num_tokens": 2749077.0,
      "step": 1585
    },
    {
      "entropy": 6.373191833496094,
      "epoch": 0.12371134020618557,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999763125326383,
      "loss": 6.171,
      "mean_token_accuracy": 0.12406648844480514,
      "num_tokens": 2758133.0,
      "step": 1590
    },
    {
      "entropy": 6.403712177276612,
      "epoch": 0.12410036957790313,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999759086687355,
      "loss": 6.0809,
      "mean_token_accuracy": 0.13498054668307305,
      "num_tokens": 2766509.0,
      "step": 1595
    },
    {
      "entropy": 6.323156118392944,
      "epoch": 0.1244893989496207,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999755013911813,
      "loss": 6.2524,
      "mean_token_accuracy": 0.12937115207314492,
      "num_tokens": 2775908.0,
      "step": 1600
    },
    {
      "entropy": 6.363321352005005,
      "epoch": 0.12487842832133826,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499975090699982,
      "loss": 6.0216,
      "mean_token_accuracy": 0.13876289501786232,
      "num_tokens": 2784803.0,
      "step": 1605
    },
    {
      "entropy": 6.361731815338135,
      "epoch": 0.12526745769305583,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999746765951438,
      "loss": 6.1877,
      "mean_token_accuracy": 0.12857512682676314,
      "num_tokens": 2793897.0,
      "step": 1610
    },
    {
      "entropy": 6.381125402450562,
      "epoch": 0.1256564870647734,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999742590766729,
      "loss": 6.0135,
      "mean_token_accuracy": 0.1335058644413948,
      "num_tokens": 2802652.0,
      "step": 1615
    },
    {
      "entropy": 6.288719320297242,
      "epoch": 0.12604551643649095,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999738381445757,
      "loss": 6.1555,
      "mean_token_accuracy": 0.12538140416145324,
      "num_tokens": 2811527.0,
      "step": 1620
    },
    {
      "entropy": 6.426501035690308,
      "epoch": 0.1264345458082085,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999734137988586,
      "loss": 6.0233,
      "mean_token_accuracy": 0.13478974401950836,
      "num_tokens": 2819410.0,
      "step": 1625
    },
    {
      "entropy": 6.1437897205352785,
      "epoch": 0.1268235751799261,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499972986039528,
      "loss": 6.0193,
      "mean_token_accuracy": 0.1394945725798607,
      "num_tokens": 2828079.0,
      "step": 1630
    },
    {
      "entropy": 6.358804178237915,
      "epoch": 0.12721260455164365,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999725548665904,
      "loss": 6.125,
      "mean_token_accuracy": 0.13202060833573342,
      "num_tokens": 2836326.0,
      "step": 1635
    },
    {
      "entropy": 6.331844520568848,
      "epoch": 0.1276016339233612,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999721202800524,
      "loss": 6.1777,
      "mean_token_accuracy": 0.12176662310957909,
      "num_tokens": 2845271.0,
      "step": 1640
    },
    {
      "entropy": 6.287081146240235,
      "epoch": 0.12799066329507877,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999716822799206,
      "loss": 6.0484,
      "mean_token_accuracy": 0.13889054954051971,
      "num_tokens": 2853688.0,
      "step": 1645
    },
    {
      "entropy": 6.34264121055603,
      "epoch": 0.12837969266679633,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999712408662016,
      "loss": 6.1345,
      "mean_token_accuracy": 0.13189240470528601,
      "num_tokens": 2862261.0,
      "step": 1650
    },
    {
      "entropy": 6.361407041549683,
      "epoch": 0.12876872203851392,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999707960389021,
      "loss": 6.1623,
      "mean_token_accuracy": 0.12860116735100746,
      "num_tokens": 2871324.0,
      "step": 1655
    },
    {
      "entropy": 6.39573564529419,
      "epoch": 0.12915775141023147,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999703477980288,
      "loss": 6.2068,
      "mean_token_accuracy": 0.12376175597310066,
      "num_tokens": 2880268.0,
      "step": 1660
    },
    {
      "entropy": 6.376931095123291,
      "epoch": 0.12954678078194903,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999698961435885,
      "loss": 6.1084,
      "mean_token_accuracy": 0.13345540538430214,
      "num_tokens": 2888510.0,
      "step": 1665
    },
    {
      "entropy": 6.306573915481567,
      "epoch": 0.1299358101536666,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999694410755882,
      "loss": 6.0144,
      "mean_token_accuracy": 0.14056088551878929,
      "num_tokens": 2897616.0,
      "step": 1670
    },
    {
      "entropy": 6.2575867652893065,
      "epoch": 0.13032483952538418,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999689825940347,
      "loss": 6.0963,
      "mean_token_accuracy": 0.13638013154268264,
      "num_tokens": 2905703.0,
      "step": 1675
    },
    {
      "entropy": 6.297283267974853,
      "epoch": 0.13071386889710174,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999685206989349,
      "loss": 5.9894,
      "mean_token_accuracy": 0.13844015896320344,
      "num_tokens": 2914881.0,
      "step": 1680
    },
    {
      "entropy": 6.246834802627563,
      "epoch": 0.1311028982688193,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999680553902959,
      "loss": 6.0484,
      "mean_token_accuracy": 0.1338697463274002,
      "num_tokens": 2923563.0,
      "step": 1685
    },
    {
      "entropy": 6.416929101943969,
      "epoch": 0.13149192764053685,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999675866681247,
      "loss": 6.2635,
      "mean_token_accuracy": 0.12606729716062545,
      "num_tokens": 2932821.0,
      "step": 1690
    },
    {
      "entropy": 6.340527868270874,
      "epoch": 0.1318809570122544,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999671145324286,
      "loss": 6.1476,
      "mean_token_accuracy": 0.13183779865503312,
      "num_tokens": 2940926.0,
      "step": 1695
    },
    {
      "entropy": 6.358092355728149,
      "epoch": 0.132269986383972,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999666389832144,
      "loss": 6.0869,
      "mean_token_accuracy": 0.13578122183680535,
      "num_tokens": 2950028.0,
      "step": 1700
    },
    {
      "entropy": 6.280818748474121,
      "epoch": 0.13265901575568956,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999661600204897,
      "loss": 6.1048,
      "mean_token_accuracy": 0.13189996257424355,
      "num_tokens": 2957636.0,
      "step": 1705
    },
    {
      "entropy": 6.183881616592407,
      "epoch": 0.13304804512740712,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999656776442615,
      "loss": 5.8677,
      "mean_token_accuracy": 0.13509365245699884,
      "num_tokens": 2966544.0,
      "step": 1710
    },
    {
      "entropy": 6.306702184677124,
      "epoch": 0.13343707449912467,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999651918545372,
      "loss": 6.1133,
      "mean_token_accuracy": 0.1262418359518051,
      "num_tokens": 2976236.0,
      "step": 1715
    },
    {
      "entropy": 6.283322811126709,
      "epoch": 0.13382610387084226,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999647026513243,
      "loss": 6.0487,
      "mean_token_accuracy": 0.1375434212386608,
      "num_tokens": 2984261.0,
      "step": 1720
    },
    {
      "entropy": 6.210408878326416,
      "epoch": 0.13421513324255982,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00049996421003463,
      "loss": 6.0494,
      "mean_token_accuracy": 0.13253966495394706,
      "num_tokens": 2992970.0,
      "step": 1725
    },
    {
      "entropy": 6.381523132324219,
      "epoch": 0.13460416261427738,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004999637140044619,
      "loss": 6.0525,
      "mean_token_accuracy": 0.13889826759696006,
      "num_tokens": 3001356.0,
      "step": 1730
    },
    {
      "entropy": 6.240667819976807,
      "epoch": 0.13499319198599494,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999632145608275,
      "loss": 6.0844,
      "mean_token_accuracy": 0.13955168649554253,
      "num_tokens": 3009509.0,
      "step": 1735
    },
    {
      "entropy": 6.309776020050049,
      "epoch": 0.1353822213577125,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999627117037343,
      "loss": 6.0914,
      "mean_token_accuracy": 0.12833233550190926,
      "num_tokens": 3017675.0,
      "step": 1740
    },
    {
      "entropy": 6.2961413860321045,
      "epoch": 0.13577125072943008,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00049996220543319,
      "loss": 6.0664,
      "mean_token_accuracy": 0.13567132800817489,
      "num_tokens": 3025527.0,
      "step": 1745
    },
    {
      "entropy": 6.275249338150024,
      "epoch": 0.13616028010114764,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999616957492025,
      "loss": 6.0121,
      "mean_token_accuracy": 0.1365427479147911,
      "num_tokens": 3034251.0,
      "step": 1750
    },
    {
      "entropy": 6.229633617401123,
      "epoch": 0.1365493094728652,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999611826517793,
      "loss": 6.2377,
      "mean_token_accuracy": 0.12501248195767403,
      "num_tokens": 3043272.0,
      "step": 1755
    },
    {
      "entropy": 6.347791910171509,
      "epoch": 0.13693833884458276,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999606661409281,
      "loss": 6.0884,
      "mean_token_accuracy": 0.1348442792892456,
      "num_tokens": 3051092.0,
      "step": 1760
    },
    {
      "entropy": 6.319974565505982,
      "epoch": 0.13732736821630034,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999601462166569,
      "loss": 6.0594,
      "mean_token_accuracy": 0.1309550181031227,
      "num_tokens": 3060584.0,
      "step": 1765
    },
    {
      "entropy": 6.226590776443482,
      "epoch": 0.1377163975880179,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999596228789736,
      "loss": 6.0635,
      "mean_token_accuracy": 0.13423156440258027,
      "num_tokens": 3069990.0,
      "step": 1770
    },
    {
      "entropy": 6.335338115692139,
      "epoch": 0.13810542695973546,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999590961278859,
      "loss": 6.0482,
      "mean_token_accuracy": 0.13068760186433792,
      "num_tokens": 3078837.0,
      "step": 1775
    },
    {
      "entropy": 6.172453451156616,
      "epoch": 0.13849445633145302,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999585659634021,
      "loss": 6.0316,
      "mean_token_accuracy": 0.1395242527127266,
      "num_tokens": 3087557.0,
      "step": 1780
    },
    {
      "entropy": 6.260955667495727,
      "epoch": 0.13888348570317058,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999580323855301,
      "loss": 6.0262,
      "mean_token_accuracy": 0.13657170236110688,
      "num_tokens": 3096524.0,
      "step": 1785
    },
    {
      "entropy": 6.351543807983399,
      "epoch": 0.13927251507488816,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999574953942781,
      "loss": 6.1139,
      "mean_token_accuracy": 0.1336659722030163,
      "num_tokens": 3105271.0,
      "step": 1790
    },
    {
      "entropy": 6.237719345092773,
      "epoch": 0.13966154444660572,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999569549896541,
      "loss": 5.9924,
      "mean_token_accuracy": 0.1303889974951744,
      "num_tokens": 3114479.0,
      "step": 1795
    },
    {
      "entropy": 6.231583595275879,
      "epoch": 0.14005057381832328,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999564111716665,
      "loss": 6.0071,
      "mean_token_accuracy": 0.14318212792277335,
      "num_tokens": 3123319.0,
      "step": 1800
    },
    {
      "entropy": 6.270225238800049,
      "epoch": 0.14043960319004084,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999558639403232,
      "loss": 6.0457,
      "mean_token_accuracy": 0.13263603895902634,
      "num_tokens": 3132532.0,
      "step": 1805
    },
    {
      "entropy": 6.358526659011841,
      "epoch": 0.1408286325617584,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999553132956328,
      "loss": 6.1067,
      "mean_token_accuracy": 0.13037939891219139,
      "num_tokens": 3140798.0,
      "step": 1810
    },
    {
      "entropy": 6.238439464569092,
      "epoch": 0.14121766193347599,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999547592376035,
      "loss": 6.0901,
      "mean_token_accuracy": 0.13772427588701247,
      "num_tokens": 3149607.0,
      "step": 1815
    },
    {
      "entropy": 6.2145038604736325,
      "epoch": 0.14160669130519354,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004999542017662438,
      "loss": 5.9226,
      "mean_token_accuracy": 0.14179680794477462,
      "num_tokens": 3157537.0,
      "step": 1820
    },
    {
      "entropy": 6.2342335224151615,
      "epoch": 0.1419957206769111,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999536408815623,
      "loss": 6.0715,
      "mean_token_accuracy": 0.1310746893286705,
      "num_tokens": 3166351.0,
      "step": 1825
    },
    {
      "entropy": 6.177899360656738,
      "epoch": 0.14238475004862866,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999530765835672,
      "loss": 5.935,
      "mean_token_accuracy": 0.14044117629528047,
      "num_tokens": 3174690.0,
      "step": 1830
    },
    {
      "entropy": 6.239687252044678,
      "epoch": 0.14277377942034625,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999525088722673,
      "loss": 6.0764,
      "mean_token_accuracy": 0.12794848904013634,
      "num_tokens": 3183460.0,
      "step": 1835
    },
    {
      "entropy": 6.237521076202393,
      "epoch": 0.1431628087920638,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999519377476711,
      "loss": 6.0105,
      "mean_token_accuracy": 0.13746723383665085,
      "num_tokens": 3192183.0,
      "step": 1840
    },
    {
      "entropy": 6.266400766372681,
      "epoch": 0.14355183816378136,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999513632097872,
      "loss": 6.0966,
      "mean_token_accuracy": 0.13392534032464026,
      "num_tokens": 3200875.0,
      "step": 1845
    },
    {
      "entropy": 6.223610925674438,
      "epoch": 0.14394086753549892,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999507852586245,
      "loss": 5.9609,
      "mean_token_accuracy": 0.13770486116409303,
      "num_tokens": 3209797.0,
      "step": 1850
    },
    {
      "entropy": 6.276148748397827,
      "epoch": 0.14432989690721648,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999502038941917,
      "loss": 6.0793,
      "mean_token_accuracy": 0.13013245537877083,
      "num_tokens": 3218612.0,
      "step": 1855
    },
    {
      "entropy": 6.2181178569793705,
      "epoch": 0.14471892627893407,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999496191164975,
      "loss": 5.9958,
      "mean_token_accuracy": 0.13566022291779517,
      "num_tokens": 3227364.0,
      "step": 1860
    },
    {
      "entropy": 6.176430511474609,
      "epoch": 0.14510795565065163,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499949030925551,
      "loss": 6.0201,
      "mean_token_accuracy": 0.13599291145801545,
      "num_tokens": 3236448.0,
      "step": 1865
    },
    {
      "entropy": 6.254899358749389,
      "epoch": 0.14549698502236919,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999484393213609,
      "loss": 5.9486,
      "mean_token_accuracy": 0.1352322556078434,
      "num_tokens": 3244957.0,
      "step": 1870
    },
    {
      "entropy": 6.109967184066773,
      "epoch": 0.14588601439408674,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999478443039363,
      "loss": 6.0408,
      "mean_token_accuracy": 0.1295456551015377,
      "num_tokens": 3253250.0,
      "step": 1875
    },
    {
      "entropy": 6.2762659072875975,
      "epoch": 0.14627504376580433,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999472458732862,
      "loss": 5.943,
      "mean_token_accuracy": 0.14542312696576118,
      "num_tokens": 3261455.0,
      "step": 1880
    },
    {
      "entropy": 6.1596604347229,
      "epoch": 0.1466640731375219,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999466440294198,
      "loss": 5.9906,
      "mean_token_accuracy": 0.13770747035741807,
      "num_tokens": 3270108.0,
      "step": 1885
    },
    {
      "entropy": 6.310276889801026,
      "epoch": 0.14705310250923945,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499946038772346,
      "loss": 6.1231,
      "mean_token_accuracy": 0.13440992087125778,
      "num_tokens": 3278425.0,
      "step": 1890
    },
    {
      "entropy": 6.235618352890015,
      "epoch": 0.147442131880957,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999454301020741,
      "loss": 6.0429,
      "mean_token_accuracy": 0.13205524757504464,
      "num_tokens": 3288358.0,
      "step": 1895
    },
    {
      "entropy": 6.265941572189331,
      "epoch": 0.14783116125267456,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999448180186133,
      "loss": 6.039,
      "mean_token_accuracy": 0.1352785974740982,
      "num_tokens": 3297199.0,
      "step": 1900
    },
    {
      "entropy": 6.18539137840271,
      "epoch": 0.14822019062439215,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499944202521973,
      "loss": 6.047,
      "mean_token_accuracy": 0.13134037628769873,
      "num_tokens": 3306350.0,
      "step": 1905
    },
    {
      "entropy": 6.20574164390564,
      "epoch": 0.1486092199961097,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999435836121624,
      "loss": 5.8873,
      "mean_token_accuracy": 0.13750338256359101,
      "num_tokens": 3314921.0,
      "step": 1910
    },
    {
      "entropy": 6.084950733184814,
      "epoch": 0.14899824936782727,
      "grad_norm": 1.140625,
      "learning_rate": 0.000499942961289191,
      "loss": 6.018,
      "mean_token_accuracy": 0.14249766021966934,
      "num_tokens": 3323324.0,
      "step": 1915
    },
    {
      "entropy": 6.381536149978638,
      "epoch": 0.14938727873954483,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999423355530683,
      "loss": 6.1469,
      "mean_token_accuracy": 0.13054955527186393,
      "num_tokens": 3331614.0,
      "step": 1920
    },
    {
      "entropy": 6.248685646057129,
      "epoch": 0.1497763081112624,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999417064038035,
      "loss": 5.9945,
      "mean_token_accuracy": 0.13594073355197905,
      "num_tokens": 3339573.0,
      "step": 1925
    },
    {
      "entropy": 6.163447618484497,
      "epoch": 0.15016533748297997,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999410738414065,
      "loss": 5.9538,
      "mean_token_accuracy": 0.14307861253619195,
      "num_tokens": 3347775.0,
      "step": 1930
    },
    {
      "entropy": 6.200021457672119,
      "epoch": 0.15055436685469753,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999404378658866,
      "loss": 5.9128,
      "mean_token_accuracy": 0.1373552069067955,
      "num_tokens": 3356947.0,
      "step": 1935
    },
    {
      "entropy": 6.167617988586426,
      "epoch": 0.1509433962264151,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999397984772536,
      "loss": 6.0098,
      "mean_token_accuracy": 0.13761954754590988,
      "num_tokens": 3365420.0,
      "step": 1940
    },
    {
      "entropy": 6.2827473163604735,
      "epoch": 0.15133242559813265,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999391556755172,
      "loss": 6.1027,
      "mean_token_accuracy": 0.13397282212972642,
      "num_tokens": 3373483.0,
      "step": 1945
    },
    {
      "entropy": 6.290587139129639,
      "epoch": 0.15172145496985023,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999385094606872,
      "loss": 6.0892,
      "mean_token_accuracy": 0.12818740382790567,
      "num_tokens": 3382557.0,
      "step": 1950
    },
    {
      "entropy": 6.261412334442139,
      "epoch": 0.1521104843415678,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999378598327733,
      "loss": 6.0347,
      "mean_token_accuracy": 0.12742510437965393,
      "num_tokens": 3392079.0,
      "step": 1955
    },
    {
      "entropy": 6.222420072555542,
      "epoch": 0.15249951371328535,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999372067917854,
      "loss": 6.1003,
      "mean_token_accuracy": 0.1310837060213089,
      "num_tokens": 3401070.0,
      "step": 1960
    },
    {
      "entropy": 6.165682315826416,
      "epoch": 0.1528885430850029,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999365503377335,
      "loss": 5.9818,
      "mean_token_accuracy": 0.13428688049316406,
      "num_tokens": 3409921.0,
      "step": 1965
    },
    {
      "entropy": 6.244135046005249,
      "epoch": 0.1532775724567205,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999358904706273,
      "loss": 5.9986,
      "mean_token_accuracy": 0.13104057684540749,
      "num_tokens": 3419149.0,
      "step": 1970
    },
    {
      "entropy": 6.156588792800903,
      "epoch": 0.15366660182843805,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999352271904771,
      "loss": 5.8891,
      "mean_token_accuracy": 0.14492207542061805,
      "num_tokens": 3427435.0,
      "step": 1975
    },
    {
      "entropy": 6.24048261642456,
      "epoch": 0.1540556312001556,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999345604972928,
      "loss": 6.0975,
      "mean_token_accuracy": 0.1309770740568638,
      "num_tokens": 3436377.0,
      "step": 1980
    },
    {
      "entropy": 6.262222719192505,
      "epoch": 0.15444466057187317,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999338903910846,
      "loss": 5.966,
      "mean_token_accuracy": 0.14030917212367058,
      "num_tokens": 3444997.0,
      "step": 1985
    },
    {
      "entropy": 6.250793886184693,
      "epoch": 0.15483368994359073,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999332168718626,
      "loss": 6.0267,
      "mean_token_accuracy": 0.13967675864696502,
      "num_tokens": 3454254.0,
      "step": 1990
    },
    {
      "entropy": 6.192072057723999,
      "epoch": 0.15522271931530832,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999325399396371,
      "loss": 5.9984,
      "mean_token_accuracy": 0.14047476574778556,
      "num_tokens": 3462853.0,
      "step": 1995
    },
    {
      "entropy": 6.162121677398682,
      "epoch": 0.15561174868702587,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999318595944183,
      "loss": 5.951,
      "mean_token_accuracy": 0.1363219663500786,
      "num_tokens": 3471288.0,
      "step": 2000
    },
    {
      "entropy": 6.123317289352417,
      "epoch": 0.15600077805874343,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999311758362166,
      "loss": 5.8464,
      "mean_token_accuracy": 0.13612058088183404,
      "num_tokens": 3479479.0,
      "step": 2005
    },
    {
      "entropy": 6.14522500038147,
      "epoch": 0.156389807430461,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999304886650422,
      "loss": 5.8684,
      "mean_token_accuracy": 0.1447673887014389,
      "num_tokens": 3488621.0,
      "step": 2010
    },
    {
      "entropy": 6.089516735076904,
      "epoch": 0.15677883680217858,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999297980809058,
      "loss": 5.9238,
      "mean_token_accuracy": 0.1373309776186943,
      "num_tokens": 3497792.0,
      "step": 2015
    },
    {
      "entropy": 6.12142162322998,
      "epoch": 0.15716786617389614,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999291040838176,
      "loss": 5.9522,
      "mean_token_accuracy": 0.13544787019491195,
      "num_tokens": 3506548.0,
      "step": 2020
    },
    {
      "entropy": 6.127046966552735,
      "epoch": 0.1575568955456137,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999284066737884,
      "loss": 5.932,
      "mean_token_accuracy": 0.14313506335020065,
      "num_tokens": 3515086.0,
      "step": 2025
    },
    {
      "entropy": 6.245706987380982,
      "epoch": 0.15794592491733125,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999277058508285,
      "loss": 5.9765,
      "mean_token_accuracy": 0.1334720067679882,
      "num_tokens": 3523985.0,
      "step": 2030
    },
    {
      "entropy": 6.193225526809693,
      "epoch": 0.1583349542890488,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999270016149486,
      "loss": 6.0321,
      "mean_token_accuracy": 0.1414737358689308,
      "num_tokens": 3532575.0,
      "step": 2035
    },
    {
      "entropy": 6.229839229583741,
      "epoch": 0.1587239836607664,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999262939661596,
      "loss": 6.0292,
      "mean_token_accuracy": 0.13414048179984092,
      "num_tokens": 3541558.0,
      "step": 2040
    },
    {
      "entropy": 6.1485107898712155,
      "epoch": 0.15911301303248396,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999255829044721,
      "loss": 5.9383,
      "mean_token_accuracy": 0.1430905818939209,
      "num_tokens": 3549697.0,
      "step": 2045
    },
    {
      "entropy": 6.240954160690308,
      "epoch": 0.15950204240420152,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999248684298968,
      "loss": 6.1118,
      "mean_token_accuracy": 0.12798432558774947,
      "num_tokens": 3558673.0,
      "step": 2050
    },
    {
      "entropy": 6.225488805770874,
      "epoch": 0.15989107177591907,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999241505424447,
      "loss": 5.963,
      "mean_token_accuracy": 0.14219792932271957,
      "num_tokens": 3567176.0,
      "step": 2055
    },
    {
      "entropy": 6.276889419555664,
      "epoch": 0.16028010114763663,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999234292421265,
      "loss": 6.0934,
      "mean_token_accuracy": 0.13421509712934493,
      "num_tokens": 3576020.0,
      "step": 2060
    },
    {
      "entropy": 6.121796798706055,
      "epoch": 0.16066913051935422,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999227045289535,
      "loss": 5.9483,
      "mean_token_accuracy": 0.13852609768509866,
      "num_tokens": 3584705.0,
      "step": 2065
    },
    {
      "entropy": 6.239097929000854,
      "epoch": 0.16105815989107178,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999219764029363,
      "loss": 6.1251,
      "mean_token_accuracy": 0.1327233374118805,
      "num_tokens": 3593339.0,
      "step": 2070
    },
    {
      "entropy": 6.17981219291687,
      "epoch": 0.16144718926278934,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999212448640861,
      "loss": 5.8296,
      "mean_token_accuracy": 0.14525919482111932,
      "num_tokens": 3601206.0,
      "step": 2075
    },
    {
      "entropy": 6.023618602752686,
      "epoch": 0.1618362186345069,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999205099124141,
      "loss": 5.9556,
      "mean_token_accuracy": 0.1340252086520195,
      "num_tokens": 3610371.0,
      "step": 2080
    },
    {
      "entropy": 6.26896104812622,
      "epoch": 0.16222524800622448,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999197715479313,
      "loss": 5.9675,
      "mean_token_accuracy": 0.13808080479502677,
      "num_tokens": 3618834.0,
      "step": 2085
    },
    {
      "entropy": 6.081180000305176,
      "epoch": 0.16261427737794204,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999190297706489,
      "loss": 5.9999,
      "mean_token_accuracy": 0.13936789706349373,
      "num_tokens": 3627426.0,
      "step": 2090
    },
    {
      "entropy": 6.239403820037841,
      "epoch": 0.1630033067496596,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999182845805783,
      "loss": 6.0013,
      "mean_token_accuracy": 0.13789541125297547,
      "num_tokens": 3636020.0,
      "step": 2095
    },
    {
      "entropy": 6.242777967453003,
      "epoch": 0.16339233612137716,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999175359777306,
      "loss": 5.9419,
      "mean_token_accuracy": 0.1419316790997982,
      "num_tokens": 3644419.0,
      "step": 2100
    },
    {
      "entropy": 6.171798038482666,
      "epoch": 0.16378136549309472,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999167839621174,
      "loss": 6.043,
      "mean_token_accuracy": 0.13091627210378648,
      "num_tokens": 3653223.0,
      "step": 2105
    },
    {
      "entropy": 6.23285026550293,
      "epoch": 0.1641703948648123,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999160285337501,
      "loss": 6.062,
      "mean_token_accuracy": 0.13183995485305786,
      "num_tokens": 3662409.0,
      "step": 2110
    },
    {
      "entropy": 6.214762258529663,
      "epoch": 0.16455942423652986,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004999152696926399,
      "loss": 6.0226,
      "mean_token_accuracy": 0.13156313672661782,
      "num_tokens": 3670635.0,
      "step": 2115
    },
    {
      "entropy": 6.076648998260498,
      "epoch": 0.16494845360824742,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999145074387985,
      "loss": 5.9493,
      "mean_token_accuracy": 0.14351119846105576,
      "num_tokens": 3679303.0,
      "step": 2120
    },
    {
      "entropy": 6.297281312942505,
      "epoch": 0.16533748297996498,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999137417722374,
      "loss": 6.0999,
      "mean_token_accuracy": 0.1333278849720955,
      "num_tokens": 3689299.0,
      "step": 2125
    },
    {
      "entropy": 6.212060928344727,
      "epoch": 0.16572651235168256,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999129726929684,
      "loss": 6.0309,
      "mean_token_accuracy": 0.141978856921196,
      "num_tokens": 3697843.0,
      "step": 2130
    },
    {
      "entropy": 6.232631540298462,
      "epoch": 0.16611554172340012,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999122002010029,
      "loss": 5.9595,
      "mean_token_accuracy": 0.13919129222631454,
      "num_tokens": 3705876.0,
      "step": 2135
    },
    {
      "entropy": 6.192243671417236,
      "epoch": 0.16650457109511768,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999114242963527,
      "loss": 6.0811,
      "mean_token_accuracy": 0.1308549664914608,
      "num_tokens": 3714835.0,
      "step": 2140
    },
    {
      "entropy": 6.1568906784057615,
      "epoch": 0.16689360046683524,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999106449790298,
      "loss": 5.9421,
      "mean_token_accuracy": 0.13271426409482956,
      "num_tokens": 3723421.0,
      "step": 2145
    },
    {
      "entropy": 6.174701166152954,
      "epoch": 0.1672826298385528,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999098622490458,
      "loss": 5.9582,
      "mean_token_accuracy": 0.1379171997308731,
      "num_tokens": 3731590.0,
      "step": 2150
    },
    {
      "entropy": 6.0976481437683105,
      "epoch": 0.16767165921027039,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999090761064125,
      "loss": 5.8819,
      "mean_token_accuracy": 0.1397905245423317,
      "num_tokens": 3740700.0,
      "step": 2155
    },
    {
      "entropy": 6.141346263885498,
      "epoch": 0.16806068858198794,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499908286551142,
      "loss": 5.9546,
      "mean_token_accuracy": 0.14155428856611252,
      "num_tokens": 3749516.0,
      "step": 2160
    },
    {
      "entropy": 6.109289598464966,
      "epoch": 0.1684497179537055,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999074935832463,
      "loss": 5.8731,
      "mean_token_accuracy": 0.14581690207123757,
      "num_tokens": 3757696.0,
      "step": 2165
    },
    {
      "entropy": 6.114305877685547,
      "epoch": 0.16883874732542306,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999066972027373,
      "loss": 5.9643,
      "mean_token_accuracy": 0.13819186612963677,
      "num_tokens": 3765786.0,
      "step": 2170
    },
    {
      "entropy": 6.129476547241211,
      "epoch": 0.16922777669714065,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999058974096271,
      "loss": 5.8694,
      "mean_token_accuracy": 0.14471860826015473,
      "num_tokens": 3773265.0,
      "step": 2175
    },
    {
      "entropy": 6.109765338897705,
      "epoch": 0.1696168060688582,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499905094203928,
      "loss": 5.9872,
      "mean_token_accuracy": 0.13433021679520607,
      "num_tokens": 3782402.0,
      "step": 2180
    },
    {
      "entropy": 6.180309247970581,
      "epoch": 0.17000583544057576,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000499904287585652,
      "loss": 6.0514,
      "mean_token_accuracy": 0.1314002774655819,
      "num_tokens": 3791901.0,
      "step": 2185
    },
    {
      "entropy": 6.139549350738525,
      "epoch": 0.17039486481229332,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999034775548113,
      "loss": 5.8778,
      "mean_token_accuracy": 0.14060187861323356,
      "num_tokens": 3799872.0,
      "step": 2190
    },
    {
      "entropy": 6.106409025192261,
      "epoch": 0.17078389418401088,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999026641114185,
      "loss": 5.8323,
      "mean_token_accuracy": 0.14609502255916595,
      "num_tokens": 3808949.0,
      "step": 2195
    },
    {
      "entropy": 6.121867275238037,
      "epoch": 0.17117292355572847,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999018472554857,
      "loss": 5.9466,
      "mean_token_accuracy": 0.13920955061912538,
      "num_tokens": 3817429.0,
      "step": 2200
    },
    {
      "entropy": 6.264400291442871,
      "epoch": 0.17156195292744603,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999010269870253,
      "loss": 6.2252,
      "mean_token_accuracy": 0.12967248037457466,
      "num_tokens": 3827101.0,
      "step": 2205
    },
    {
      "entropy": 6.248671436309815,
      "epoch": 0.17195098229916359,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999002033060498,
      "loss": 5.9701,
      "mean_token_accuracy": 0.13611226081848143,
      "num_tokens": 3836147.0,
      "step": 2210
    },
    {
      "entropy": 6.070646286010742,
      "epoch": 0.17234001167088114,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998993762125716,
      "loss": 5.9555,
      "mean_token_accuracy": 0.1376994989812374,
      "num_tokens": 3845300.0,
      "step": 2215
    },
    {
      "entropy": 6.248652124404908,
      "epoch": 0.17272904104259873,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004998985457066034,
      "loss": 5.9556,
      "mean_token_accuracy": 0.14077395275235177,
      "num_tokens": 3853620.0,
      "step": 2220
    },
    {
      "entropy": 6.078034114837647,
      "epoch": 0.1731180704143163,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998977117881576,
      "loss": 5.952,
      "mean_token_accuracy": 0.14219790250062941,
      "num_tokens": 3861734.0,
      "step": 2225
    },
    {
      "entropy": 6.147058439254761,
      "epoch": 0.17350709978603385,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998968744572472,
      "loss": 5.9706,
      "mean_token_accuracy": 0.13935380205512046,
      "num_tokens": 3870753.0,
      "step": 2230
    },
    {
      "entropy": 6.128642892837524,
      "epoch": 0.1738961291577514,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998960337138845,
      "loss": 5.852,
      "mean_token_accuracy": 0.14278298988938332,
      "num_tokens": 3879382.0,
      "step": 2235
    },
    {
      "entropy": 6.17204442024231,
      "epoch": 0.17428515852946896,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998951895580826,
      "loss": 6.0249,
      "mean_token_accuracy": 0.13453888818621634,
      "num_tokens": 3888417.0,
      "step": 2240
    },
    {
      "entropy": 6.054234218597412,
      "epoch": 0.17467418790118655,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998943419898542,
      "loss": 5.8813,
      "mean_token_accuracy": 0.1384780913591385,
      "num_tokens": 3897083.0,
      "step": 2245
    },
    {
      "entropy": 6.077237939834594,
      "epoch": 0.1750632172729041,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998934910092119,
      "loss": 5.8526,
      "mean_token_accuracy": 0.14408235251903534,
      "num_tokens": 3905564.0,
      "step": 2250
    },
    {
      "entropy": 6.146784496307373,
      "epoch": 0.17545224664462167,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998926366161689,
      "loss": 5.9942,
      "mean_token_accuracy": 0.13211930319666862,
      "num_tokens": 3914896.0,
      "step": 2255
    },
    {
      "entropy": 6.1404224872589115,
      "epoch": 0.17584127601633923,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998917788107383,
      "loss": 5.9972,
      "mean_token_accuracy": 0.13836849108338356,
      "num_tokens": 3924151.0,
      "step": 2260
    },
    {
      "entropy": 6.213042545318603,
      "epoch": 0.17623030538805678,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998909175929326,
      "loss": 5.9462,
      "mean_token_accuracy": 0.13971734270453454,
      "num_tokens": 3933163.0,
      "step": 2265
    },
    {
      "entropy": 6.098297929763794,
      "epoch": 0.17661933475977437,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998900529627653,
      "loss": 5.8671,
      "mean_token_accuracy": 0.14396204575896263,
      "num_tokens": 3941120.0,
      "step": 2270
    },
    {
      "entropy": 5.991336631774902,
      "epoch": 0.17700836413149193,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998891849202495,
      "loss": 5.7531,
      "mean_token_accuracy": 0.1438086897134781,
      "num_tokens": 3949896.0,
      "step": 2275
    },
    {
      "entropy": 6.120736980438233,
      "epoch": 0.1773973935032095,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499888313465398,
      "loss": 5.9127,
      "mean_token_accuracy": 0.13862551301717757,
      "num_tokens": 3958794.0,
      "step": 2280
    },
    {
      "entropy": 6.140617322921753,
      "epoch": 0.17778642287492705,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998874385982247,
      "loss": 5.9773,
      "mean_token_accuracy": 0.14108967706561087,
      "num_tokens": 3967267.0,
      "step": 2285
    },
    {
      "entropy": 6.083899354934692,
      "epoch": 0.17817545224664463,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998865603187421,
      "loss": 5.892,
      "mean_token_accuracy": 0.14135736748576164,
      "num_tokens": 3976536.0,
      "step": 2290
    },
    {
      "entropy": 6.207363796234131,
      "epoch": 0.1785644816183622,
      "grad_norm": 1.0,
      "learning_rate": 0.000499885678626964,
      "loss": 5.9564,
      "mean_token_accuracy": 0.13729356452822686,
      "num_tokens": 3986006.0,
      "step": 2295
    },
    {
      "entropy": 6.016284608840943,
      "epoch": 0.17895351099007975,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998847935229038,
      "loss": 5.9038,
      "mean_token_accuracy": 0.1368868224322796,
      "num_tokens": 3994478.0,
      "step": 2300
    },
    {
      "entropy": 6.116120862960815,
      "epoch": 0.1793425403617973,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998839050065746,
      "loss": 5.9516,
      "mean_token_accuracy": 0.1373460717499256,
      "num_tokens": 4003190.0,
      "step": 2305
    },
    {
      "entropy": 6.196177530288696,
      "epoch": 0.17973156973351487,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004998830130779902,
      "loss": 5.8728,
      "mean_token_accuracy": 0.1443126007914543,
      "num_tokens": 4011886.0,
      "step": 2310
    },
    {
      "entropy": 5.965580034255981,
      "epoch": 0.18012059910523245,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499882117737164,
      "loss": 5.7727,
      "mean_token_accuracy": 0.1477213680744171,
      "num_tokens": 4020792.0,
      "step": 2315
    },
    {
      "entropy": 5.96843729019165,
      "epoch": 0.18050962847695,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998812189841096,
      "loss": 5.7491,
      "mean_token_accuracy": 0.14528555423021317,
      "num_tokens": 4029235.0,
      "step": 2320
    },
    {
      "entropy": 6.036883306503296,
      "epoch": 0.18089865784866757,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998803168188407,
      "loss": 5.9845,
      "mean_token_accuracy": 0.14515455886721612,
      "num_tokens": 4037341.0,
      "step": 2325
    },
    {
      "entropy": 6.072755193710327,
      "epoch": 0.18128768722038513,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998794112413708,
      "loss": 5.8554,
      "mean_token_accuracy": 0.14166971370577813,
      "num_tokens": 4046324.0,
      "step": 2330
    },
    {
      "entropy": 6.041338300704956,
      "epoch": 0.18167671659210272,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998785022517137,
      "loss": 5.9319,
      "mean_token_accuracy": 0.1387798421084881,
      "num_tokens": 4054579.0,
      "step": 2335
    },
    {
      "entropy": 6.153178119659424,
      "epoch": 0.18206574596382027,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998775898498835,
      "loss": 5.8939,
      "mean_token_accuracy": 0.1421814054250717,
      "num_tokens": 4063087.0,
      "step": 2340
    },
    {
      "entropy": 6.121365165710449,
      "epoch": 0.18245477533553783,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998766740358936,
      "loss": 5.9531,
      "mean_token_accuracy": 0.14040177017450334,
      "num_tokens": 4072309.0,
      "step": 2345
    },
    {
      "entropy": 6.092164039611816,
      "epoch": 0.1828438047072554,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998757548097582,
      "loss": 5.9948,
      "mean_token_accuracy": 0.13580115213990213,
      "num_tokens": 4081017.0,
      "step": 2350
    },
    {
      "entropy": 6.170856666564942,
      "epoch": 0.18323283407897295,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998748321714911,
      "loss": 6.0183,
      "mean_token_accuracy": 0.13338493853807448,
      "num_tokens": 4089799.0,
      "step": 2355
    },
    {
      "entropy": 6.063455200195312,
      "epoch": 0.18362186345069054,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004998739061211065,
      "loss": 5.9271,
      "mean_token_accuracy": 0.14566840901970862,
      "num_tokens": 4097948.0,
      "step": 2360
    },
    {
      "entropy": 6.108722352981568,
      "epoch": 0.1840108928224081,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998729766586181,
      "loss": 5.8152,
      "mean_token_accuracy": 0.14332541227340698,
      "num_tokens": 4107050.0,
      "step": 2365
    },
    {
      "entropy": 6.023930311203003,
      "epoch": 0.18439992219412565,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998720437840403,
      "loss": 5.8309,
      "mean_token_accuracy": 0.13776944130659102,
      "num_tokens": 4115547.0,
      "step": 2370
    },
    {
      "entropy": 6.079334878921509,
      "epoch": 0.1847889515658432,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998711074973871,
      "loss": 5.8328,
      "mean_token_accuracy": 0.14491757974028588,
      "num_tokens": 4124047.0,
      "step": 2375
    },
    {
      "entropy": 5.999879503250122,
      "epoch": 0.1851779809375608,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998701677986728,
      "loss": 5.8029,
      "mean_token_accuracy": 0.14852844551205635,
      "num_tokens": 4132005.0,
      "step": 2380
    },
    {
      "entropy": 6.126709222793579,
      "epoch": 0.18556701030927836,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998692246879116,
      "loss": 5.9743,
      "mean_token_accuracy": 0.13506074547767638,
      "num_tokens": 4140199.0,
      "step": 2385
    },
    {
      "entropy": 6.115787029266357,
      "epoch": 0.18595603968099592,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998682781651178,
      "loss": 5.9031,
      "mean_token_accuracy": 0.1383533462882042,
      "num_tokens": 4149353.0,
      "step": 2390
    },
    {
      "entropy": 5.986810922622681,
      "epoch": 0.18634506905271347,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998673282303059,
      "loss": 5.8724,
      "mean_token_accuracy": 0.1399594560265541,
      "num_tokens": 4158531.0,
      "step": 2395
    },
    {
      "entropy": 5.963578128814698,
      "epoch": 0.18673409842443103,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998663748834901,
      "loss": 5.6698,
      "mean_token_accuracy": 0.15333113372325896,
      "num_tokens": 4166743.0,
      "step": 2400
    },
    {
      "entropy": 6.126782751083374,
      "epoch": 0.18712312779614862,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998654181246851,
      "loss": 6.0205,
      "mean_token_accuracy": 0.13510823845863343,
      "num_tokens": 4175416.0,
      "step": 2405
    },
    {
      "entropy": 6.130587100982666,
      "epoch": 0.18751215716786618,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998644579539052,
      "loss": 5.9563,
      "mean_token_accuracy": 0.13990581631660462,
      "num_tokens": 4185324.0,
      "step": 2410
    },
    {
      "entropy": 6.074210596084595,
      "epoch": 0.18790118653958374,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499863494371165,
      "loss": 5.8842,
      "mean_token_accuracy": 0.14497815519571305,
      "num_tokens": 4193361.0,
      "step": 2415
    },
    {
      "entropy": 6.126933288574219,
      "epoch": 0.1882902159113013,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998625273764793,
      "loss": 5.9397,
      "mean_token_accuracy": 0.13277981653809548,
      "num_tokens": 4202553.0,
      "step": 2420
    },
    {
      "entropy": 6.058669185638427,
      "epoch": 0.18867924528301888,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998615569698626,
      "loss": 5.8727,
      "mean_token_accuracy": 0.14963430315256118,
      "num_tokens": 4211585.0,
      "step": 2425
    },
    {
      "entropy": 5.997480392456055,
      "epoch": 0.18906827465473644,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998605831513296,
      "loss": 5.9164,
      "mean_token_accuracy": 0.13643818199634553,
      "num_tokens": 4220829.0,
      "step": 2430
    },
    {
      "entropy": 6.057970905303955,
      "epoch": 0.189457304026454,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998596059208953,
      "loss": 5.8181,
      "mean_token_accuracy": 0.14352203160524368,
      "num_tokens": 4228658.0,
      "step": 2435
    },
    {
      "entropy": 6.13070707321167,
      "epoch": 0.18984633339817156,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998586252785743,
      "loss": 5.9049,
      "mean_token_accuracy": 0.1442067116498947,
      "num_tokens": 4237650.0,
      "step": 2440
    },
    {
      "entropy": 6.030512046813965,
      "epoch": 0.19023536276988912,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998576412243816,
      "loss": 5.8831,
      "mean_token_accuracy": 0.14705661460757255,
      "num_tokens": 4246701.0,
      "step": 2445
    },
    {
      "entropy": 6.080326223373413,
      "epoch": 0.1906243921416067,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998566537583321,
      "loss": 5.8482,
      "mean_token_accuracy": 0.14321040436625482,
      "num_tokens": 4254541.0,
      "step": 2450
    },
    {
      "entropy": 6.082454824447632,
      "epoch": 0.19101342151332426,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998556628804408,
      "loss": 5.8831,
      "mean_token_accuracy": 0.14139181822538377,
      "num_tokens": 4263111.0,
      "step": 2455
    },
    {
      "entropy": 6.04579610824585,
      "epoch": 0.19140245088504182,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998546685907225,
      "loss": 5.8954,
      "mean_token_accuracy": 0.14221663847565652,
      "num_tokens": 4272796.0,
      "step": 2460
    },
    {
      "entropy": 6.009330224990845,
      "epoch": 0.19179148025675938,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998536708891927,
      "loss": 5.8127,
      "mean_token_accuracy": 0.15046731382608414,
      "num_tokens": 4281361.0,
      "step": 2465
    },
    {
      "entropy": 6.070437002182007,
      "epoch": 0.19218050962847694,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998526697758663,
      "loss": 5.8705,
      "mean_token_accuracy": 0.1421242132782936,
      "num_tokens": 4290470.0,
      "step": 2470
    },
    {
      "entropy": 6.059012508392334,
      "epoch": 0.19256953900019452,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998516652507585,
      "loss": 5.9169,
      "mean_token_accuracy": 0.13801453858613968,
      "num_tokens": 4299087.0,
      "step": 2475
    },
    {
      "entropy": 6.017315196990967,
      "epoch": 0.19295856837191208,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998506573138846,
      "loss": 5.7824,
      "mean_token_accuracy": 0.15042203813791274,
      "num_tokens": 4307572.0,
      "step": 2480
    },
    {
      "entropy": 6.000125503540039,
      "epoch": 0.19334759774362964,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998496459652598,
      "loss": 5.8715,
      "mean_token_accuracy": 0.14248882085084916,
      "num_tokens": 4316337.0,
      "step": 2485
    },
    {
      "entropy": 6.117257308959961,
      "epoch": 0.1937366271153472,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998486312048995,
      "loss": 5.9536,
      "mean_token_accuracy": 0.14570799991488456,
      "num_tokens": 4326258.0,
      "step": 2490
    },
    {
      "entropy": 6.035017681121826,
      "epoch": 0.19412565648706478,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998476130328193,
      "loss": 5.7979,
      "mean_token_accuracy": 0.14963613897562028,
      "num_tokens": 4334711.0,
      "step": 2495
    },
    {
      "entropy": 6.051698970794678,
      "epoch": 0.19451468585878234,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004998465914490343,
      "loss": 5.8939,
      "mean_token_accuracy": 0.14221276044845582,
      "num_tokens": 4343739.0,
      "step": 2500
    },
    {
      "entropy": 6.042819595336914,
      "epoch": 0.1949037152304999,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00049984556645356,
      "loss": 5.8445,
      "mean_token_accuracy": 0.1475151002407074,
      "num_tokens": 4352905.0,
      "step": 2505
    },
    {
      "entropy": 6.109594774246216,
      "epoch": 0.19529274460221746,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998445380464124,
      "loss": 5.8667,
      "mean_token_accuracy": 0.14822212606668472,
      "num_tokens": 4361333.0,
      "step": 2510
    },
    {
      "entropy": 5.941962146759034,
      "epoch": 0.19568177397393502,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998435062276066,
      "loss": 5.8167,
      "mean_token_accuracy": 0.14410733431577682,
      "num_tokens": 4369872.0,
      "step": 2515
    },
    {
      "entropy": 6.0431900978088375,
      "epoch": 0.1960708033456526,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998424709971586,
      "loss": 5.894,
      "mean_token_accuracy": 0.14601442515850066,
      "num_tokens": 4378097.0,
      "step": 2520
    },
    {
      "entropy": 6.025267696380615,
      "epoch": 0.19645983271737016,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998414323550839,
      "loss": 5.8153,
      "mean_token_accuracy": 0.14793820455670356,
      "num_tokens": 4387414.0,
      "step": 2525
    },
    {
      "entropy": 6.09251389503479,
      "epoch": 0.19684886208908772,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004998403903013984,
      "loss": 5.9085,
      "mean_token_accuracy": 0.14772322848439218,
      "num_tokens": 4395922.0,
      "step": 2530
    },
    {
      "entropy": 5.9777669429779055,
      "epoch": 0.19723789146080528,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998393448361179,
      "loss": 5.8766,
      "mean_token_accuracy": 0.14107875004410744,
      "num_tokens": 4404500.0,
      "step": 2535
    },
    {
      "entropy": 5.933836650848389,
      "epoch": 0.19762692083252287,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499838295959258,
      "loss": 5.8752,
      "mean_token_accuracy": 0.14383502304553986,
      "num_tokens": 4413433.0,
      "step": 2540
    },
    {
      "entropy": 6.087890243530273,
      "epoch": 0.19801595020424043,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499837243670835,
      "loss": 5.8942,
      "mean_token_accuracy": 0.14684352204203605,
      "num_tokens": 4421789.0,
      "step": 2545
    },
    {
      "entropy": 5.997798299789428,
      "epoch": 0.19840497957595798,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998361879708646,
      "loss": 5.7573,
      "mean_token_accuracy": 0.15463868230581285,
      "num_tokens": 4430256.0,
      "step": 2550
    },
    {
      "entropy": 5.990434455871582,
      "epoch": 0.19879400894767554,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499835128859363,
      "loss": 5.8106,
      "mean_token_accuracy": 0.14735610634088517,
      "num_tokens": 4438431.0,
      "step": 2555
    },
    {
      "entropy": 6.06252908706665,
      "epoch": 0.1991830383193931,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998340663363461,
      "loss": 5.8695,
      "mean_token_accuracy": 0.1388321965932846,
      "num_tokens": 4447150.0,
      "step": 2560
    },
    {
      "entropy": 5.991088819503784,
      "epoch": 0.1995720676911107,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998330004018301,
      "loss": 5.7606,
      "mean_token_accuracy": 0.14833912998437881,
      "num_tokens": 4455181.0,
      "step": 2565
    },
    {
      "entropy": 5.980442047119141,
      "epoch": 0.19996109706282825,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998319310558312,
      "loss": 5.7811,
      "mean_token_accuracy": 0.14660090953111649,
      "num_tokens": 4463361.0,
      "step": 2570
    },
    {
      "entropy": 6.045340967178345,
      "epoch": 0.2003501264345458,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998308582983657,
      "loss": 5.8407,
      "mean_token_accuracy": 0.1461796782910824,
      "num_tokens": 4471853.0,
      "step": 2575
    },
    {
      "entropy": 5.949068546295166,
      "epoch": 0.20073915580626336,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998297821294497,
      "loss": 5.774,
      "mean_token_accuracy": 0.1478642612695694,
      "num_tokens": 4480812.0,
      "step": 2580
    },
    {
      "entropy": 6.020922899246216,
      "epoch": 0.20112818517798095,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998287025490995,
      "loss": 5.8249,
      "mean_token_accuracy": 0.14373451471328735,
      "num_tokens": 4489180.0,
      "step": 2585
    },
    {
      "entropy": 5.996046781539917,
      "epoch": 0.2015172145496985,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998276195573317,
      "loss": 5.8982,
      "mean_token_accuracy": 0.1447764255106449,
      "num_tokens": 4498013.0,
      "step": 2590
    },
    {
      "entropy": 6.1362189769744875,
      "epoch": 0.20190624392141607,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998265331541627,
      "loss": 5.9291,
      "mean_token_accuracy": 0.1334374077618122,
      "num_tokens": 4506767.0,
      "step": 2595
    },
    {
      "entropy": 6.044833755493164,
      "epoch": 0.20229527329313363,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998254433396088,
      "loss": 5.9371,
      "mean_token_accuracy": 0.14291547536849974,
      "num_tokens": 4516657.0,
      "step": 2600
    },
    {
      "entropy": 5.959131669998169,
      "epoch": 0.20268430266485118,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998243501136867,
      "loss": 5.7745,
      "mean_token_accuracy": 0.15024542286992074,
      "num_tokens": 4524980.0,
      "step": 2605
    },
    {
      "entropy": 5.998394441604614,
      "epoch": 0.20307333203656877,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998232534764129,
      "loss": 5.8389,
      "mean_token_accuracy": 0.14741490185260772,
      "num_tokens": 4533236.0,
      "step": 2610
    },
    {
      "entropy": 5.903755569458008,
      "epoch": 0.20346236140828633,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998221534278041,
      "loss": 5.7399,
      "mean_token_accuracy": 0.14562903121113777,
      "num_tokens": 4541778.0,
      "step": 2615
    },
    {
      "entropy": 6.064036464691162,
      "epoch": 0.2038513907800039,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998210499678769,
      "loss": 5.8538,
      "mean_token_accuracy": 0.1419381931424141,
      "num_tokens": 4551131.0,
      "step": 2620
    },
    {
      "entropy": 6.035385370254517,
      "epoch": 0.20424042015172145,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998199430966481,
      "loss": 5.8297,
      "mean_token_accuracy": 0.14448252841830253,
      "num_tokens": 4559530.0,
      "step": 2625
    },
    {
      "entropy": 5.848511409759522,
      "epoch": 0.20462944952343903,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998188328141346,
      "loss": 5.7181,
      "mean_token_accuracy": 0.1512894958257675,
      "num_tokens": 4567533.0,
      "step": 2630
    },
    {
      "entropy": 6.118367338180542,
      "epoch": 0.2050184788951566,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998177191203531,
      "loss": 5.8612,
      "mean_token_accuracy": 0.14240461885929107,
      "num_tokens": 4576414.0,
      "step": 2635
    },
    {
      "entropy": 5.999768924713135,
      "epoch": 0.20540750826687415,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998166020153204,
      "loss": 5.7924,
      "mean_token_accuracy": 0.1489753618836403,
      "num_tokens": 4585331.0,
      "step": 2640
    },
    {
      "entropy": 6.012185764312744,
      "epoch": 0.2057965376385917,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998154814990538,
      "loss": 5.8307,
      "mean_token_accuracy": 0.14780088663101196,
      "num_tokens": 4593526.0,
      "step": 2645
    },
    {
      "entropy": 5.972072982788086,
      "epoch": 0.20618556701030927,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00049981435757157,
      "loss": 5.8305,
      "mean_token_accuracy": 0.1450671598315239,
      "num_tokens": 4602400.0,
      "step": 2650
    },
    {
      "entropy": 5.94981198310852,
      "epoch": 0.20657459638202685,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998132302328862,
      "loss": 5.7735,
      "mean_token_accuracy": 0.14710604920983314,
      "num_tokens": 4611037.0,
      "step": 2655
    },
    {
      "entropy": 6.004731702804565,
      "epoch": 0.2069636257537444,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998120994830195,
      "loss": 5.7756,
      "mean_token_accuracy": 0.14985849261283873,
      "num_tokens": 4619030.0,
      "step": 2660
    },
    {
      "entropy": 5.977835988998413,
      "epoch": 0.20735265512546197,
      "grad_norm": 0.984375,
      "learning_rate": 0.000499810965321987,
      "loss": 5.7949,
      "mean_token_accuracy": 0.14194909781217574,
      "num_tokens": 4628649.0,
      "step": 2665
    },
    {
      "entropy": 5.940742921829224,
      "epoch": 0.20774168449717953,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998098277498059,
      "loss": 5.6989,
      "mean_token_accuracy": 0.14827831983566284,
      "num_tokens": 4636935.0,
      "step": 2670
    },
    {
      "entropy": 5.9714634895324705,
      "epoch": 0.2081307138688971,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998086867664934,
      "loss": 5.8852,
      "mean_token_accuracy": 0.1452195018529892,
      "num_tokens": 4645532.0,
      "step": 2675
    },
    {
      "entropy": 6.067996883392334,
      "epoch": 0.20851974324061467,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499807542372067,
      "loss": 5.958,
      "mean_token_accuracy": 0.1363249361515045,
      "num_tokens": 4654433.0,
      "step": 2680
    },
    {
      "entropy": 6.098356103897094,
      "epoch": 0.20890877261233223,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998063945665439,
      "loss": 5.9518,
      "mean_token_accuracy": 0.1396227262914181,
      "num_tokens": 4662913.0,
      "step": 2685
    },
    {
      "entropy": 6.0581354141235355,
      "epoch": 0.2092978019840498,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004998052433499416,
      "loss": 5.9174,
      "mean_token_accuracy": 0.13971545100212096,
      "num_tokens": 4672270.0,
      "step": 2690
    },
    {
      "entropy": 6.120678234100342,
      "epoch": 0.20968683135576735,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998040887222776,
      "loss": 5.9615,
      "mean_token_accuracy": 0.14097035080194473,
      "num_tokens": 4681428.0,
      "step": 2695
    },
    {
      "entropy": 6.031071949005127,
      "epoch": 0.21007586072748494,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998029306835693,
      "loss": 5.7864,
      "mean_token_accuracy": 0.14907388389110565,
      "num_tokens": 4690492.0,
      "step": 2700
    },
    {
      "entropy": 6.025369882583618,
      "epoch": 0.2104648900992025,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998017692338344,
      "loss": 5.9018,
      "mean_token_accuracy": 0.14205559343099594,
      "num_tokens": 4699667.0,
      "step": 2705
    },
    {
      "entropy": 5.896922922134399,
      "epoch": 0.21085391947092005,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998006043730905,
      "loss": 5.5889,
      "mean_token_accuracy": 0.156715589761734,
      "num_tokens": 4708529.0,
      "step": 2710
    },
    {
      "entropy": 5.875597381591797,
      "epoch": 0.2112429488426376,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997994361013551,
      "loss": 5.8478,
      "mean_token_accuracy": 0.1417768731713295,
      "num_tokens": 4717684.0,
      "step": 2715
    },
    {
      "entropy": 5.999152946472168,
      "epoch": 0.21163197821435517,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004997982644186461,
      "loss": 5.7645,
      "mean_token_accuracy": 0.15027033016085625,
      "num_tokens": 4727067.0,
      "step": 2720
    },
    {
      "entropy": 5.941830015182495,
      "epoch": 0.21202100758607276,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997970893249813,
      "loss": 5.7409,
      "mean_token_accuracy": 0.15033891201019287,
      "num_tokens": 4735434.0,
      "step": 2725
    },
    {
      "entropy": 5.998355436325073,
      "epoch": 0.21241003695779032,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997959108203785,
      "loss": 5.7831,
      "mean_token_accuracy": 0.15076861307024955,
      "num_tokens": 4743843.0,
      "step": 2730
    },
    {
      "entropy": 6.021811771392822,
      "epoch": 0.21279906632950787,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004997947289048554,
      "loss": 5.8298,
      "mean_token_accuracy": 0.14842365086078643,
      "num_tokens": 4753228.0,
      "step": 2735
    },
    {
      "entropy": 6.014471864700317,
      "epoch": 0.21318809570122543,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997935435784302,
      "loss": 5.9131,
      "mean_token_accuracy": 0.14402995258569717,
      "num_tokens": 4763401.0,
      "step": 2740
    },
    {
      "entropy": 5.965554714202881,
      "epoch": 0.21357712507294302,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997923548411208,
      "loss": 5.7771,
      "mean_token_accuracy": 0.1451819881796837,
      "num_tokens": 4771812.0,
      "step": 2745
    },
    {
      "entropy": 5.930664920806885,
      "epoch": 0.21396615444466058,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499791162692945,
      "loss": 5.799,
      "mean_token_accuracy": 0.14840952157974244,
      "num_tokens": 4780616.0,
      "step": 2750
    },
    {
      "entropy": 5.959417247772217,
      "epoch": 0.21435518381637814,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997899671339214,
      "loss": 5.7985,
      "mean_token_accuracy": 0.14958962947130203,
      "num_tokens": 4789886.0,
      "step": 2755
    },
    {
      "entropy": 5.933303451538086,
      "epoch": 0.2147442131880957,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997887681640675,
      "loss": 5.7753,
      "mean_token_accuracy": 0.14589939936995505,
      "num_tokens": 4799189.0,
      "step": 2760
    },
    {
      "entropy": 6.007724142074585,
      "epoch": 0.21513324255981325,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997875657834021,
      "loss": 5.821,
      "mean_token_accuracy": 0.15184845104813577,
      "num_tokens": 4807850.0,
      "step": 2765
    },
    {
      "entropy": 6.024917030334473,
      "epoch": 0.21552227193153084,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997863599919432,
      "loss": 5.7387,
      "mean_token_accuracy": 0.15238860324025155,
      "num_tokens": 4815742.0,
      "step": 2770
    },
    {
      "entropy": 5.997466993331909,
      "epoch": 0.2159113013032484,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997851507897089,
      "loss": 5.8285,
      "mean_token_accuracy": 0.14630141407251357,
      "num_tokens": 4824761.0,
      "step": 2775
    },
    {
      "entropy": 6.0686828136444095,
      "epoch": 0.21630033067496596,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997839381767178,
      "loss": 5.9577,
      "mean_token_accuracy": 0.1376100316643715,
      "num_tokens": 4832805.0,
      "step": 2780
    },
    {
      "entropy": 5.969825267791748,
      "epoch": 0.21668936004668352,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997827221529882,
      "loss": 5.8107,
      "mean_token_accuracy": 0.14742545261979104,
      "num_tokens": 4841938.0,
      "step": 2785
    },
    {
      "entropy": 5.980387020111084,
      "epoch": 0.2170783894184011,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997815027185386,
      "loss": 5.8302,
      "mean_token_accuracy": 0.1437895864248276,
      "num_tokens": 4849932.0,
      "step": 2790
    },
    {
      "entropy": 5.921569156646728,
      "epoch": 0.21746741879011866,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997802798733874,
      "loss": 5.6874,
      "mean_token_accuracy": 0.15188782587647437,
      "num_tokens": 4859054.0,
      "step": 2795
    },
    {
      "entropy": 6.0102729320526125,
      "epoch": 0.21785644816183622,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997790536175534,
      "loss": 5.8669,
      "mean_token_accuracy": 0.14981231167912484,
      "num_tokens": 4867415.0,
      "step": 2800
    },
    {
      "entropy": 6.026336431503296,
      "epoch": 0.21824547753355378,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997778239510548,
      "loss": 5.8956,
      "mean_token_accuracy": 0.1462355837225914,
      "num_tokens": 4875965.0,
      "step": 2805
    },
    {
      "entropy": 5.998291397094727,
      "epoch": 0.21863450690527134,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997765908739106,
      "loss": 5.7996,
      "mean_token_accuracy": 0.14824215620756148,
      "num_tokens": 4884629.0,
      "step": 2810
    },
    {
      "entropy": 5.994702625274658,
      "epoch": 0.21902353627698892,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997753543861395,
      "loss": 5.8435,
      "mean_token_accuracy": 0.1457155279815197,
      "num_tokens": 4893573.0,
      "step": 2815
    },
    {
      "entropy": 5.932880544662476,
      "epoch": 0.21941256564870648,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00049977411448776,
      "loss": 5.7718,
      "mean_token_accuracy": 0.1455760970711708,
      "num_tokens": 4902831.0,
      "step": 2820
    },
    {
      "entropy": 6.032305574417114,
      "epoch": 0.21980159502042404,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997728711787912,
      "loss": 5.8263,
      "mean_token_accuracy": 0.14247847720980644,
      "num_tokens": 4911360.0,
      "step": 2825
    },
    {
      "entropy": 5.986047887802124,
      "epoch": 0.2201906243921416,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997716244592518,
      "loss": 5.7947,
      "mean_token_accuracy": 0.142076013982296,
      "num_tokens": 4919758.0,
      "step": 2830
    },
    {
      "entropy": 5.846888208389283,
      "epoch": 0.22057965376385918,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997703743291607,
      "loss": 5.6693,
      "mean_token_accuracy": 0.15803906843066215,
      "num_tokens": 4928934.0,
      "step": 2835
    },
    {
      "entropy": 5.908144187927246,
      "epoch": 0.22096868313557674,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499769120788537,
      "loss": 5.7258,
      "mean_token_accuracy": 0.1486142821609974,
      "num_tokens": 4937780.0,
      "step": 2840
    },
    {
      "entropy": 6.082314538955688,
      "epoch": 0.2213577125072943,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997678638373995,
      "loss": 5.8261,
      "mean_token_accuracy": 0.1448799580335617,
      "num_tokens": 4946862.0,
      "step": 2845
    },
    {
      "entropy": 5.94984769821167,
      "epoch": 0.22174674187901186,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997666034757676,
      "loss": 5.8049,
      "mean_token_accuracy": 0.1516811653971672,
      "num_tokens": 4954883.0,
      "step": 2850
    },
    {
      "entropy": 5.85023832321167,
      "epoch": 0.22213577125072942,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997653397036603,
      "loss": 5.6035,
      "mean_token_accuracy": 0.15844471007585526,
      "num_tokens": 4962840.0,
      "step": 2855
    },
    {
      "entropy": 5.935201215744018,
      "epoch": 0.222524800622447,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997640725210965,
      "loss": 5.8358,
      "mean_token_accuracy": 0.14835945665836334,
      "num_tokens": 4972334.0,
      "step": 2860
    },
    {
      "entropy": 5.970633888244629,
      "epoch": 0.22291382999416456,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997628019280958,
      "loss": 5.8008,
      "mean_token_accuracy": 0.1477268561720848,
      "num_tokens": 4981032.0,
      "step": 2865
    },
    {
      "entropy": 5.974843454360962,
      "epoch": 0.22330285936588212,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997615279246773,
      "loss": 5.8054,
      "mean_token_accuracy": 0.14543402120471,
      "num_tokens": 4990502.0,
      "step": 2870
    },
    {
      "entropy": 6.052269506454468,
      "epoch": 0.22369188873759968,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997602505108603,
      "loss": 5.9916,
      "mean_token_accuracy": 0.13957195803523065,
      "num_tokens": 4999121.0,
      "step": 2875
    },
    {
      "entropy": 5.986846828460694,
      "epoch": 0.22408091810931724,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997589696866644,
      "loss": 5.7363,
      "mean_token_accuracy": 0.15012581199407576,
      "num_tokens": 5007271.0,
      "step": 2880
    },
    {
      "entropy": 5.880199956893921,
      "epoch": 0.22446994748103483,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997576854521088,
      "loss": 5.7994,
      "mean_token_accuracy": 0.143898157030344,
      "num_tokens": 5015940.0,
      "step": 2885
    },
    {
      "entropy": 5.899425983428955,
      "epoch": 0.22485897685275238,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499756397807213,
      "loss": 5.7234,
      "mean_token_accuracy": 0.1511826515197754,
      "num_tokens": 5024294.0,
      "step": 2890
    },
    {
      "entropy": 5.997026634216309,
      "epoch": 0.22524800622446994,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997551067519966,
      "loss": 5.6722,
      "mean_token_accuracy": 0.1542857527732849,
      "num_tokens": 5033401.0,
      "step": 2895
    },
    {
      "entropy": 5.861394643783569,
      "epoch": 0.2256370355961875,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997538122864792,
      "loss": 5.7638,
      "mean_token_accuracy": 0.14565517976880074,
      "num_tokens": 5041955.0,
      "step": 2900
    },
    {
      "entropy": 5.918551397323609,
      "epoch": 0.2260260649679051,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997525144106804,
      "loss": 5.8115,
      "mean_token_accuracy": 0.14099842831492423,
      "num_tokens": 5051132.0,
      "step": 2905
    },
    {
      "entropy": 5.935513830184936,
      "epoch": 0.22641509433962265,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00049975121312462,
      "loss": 5.7359,
      "mean_token_accuracy": 0.15899298340082169,
      "num_tokens": 5059492.0,
      "step": 2910
    },
    {
      "entropy": 5.87189130783081,
      "epoch": 0.2268041237113402,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997499084283177,
      "loss": 5.7837,
      "mean_token_accuracy": 0.1493110865354538,
      "num_tokens": 5069015.0,
      "step": 2915
    },
    {
      "entropy": 5.9538294792175295,
      "epoch": 0.22719315308305776,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997486003217932,
      "loss": 5.7948,
      "mean_token_accuracy": 0.14656500220298768,
      "num_tokens": 5077614.0,
      "step": 2920
    },
    {
      "entropy": 5.937190008163452,
      "epoch": 0.22758218245477532,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997472888050664,
      "loss": 5.724,
      "mean_token_accuracy": 0.15037651658058165,
      "num_tokens": 5086300.0,
      "step": 2925
    },
    {
      "entropy": 5.855981540679932,
      "epoch": 0.2279712118264929,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997459738781573,
      "loss": 5.7738,
      "mean_token_accuracy": 0.1492456465959549,
      "num_tokens": 5095967.0,
      "step": 2930
    },
    {
      "entropy": 5.961302804946899,
      "epoch": 0.22836024119821047,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997446555410857,
      "loss": 5.7319,
      "mean_token_accuracy": 0.148114612698555,
      "num_tokens": 5104611.0,
      "step": 2935
    },
    {
      "entropy": 5.902946662902832,
      "epoch": 0.22874927056992803,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004997433337938716,
      "loss": 5.6867,
      "mean_token_accuracy": 0.15346912294626236,
      "num_tokens": 5113452.0,
      "step": 2940
    },
    {
      "entropy": 5.822092819213867,
      "epoch": 0.22913829994164558,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997420086365351,
      "loss": 5.6718,
      "mean_token_accuracy": 0.15292285978794098,
      "num_tokens": 5121558.0,
      "step": 2945
    },
    {
      "entropy": 5.884310340881347,
      "epoch": 0.22952732931336317,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997406800690966,
      "loss": 5.6482,
      "mean_token_accuracy": 0.16208126842975618,
      "num_tokens": 5129942.0,
      "step": 2950
    },
    {
      "entropy": 5.936553859710694,
      "epoch": 0.22991635868508073,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997393480915758,
      "loss": 5.7766,
      "mean_token_accuracy": 0.1436438135802746,
      "num_tokens": 5139012.0,
      "step": 2955
    },
    {
      "entropy": 5.977424049377442,
      "epoch": 0.2303053880567983,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997380127039931,
      "loss": 5.7831,
      "mean_token_accuracy": 0.14895185455679893,
      "num_tokens": 5148185.0,
      "step": 2960
    },
    {
      "entropy": 5.966456031799316,
      "epoch": 0.23069441742851585,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997366739063688,
      "loss": 5.808,
      "mean_token_accuracy": 0.14403805360198021,
      "num_tokens": 5157086.0,
      "step": 2965
    },
    {
      "entropy": 6.039245176315307,
      "epoch": 0.2310834468002334,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997353316987231,
      "loss": 5.871,
      "mean_token_accuracy": 0.1416305460035801,
      "num_tokens": 5165842.0,
      "step": 2970
    },
    {
      "entropy": 5.9533428192138675,
      "epoch": 0.231472476171951,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997339860810765,
      "loss": 5.8666,
      "mean_token_accuracy": 0.14629089161753656,
      "num_tokens": 5174161.0,
      "step": 2975
    },
    {
      "entropy": 5.932276821136474,
      "epoch": 0.23186150554366855,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997326370534495,
      "loss": 5.8232,
      "mean_token_accuracy": 0.144809553027153,
      "num_tokens": 5183830.0,
      "step": 2980
    },
    {
      "entropy": 5.911020135879516,
      "epoch": 0.2322505349153861,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997312846158622,
      "loss": 5.7319,
      "mean_token_accuracy": 0.14793932288885117,
      "num_tokens": 5192248.0,
      "step": 2985
    },
    {
      "entropy": 5.829217910766602,
      "epoch": 0.23263956428710367,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997299287683355,
      "loss": 5.7053,
      "mean_token_accuracy": 0.15378875881433487,
      "num_tokens": 5201212.0,
      "step": 2990
    },
    {
      "entropy": 5.876448011398315,
      "epoch": 0.23302859365882125,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997285695108898,
      "loss": 5.7687,
      "mean_token_accuracy": 0.1468915194272995,
      "num_tokens": 5209338.0,
      "step": 2995
    },
    {
      "entropy": 6.041135358810425,
      "epoch": 0.2334176230305388,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997272068435458,
      "loss": 5.7828,
      "mean_token_accuracy": 0.147082382440567,
      "num_tokens": 5217590.0,
      "step": 3000
    },
    {
      "epoch": 0.2334176230305388,
      "eval_entropy": 5.745479221086988,
      "eval_loss": 5.805346488952637,
      "eval_mean_token_accuracy": 0.15537624003812026,
      "eval_num_tokens": 5217590.0,
      "eval_runtime": 21.6146,
      "eval_samples_per_second": 1922.684,
      "eval_steps_per_second": 240.347,
      "step": 3000
    },
    {
      "entropy": 5.862835884094238,
      "epoch": 0.23380665240225637,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000499725840766324,
      "loss": 5.6977,
      "mean_token_accuracy": 0.15283167362213135,
      "num_tokens": 5226383.0,
      "step": 3005
    },
    {
      "entropy": 5.940398120880127,
      "epoch": 0.23419568177397393,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004997244712792453,
      "loss": 5.7304,
      "mean_token_accuracy": 0.15186440497636794,
      "num_tokens": 5235517.0,
      "step": 3010
    },
    {
      "entropy": 5.985174703598022,
      "epoch": 0.2345847111456915,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997230983823305,
      "loss": 5.7156,
      "mean_token_accuracy": 0.15750928670167924,
      "num_tokens": 5243812.0,
      "step": 3015
    },
    {
      "entropy": 5.887314510345459,
      "epoch": 0.23497374051740907,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004997217220756003,
      "loss": 5.7189,
      "mean_token_accuracy": 0.15049459934234619,
      "num_tokens": 5252399.0,
      "step": 3020
    },
    {
      "entropy": 5.908802223205567,
      "epoch": 0.23536276988912663,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997203423590757,
      "loss": 5.7499,
      "mean_token_accuracy": 0.1497652493417263,
      "num_tokens": 5260847.0,
      "step": 3025
    },
    {
      "entropy": 5.948167753219605,
      "epoch": 0.2357517992608442,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997189592327775,
      "loss": 5.8329,
      "mean_token_accuracy": 0.1460947275161743,
      "num_tokens": 5268847.0,
      "step": 3030
    },
    {
      "entropy": 5.982191753387451,
      "epoch": 0.23614082863256175,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997175726967268,
      "loss": 5.7558,
      "mean_token_accuracy": 0.14891543239355087,
      "num_tokens": 5277989.0,
      "step": 3035
    },
    {
      "entropy": 5.815080404281616,
      "epoch": 0.23652985800427934,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997161827509447,
      "loss": 5.7623,
      "mean_token_accuracy": 0.15036816596984864,
      "num_tokens": 5286272.0,
      "step": 3040
    },
    {
      "entropy": 5.897141027450561,
      "epoch": 0.2369188873759969,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997147893954521,
      "loss": 5.7553,
      "mean_token_accuracy": 0.15387040227651597,
      "num_tokens": 5295281.0,
      "step": 3045
    },
    {
      "entropy": 5.974369812011719,
      "epoch": 0.23730791674771445,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997133926302702,
      "loss": 5.7105,
      "mean_token_accuracy": 0.14991148561239243,
      "num_tokens": 5303905.0,
      "step": 3050
    },
    {
      "entropy": 5.873412418365478,
      "epoch": 0.237696946119432,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997119924554204,
      "loss": 5.687,
      "mean_token_accuracy": 0.1482005760073662,
      "num_tokens": 5313146.0,
      "step": 3055
    },
    {
      "entropy": 5.887405920028686,
      "epoch": 0.23808597549114957,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997105888709236,
      "loss": 5.7235,
      "mean_token_accuracy": 0.15186246931552888,
      "num_tokens": 5320838.0,
      "step": 3060
    },
    {
      "entropy": 5.908910322189331,
      "epoch": 0.23847500486286716,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997091818768015,
      "loss": 5.7509,
      "mean_token_accuracy": 0.1518412262201309,
      "num_tokens": 5329312.0,
      "step": 3065
    },
    {
      "entropy": 5.869619989395142,
      "epoch": 0.23886403423458472,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499707771473075,
      "loss": 5.7412,
      "mean_token_accuracy": 0.15202742367982863,
      "num_tokens": 5337747.0,
      "step": 3070
    },
    {
      "entropy": 5.9338490009307865,
      "epoch": 0.23925306360630227,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997063576597659,
      "loss": 5.7333,
      "mean_token_accuracy": 0.15094367563724517,
      "num_tokens": 5346928.0,
      "step": 3075
    },
    {
      "entropy": 5.900696659088135,
      "epoch": 0.23964209297801983,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997049404368954,
      "loss": 5.7581,
      "mean_token_accuracy": 0.1485590822994709,
      "num_tokens": 5355827.0,
      "step": 3080
    },
    {
      "entropy": 5.995998096466065,
      "epoch": 0.2400311223497374,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499703519804485,
      "loss": 5.8744,
      "mean_token_accuracy": 0.14193669930100442,
      "num_tokens": 5364309.0,
      "step": 3085
    },
    {
      "entropy": 5.94293532371521,
      "epoch": 0.24042015172145498,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997020957625564,
      "loss": 5.7683,
      "mean_token_accuracy": 0.15073340982198716,
      "num_tokens": 5372979.0,
      "step": 3090
    },
    {
      "entropy": 5.888595533370972,
      "epoch": 0.24080918109317254,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004997006683111312,
      "loss": 5.7505,
      "mean_token_accuracy": 0.14528402984142302,
      "num_tokens": 5381809.0,
      "step": 3095
    },
    {
      "entropy": 5.973602533340454,
      "epoch": 0.2411982104648901,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499699237450231,
      "loss": 5.7928,
      "mean_token_accuracy": 0.1505873367190361,
      "num_tokens": 5390036.0,
      "step": 3100
    },
    {
      "entropy": 6.0370574474334715,
      "epoch": 0.24158723983660765,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996978031798774,
      "loss": 5.8516,
      "mean_token_accuracy": 0.14484131932258607,
      "num_tokens": 5398739.0,
      "step": 3105
    },
    {
      "entropy": 5.894684076309204,
      "epoch": 0.24197626920832524,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004996963655000924,
      "loss": 5.7647,
      "mean_token_accuracy": 0.149184250831604,
      "num_tokens": 5407441.0,
      "step": 3110
    },
    {
      "entropy": 5.943358516693115,
      "epoch": 0.2423652985800428,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996949244108977,
      "loss": 5.7116,
      "mean_token_accuracy": 0.14837338477373124,
      "num_tokens": 5416534.0,
      "step": 3115
    },
    {
      "entropy": 5.914983034133911,
      "epoch": 0.24275432795176036,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499693479912315,
      "loss": 5.7721,
      "mean_token_accuracy": 0.1526084750890732,
      "num_tokens": 5425170.0,
      "step": 3120
    },
    {
      "entropy": 5.8383172988891605,
      "epoch": 0.24314335732347792,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996920320043666,
      "loss": 5.7394,
      "mean_token_accuracy": 0.1426021859049797,
      "num_tokens": 5433721.0,
      "step": 3125
    },
    {
      "entropy": 5.985742807388306,
      "epoch": 0.24353238669519547,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996905806870742,
      "loss": 5.7984,
      "mean_token_accuracy": 0.14674716889858247,
      "num_tokens": 5441708.0,
      "step": 3130
    },
    {
      "entropy": 5.8728574275970455,
      "epoch": 0.24392141606691306,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996891259604598,
      "loss": 5.6953,
      "mean_token_accuracy": 0.15019447654485701,
      "num_tokens": 5449942.0,
      "step": 3135
    },
    {
      "entropy": 5.941146516799927,
      "epoch": 0.24431044543863062,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996876678245455,
      "loss": 5.7229,
      "mean_token_accuracy": 0.15215176492929458,
      "num_tokens": 5458004.0,
      "step": 3140
    },
    {
      "entropy": 5.9605811595916744,
      "epoch": 0.24469947481034818,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996862062793536,
      "loss": 5.8024,
      "mean_token_accuracy": 0.14993125721812248,
      "num_tokens": 5467071.0,
      "step": 3145
    },
    {
      "entropy": 5.911229848861694,
      "epoch": 0.24508850418206574,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004996847413249061,
      "loss": 5.7739,
      "mean_token_accuracy": 0.15182917267084123,
      "num_tokens": 5476221.0,
      "step": 3150
    },
    {
      "entropy": 5.919036197662353,
      "epoch": 0.24547753355378332,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996832729612252,
      "loss": 5.7022,
      "mean_token_accuracy": 0.15160031914710997,
      "num_tokens": 5484549.0,
      "step": 3155
    },
    {
      "entropy": 5.81336932182312,
      "epoch": 0.24586656292550088,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996818011883333,
      "loss": 5.7017,
      "mean_token_accuracy": 0.1576054573059082,
      "num_tokens": 5492909.0,
      "step": 3160
    },
    {
      "entropy": 5.887171792984009,
      "epoch": 0.24625559229721844,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996803260062527,
      "loss": 5.7394,
      "mean_token_accuracy": 0.15224794745445253,
      "num_tokens": 5501418.0,
      "step": 3165
    },
    {
      "entropy": 5.92157883644104,
      "epoch": 0.246644621668936,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996788474150057,
      "loss": 5.7461,
      "mean_token_accuracy": 0.14818776100873948,
      "num_tokens": 5510609.0,
      "step": 3170
    },
    {
      "entropy": 5.827919673919678,
      "epoch": 0.24703365104065356,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004996773654146149,
      "loss": 5.5989,
      "mean_token_accuracy": 0.1590520143508911,
      "num_tokens": 5518349.0,
      "step": 3175
    },
    {
      "entropy": 5.784766435623169,
      "epoch": 0.24742268041237114,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996758800051026,
      "loss": 5.5514,
      "mean_token_accuracy": 0.16138913482427597,
      "num_tokens": 5526972.0,
      "step": 3180
    },
    {
      "entropy": 5.846664333343506,
      "epoch": 0.2478117097840887,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996743911864916,
      "loss": 5.7182,
      "mean_token_accuracy": 0.14989619702100754,
      "num_tokens": 5535683.0,
      "step": 3185
    },
    {
      "entropy": 5.819127082824707,
      "epoch": 0.24820073915580626,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996728989588041,
      "loss": 5.6818,
      "mean_token_accuracy": 0.1583881087601185,
      "num_tokens": 5544258.0,
      "step": 3190
    },
    {
      "entropy": 5.786135149002075,
      "epoch": 0.24858976852752382,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996714033220632,
      "loss": 5.6991,
      "mean_token_accuracy": 0.14900739789009093,
      "num_tokens": 5552608.0,
      "step": 3195
    },
    {
      "entropy": 5.917917776107788,
      "epoch": 0.2489787978992414,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004996699042762911,
      "loss": 5.8147,
      "mean_token_accuracy": 0.1450411006808281,
      "num_tokens": 5561357.0,
      "step": 3200
    },
    {
      "entropy": 5.957511854171753,
      "epoch": 0.24936782727095896,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996684018215111,
      "loss": 5.6472,
      "mean_token_accuracy": 0.16130171865224838,
      "num_tokens": 5569862.0,
      "step": 3205
    },
    {
      "entropy": 5.823313760757446,
      "epoch": 0.24975685664267652,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996668959577453,
      "loss": 5.7183,
      "mean_token_accuracy": 0.15448436588048936,
      "num_tokens": 5578485.0,
      "step": 3210
    },
    {
      "entropy": 5.858618497848511,
      "epoch": 0.2501458860143941,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004996653866850173,
      "loss": 5.6987,
      "mean_token_accuracy": 0.1483453944325447,
      "num_tokens": 5587713.0,
      "step": 3215
    },
    {
      "entropy": 5.905856943130493,
      "epoch": 0.25053491538611167,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996638740033495,
      "loss": 5.7683,
      "mean_token_accuracy": 0.15061864107847214,
      "num_tokens": 5596541.0,
      "step": 3220
    },
    {
      "entropy": 5.824166011810303,
      "epoch": 0.2509239447578292,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004996623579127651,
      "loss": 5.6473,
      "mean_token_accuracy": 0.15734103843569755,
      "num_tokens": 5605573.0,
      "step": 3225
    },
    {
      "entropy": 5.868488788604736,
      "epoch": 0.2513129741295468,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996608384132868,
      "loss": 5.6594,
      "mean_token_accuracy": 0.15387704223394394,
      "num_tokens": 5614016.0,
      "step": 3230
    },
    {
      "entropy": 5.972904205322266,
      "epoch": 0.25170200350126437,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499659315504938,
      "loss": 5.8798,
      "mean_token_accuracy": 0.1461018592119217,
      "num_tokens": 5622451.0,
      "step": 3235
    },
    {
      "entropy": 5.87816424369812,
      "epoch": 0.2520910328729819,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004996577891877417,
      "loss": 5.719,
      "mean_token_accuracy": 0.15081416219472885,
      "num_tokens": 5631384.0,
      "step": 3240
    },
    {
      "entropy": 5.922952270507812,
      "epoch": 0.2524800622446995,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996562594617209,
      "loss": 5.6895,
      "mean_token_accuracy": 0.15270936638116836,
      "num_tokens": 5640726.0,
      "step": 3245
    },
    {
      "entropy": 5.879611396789551,
      "epoch": 0.252869091616417,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996547263268991,
      "loss": 5.6954,
      "mean_token_accuracy": 0.15027550160884856,
      "num_tokens": 5648948.0,
      "step": 3250
    },
    {
      "entropy": 5.7275628566741945,
      "epoch": 0.2532581209881346,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996531897832993,
      "loss": 5.65,
      "mean_token_accuracy": 0.1583116427063942,
      "num_tokens": 5657951.0,
      "step": 3255
    },
    {
      "entropy": 5.944765377044678,
      "epoch": 0.2536471503598522,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499651649830945,
      "loss": 5.7658,
      "mean_token_accuracy": 0.15232782810926437,
      "num_tokens": 5667476.0,
      "step": 3260
    },
    {
      "entropy": 5.905933570861817,
      "epoch": 0.2540361797315697,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004996501064698593,
      "loss": 5.77,
      "mean_token_accuracy": 0.1492228925228119,
      "num_tokens": 5676627.0,
      "step": 3265
    },
    {
      "entropy": 5.85449366569519,
      "epoch": 0.2544252091032873,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000499648559700066,
      "loss": 5.7119,
      "mean_token_accuracy": 0.15356220155954362,
      "num_tokens": 5684376.0,
      "step": 3270
    },
    {
      "entropy": 5.971212577819824,
      "epoch": 0.25481423847500484,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996470095215884,
      "loss": 5.8714,
      "mean_token_accuracy": 0.1364065058529377,
      "num_tokens": 5693629.0,
      "step": 3275
    },
    {
      "entropy": 5.968133783340454,
      "epoch": 0.2552032678467224,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004996454559344498,
      "loss": 5.7528,
      "mean_token_accuracy": 0.1514081761240959,
      "num_tokens": 5702010.0,
      "step": 3280
    },
    {
      "entropy": 5.786150693893433,
      "epoch": 0.25559229721844,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996438989386741,
      "loss": 5.6462,
      "mean_token_accuracy": 0.15590638667345047,
      "num_tokens": 5710823.0,
      "step": 3285
    },
    {
      "entropy": 5.8355076789855955,
      "epoch": 0.25598132659015754,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996423385342847,
      "loss": 5.7047,
      "mean_token_accuracy": 0.15430673509836196,
      "num_tokens": 5719389.0,
      "step": 3290
    },
    {
      "entropy": 5.8981376647949215,
      "epoch": 0.25637035596187513,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996407747213055,
      "loss": 5.7159,
      "mean_token_accuracy": 0.15026010125875472,
      "num_tokens": 5727327.0,
      "step": 3295
    },
    {
      "entropy": 5.861189365386963,
      "epoch": 0.25675938533359266,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00049963920749976,
      "loss": 5.609,
      "mean_token_accuracy": 0.16035784035921097,
      "num_tokens": 5735886.0,
      "step": 3300
    },
    {
      "entropy": 5.825350904464722,
      "epoch": 0.25714841470531025,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004996376368696721,
      "loss": 5.649,
      "mean_token_accuracy": 0.15595559626817704,
      "num_tokens": 5744656.0,
      "step": 3305
    },
    {
      "entropy": 5.91638069152832,
      "epoch": 0.25753744407702783,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996360628310656,
      "loss": 5.8249,
      "mean_token_accuracy": 0.14670005738735198,
      "num_tokens": 5753026.0,
      "step": 3310
    },
    {
      "entropy": 5.902420473098755,
      "epoch": 0.25792647344874536,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996344853839644,
      "loss": 5.7532,
      "mean_token_accuracy": 0.1508739858865738,
      "num_tokens": 5762771.0,
      "step": 3315
    },
    {
      "entropy": 5.813096714019776,
      "epoch": 0.25831550282046295,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996329045283924,
      "loss": 5.7379,
      "mean_token_accuracy": 0.1582916796207428,
      "num_tokens": 5771151.0,
      "step": 3320
    },
    {
      "entropy": 5.851997327804566,
      "epoch": 0.25870453219218054,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996313202643737,
      "loss": 5.6458,
      "mean_token_accuracy": 0.15273822695016862,
      "num_tokens": 5780648.0,
      "step": 3325
    },
    {
      "entropy": 5.861184453964233,
      "epoch": 0.25909356156389807,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004996297325919321,
      "loss": 5.6307,
      "mean_token_accuracy": 0.15040383338928223,
      "num_tokens": 5789698.0,
      "step": 3330
    },
    {
      "entropy": 5.891029977798462,
      "epoch": 0.25948259093561565,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996281415110919,
      "loss": 5.8184,
      "mean_token_accuracy": 0.14279821440577506,
      "num_tokens": 5799218.0,
      "step": 3335
    },
    {
      "entropy": 5.821232509613037,
      "epoch": 0.2598716203073332,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004996265470218773,
      "loss": 5.6975,
      "mean_token_accuracy": 0.15069891512393951,
      "num_tokens": 5808455.0,
      "step": 3340
    },
    {
      "entropy": 5.908162641525268,
      "epoch": 0.26026064967905077,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004996249491243122,
      "loss": 5.7226,
      "mean_token_accuracy": 0.14941658973693847,
      "num_tokens": 5816464.0,
      "step": 3345
    },
    {
      "entropy": 5.841771650314331,
      "epoch": 0.26064967905076836,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996233478184211,
      "loss": 5.6819,
      "mean_token_accuracy": 0.15169447436928749,
      "num_tokens": 5825188.0,
      "step": 3350
    },
    {
      "entropy": 5.898590850830078,
      "epoch": 0.2610387084224859,
      "grad_norm": 1.125,
      "learning_rate": 0.0004996217431042282,
      "loss": 5.7252,
      "mean_token_accuracy": 0.15503444373607636,
      "num_tokens": 5834150.0,
      "step": 3355
    },
    {
      "entropy": 5.847413158416748,
      "epoch": 0.2614277377942035,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499620134981758,
      "loss": 5.5782,
      "mean_token_accuracy": 0.1629167377948761,
      "num_tokens": 5842801.0,
      "step": 3360
    },
    {
      "entropy": 5.793002605438232,
      "epoch": 0.261816767165921,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996185234510346,
      "loss": 5.6255,
      "mean_token_accuracy": 0.1601943254470825,
      "num_tokens": 5851547.0,
      "step": 3365
    },
    {
      "entropy": 5.932187223434449,
      "epoch": 0.2622057965376386,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996169085120828,
      "loss": 5.7676,
      "mean_token_accuracy": 0.15287947207689284,
      "num_tokens": 5860430.0,
      "step": 3370
    },
    {
      "entropy": 5.81215090751648,
      "epoch": 0.2625948259093562,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004996152901649268,
      "loss": 5.6576,
      "mean_token_accuracy": 0.15576451867818833,
      "num_tokens": 5869648.0,
      "step": 3375
    },
    {
      "entropy": 5.932130193710327,
      "epoch": 0.2629838552810737,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996136684095913,
      "loss": 5.6687,
      "mean_token_accuracy": 0.15645158290863037,
      "num_tokens": 5878606.0,
      "step": 3380
    },
    {
      "entropy": 5.818367958068848,
      "epoch": 0.2633728846527913,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004996120432461009,
      "loss": 5.7468,
      "mean_token_accuracy": 0.15218595266342164,
      "num_tokens": 5886772.0,
      "step": 3385
    },
    {
      "entropy": 5.805695343017578,
      "epoch": 0.2637619140245088,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996104146744804,
      "loss": 5.6477,
      "mean_token_accuracy": 0.1543953076004982,
      "num_tokens": 5895007.0,
      "step": 3390
    },
    {
      "entropy": 5.982627630233765,
      "epoch": 0.2641509433962264,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996087826947541,
      "loss": 5.775,
      "mean_token_accuracy": 0.14896861910820008,
      "num_tokens": 5904249.0,
      "step": 3395
    },
    {
      "entropy": 5.806077241897583,
      "epoch": 0.264539972767944,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004996071473069471,
      "loss": 5.729,
      "mean_token_accuracy": 0.15165796950459481,
      "num_tokens": 5911416.0,
      "step": 3400
    },
    {
      "entropy": 5.891200923919678,
      "epoch": 0.26492900213966153,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004996055085110842,
      "loss": 5.7745,
      "mean_token_accuracy": 0.14715634509921074,
      "num_tokens": 5920370.0,
      "step": 3405
    },
    {
      "entropy": 5.907678842544556,
      "epoch": 0.2653180315113791,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004996038663071902,
      "loss": 5.7279,
      "mean_token_accuracy": 0.1560338944196701,
      "num_tokens": 5928884.0,
      "step": 3410
    },
    {
      "entropy": 5.8364808559417725,
      "epoch": 0.26570706088309665,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996022206952901,
      "loss": 5.7516,
      "mean_token_accuracy": 0.142772651463747,
      "num_tokens": 5938456.0,
      "step": 3415
    },
    {
      "entropy": 5.983996152877808,
      "epoch": 0.26609609025481423,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996005716754086,
      "loss": 5.7893,
      "mean_token_accuracy": 0.15027910396456717,
      "num_tokens": 5947157.0,
      "step": 3420
    },
    {
      "entropy": 5.902491044998169,
      "epoch": 0.2664851196265318,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499598919247571,
      "loss": 5.6287,
      "mean_token_accuracy": 0.16299375891685486,
      "num_tokens": 5956084.0,
      "step": 3425
    },
    {
      "entropy": 5.794814205169677,
      "epoch": 0.26687414899824935,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995972634118023,
      "loss": 5.6765,
      "mean_token_accuracy": 0.15660672038793563,
      "num_tokens": 5964782.0,
      "step": 3430
    },
    {
      "entropy": 5.972515344619751,
      "epoch": 0.26726317836996694,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004995956041681277,
      "loss": 5.8994,
      "mean_token_accuracy": 0.14056409224867822,
      "num_tokens": 5974013.0,
      "step": 3435
    },
    {
      "entropy": 5.953153657913208,
      "epoch": 0.2676522077416845,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004995939415165722,
      "loss": 5.7584,
      "mean_token_accuracy": 0.14940284192562103,
      "num_tokens": 5983351.0,
      "step": 3440
    },
    {
      "entropy": 5.763549518585205,
      "epoch": 0.26804123711340205,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995922754571611,
      "loss": 5.5958,
      "mean_token_accuracy": 0.16312941908836365,
      "num_tokens": 5992084.0,
      "step": 3445
    },
    {
      "entropy": 5.723555326461792,
      "epoch": 0.26843026648511964,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004995906059899197,
      "loss": 5.61,
      "mean_token_accuracy": 0.1542106583714485,
      "num_tokens": 6001043.0,
      "step": 3450
    },
    {
      "entropy": 5.816799783706665,
      "epoch": 0.26881929585683717,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004995889331148733,
      "loss": 5.6236,
      "mean_token_accuracy": 0.15134564638137818,
      "num_tokens": 6009620.0,
      "step": 3455
    },
    {
      "entropy": 5.787903308868408,
      "epoch": 0.26920832522855476,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004995872568320474,
      "loss": 5.6296,
      "mean_token_accuracy": 0.15400015860795974,
      "num_tokens": 6018278.0,
      "step": 3460
    },
    {
      "entropy": 5.747702503204346,
      "epoch": 0.26959735460027234,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995855771414673,
      "loss": 5.6716,
      "mean_token_accuracy": 0.15881235152482986,
      "num_tokens": 6026741.0,
      "step": 3465
    },
    {
      "entropy": 5.963505649566651,
      "epoch": 0.2699863839719899,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995838940431585,
      "loss": 5.6997,
      "mean_token_accuracy": 0.15211729407310487,
      "num_tokens": 6035370.0,
      "step": 3470
    },
    {
      "entropy": 5.8460045337677,
      "epoch": 0.27037541334370746,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995822075371466,
      "loss": 5.6696,
      "mean_token_accuracy": 0.15098004043102264,
      "num_tokens": 6044372.0,
      "step": 3475
    },
    {
      "entropy": 5.772753095626831,
      "epoch": 0.270764442715425,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995805176234571,
      "loss": 5.6944,
      "mean_token_accuracy": 0.1576540380716324,
      "num_tokens": 6052940.0,
      "step": 3480
    },
    {
      "entropy": 5.886135625839233,
      "epoch": 0.2711534720871426,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004995788243021158,
      "loss": 5.6215,
      "mean_token_accuracy": 0.15520101338624953,
      "num_tokens": 6062048.0,
      "step": 3485
    },
    {
      "entropy": 5.798374605178833,
      "epoch": 0.27154250145886016,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004995771275731483,
      "loss": 5.6319,
      "mean_token_accuracy": 0.1564753457903862,
      "num_tokens": 6071478.0,
      "step": 3490
    },
    {
      "entropy": 5.730036306381225,
      "epoch": 0.2719315308305777,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995754274365802,
      "loss": 5.5841,
      "mean_token_accuracy": 0.16150409132242202,
      "num_tokens": 6080495.0,
      "step": 3495
    },
    {
      "entropy": 5.864993762969971,
      "epoch": 0.2723205602022953,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004995737238924376,
      "loss": 5.7258,
      "mean_token_accuracy": 0.14979194104671478,
      "num_tokens": 6090141.0,
      "step": 3500
    },
    {
      "entropy": 5.888671731948852,
      "epoch": 0.2727095895740128,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995720169407461,
      "loss": 5.7003,
      "mean_token_accuracy": 0.1482359305024147,
      "num_tokens": 6099215.0,
      "step": 3505
    },
    {
      "entropy": 5.928614854812622,
      "epoch": 0.2730986189457304,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004995703065815317,
      "loss": 5.781,
      "mean_token_accuracy": 0.1452876403927803,
      "num_tokens": 6108802.0,
      "step": 3510
    },
    {
      "entropy": 5.899406576156617,
      "epoch": 0.273487648317448,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004995685928148203,
      "loss": 5.7651,
      "mean_token_accuracy": 0.1519528865814209,
      "num_tokens": 6117486.0,
      "step": 3515
    },
    {
      "entropy": 5.771115303039551,
      "epoch": 0.2738766776891655,
      "grad_norm": 1.125,
      "learning_rate": 0.000499566875640638,
      "loss": 5.5847,
      "mean_token_accuracy": 0.16556673794984816,
      "num_tokens": 6125248.0,
      "step": 3520
    },
    {
      "entropy": 5.865322780609131,
      "epoch": 0.2742657070608831,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004995651550590108,
      "loss": 5.6499,
      "mean_token_accuracy": 0.1568447843194008,
      "num_tokens": 6134831.0,
      "step": 3525
    },
    {
      "entropy": 5.824641561508178,
      "epoch": 0.2746547364326007,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004995634310699647,
      "loss": 5.7241,
      "mean_token_accuracy": 0.15021293833851815,
      "num_tokens": 6143248.0,
      "step": 3530
    },
    {
      "entropy": 5.92924017906189,
      "epoch": 0.2750437658043182,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004995617036735261,
      "loss": 5.7329,
      "mean_token_accuracy": 0.1525756947696209,
      "num_tokens": 6151768.0,
      "step": 3535
    },
    {
      "entropy": 5.816738557815552,
      "epoch": 0.2754327951760358,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995599728697211,
      "loss": 5.614,
      "mean_token_accuracy": 0.16288287937641144,
      "num_tokens": 6160091.0,
      "step": 3540
    },
    {
      "entropy": 5.843801975250244,
      "epoch": 0.27582182454775334,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995582386585759,
      "loss": 5.7373,
      "mean_token_accuracy": 0.1442616179585457,
      "num_tokens": 6169440.0,
      "step": 3545
    },
    {
      "entropy": 5.833621549606323,
      "epoch": 0.2762108539194709,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995565010401167,
      "loss": 5.6777,
      "mean_token_accuracy": 0.15748906433582305,
      "num_tokens": 6177604.0,
      "step": 3550
    },
    {
      "entropy": 5.864107942581176,
      "epoch": 0.2765998832911885,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995547600143702,
      "loss": 5.7003,
      "mean_token_accuracy": 0.14799479395151138,
      "num_tokens": 6186252.0,
      "step": 3555
    },
    {
      "entropy": 5.782134962081909,
      "epoch": 0.27698891266290604,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995530155813625,
      "loss": 5.647,
      "mean_token_accuracy": 0.15824104249477386,
      "num_tokens": 6194641.0,
      "step": 3560
    },
    {
      "entropy": 5.902602052688598,
      "epoch": 0.2773779420346236,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004995512677411203,
      "loss": 5.7762,
      "mean_token_accuracy": 0.14909671768546104,
      "num_tokens": 6203925.0,
      "step": 3565
    },
    {
      "entropy": 5.928490161895752,
      "epoch": 0.27776697140634116,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004995495164936698,
      "loss": 5.6625,
      "mean_token_accuracy": 0.15092431008815765,
      "num_tokens": 6213001.0,
      "step": 3570
    },
    {
      "entropy": 5.808505439758301,
      "epoch": 0.27815600077805874,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004995477618390381,
      "loss": 5.6875,
      "mean_token_accuracy": 0.14438140839338304,
      "num_tokens": 6221754.0,
      "step": 3575
    },
    {
      "entropy": 5.779090499877929,
      "epoch": 0.27854503014977633,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004995460037772513,
      "loss": 5.5788,
      "mean_token_accuracy": 0.1568375810980797,
      "num_tokens": 6229792.0,
      "step": 3580
    },
    {
      "entropy": 5.779662370681763,
      "epoch": 0.27893405952149386,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004995442423083365,
      "loss": 5.6599,
      "mean_token_accuracy": 0.15357850790023803,
      "num_tokens": 6238839.0,
      "step": 3585
    },
    {
      "entropy": 5.85211501121521,
      "epoch": 0.27932308889321145,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995424774323201,
      "loss": 5.5682,
      "mean_token_accuracy": 0.16091491729021073,
      "num_tokens": 6246939.0,
      "step": 3590
    },
    {
      "entropy": 5.865459680557251,
      "epoch": 0.279712118264929,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000499540709149229,
      "loss": 5.7006,
      "mean_token_accuracy": 0.15096115618944167,
      "num_tokens": 6255858.0,
      "step": 3595
    },
    {
      "entropy": 5.769324588775635,
      "epoch": 0.28010114763664656,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995389374590901,
      "loss": 5.625,
      "mean_token_accuracy": 0.1556652620434761,
      "num_tokens": 6265137.0,
      "step": 3600
    },
    {
      "entropy": 5.815653467178345,
      "epoch": 0.28049017700836415,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995371623619301,
      "loss": 5.6376,
      "mean_token_accuracy": 0.15269655138254165,
      "num_tokens": 6273286.0,
      "step": 3605
    },
    {
      "entropy": 5.734041976928711,
      "epoch": 0.2808792063800817,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499535383857776,
      "loss": 5.6569,
      "mean_token_accuracy": 0.15289225429296494,
      "num_tokens": 6282477.0,
      "step": 3610
    },
    {
      "entropy": 5.840460300445557,
      "epoch": 0.28126823575179927,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499533601946655,
      "loss": 5.6878,
      "mean_token_accuracy": 0.15312220752239228,
      "num_tokens": 6291756.0,
      "step": 3615
    },
    {
      "entropy": 5.885938024520874,
      "epoch": 0.2816572651235168,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995318166285938,
      "loss": 5.7617,
      "mean_token_accuracy": 0.14861457124352456,
      "num_tokens": 6301093.0,
      "step": 3620
    },
    {
      "entropy": 5.819167995452881,
      "epoch": 0.2820462944952344,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004995300279036199,
      "loss": 5.5723,
      "mean_token_accuracy": 0.1582958698272705,
      "num_tokens": 6310379.0,
      "step": 3625
    },
    {
      "entropy": 5.771732759475708,
      "epoch": 0.28243532386695197,
      "grad_norm": 1.0625,
      "learning_rate": 0.00049952823577176,
      "loss": 5.6152,
      "mean_token_accuracy": 0.15469675064086913,
      "num_tokens": 6318983.0,
      "step": 3630
    },
    {
      "entropy": 5.784991931915283,
      "epoch": 0.2828243532386695,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995264402330416,
      "loss": 5.6046,
      "mean_token_accuracy": 0.15919178277254104,
      "num_tokens": 6327816.0,
      "step": 3635
    },
    {
      "entropy": 5.768844318389893,
      "epoch": 0.2832133826103871,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995246412874919,
      "loss": 5.6904,
      "mean_token_accuracy": 0.15365249216556548,
      "num_tokens": 6336693.0,
      "step": 3640
    },
    {
      "entropy": 5.82958459854126,
      "epoch": 0.2836024119821047,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499522838935138,
      "loss": 5.6174,
      "mean_token_accuracy": 0.15765645503997802,
      "num_tokens": 6345456.0,
      "step": 3645
    },
    {
      "entropy": 5.795063829421997,
      "epoch": 0.2839914413538222,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004995210331760074,
      "loss": 5.6223,
      "mean_token_accuracy": 0.16065620630979538,
      "num_tokens": 6353584.0,
      "step": 3650
    },
    {
      "entropy": 5.761292266845703,
      "epoch": 0.2843804707255398,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004995192240101275,
      "loss": 5.6538,
      "mean_token_accuracy": 0.15679923743009566,
      "num_tokens": 6362258.0,
      "step": 3655
    },
    {
      "entropy": 5.786276054382324,
      "epoch": 0.2847695000972573,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004995174114375258,
      "loss": 5.5671,
      "mean_token_accuracy": 0.1589123472571373,
      "num_tokens": 6371373.0,
      "step": 3660
    },
    {
      "entropy": 5.836294841766358,
      "epoch": 0.2851585294689749,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995155954582297,
      "loss": 5.6458,
      "mean_token_accuracy": 0.1591172069311142,
      "num_tokens": 6380431.0,
      "step": 3665
    },
    {
      "entropy": 5.8760663032531735,
      "epoch": 0.2855475588406925,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995137760722668,
      "loss": 5.7963,
      "mean_token_accuracy": 0.15557733029127122,
      "num_tokens": 6389305.0,
      "step": 3670
    },
    {
      "entropy": 5.843765306472778,
      "epoch": 0.28593658821241,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004995119532796646,
      "loss": 5.6244,
      "mean_token_accuracy": 0.16239040046930314,
      "num_tokens": 6397065.0,
      "step": 3675
    },
    {
      "entropy": 5.798708438873291,
      "epoch": 0.2863256175841276,
      "grad_norm": 0.96875,
      "learning_rate": 0.000499510127080451,
      "loss": 5.7606,
      "mean_token_accuracy": 0.14602240175008774,
      "num_tokens": 6407053.0,
      "step": 3680
    },
    {
      "entropy": 5.877055644989014,
      "epoch": 0.28671464695584514,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995082974746535,
      "loss": 5.7204,
      "mean_token_accuracy": 0.15210573226213456,
      "num_tokens": 6415165.0,
      "step": 3685
    },
    {
      "entropy": 5.877712726593018,
      "epoch": 0.28710367632756273,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004995064644622999,
      "loss": 5.6941,
      "mean_token_accuracy": 0.1575799763202667,
      "num_tokens": 6423646.0,
      "step": 3690
    },
    {
      "entropy": 5.808514022827149,
      "epoch": 0.2874927056992803,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004995046280434181,
      "loss": 5.7094,
      "mean_token_accuracy": 0.15264169871807098,
      "num_tokens": 6432134.0,
      "step": 3695
    },
    {
      "entropy": 5.832041025161743,
      "epoch": 0.28788173507099785,
      "grad_norm": 1.0,
      "learning_rate": 0.0004995027882180358,
      "loss": 5.6063,
      "mean_token_accuracy": 0.1572796657681465,
      "num_tokens": 6440456.0,
      "step": 3700
    },
    {
      "entropy": 5.835405588150024,
      "epoch": 0.28827076444271543,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995009449861812,
      "loss": 5.6916,
      "mean_token_accuracy": 0.14945697486400605,
      "num_tokens": 6448952.0,
      "step": 3705
    },
    {
      "entropy": 5.773239278793335,
      "epoch": 0.28865979381443296,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000499499098347882,
      "loss": 5.5347,
      "mean_token_accuracy": 0.16946829110383987,
      "num_tokens": 6457225.0,
      "step": 3710
    },
    {
      "entropy": 5.796914863586426,
      "epoch": 0.28904882318615055,
      "grad_norm": 1.0,
      "learning_rate": 0.0004994972483031662,
      "loss": 5.6842,
      "mean_token_accuracy": 0.1543801799416542,
      "num_tokens": 6465450.0,
      "step": 3715
    },
    {
      "entropy": 5.822849369049072,
      "epoch": 0.28943785255786814,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499495394852062,
      "loss": 5.7131,
      "mean_token_accuracy": 0.14737720638513566,
      "num_tokens": 6473736.0,
      "step": 3720
    },
    {
      "entropy": 5.834820890426636,
      "epoch": 0.28982688192958567,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004994935379945977,
      "loss": 5.6667,
      "mean_token_accuracy": 0.15716438889503478,
      "num_tokens": 6482092.0,
      "step": 3725
    },
    {
      "entropy": 5.822330474853516,
      "epoch": 0.29021591130130325,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004994916777308012,
      "loss": 5.6749,
      "mean_token_accuracy": 0.1561664879322052,
      "num_tokens": 6490704.0,
      "step": 3730
    },
    {
      "entropy": 5.91683177947998,
      "epoch": 0.29060494067302084,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004994898140607007,
      "loss": 5.6009,
      "mean_token_accuracy": 0.1599688485264778,
      "num_tokens": 6499634.0,
      "step": 3735
    },
    {
      "entropy": 5.731499719619751,
      "epoch": 0.29099397004473837,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004994879469843247,
      "loss": 5.7682,
      "mean_token_accuracy": 0.14854224994778634,
      "num_tokens": 6509261.0,
      "step": 3740
    },
    {
      "entropy": 5.84810438156128,
      "epoch": 0.29138299941645596,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004994860765017013,
      "loss": 5.6243,
      "mean_token_accuracy": 0.15548288971185684,
      "num_tokens": 6517542.0,
      "step": 3745
    },
    {
      "entropy": 5.8257911682128904,
      "epoch": 0.2917720287881735,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499484202612859,
      "loss": 5.6009,
      "mean_token_accuracy": 0.1555945947766304,
      "num_tokens": 6525945.0,
      "step": 3750
    },
    {
      "entropy": 5.738781452178955,
      "epoch": 0.2921610581598911,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004994823253178264,
      "loss": 5.628,
      "mean_token_accuracy": 0.1553563207387924,
      "num_tokens": 6534744.0,
      "step": 3755
    },
    {
      "entropy": 5.771017980575562,
      "epoch": 0.29255008753160866,
      "grad_norm": 1.125,
      "learning_rate": 0.0004994804446166317,
      "loss": 5.5867,
      "mean_token_accuracy": 0.16410646140575408,
      "num_tokens": 6542640.0,
      "step": 3760
    },
    {
      "entropy": 5.816923141479492,
      "epoch": 0.2929391169033262,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004994785605093035,
      "loss": 5.6569,
      "mean_token_accuracy": 0.16179122030735016,
      "num_tokens": 6551292.0,
      "step": 3765
    },
    {
      "entropy": 5.832165241241455,
      "epoch": 0.2933281462750438,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004994766729958705,
      "loss": 5.7297,
      "mean_token_accuracy": 0.15468698143959045,
      "num_tokens": 6560760.0,
      "step": 3770
    },
    {
      "entropy": 5.898743486404419,
      "epoch": 0.2937171756467613,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004994747820763612,
      "loss": 5.7414,
      "mean_token_accuracy": 0.15264707952737808,
      "num_tokens": 6570618.0,
      "step": 3775
    },
    {
      "entropy": 5.82996940612793,
      "epoch": 0.2941062050184789,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004994728877508046,
      "loss": 5.6129,
      "mean_token_accuracy": 0.15315211713314056,
      "num_tokens": 6579484.0,
      "step": 3780
    },
    {
      "entropy": 5.785426616668701,
      "epoch": 0.2944952343901965,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004994709900192289,
      "loss": 5.6532,
      "mean_token_accuracy": 0.15257718414068222,
      "num_tokens": 6588531.0,
      "step": 3785
    },
    {
      "entropy": 5.844866418838501,
      "epoch": 0.294884263761914,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004994690888816635,
      "loss": 5.7293,
      "mean_token_accuracy": 0.15567389279603958,
      "num_tokens": 6596862.0,
      "step": 3790
    },
    {
      "entropy": 5.764641141891479,
      "epoch": 0.2952732931336316,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004994671843381367,
      "loss": 5.6292,
      "mean_token_accuracy": 0.15766787976026536,
      "num_tokens": 6605594.0,
      "step": 3795
    },
    {
      "entropy": 5.838379144668579,
      "epoch": 0.29566232250534913,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004994652763886778,
      "loss": 5.6635,
      "mean_token_accuracy": 0.16003731936216353,
      "num_tokens": 6613971.0,
      "step": 3800
    },
    {
      "entropy": 5.8031247615814205,
      "epoch": 0.2960513518770667,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004994633650333155,
      "loss": 5.7493,
      "mean_token_accuracy": 0.14524856358766555,
      "num_tokens": 6623116.0,
      "step": 3805
    },
    {
      "entropy": 5.873272895812988,
      "epoch": 0.2964403812487843,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004994614502720792,
      "loss": 5.6158,
      "mean_token_accuracy": 0.16138359159231186,
      "num_tokens": 6631910.0,
      "step": 3810
    },
    {
      "entropy": 5.811766481399536,
      "epoch": 0.29682941062050183,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004994595321049974,
      "loss": 5.6467,
      "mean_token_accuracy": 0.16312259137630464,
      "num_tokens": 6640882.0,
      "step": 3815
    },
    {
      "entropy": 5.884197998046875,
      "epoch": 0.2972184399922194,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004994576105320994,
      "loss": 5.7315,
      "mean_token_accuracy": 0.1533009871840477,
      "num_tokens": 6649665.0,
      "step": 3820
    },
    {
      "entropy": 5.775503635406494,
      "epoch": 0.29760746936393695,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004994556855534146,
      "loss": 5.4815,
      "mean_token_accuracy": 0.1663750559091568,
      "num_tokens": 6658725.0,
      "step": 3825
    },
    {
      "entropy": 5.686539030075073,
      "epoch": 0.29799649873565454,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004994537571689719,
      "loss": 5.551,
      "mean_token_accuracy": 0.1626267284154892,
      "num_tokens": 6667403.0,
      "step": 3830
    },
    {
      "entropy": 5.8347007751464846,
      "epoch": 0.2983855281073721,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004994518253788007,
      "loss": 5.6452,
      "mean_token_accuracy": 0.15373515188694,
      "num_tokens": 6676673.0,
      "step": 3835
    },
    {
      "entropy": 5.848038959503174,
      "epoch": 0.29877455747908965,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004994498901829304,
      "loss": 5.6342,
      "mean_token_accuracy": 0.1482851818203926,
      "num_tokens": 6686212.0,
      "step": 3840
    },
    {
      "entropy": 5.783165359497071,
      "epoch": 0.29916358685080724,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004994479515813903,
      "loss": 5.6406,
      "mean_token_accuracy": 0.15525024384260178,
      "num_tokens": 6694812.0,
      "step": 3845
    },
    {
      "entropy": 5.783064556121826,
      "epoch": 0.2995526162225248,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004994460095742096,
      "loss": 5.6195,
      "mean_token_accuracy": 0.15523335412144662,
      "num_tokens": 6703439.0,
      "step": 3850
    },
    {
      "entropy": 5.826699924468994,
      "epoch": 0.29994164559424236,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499444064161418,
      "loss": 5.5879,
      "mean_token_accuracy": 0.15931765511631965,
      "num_tokens": 6711465.0,
      "step": 3855
    },
    {
      "entropy": 5.7629138946533205,
      "epoch": 0.30033067496595994,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000499442115343045,
      "loss": 5.6205,
      "mean_token_accuracy": 0.15265668630599977,
      "num_tokens": 6720526.0,
      "step": 3860
    },
    {
      "entropy": 5.88586049079895,
      "epoch": 0.3007197043376775,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004994401631191202,
      "loss": 5.806,
      "mean_token_accuracy": 0.14357833117246627,
      "num_tokens": 6728696.0,
      "step": 3865
    },
    {
      "entropy": 5.805546855926513,
      "epoch": 0.30110873370939506,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004994382074896731,
      "loss": 5.6646,
      "mean_token_accuracy": 0.1538841776549816,
      "num_tokens": 6737955.0,
      "step": 3870
    },
    {
      "entropy": 5.837248516082764,
      "epoch": 0.30149776308111265,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004994362484547335,
      "loss": 5.7355,
      "mean_token_accuracy": 0.14724683314561843,
      "num_tokens": 6746413.0,
      "step": 3875
    },
    {
      "entropy": 5.749407148361206,
      "epoch": 0.3018867924528302,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004994342860143309,
      "loss": 5.5684,
      "mean_token_accuracy": 0.16013670861721038,
      "num_tokens": 6754341.0,
      "step": 3880
    },
    {
      "entropy": 5.644632148742676,
      "epoch": 0.30227582182454776,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004994323201684953,
      "loss": 5.4565,
      "mean_token_accuracy": 0.1645071476697922,
      "num_tokens": 6762623.0,
      "step": 3885
    },
    {
      "entropy": 5.780033159255981,
      "epoch": 0.3026648511962653,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004994303509172566,
      "loss": 5.6016,
      "mean_token_accuracy": 0.1589878648519516,
      "num_tokens": 6771270.0,
      "step": 3890
    },
    {
      "entropy": 5.81092095375061,
      "epoch": 0.3030538805679829,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004994283782606444,
      "loss": 5.6675,
      "mean_token_accuracy": 0.15384138971567154,
      "num_tokens": 6780292.0,
      "step": 3895
    },
    {
      "entropy": 5.825908756256103,
      "epoch": 0.30344290993970047,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004994264021986888,
      "loss": 5.5621,
      "mean_token_accuracy": 0.15870441645383834,
      "num_tokens": 6788580.0,
      "step": 3900
    },
    {
      "entropy": 5.66625018119812,
      "epoch": 0.303831939311418,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004994244227314197,
      "loss": 5.5613,
      "mean_token_accuracy": 0.1561402976512909,
      "num_tokens": 6796801.0,
      "step": 3905
    },
    {
      "entropy": 5.754960536956787,
      "epoch": 0.3042209686831356,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004994224398588672,
      "loss": 5.5446,
      "mean_token_accuracy": 0.16153970509767532,
      "num_tokens": 6805630.0,
      "step": 3910
    },
    {
      "entropy": 5.659200811386109,
      "epoch": 0.3046099980548531,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004994204535810615,
      "loss": 5.5523,
      "mean_token_accuracy": 0.16061975955963134,
      "num_tokens": 6814316.0,
      "step": 3915
    },
    {
      "entropy": 5.887920093536377,
      "epoch": 0.3049990274265707,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004994184638980326,
      "loss": 5.7088,
      "mean_token_accuracy": 0.15016086995601655,
      "num_tokens": 6823224.0,
      "step": 3920
    },
    {
      "entropy": 5.86029953956604,
      "epoch": 0.3053880567982883,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004994164708098107,
      "loss": 5.7397,
      "mean_token_accuracy": 0.14942816495895386,
      "num_tokens": 6832104.0,
      "step": 3925
    },
    {
      "entropy": 5.820219278335571,
      "epoch": 0.3057770861700058,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000499414474316426,
      "loss": 5.6379,
      "mean_token_accuracy": 0.15939020812511445,
      "num_tokens": 6841208.0,
      "step": 3930
    },
    {
      "entropy": 5.628757429122925,
      "epoch": 0.3061661155417234,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004994124744179088,
      "loss": 5.4482,
      "mean_token_accuracy": 0.16597717478871346,
      "num_tokens": 6849622.0,
      "step": 3935
    },
    {
      "entropy": 5.761292362213135,
      "epoch": 0.306555144913441,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004994104711142896,
      "loss": 5.6437,
      "mean_token_accuracy": 0.16336634159088134,
      "num_tokens": 6857355.0,
      "step": 3940
    },
    {
      "entropy": 5.810007333755493,
      "epoch": 0.3069441742851585,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994084644055986,
      "loss": 5.6582,
      "mean_token_accuracy": 0.1595444455742836,
      "num_tokens": 6866173.0,
      "step": 3945
    },
    {
      "entropy": 5.7510986328125,
      "epoch": 0.3073332036568761,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004994064542918664,
      "loss": 5.6718,
      "mean_token_accuracy": 0.1521872892975807,
      "num_tokens": 6874438.0,
      "step": 3950
    },
    {
      "entropy": 5.8026892185211185,
      "epoch": 0.30772223302859364,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004994044407731235,
      "loss": 5.66,
      "mean_token_accuracy": 0.1539344683289528,
      "num_tokens": 6882790.0,
      "step": 3955
    },
    {
      "entropy": 5.790673875808716,
      "epoch": 0.3081112624003112,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004994024238494002,
      "loss": 5.671,
      "mean_token_accuracy": 0.14898976385593415,
      "num_tokens": 6891375.0,
      "step": 3960
    },
    {
      "entropy": 5.692372560501099,
      "epoch": 0.3085002917720288,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004994004035207274,
      "loss": 5.5891,
      "mean_token_accuracy": 0.1600603513419628,
      "num_tokens": 6899510.0,
      "step": 3965
    },
    {
      "entropy": 5.840362453460694,
      "epoch": 0.30888932114374634,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004993983797871356,
      "loss": 5.6485,
      "mean_token_accuracy": 0.1631278231739998,
      "num_tokens": 6908311.0,
      "step": 3970
    },
    {
      "entropy": 5.738107919692993,
      "epoch": 0.30927835051546393,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004993963526486555,
      "loss": 5.5953,
      "mean_token_accuracy": 0.15895417034626008,
      "num_tokens": 6917064.0,
      "step": 3975
    },
    {
      "entropy": 5.74478964805603,
      "epoch": 0.30966737988718146,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499394322105318,
      "loss": 5.6,
      "mean_token_accuracy": 0.16004132479429245,
      "num_tokens": 6925705.0,
      "step": 3980
    },
    {
      "entropy": 5.891635417938232,
      "epoch": 0.31005640925889905,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004993922881571538,
      "loss": 5.74,
      "mean_token_accuracy": 0.15065352767705917,
      "num_tokens": 6934232.0,
      "step": 3985
    },
    {
      "entropy": 5.842127132415771,
      "epoch": 0.31044543863061663,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004993902508041939,
      "loss": 5.6878,
      "mean_token_accuracy": 0.1538147211074829,
      "num_tokens": 6943143.0,
      "step": 3990
    },
    {
      "entropy": 5.768838930130005,
      "epoch": 0.31083446800233416,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000499388210046469,
      "loss": 5.5417,
      "mean_token_accuracy": 0.16123594641685485,
      "num_tokens": 6951613.0,
      "step": 3995
    },
    {
      "entropy": 5.795172452926636,
      "epoch": 0.31122349737405175,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004993861658840102,
      "loss": 5.74,
      "mean_token_accuracy": 0.152002190053463,
      "num_tokens": 6960531.0,
      "step": 4000
    },
    {
      "entropy": 5.8380537033081055,
      "epoch": 0.3116125267457693,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993841183168484,
      "loss": 5.6748,
      "mean_token_accuracy": 0.15421621799468993,
      "num_tokens": 6969211.0,
      "step": 4005
    },
    {
      "entropy": 5.900531911849976,
      "epoch": 0.31200155611748687,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004993820673450148,
      "loss": 5.6941,
      "mean_token_accuracy": 0.15220338851213455,
      "num_tokens": 6978414.0,
      "step": 4010
    },
    {
      "entropy": 5.698403549194336,
      "epoch": 0.31239058548920445,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004993800129685404,
      "loss": 5.4602,
      "mean_token_accuracy": 0.16010657995939254,
      "num_tokens": 6987350.0,
      "step": 4015
    },
    {
      "entropy": 5.745885181427002,
      "epoch": 0.312779614860922,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004993779551874565,
      "loss": 5.5843,
      "mean_token_accuracy": 0.16465893834829332,
      "num_tokens": 6995817.0,
      "step": 4020
    },
    {
      "entropy": 5.789159774780273,
      "epoch": 0.31316864423263957,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004993758940017943,
      "loss": 5.6774,
      "mean_token_accuracy": 0.1573045790195465,
      "num_tokens": 7004454.0,
      "step": 4025
    },
    {
      "entropy": 5.754794311523438,
      "epoch": 0.31355767360435716,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499373829411585,
      "loss": 5.5896,
      "mean_token_accuracy": 0.15946647599339486,
      "num_tokens": 7012126.0,
      "step": 4030
    },
    {
      "entropy": 5.767265462875367,
      "epoch": 0.3139467029760747,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00049937176141686,
      "loss": 5.682,
      "mean_token_accuracy": 0.1577837847173214,
      "num_tokens": 7021232.0,
      "step": 4035
    },
    {
      "entropy": 5.881110000610351,
      "epoch": 0.3143357323477923,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004993696900176506,
      "loss": 5.6477,
      "mean_token_accuracy": 0.15886460542678832,
      "num_tokens": 7029488.0,
      "step": 4040
    },
    {
      "entropy": 5.776513290405274,
      "epoch": 0.3147247617195098,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004993676152139883,
      "loss": 5.6851,
      "mean_token_accuracy": 0.1527479648590088,
      "num_tokens": 7037323.0,
      "step": 4045
    },
    {
      "entropy": 5.753616714477539,
      "epoch": 0.3151137910912274,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004993655370059046,
      "loss": 5.6355,
      "mean_token_accuracy": 0.15245578587055206,
      "num_tokens": 7045754.0,
      "step": 4050
    },
    {
      "entropy": 5.798805236816406,
      "epoch": 0.315502820462945,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004993634553934309,
      "loss": 5.5769,
      "mean_token_accuracy": 0.15374290943145752,
      "num_tokens": 7054026.0,
      "step": 4055
    },
    {
      "entropy": 5.756967258453369,
      "epoch": 0.3158918498346625,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004993613703765989,
      "loss": 5.6312,
      "mean_token_accuracy": 0.15930403470993043,
      "num_tokens": 7063383.0,
      "step": 4060
    },
    {
      "entropy": 5.731030797958374,
      "epoch": 0.3162808792063801,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004993592819554402,
      "loss": 5.5773,
      "mean_token_accuracy": 0.1582264557480812,
      "num_tokens": 7071952.0,
      "step": 4065
    },
    {
      "entropy": 5.855757522583008,
      "epoch": 0.3166699085780976,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004993571901299866,
      "loss": 5.6446,
      "mean_token_accuracy": 0.15414991676807405,
      "num_tokens": 7080283.0,
      "step": 4070
    },
    {
      "entropy": 5.688795566558838,
      "epoch": 0.3170589379498152,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993550949002697,
      "loss": 5.569,
      "mean_token_accuracy": 0.15824397653341293,
      "num_tokens": 7088669.0,
      "step": 4075
    },
    {
      "entropy": 5.705701732635498,
      "epoch": 0.3174479673215328,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004993529962663213,
      "loss": 5.4657,
      "mean_token_accuracy": 0.16952413469552993,
      "num_tokens": 7097201.0,
      "step": 4080
    },
    {
      "entropy": 5.713267803192139,
      "epoch": 0.31783699669325033,
      "grad_norm": 1.0,
      "learning_rate": 0.0004993508942281732,
      "loss": 5.638,
      "mean_token_accuracy": 0.15605451464653014,
      "num_tokens": 7106509.0,
      "step": 4085
    },
    {
      "entropy": 5.763213062286377,
      "epoch": 0.3182260260649679,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004993487887858575,
      "loss": 5.5622,
      "mean_token_accuracy": 0.15604040771722794,
      "num_tokens": 7114550.0,
      "step": 4090
    },
    {
      "entropy": 5.821378469467163,
      "epoch": 0.31861505543668545,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000499346679939406,
      "loss": 5.6685,
      "mean_token_accuracy": 0.15556033849716186,
      "num_tokens": 7122875.0,
      "step": 4095
    },
    {
      "entropy": 5.732659769058228,
      "epoch": 0.31900408480840303,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993445676888507,
      "loss": 5.4924,
      "mean_token_accuracy": 0.15718897432088852,
      "num_tokens": 7132263.0,
      "step": 4100
    },
    {
      "entropy": 5.739199256896972,
      "epoch": 0.3193931141801206,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004993424520342236,
      "loss": 5.6961,
      "mean_token_accuracy": 0.15194219574332238,
      "num_tokens": 7141448.0,
      "step": 4105
    },
    {
      "entropy": 5.737508296966553,
      "epoch": 0.31978214355183815,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004993403329755569,
      "loss": 5.7148,
      "mean_token_accuracy": 0.1534551165997982,
      "num_tokens": 7151021.0,
      "step": 4110
    },
    {
      "entropy": 5.86180305480957,
      "epoch": 0.32017117292355574,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004993382105128828,
      "loss": 5.6588,
      "mean_token_accuracy": 0.1554321050643921,
      "num_tokens": 7160103.0,
      "step": 4115
    },
    {
      "entropy": 5.833527898788452,
      "epoch": 0.32056020229527327,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004993360846462333,
      "loss": 5.6423,
      "mean_token_accuracy": 0.14571528136730194,
      "num_tokens": 7169003.0,
      "step": 4120
    },
    {
      "entropy": 5.811704015731811,
      "epoch": 0.32094923166699085,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004993339553756408,
      "loss": 5.6705,
      "mean_token_accuracy": 0.15842568203806878,
      "num_tokens": 7177086.0,
      "step": 4125
    },
    {
      "entropy": 5.772511434555054,
      "epoch": 0.32133826103870844,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004993318227011378,
      "loss": 5.5535,
      "mean_token_accuracy": 0.15517180562019348,
      "num_tokens": 7185772.0,
      "step": 4130
    },
    {
      "entropy": 5.821723508834839,
      "epoch": 0.32172729041042597,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004993296866227562,
      "loss": 5.6943,
      "mean_token_accuracy": 0.15868849605321883,
      "num_tokens": 7194394.0,
      "step": 4135
    },
    {
      "entropy": 5.725718975067139,
      "epoch": 0.32211631978214356,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004993275471405288,
      "loss": 5.5055,
      "mean_token_accuracy": 0.1687486931681633,
      "num_tokens": 7202143.0,
      "step": 4140
    },
    {
      "entropy": 5.703480577468872,
      "epoch": 0.32250534915386114,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004993254042544879,
      "loss": 5.6392,
      "mean_token_accuracy": 0.15573893785476683,
      "num_tokens": 7210645.0,
      "step": 4145
    },
    {
      "entropy": 5.810609340667725,
      "epoch": 0.3228943785255787,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000499323257964666,
      "loss": 5.5839,
      "mean_token_accuracy": 0.1578369915485382,
      "num_tokens": 7218872.0,
      "step": 4150
    },
    {
      "entropy": 5.827525615692139,
      "epoch": 0.32328340789729626,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004993211082710957,
      "loss": 5.7357,
      "mean_token_accuracy": 0.15549167096614838,
      "num_tokens": 7226977.0,
      "step": 4155
    },
    {
      "entropy": 5.805359888076782,
      "epoch": 0.3236724372690138,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993189551738097,
      "loss": 5.5939,
      "mean_token_accuracy": 0.1561035618185997,
      "num_tokens": 7235635.0,
      "step": 4160
    },
    {
      "entropy": 5.752389812469483,
      "epoch": 0.3240614666407314,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993167986728405,
      "loss": 5.5851,
      "mean_token_accuracy": 0.1615038052201271,
      "num_tokens": 7244506.0,
      "step": 4165
    },
    {
      "entropy": 5.71797547340393,
      "epoch": 0.32445049601244896,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499314638768221,
      "loss": 5.6141,
      "mean_token_accuracy": 0.15862663090229034,
      "num_tokens": 7252883.0,
      "step": 4170
    },
    {
      "entropy": 5.817626667022705,
      "epoch": 0.3248395253841665,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004993124754599839,
      "loss": 5.559,
      "mean_token_accuracy": 0.1573678582906723,
      "num_tokens": 7261580.0,
      "step": 4175
    },
    {
      "entropy": 5.788261413574219,
      "epoch": 0.3252285547558841,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004993103087481619,
      "loss": 5.6593,
      "mean_token_accuracy": 0.15661897361278534,
      "num_tokens": 7270346.0,
      "step": 4180
    },
    {
      "entropy": 5.83748345375061,
      "epoch": 0.3256175841276016,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993081386327881,
      "loss": 5.729,
      "mean_token_accuracy": 0.1532787561416626,
      "num_tokens": 7279393.0,
      "step": 4185
    },
    {
      "entropy": 5.756169605255127,
      "epoch": 0.3260066134993192,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004993059651138953,
      "loss": 5.5584,
      "mean_token_accuracy": 0.15885259956121445,
      "num_tokens": 7287929.0,
      "step": 4190
    },
    {
      "entropy": 5.778932189941406,
      "epoch": 0.3263956428710368,
      "grad_norm": 1.0,
      "learning_rate": 0.0004993037881915165,
      "loss": 5.6158,
      "mean_token_accuracy": 0.1615328386425972,
      "num_tokens": 7296336.0,
      "step": 4195
    },
    {
      "entropy": 5.6543677806854244,
      "epoch": 0.3267846722427543,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004993016078656847,
      "loss": 5.4912,
      "mean_token_accuracy": 0.15942687541246414,
      "num_tokens": 7304624.0,
      "step": 4200
    },
    {
      "entropy": 5.716690254211426,
      "epoch": 0.3271737016144719,
      "grad_norm": 0.96875,
      "learning_rate": 0.000499299424136433,
      "loss": 5.6219,
      "mean_token_accuracy": 0.15619123354554176,
      "num_tokens": 7312945.0,
      "step": 4205
    },
    {
      "entropy": 5.769815158843994,
      "epoch": 0.32756273098618943,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004992972370037946,
      "loss": 5.554,
      "mean_token_accuracy": 0.1645971119403839,
      "num_tokens": 7321273.0,
      "step": 4210
    },
    {
      "entropy": 5.797748136520386,
      "epoch": 0.327951760357907,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004992950464678026,
      "loss": 5.6568,
      "mean_token_accuracy": 0.15764952600002288,
      "num_tokens": 7330446.0,
      "step": 4215
    },
    {
      "entropy": 5.675560712814331,
      "epoch": 0.3283407897296246,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004992928525284903,
      "loss": 5.5958,
      "mean_token_accuracy": 0.161003540456295,
      "num_tokens": 7339282.0,
      "step": 4220
    },
    {
      "entropy": 5.764131212234497,
      "epoch": 0.32872981910134214,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499290655185891,
      "loss": 5.5609,
      "mean_token_accuracy": 0.1540660962462425,
      "num_tokens": 7347848.0,
      "step": 4225
    },
    {
      "entropy": 5.696079444885254,
      "epoch": 0.3291188484730597,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000499288454440038,
      "loss": 5.5608,
      "mean_token_accuracy": 0.16451414823532104,
      "num_tokens": 7356678.0,
      "step": 4230
    },
    {
      "entropy": 5.768759489059448,
      "epoch": 0.3295078778447773,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004992862502909645,
      "loss": 5.6019,
      "mean_token_accuracy": 0.1615913599729538,
      "num_tokens": 7365411.0,
      "step": 4235
    },
    {
      "entropy": 5.6910933494567875,
      "epoch": 0.32989690721649484,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004992840427387043,
      "loss": 5.5292,
      "mean_token_accuracy": 0.164210769534111,
      "num_tokens": 7373635.0,
      "step": 4240
    },
    {
      "entropy": 5.741477632522583,
      "epoch": 0.3302859365882124,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004992818317832908,
      "loss": 5.6588,
      "mean_token_accuracy": 0.15966336578130721,
      "num_tokens": 7382469.0,
      "step": 4245
    },
    {
      "entropy": 5.802251148223877,
      "epoch": 0.33067496595992996,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004992796174247574,
      "loss": 5.6306,
      "mean_token_accuracy": 0.15575413405895233,
      "num_tokens": 7390712.0,
      "step": 4250
    },
    {
      "entropy": 5.818617582321167,
      "epoch": 0.33106399533164754,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004992773996631378,
      "loss": 5.5586,
      "mean_token_accuracy": 0.1632613092660904,
      "num_tokens": 7398905.0,
      "step": 4255
    },
    {
      "entropy": 5.753144216537476,
      "epoch": 0.33145302470336513,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004992751784984656,
      "loss": 5.6413,
      "mean_token_accuracy": 0.15888381749391556,
      "num_tokens": 7407242.0,
      "step": 4260
    },
    {
      "entropy": 5.742626333236695,
      "epoch": 0.33184205407508266,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004992729539307745,
      "loss": 5.6662,
      "mean_token_accuracy": 0.15572152584791182,
      "num_tokens": 7416489.0,
      "step": 4265
    },
    {
      "entropy": 5.865239953994751,
      "epoch": 0.33223108344680025,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004992707259600984,
      "loss": 5.657,
      "mean_token_accuracy": 0.1557759702205658,
      "num_tokens": 7424507.0,
      "step": 4270
    },
    {
      "entropy": 5.741849136352539,
      "epoch": 0.3326201128185178,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004992684945864709,
      "loss": 5.6318,
      "mean_token_accuracy": 0.1520262397825718,
      "num_tokens": 7433343.0,
      "step": 4275
    },
    {
      "entropy": 5.688818788528442,
      "epoch": 0.33300914219023536,
      "grad_norm": 1.0,
      "learning_rate": 0.0004992662598099261,
      "loss": 5.5601,
      "mean_token_accuracy": 0.16512439996004105,
      "num_tokens": 7441452.0,
      "step": 4280
    },
    {
      "entropy": 5.804032039642334,
      "epoch": 0.33339817156195295,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004992640216304975,
      "loss": 5.6298,
      "mean_token_accuracy": 0.16103580743074417,
      "num_tokens": 7449541.0,
      "step": 4285
    },
    {
      "entropy": 5.765760898590088,
      "epoch": 0.3337872009336705,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004992617800482196,
      "loss": 5.642,
      "mean_token_accuracy": 0.15369921773672104,
      "num_tokens": 7458812.0,
      "step": 4290
    },
    {
      "entropy": 5.7393537044525145,
      "epoch": 0.33417623030538807,
      "grad_norm": 0.953125,
      "learning_rate": 0.000499259535063126,
      "loss": 5.5783,
      "mean_token_accuracy": 0.15623634457588195,
      "num_tokens": 7467607.0,
      "step": 4295
    },
    {
      "entropy": 5.776077842712402,
      "epoch": 0.3345652596771056,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499257286675251,
      "loss": 5.5016,
      "mean_token_accuracy": 0.16391390860080718,
      "num_tokens": 7476047.0,
      "step": 4300
    },
    {
      "entropy": 5.741789388656616,
      "epoch": 0.3349542890488232,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004992550348846285,
      "loss": 5.5719,
      "mean_token_accuracy": 0.15823668837547303,
      "num_tokens": 7484496.0,
      "step": 4305
    },
    {
      "entropy": 5.672589492797852,
      "epoch": 0.33534331842054077,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004992527796912928,
      "loss": 5.5001,
      "mean_token_accuracy": 0.16356348618865013,
      "num_tokens": 7493161.0,
      "step": 4310
    },
    {
      "entropy": 5.758045291900634,
      "epoch": 0.3357323477922583,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004992505210952782,
      "loss": 5.6336,
      "mean_token_accuracy": 0.156315216422081,
      "num_tokens": 7502122.0,
      "step": 4315
    },
    {
      "entropy": 5.763626956939698,
      "epoch": 0.3361213771639759,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004992482590966188,
      "loss": 5.6132,
      "mean_token_accuracy": 0.1597306936979294,
      "num_tokens": 7510369.0,
      "step": 4320
    },
    {
      "entropy": 5.685847663879395,
      "epoch": 0.3365104065356934,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004992459936953489,
      "loss": 5.4874,
      "mean_token_accuracy": 0.16831238120794295,
      "num_tokens": 7519321.0,
      "step": 4325
    },
    {
      "entropy": 5.685293483734131,
      "epoch": 0.336899435907411,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004992437248915032,
      "loss": 5.5365,
      "mean_token_accuracy": 0.16498717069625854,
      "num_tokens": 7528196.0,
      "step": 4330
    },
    {
      "entropy": 5.687713193893432,
      "epoch": 0.3372884652791286,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004992414526851157,
      "loss": 5.6024,
      "mean_token_accuracy": 0.15667464807629586,
      "num_tokens": 7536124.0,
      "step": 4335
    },
    {
      "entropy": 5.829200553894043,
      "epoch": 0.3376774946508461,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004992391770762212,
      "loss": 5.6934,
      "mean_token_accuracy": 0.1513142079114914,
      "num_tokens": 7544409.0,
      "step": 4340
    },
    {
      "entropy": 5.788577079772949,
      "epoch": 0.3380665240225637,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000499236898064854,
      "loss": 5.5907,
      "mean_token_accuracy": 0.1621900752186775,
      "num_tokens": 7552803.0,
      "step": 4345
    },
    {
      "entropy": 5.67437744140625,
      "epoch": 0.3384555533942813,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004992346156510488,
      "loss": 5.5643,
      "mean_token_accuracy": 0.15769924521446227,
      "num_tokens": 7561909.0,
      "step": 4350
    },
    {
      "entropy": 5.832115030288696,
      "epoch": 0.3388445827659988,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004992323298348403,
      "loss": 5.6147,
      "mean_token_accuracy": 0.15296869426965715,
      "num_tokens": 7571759.0,
      "step": 4355
    },
    {
      "entropy": 5.784440565109253,
      "epoch": 0.3392336121377164,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499230040616263,
      "loss": 5.7427,
      "mean_token_accuracy": 0.1566373661160469,
      "num_tokens": 7580010.0,
      "step": 4360
    },
    {
      "entropy": 5.756098937988281,
      "epoch": 0.33962264150943394,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004992277479953518,
      "loss": 5.5451,
      "mean_token_accuracy": 0.15730045288801192,
      "num_tokens": 7587841.0,
      "step": 4365
    },
    {
      "entropy": 5.758036184310913,
      "epoch": 0.34001167088115153,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004992254519721414,
      "loss": 5.5907,
      "mean_token_accuracy": 0.1559063196182251,
      "num_tokens": 7596187.0,
      "step": 4370
    },
    {
      "entropy": 5.651873350143433,
      "epoch": 0.3404007002528691,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004992231525466666,
      "loss": 5.4951,
      "mean_token_accuracy": 0.16344073712825774,
      "num_tokens": 7604361.0,
      "step": 4375
    },
    {
      "entropy": 5.770703601837158,
      "epoch": 0.34078972962458665,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004992208497189624,
      "loss": 5.5845,
      "mean_token_accuracy": 0.1653784215450287,
      "num_tokens": 7613173.0,
      "step": 4380
    },
    {
      "entropy": 5.727901792526245,
      "epoch": 0.34117875899630423,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004992185434890637,
      "loss": 5.6743,
      "mean_token_accuracy": 0.1515727773308754,
      "num_tokens": 7621730.0,
      "step": 4385
    },
    {
      "entropy": 5.803450727462769,
      "epoch": 0.34156778836802176,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004992162338570055,
      "loss": 5.6471,
      "mean_token_accuracy": 0.157301127910614,
      "num_tokens": 7630362.0,
      "step": 4390
    },
    {
      "entropy": 5.774415397644043,
      "epoch": 0.34195681773973935,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004992139208228227,
      "loss": 5.5894,
      "mean_token_accuracy": 0.15711733400821687,
      "num_tokens": 7639368.0,
      "step": 4395
    },
    {
      "entropy": 5.756467247009278,
      "epoch": 0.34234584711145694,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004992116043865506,
      "loss": 5.6205,
      "mean_token_accuracy": 0.15912013053894042,
      "num_tokens": 7648399.0,
      "step": 4400
    },
    {
      "entropy": 5.696918106079101,
      "epoch": 0.34273487648317447,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004992092845482244,
      "loss": 5.4518,
      "mean_token_accuracy": 0.16868401318788528,
      "num_tokens": 7657298.0,
      "step": 4405
    },
    {
      "entropy": 5.810102844238282,
      "epoch": 0.34312390585489205,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004992069613078791,
      "loss": 5.6098,
      "mean_token_accuracy": 0.16102852821350097,
      "num_tokens": 7665328.0,
      "step": 4410
    },
    {
      "entropy": 5.696934795379638,
      "epoch": 0.3435129352266096,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004992046346655499,
      "loss": 5.5544,
      "mean_token_accuracy": 0.1619025468826294,
      "num_tokens": 7674610.0,
      "step": 4415
    },
    {
      "entropy": 5.747613096237183,
      "epoch": 0.34390196459832717,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004992023046212723,
      "loss": 5.5785,
      "mean_token_accuracy": 0.15795013606548308,
      "num_tokens": 7682727.0,
      "step": 4420
    },
    {
      "entropy": 5.72398157119751,
      "epoch": 0.34429099397004476,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004991999711750816,
      "loss": 5.5445,
      "mean_token_accuracy": 0.17041088342666627,
      "num_tokens": 7691547.0,
      "step": 4425
    },
    {
      "entropy": 5.679721450805664,
      "epoch": 0.3446800233417623,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004991976343270133,
      "loss": 5.5176,
      "mean_token_accuracy": 0.1569729596376419,
      "num_tokens": 7699987.0,
      "step": 4430
    },
    {
      "entropy": 5.679746198654175,
      "epoch": 0.3450690527134799,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004991952940771026,
      "loss": 5.5679,
      "mean_token_accuracy": 0.16395378410816192,
      "num_tokens": 7708397.0,
      "step": 4435
    },
    {
      "entropy": 5.719892168045044,
      "epoch": 0.34545808208519746,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004991929504253852,
      "loss": 5.5428,
      "mean_token_accuracy": 0.15749787986278535,
      "num_tokens": 7716795.0,
      "step": 4440
    },
    {
      "entropy": 5.704584217071533,
      "epoch": 0.345847111456915,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004991906033718966,
      "loss": 5.5675,
      "mean_token_accuracy": 0.1568581983447075,
      "num_tokens": 7725663.0,
      "step": 4445
    },
    {
      "entropy": 5.702336406707763,
      "epoch": 0.3462361408286326,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004991882529166724,
      "loss": 5.6945,
      "mean_token_accuracy": 0.15824410766363145,
      "num_tokens": 7735175.0,
      "step": 4450
    },
    {
      "entropy": 5.787890100479126,
      "epoch": 0.3466251702003501,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004991858990597483,
      "loss": 5.5482,
      "mean_token_accuracy": 0.16432606428861618,
      "num_tokens": 7744041.0,
      "step": 4455
    },
    {
      "entropy": 5.751737403869629,
      "epoch": 0.3470141995720677,
      "grad_norm": 1.0,
      "learning_rate": 0.0004991835418011601,
      "loss": 5.6157,
      "mean_token_accuracy": 0.15478222519159318,
      "num_tokens": 7752168.0,
      "step": 4460
    },
    {
      "entropy": 5.70343656539917,
      "epoch": 0.3474032289437853,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004991811811409434,
      "loss": 5.5576,
      "mean_token_accuracy": 0.16113696992397308,
      "num_tokens": 7760746.0,
      "step": 4465
    },
    {
      "entropy": 5.749729156494141,
      "epoch": 0.3477922583155028,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004991788170791341,
      "loss": 5.5513,
      "mean_token_accuracy": 0.1597917929291725,
      "num_tokens": 7769960.0,
      "step": 4470
    },
    {
      "entropy": 5.750726318359375,
      "epoch": 0.3481812876872204,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004991764496157682,
      "loss": 5.5712,
      "mean_token_accuracy": 0.16283269673585893,
      "num_tokens": 7778475.0,
      "step": 4475
    },
    {
      "entropy": 5.662684679031372,
      "epoch": 0.34857031705893793,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004991740787508814,
      "loss": 5.4511,
      "mean_token_accuracy": 0.16975423246622084,
      "num_tokens": 7787590.0,
      "step": 4480
    },
    {
      "entropy": 5.736890506744385,
      "epoch": 0.3489593464306555,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004991717044845097,
      "loss": 5.62,
      "mean_token_accuracy": 0.15432989001274108,
      "num_tokens": 7796841.0,
      "step": 4485
    },
    {
      "entropy": 5.816498899459839,
      "epoch": 0.3493483758023731,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004991693268166892,
      "loss": 5.6291,
      "mean_token_accuracy": 0.15610261186957358,
      "num_tokens": 7804889.0,
      "step": 4490
    },
    {
      "entropy": 5.769539308547974,
      "epoch": 0.34973740517409063,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004991669457474559,
      "loss": 5.6189,
      "mean_token_accuracy": 0.15933216661214827,
      "num_tokens": 7813185.0,
      "step": 4495
    },
    {
      "entropy": 5.716590642929077,
      "epoch": 0.3501264345458082,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004991645612768461,
      "loss": 5.5946,
      "mean_token_accuracy": 0.15529540926218033,
      "num_tokens": 7821630.0,
      "step": 4500
    },
    {
      "entropy": 5.784337091445923,
      "epoch": 0.35051546391752575,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004991621734048958,
      "loss": 5.6226,
      "mean_token_accuracy": 0.16395158022642137,
      "num_tokens": 7829941.0,
      "step": 4505
    },
    {
      "entropy": 5.713160514831543,
      "epoch": 0.35090449328924334,
      "grad_norm": 1.0,
      "learning_rate": 0.0004991597821316415,
      "loss": 5.5748,
      "mean_token_accuracy": 0.16333524584770204,
      "num_tokens": 7838515.0,
      "step": 4510
    },
    {
      "entropy": 5.789814758300781,
      "epoch": 0.3512935226609609,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004991573874571192,
      "loss": 5.5708,
      "mean_token_accuracy": 0.1628706246614456,
      "num_tokens": 7846575.0,
      "step": 4515
    },
    {
      "entropy": 5.653206491470337,
      "epoch": 0.35168255203267845,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004991549893813653,
      "loss": 5.4598,
      "mean_token_accuracy": 0.16667032688856126,
      "num_tokens": 7854970.0,
      "step": 4520
    },
    {
      "entropy": 5.701358604431152,
      "epoch": 0.35207158140439604,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004991525879044162,
      "loss": 5.5581,
      "mean_token_accuracy": 0.16041914373636246,
      "num_tokens": 7863293.0,
      "step": 4525
    },
    {
      "entropy": 5.754718875885009,
      "epoch": 0.35246061077611357,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004991501830263084,
      "loss": 5.5569,
      "mean_token_accuracy": 0.16397467255592346,
      "num_tokens": 7871029.0,
      "step": 4530
    },
    {
      "entropy": 5.646997356414795,
      "epoch": 0.35284964014783116,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004991477747470783,
      "loss": 5.42,
      "mean_token_accuracy": 0.1670575261116028,
      "num_tokens": 7879797.0,
      "step": 4535
    },
    {
      "entropy": 5.8155810832977295,
      "epoch": 0.35323866951954874,
      "grad_norm": 1.0,
      "learning_rate": 0.0004991453630667625,
      "loss": 5.7321,
      "mean_token_accuracy": 0.15194890201091765,
      "num_tokens": 7888600.0,
      "step": 4540
    },
    {
      "entropy": 5.701278352737427,
      "epoch": 0.3536276988912663,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004991429479853976,
      "loss": 5.552,
      "mean_token_accuracy": 0.16275644749403,
      "num_tokens": 7897794.0,
      "step": 4545
    },
    {
      "entropy": 5.757588911056518,
      "epoch": 0.35401672826298386,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004991405295030202,
      "loss": 5.5356,
      "mean_token_accuracy": 0.16324365884065628,
      "num_tokens": 7905822.0,
      "step": 4550
    },
    {
      "entropy": 5.6449367046356205,
      "epoch": 0.35440575763470145,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004991381076196671,
      "loss": 5.622,
      "mean_token_accuracy": 0.16375057995319367,
      "num_tokens": 7914560.0,
      "step": 4555
    },
    {
      "entropy": 5.6144177436828615,
      "epoch": 0.354794787006419,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004991356823353748,
      "loss": 5.4537,
      "mean_token_accuracy": 0.171008837223053,
      "num_tokens": 7924058.0,
      "step": 4560
    },
    {
      "entropy": 5.7179710388183596,
      "epoch": 0.35518381637813656,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004991332536501804,
      "loss": 5.599,
      "mean_token_accuracy": 0.15646698474884033,
      "num_tokens": 7932979.0,
      "step": 4565
    },
    {
      "entropy": 5.646498298645019,
      "epoch": 0.3555728457498541,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004991308215641205,
      "loss": 5.385,
      "mean_token_accuracy": 0.17703631222248079,
      "num_tokens": 7941357.0,
      "step": 4570
    },
    {
      "entropy": 5.65172176361084,
      "epoch": 0.3559618751215717,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004991283860772323,
      "loss": 5.5607,
      "mean_token_accuracy": 0.1630801796913147,
      "num_tokens": 7949997.0,
      "step": 4575
    },
    {
      "entropy": 5.711667251586914,
      "epoch": 0.35635090449328927,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004991259471895525,
      "loss": 5.5028,
      "mean_token_accuracy": 0.16014036685228347,
      "num_tokens": 7958246.0,
      "step": 4580
    },
    {
      "entropy": 5.708824253082275,
      "epoch": 0.3567399338650068,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004991235049011182,
      "loss": 5.5885,
      "mean_token_accuracy": 0.1583796262741089,
      "num_tokens": 7967176.0,
      "step": 4585
    },
    {
      "entropy": 5.70987868309021,
      "epoch": 0.3571289632367244,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004991210592119663,
      "loss": 5.4486,
      "mean_token_accuracy": 0.16676732301712036,
      "num_tokens": 7974761.0,
      "step": 4590
    },
    {
      "entropy": 5.630984497070313,
      "epoch": 0.3575179926084419,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004991186101221342,
      "loss": 5.5267,
      "mean_token_accuracy": 0.16331609338521957,
      "num_tokens": 7983155.0,
      "step": 4595
    },
    {
      "entropy": 5.699593448638916,
      "epoch": 0.3579070219801595,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000499116157631659,
      "loss": 5.5458,
      "mean_token_accuracy": 0.15738096237182617,
      "num_tokens": 7991276.0,
      "step": 4600
    },
    {
      "entropy": 5.7626330852508545,
      "epoch": 0.3582960513518771,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004991137017405777,
      "loss": 5.6711,
      "mean_token_accuracy": 0.15599002540111542,
      "num_tokens": 7999488.0,
      "step": 4605
    },
    {
      "entropy": 5.720730113983154,
      "epoch": 0.3586850807235946,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004991112424489277,
      "loss": 5.5315,
      "mean_token_accuracy": 0.16649463176727294,
      "num_tokens": 8008553.0,
      "step": 4610
    },
    {
      "entropy": 5.61679973602295,
      "epoch": 0.3590741100953122,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004991087797567462,
      "loss": 5.5026,
      "mean_token_accuracy": 0.16620901077985764,
      "num_tokens": 8017890.0,
      "step": 4615
    },
    {
      "entropy": 5.737289190292358,
      "epoch": 0.35946313946702974,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004991063136640709,
      "loss": 5.6677,
      "mean_token_accuracy": 0.14980827048420906,
      "num_tokens": 8025782.0,
      "step": 4620
    },
    {
      "entropy": 5.7484447956085205,
      "epoch": 0.3598521688387473,
      "grad_norm": 1.0,
      "learning_rate": 0.0004991038441709388,
      "loss": 5.5291,
      "mean_token_accuracy": 0.16813595592975616,
      "num_tokens": 8034623.0,
      "step": 4625
    },
    {
      "entropy": 5.672695159912109,
      "epoch": 0.3602411982104649,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004991013712773877,
      "loss": 5.5368,
      "mean_token_accuracy": 0.16356041878461838,
      "num_tokens": 8043789.0,
      "step": 4630
    },
    {
      "entropy": 5.66884446144104,
      "epoch": 0.36063022758218244,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004990988949834549,
      "loss": 5.5089,
      "mean_token_accuracy": 0.16036368310451507,
      "num_tokens": 8053456.0,
      "step": 4635
    },
    {
      "entropy": 5.6446342945098875,
      "epoch": 0.3610192569539,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499096415289178,
      "loss": 5.4861,
      "mean_token_accuracy": 0.16083667427301407,
      "num_tokens": 8061692.0,
      "step": 4640
    },
    {
      "entropy": 5.5944324970245365,
      "epoch": 0.3614082863256176,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004990939321945948,
      "loss": 5.4375,
      "mean_token_accuracy": 0.16898863017559052,
      "num_tokens": 8069564.0,
      "step": 4645
    },
    {
      "entropy": 5.7000421524047855,
      "epoch": 0.36179731569733514,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004990914456997427,
      "loss": 5.5035,
      "mean_token_accuracy": 0.16119355112314224,
      "num_tokens": 8078490.0,
      "step": 4650
    },
    {
      "entropy": 5.7108453750610355,
      "epoch": 0.36218634506905273,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004990889558046598,
      "loss": 5.6043,
      "mean_token_accuracy": 0.158635513484478,
      "num_tokens": 8086858.0,
      "step": 4655
    },
    {
      "entropy": 5.704522180557251,
      "epoch": 0.36257537444077026,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004990864625093836,
      "loss": 5.5605,
      "mean_token_accuracy": 0.16476080417633057,
      "num_tokens": 8094542.0,
      "step": 4660
    },
    {
      "entropy": 5.746029615402222,
      "epoch": 0.36296440381248785,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004990839658139519,
      "loss": 5.5847,
      "mean_token_accuracy": 0.16405946165323257,
      "num_tokens": 8102756.0,
      "step": 4665
    },
    {
      "entropy": 5.765752935409546,
      "epoch": 0.36335343318420543,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004990814657184028,
      "loss": 5.615,
      "mean_token_accuracy": 0.16670868843793868,
      "num_tokens": 8111258.0,
      "step": 4670
    },
    {
      "entropy": 5.8236072063446045,
      "epoch": 0.36374246255592296,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004990789622227741,
      "loss": 5.7199,
      "mean_token_accuracy": 0.1491565763950348,
      "num_tokens": 8119624.0,
      "step": 4675
    },
    {
      "entropy": 5.668383979797364,
      "epoch": 0.36413149192764055,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004990764553271038,
      "loss": 5.5055,
      "mean_token_accuracy": 0.1655653491616249,
      "num_tokens": 8128156.0,
      "step": 4680
    },
    {
      "entropy": 5.650137901306152,
      "epoch": 0.3645205212993581,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004990739450314299,
      "loss": 5.5584,
      "mean_token_accuracy": 0.15826979875564576,
      "num_tokens": 8136080.0,
      "step": 4685
    },
    {
      "entropy": 5.745880222320556,
      "epoch": 0.36490955067107567,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004990714313357906,
      "loss": 5.5416,
      "mean_token_accuracy": 0.1668192923069,
      "num_tokens": 8144742.0,
      "step": 4690
    },
    {
      "entropy": 5.750147914886474,
      "epoch": 0.36529858004279325,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499068914240224,
      "loss": 5.6008,
      "mean_token_accuracy": 0.16182359755039216,
      "num_tokens": 8153377.0,
      "step": 4695
    },
    {
      "entropy": 5.7304154396057125,
      "epoch": 0.3656876094145108,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004990663937447682,
      "loss": 5.534,
      "mean_token_accuracy": 0.16247955858707427,
      "num_tokens": 8162169.0,
      "step": 4700
    },
    {
      "entropy": 5.65903868675232,
      "epoch": 0.36607663878622837,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004990638698494615,
      "loss": 5.4948,
      "mean_token_accuracy": 0.16574820280075073,
      "num_tokens": 8170727.0,
      "step": 4705
    },
    {
      "entropy": 5.609049987792969,
      "epoch": 0.3664656681579459,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004990613425543423,
      "loss": 5.4432,
      "mean_token_accuracy": 0.16713500916957855,
      "num_tokens": 8179013.0,
      "step": 4710
    },
    {
      "entropy": 5.595587873458863,
      "epoch": 0.3668546975296635,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004990588118594487,
      "loss": 5.4148,
      "mean_token_accuracy": 0.16975449919700622,
      "num_tokens": 8187030.0,
      "step": 4715
    },
    {
      "entropy": 5.610220909118652,
      "epoch": 0.3672437269013811,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004990562777648194,
      "loss": 5.3988,
      "mean_token_accuracy": 0.16739748567342758,
      "num_tokens": 8195073.0,
      "step": 4720
    },
    {
      "entropy": 5.633205413818359,
      "epoch": 0.3676327562730986,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004990537402704928,
      "loss": 5.57,
      "mean_token_accuracy": 0.16305485963821412,
      "num_tokens": 8203626.0,
      "step": 4725
    },
    {
      "entropy": 5.68911452293396,
      "epoch": 0.3680217856448162,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004990511993765071,
      "loss": 5.5415,
      "mean_token_accuracy": 0.16565153896808624,
      "num_tokens": 8212750.0,
      "step": 4730
    },
    {
      "entropy": 5.757883167266845,
      "epoch": 0.3684108150165337,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004990486550829011,
      "loss": 5.5506,
      "mean_token_accuracy": 0.15366293638944625,
      "num_tokens": 8221542.0,
      "step": 4735
    },
    {
      "entropy": 5.560074138641357,
      "epoch": 0.3687998443882513,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004990461073897134,
      "loss": 5.4779,
      "mean_token_accuracy": 0.16144995987415314,
      "num_tokens": 8230574.0,
      "step": 4740
    },
    {
      "entropy": 5.744259548187256,
      "epoch": 0.3691888737599689,
      "grad_norm": 1.0,
      "learning_rate": 0.0004990435562969827,
      "loss": 5.573,
      "mean_token_accuracy": 0.15981101393699645,
      "num_tokens": 8239400.0,
      "step": 4745
    },
    {
      "entropy": 5.690568304061889,
      "epoch": 0.3695779031316864,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004990410018047475,
      "loss": 5.4074,
      "mean_token_accuracy": 0.16521767675876617,
      "num_tokens": 8247537.0,
      "step": 4750
    },
    {
      "entropy": 5.611506748199463,
      "epoch": 0.369966932503404,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004990384439130467,
      "loss": 5.5222,
      "mean_token_accuracy": 0.15909540504217148,
      "num_tokens": 8256551.0,
      "step": 4755
    },
    {
      "entropy": 5.7253063201904295,
      "epoch": 0.3703559618751216,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004990358826219192,
      "loss": 5.5599,
      "mean_token_accuracy": 0.16011908501386643,
      "num_tokens": 8265492.0,
      "step": 4760
    },
    {
      "entropy": 5.712846469879151,
      "epoch": 0.37074499124683913,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004990333179314038,
      "loss": 5.4861,
      "mean_token_accuracy": 0.16490963697433472,
      "num_tokens": 8273798.0,
      "step": 4765
    },
    {
      "entropy": 5.616565990447998,
      "epoch": 0.3711340206185567,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004990307498415393,
      "loss": 5.4606,
      "mean_token_accuracy": 0.16989797204732895,
      "num_tokens": 8281998.0,
      "step": 4770
    },
    {
      "entropy": 5.683922147750854,
      "epoch": 0.37152304999027425,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004990281783523648,
      "loss": 5.6,
      "mean_token_accuracy": 0.16016341298818587,
      "num_tokens": 8289833.0,
      "step": 4775
    },
    {
      "entropy": 5.606426763534546,
      "epoch": 0.37191207936199183,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004990256034639192,
      "loss": 5.5551,
      "mean_token_accuracy": 0.16734271496534348,
      "num_tokens": 8297562.0,
      "step": 4780
    },
    {
      "entropy": 5.742149448394775,
      "epoch": 0.3723011087337094,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004990230251762418,
      "loss": 5.5653,
      "mean_token_accuracy": 0.16116626635193826,
      "num_tokens": 8305262.0,
      "step": 4785
    },
    {
      "entropy": 5.674399423599243,
      "epoch": 0.37269013810542695,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004990204434893713,
      "loss": 5.4741,
      "mean_token_accuracy": 0.16709145307540893,
      "num_tokens": 8313949.0,
      "step": 4790
    },
    {
      "entropy": 5.641895389556884,
      "epoch": 0.37307916747714454,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004990178584033474,
      "loss": 5.5143,
      "mean_token_accuracy": 0.16286983639001845,
      "num_tokens": 8322801.0,
      "step": 4795
    },
    {
      "entropy": 5.716209983825683,
      "epoch": 0.37346819684886207,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004990152699182089,
      "loss": 5.5031,
      "mean_token_accuracy": 0.16409722119569778,
      "num_tokens": 8331118.0,
      "step": 4800
    },
    {
      "entropy": 5.6461936950683596,
      "epoch": 0.37385722622057965,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004990126780339953,
      "loss": 5.5746,
      "mean_token_accuracy": 0.16201503276824952,
      "num_tokens": 8340498.0,
      "step": 4805
    },
    {
      "entropy": 5.795285844802857,
      "epoch": 0.37424625559229724,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004990100827507459,
      "loss": 5.6856,
      "mean_token_accuracy": 0.15614394545555116,
      "num_tokens": 8349235.0,
      "step": 4810
    },
    {
      "entropy": 5.668808126449585,
      "epoch": 0.37463528496401477,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004990074840684999,
      "loss": 5.5826,
      "mean_token_accuracy": 0.16042376458644866,
      "num_tokens": 8358484.0,
      "step": 4815
    },
    {
      "entropy": 5.689554738998413,
      "epoch": 0.37502431433573236,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004990048819872969,
      "loss": 5.4939,
      "mean_token_accuracy": 0.1701010376214981,
      "num_tokens": 8366894.0,
      "step": 4820
    },
    {
      "entropy": 5.618418550491333,
      "epoch": 0.3754133437074499,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004990022765071765,
      "loss": 5.5087,
      "mean_token_accuracy": 0.15820858031511306,
      "num_tokens": 8375271.0,
      "step": 4825
    },
    {
      "entropy": 5.691072702407837,
      "epoch": 0.3758023730791675,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000498999667628178,
      "loss": 5.452,
      "mean_token_accuracy": 0.1625892162322998,
      "num_tokens": 8383714.0,
      "step": 4830
    },
    {
      "entropy": 5.630849313735962,
      "epoch": 0.37619140245088506,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004989970553503411,
      "loss": 5.552,
      "mean_token_accuracy": 0.15851999670267106,
      "num_tokens": 8392022.0,
      "step": 4835
    },
    {
      "entropy": 5.704822587966919,
      "epoch": 0.3765804318226026,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004989944396737054,
      "loss": 5.5144,
      "mean_token_accuracy": 0.1620304763317108,
      "num_tokens": 8401282.0,
      "step": 4840
    },
    {
      "entropy": 5.679300594329834,
      "epoch": 0.3769694611943202,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004989918205983106,
      "loss": 5.5913,
      "mean_token_accuracy": 0.16058251559734343,
      "num_tokens": 8411147.0,
      "step": 4845
    },
    {
      "entropy": 5.80027208328247,
      "epoch": 0.37735849056603776,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004989891981241964,
      "loss": 5.5855,
      "mean_token_accuracy": 0.16146292835474013,
      "num_tokens": 8420434.0,
      "step": 4850
    },
    {
      "entropy": 5.73548994064331,
      "epoch": 0.3777475199377553,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004989865722514027,
      "loss": 5.5538,
      "mean_token_accuracy": 0.16584563851356507,
      "num_tokens": 8430022.0,
      "step": 4855
    },
    {
      "entropy": 5.528020763397217,
      "epoch": 0.3781365493094729,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004989839429799692,
      "loss": 5.3401,
      "mean_token_accuracy": 0.17489640563726425,
      "num_tokens": 8438898.0,
      "step": 4860
    },
    {
      "entropy": 5.590051126480103,
      "epoch": 0.3785255786811904,
      "grad_norm": 1.078125,
      "learning_rate": 0.000498981310309936,
      "loss": 5.4349,
      "mean_token_accuracy": 0.17143376916646957,
      "num_tokens": 8446931.0,
      "step": 4865
    },
    {
      "entropy": 5.696088552474976,
      "epoch": 0.378914608052908,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004989786742413428,
      "loss": 5.5092,
      "mean_token_accuracy": 0.16275201737880707,
      "num_tokens": 8455515.0,
      "step": 4870
    },
    {
      "entropy": 5.677388048171997,
      "epoch": 0.3793036374246256,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004989760347742298,
      "loss": 5.5269,
      "mean_token_accuracy": 0.1604243353009224,
      "num_tokens": 8464135.0,
      "step": 4875
    },
    {
      "entropy": 5.745461559295654,
      "epoch": 0.3796926667963431,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004989733919086369,
      "loss": 5.6083,
      "mean_token_accuracy": 0.16428772658109664,
      "num_tokens": 8473222.0,
      "step": 4880
    },
    {
      "entropy": 5.6897931575775145,
      "epoch": 0.3800816961680607,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004989707456446043,
      "loss": 5.5862,
      "mean_token_accuracy": 0.169316303730011,
      "num_tokens": 8481966.0,
      "step": 4885
    },
    {
      "entropy": 5.693439865112305,
      "epoch": 0.38047072553977823,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004989680959821721,
      "loss": 5.5924,
      "mean_token_accuracy": 0.15563373863697053,
      "num_tokens": 8491276.0,
      "step": 4890
    },
    {
      "entropy": 5.698636674880982,
      "epoch": 0.3808597549114958,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004989654429213805,
      "loss": 5.5523,
      "mean_token_accuracy": 0.1644874855875969,
      "num_tokens": 8499926.0,
      "step": 4895
    },
    {
      "entropy": 5.647000408172607,
      "epoch": 0.3812487842832134,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004989627864622699,
      "loss": 5.4189,
      "mean_token_accuracy": 0.16593925654888153,
      "num_tokens": 8508544.0,
      "step": 4900
    },
    {
      "entropy": 5.704534101486206,
      "epoch": 0.38163781365493094,
      "grad_norm": 1.125,
      "learning_rate": 0.0004989601266048804,
      "loss": 5.5713,
      "mean_token_accuracy": 0.1553783245384693,
      "num_tokens": 8517326.0,
      "step": 4905
    },
    {
      "entropy": 5.6575311660766605,
      "epoch": 0.3820268430266485,
      "grad_norm": 1.0,
      "learning_rate": 0.0004989574633492525,
      "loss": 5.5284,
      "mean_token_accuracy": 0.15839311182498933,
      "num_tokens": 8526218.0,
      "step": 4910
    },
    {
      "entropy": 5.743776273727417,
      "epoch": 0.38241587239836605,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004989547966954266,
      "loss": 5.5686,
      "mean_token_accuracy": 0.1592984326183796,
      "num_tokens": 8534708.0,
      "step": 4915
    },
    {
      "entropy": 5.61528787612915,
      "epoch": 0.38280490177008364,
      "grad_norm": 0.953125,
      "learning_rate": 0.000498952126643443,
      "loss": 5.5106,
      "mean_token_accuracy": 0.16692230254411697,
      "num_tokens": 8543302.0,
      "step": 4920
    },
    {
      "entropy": 5.602381849288941,
      "epoch": 0.3831939311418012,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004989494531933424,
      "loss": 5.4373,
      "mean_token_accuracy": 0.16146207898855208,
      "num_tokens": 8552586.0,
      "step": 4925
    },
    {
      "entropy": 5.617420196533203,
      "epoch": 0.38358296051351876,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004989467763451652,
      "loss": 5.3988,
      "mean_token_accuracy": 0.1710638538002968,
      "num_tokens": 8561720.0,
      "step": 4930
    },
    {
      "entropy": 5.663424062728882,
      "epoch": 0.38397198988523634,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004989440960989523,
      "loss": 5.4109,
      "mean_token_accuracy": 0.16861876845359802,
      "num_tokens": 8570510.0,
      "step": 4935
    },
    {
      "entropy": 5.665288877487183,
      "epoch": 0.3843610192569539,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000498941412454744,
      "loss": 5.538,
      "mean_token_accuracy": 0.16042274087667466,
      "num_tokens": 8579986.0,
      "step": 4940
    },
    {
      "entropy": 5.662878370285034,
      "epoch": 0.38475004862867146,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004989387254125813,
      "loss": 5.5009,
      "mean_token_accuracy": 0.1592758759856224,
      "num_tokens": 8588507.0,
      "step": 4945
    },
    {
      "entropy": 5.637141275405884,
      "epoch": 0.38513907800038905,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004989360349725049,
      "loss": 5.4032,
      "mean_token_accuracy": 0.16547817438840867,
      "num_tokens": 8596651.0,
      "step": 4950
    },
    {
      "entropy": 5.6403545379638675,
      "epoch": 0.3855281073721066,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004989333411345555,
      "loss": 5.4617,
      "mean_token_accuracy": 0.1652302175760269,
      "num_tokens": 8605080.0,
      "step": 4955
    },
    {
      "entropy": 5.5624915599823,
      "epoch": 0.38591713674382416,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004989306438987742,
      "loss": 5.4226,
      "mean_token_accuracy": 0.16806283146142958,
      "num_tokens": 8613017.0,
      "step": 4960
    },
    {
      "entropy": 5.68736834526062,
      "epoch": 0.38630616611554175,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004989279432652018,
      "loss": 5.5976,
      "mean_token_accuracy": 0.1638832852244377,
      "num_tokens": 8621954.0,
      "step": 4965
    },
    {
      "entropy": 5.71930775642395,
      "epoch": 0.3866951954872593,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004989252392338791,
      "loss": 5.5433,
      "mean_token_accuracy": 0.1683834195137024,
      "num_tokens": 8630213.0,
      "step": 4970
    },
    {
      "entropy": 5.727107429504395,
      "epoch": 0.38708422485897687,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004989225318048475,
      "loss": 5.5426,
      "mean_token_accuracy": 0.16103035658597947,
      "num_tokens": 8638360.0,
      "step": 4975
    },
    {
      "entropy": 5.720944547653199,
      "epoch": 0.3874732542306944,
      "grad_norm": 1.0,
      "learning_rate": 0.0004989198209781478,
      "loss": 5.6003,
      "mean_token_accuracy": 0.16013864651322365,
      "num_tokens": 8647778.0,
      "step": 4980
    },
    {
      "entropy": 5.692323637008667,
      "epoch": 0.387862283602412,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004989171067538211,
      "loss": 5.524,
      "mean_token_accuracy": 0.16101902723312378,
      "num_tokens": 8655584.0,
      "step": 4985
    },
    {
      "entropy": 5.648969650268555,
      "epoch": 0.38825131297412957,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000498914389131909,
      "loss": 5.5557,
      "mean_token_accuracy": 0.1625164344906807,
      "num_tokens": 8664159.0,
      "step": 4990
    },
    {
      "entropy": 5.676162624359131,
      "epoch": 0.3886403423458471,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004989116681124523,
      "loss": 5.4414,
      "mean_token_accuracy": 0.16503534018993377,
      "num_tokens": 8672488.0,
      "step": 4995
    },
    {
      "entropy": 5.730154418945313,
      "epoch": 0.3890293717175647,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004989089436954924,
      "loss": 5.5977,
      "mean_token_accuracy": 0.1612487852573395,
      "num_tokens": 8681492.0,
      "step": 5000
    },
    {
      "entropy": 5.601837539672852,
      "epoch": 0.3894184010892822,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004989062158810706,
      "loss": 5.4851,
      "mean_token_accuracy": 0.16473744660615922,
      "num_tokens": 8690453.0,
      "step": 5005
    },
    {
      "entropy": 5.719622421264648,
      "epoch": 0.3898074304609998,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004989034846692284,
      "loss": 5.5561,
      "mean_token_accuracy": 0.1666879713535309,
      "num_tokens": 8699349.0,
      "step": 5010
    },
    {
      "entropy": 5.679425525665283,
      "epoch": 0.3901964598327174,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004989007500600073,
      "loss": 5.5617,
      "mean_token_accuracy": 0.16023534089326857,
      "num_tokens": 8707512.0,
      "step": 5015
    },
    {
      "entropy": 5.760986375808716,
      "epoch": 0.3905854892044349,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004988980120534486,
      "loss": 5.5866,
      "mean_token_accuracy": 0.16257491409778596,
      "num_tokens": 8715847.0,
      "step": 5020
    },
    {
      "entropy": 5.6762927055358885,
      "epoch": 0.3909745185761525,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000498895270649594,
      "loss": 5.4819,
      "mean_token_accuracy": 0.1681723803281784,
      "num_tokens": 8725548.0,
      "step": 5025
    },
    {
      "entropy": 5.57688341140747,
      "epoch": 0.39136354794787004,
      "grad_norm": 1.046875,
      "learning_rate": 0.000498892525848485,
      "loss": 5.4595,
      "mean_token_accuracy": 0.16113873124122619,
      "num_tokens": 8733935.0,
      "step": 5030
    },
    {
      "entropy": 5.660002040863037,
      "epoch": 0.3917525773195876,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004988897776501633,
      "loss": 5.5672,
      "mean_token_accuracy": 0.16277794241905214,
      "num_tokens": 8743110.0,
      "step": 5035
    },
    {
      "entropy": 5.650880193710327,
      "epoch": 0.3921416066913052,
      "grad_norm": 1.0,
      "learning_rate": 0.0004988870260546705,
      "loss": 5.377,
      "mean_token_accuracy": 0.16757228821516038,
      "num_tokens": 8751166.0,
      "step": 5040
    },
    {
      "entropy": 5.676319599151611,
      "epoch": 0.39253063606302274,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004988842710620486,
      "loss": 5.5755,
      "mean_token_accuracy": 0.16439453810453414,
      "num_tokens": 8759929.0,
      "step": 5045
    },
    {
      "entropy": 5.710107803344727,
      "epoch": 0.39291966543474033,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004988815126723391,
      "loss": 5.5927,
      "mean_token_accuracy": 0.15978324115276338,
      "num_tokens": 8768541.0,
      "step": 5050
    },
    {
      "entropy": 5.739083576202392,
      "epoch": 0.3933086948064579,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004988787508855841,
      "loss": 5.4913,
      "mean_token_accuracy": 0.161235274374485,
      "num_tokens": 8776803.0,
      "step": 5055
    },
    {
      "entropy": 5.619376373291016,
      "epoch": 0.39369772417817545,
      "grad_norm": 1.125,
      "learning_rate": 0.0004988759857018253,
      "loss": 5.4953,
      "mean_token_accuracy": 0.16697628945112228,
      "num_tokens": 8784773.0,
      "step": 5060
    },
    {
      "entropy": 5.731350326538086,
      "epoch": 0.39408675354989303,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004988732171211048,
      "loss": 5.5227,
      "mean_token_accuracy": 0.16417534202337264,
      "num_tokens": 8793797.0,
      "step": 5065
    },
    {
      "entropy": 5.592066097259521,
      "epoch": 0.39447578292161056,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004988704451434644,
      "loss": 5.4019,
      "mean_token_accuracy": 0.16893110424280167,
      "num_tokens": 8802366.0,
      "step": 5070
    },
    {
      "entropy": 5.660155868530273,
      "epoch": 0.39486481229332815,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004988676697689465,
      "loss": 5.5391,
      "mean_token_accuracy": 0.16272079348564147,
      "num_tokens": 8810975.0,
      "step": 5075
    },
    {
      "entropy": 5.680756092071533,
      "epoch": 0.39525384166504574,
      "grad_norm": 1.046875,
      "learning_rate": 0.000498864890997593,
      "loss": 5.5829,
      "mean_token_accuracy": 0.15348358899354936,
      "num_tokens": 8819137.0,
      "step": 5080
    },
    {
      "entropy": 5.720478916168213,
      "epoch": 0.39564287103676327,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004988621088294461,
      "loss": 5.567,
      "mean_token_accuracy": 0.15802430659532546,
      "num_tokens": 8827676.0,
      "step": 5085
    },
    {
      "entropy": 5.639574909210205,
      "epoch": 0.39603190040848085,
      "grad_norm": 1.015625,
      "learning_rate": 0.000498859323264548,
      "loss": 5.5193,
      "mean_token_accuracy": 0.16178436428308487,
      "num_tokens": 8836005.0,
      "step": 5090
    },
    {
      "entropy": 5.609964084625244,
      "epoch": 0.3964209297801984,
      "grad_norm": 1.015625,
      "learning_rate": 0.000498856534302941,
      "loss": 5.4125,
      "mean_token_accuracy": 0.1722085103392601,
      "num_tokens": 8844153.0,
      "step": 5095
    },
    {
      "entropy": 5.700956249237061,
      "epoch": 0.39680995915191597,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004988537419446673,
      "loss": 5.517,
      "mean_token_accuracy": 0.16102592051029205,
      "num_tokens": 8852520.0,
      "step": 5100
    },
    {
      "entropy": 5.665908241271973,
      "epoch": 0.39719898852363356,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004988509461897694,
      "loss": 5.364,
      "mean_token_accuracy": 0.17337042838335037,
      "num_tokens": 8860609.0,
      "step": 5105
    },
    {
      "entropy": 5.617171764373779,
      "epoch": 0.3975880178953511,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004988481470382897,
      "loss": 5.5373,
      "mean_token_accuracy": 0.16442073434591292,
      "num_tokens": 8869185.0,
      "step": 5110
    },
    {
      "entropy": 5.66382327079773,
      "epoch": 0.3979770472670687,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004988453444902708,
      "loss": 5.4023,
      "mean_token_accuracy": 0.16232824474573135,
      "num_tokens": 8877758.0,
      "step": 5115
    },
    {
      "entropy": 5.666703510284424,
      "epoch": 0.3983660766387862,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004988425385457549,
      "loss": 5.5814,
      "mean_token_accuracy": 0.16767374575138091,
      "num_tokens": 8887236.0,
      "step": 5120
    },
    {
      "entropy": 5.727814960479736,
      "epoch": 0.3987551060105038,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004988397292047848,
      "loss": 5.5315,
      "mean_token_accuracy": 0.16686426103115082,
      "num_tokens": 8895808.0,
      "step": 5125
    },
    {
      "entropy": 5.680415868759155,
      "epoch": 0.3991441353822214,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004988369164674032,
      "loss": 5.5185,
      "mean_token_accuracy": 0.16700370758771896,
      "num_tokens": 8904091.0,
      "step": 5130
    },
    {
      "entropy": 5.540524578094482,
      "epoch": 0.3995331647539389,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004988341003336526,
      "loss": 5.3377,
      "mean_token_accuracy": 0.16943544745445252,
      "num_tokens": 8912046.0,
      "step": 5135
    },
    {
      "entropy": 5.691886138916016,
      "epoch": 0.3999221941256565,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004988312808035757,
      "loss": 5.5576,
      "mean_token_accuracy": 0.1542545847594738,
      "num_tokens": 8921060.0,
      "step": 5140
    },
    {
      "entropy": 5.678117561340332,
      "epoch": 0.400311223497374,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004988284578772155,
      "loss": 5.493,
      "mean_token_accuracy": 0.16438425034284593,
      "num_tokens": 8930757.0,
      "step": 5145
    },
    {
      "entropy": 5.776554250717163,
      "epoch": 0.4007002528690916,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004988256315546146,
      "loss": 5.6521,
      "mean_token_accuracy": 0.1632249116897583,
      "num_tokens": 8939544.0,
      "step": 5150
    },
    {
      "entropy": 5.60939998626709,
      "epoch": 0.4010892822408092,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004988228018358161,
      "loss": 5.4761,
      "mean_token_accuracy": 0.1680803343653679,
      "num_tokens": 8949347.0,
      "step": 5155
    },
    {
      "entropy": 5.6522157192230225,
      "epoch": 0.40147831161252673,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004988199687208628,
      "loss": 5.5029,
      "mean_token_accuracy": 0.16580344885587692,
      "num_tokens": 8957697.0,
      "step": 5160
    },
    {
      "entropy": 5.667740964889527,
      "epoch": 0.4018673409842443,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004988171322097977,
      "loss": 5.5452,
      "mean_token_accuracy": 0.15877031087875365,
      "num_tokens": 8966107.0,
      "step": 5165
    },
    {
      "entropy": 5.63241982460022,
      "epoch": 0.4022563703559619,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004988142923026638,
      "loss": 5.4586,
      "mean_token_accuracy": 0.16741299033164977,
      "num_tokens": 8974128.0,
      "step": 5170
    },
    {
      "entropy": 5.608824968338013,
      "epoch": 0.40264539972767943,
      "grad_norm": 1.0,
      "learning_rate": 0.0004988114489995044,
      "loss": 5.4587,
      "mean_token_accuracy": 0.16279011368751525,
      "num_tokens": 8982564.0,
      "step": 5175
    },
    {
      "entropy": 5.594621706008911,
      "epoch": 0.403034429099397,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004988086023003624,
      "loss": 5.4747,
      "mean_token_accuracy": 0.16702579110860824,
      "num_tokens": 8990924.0,
      "step": 5180
    },
    {
      "entropy": 5.7941367626190186,
      "epoch": 0.40342345847111455,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004988057522052811,
      "loss": 5.5863,
      "mean_token_accuracy": 0.15846182107925416,
      "num_tokens": 8999827.0,
      "step": 5185
    },
    {
      "entropy": 5.667945146560669,
      "epoch": 0.40381248784283214,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004988028987143037,
      "loss": 5.4127,
      "mean_token_accuracy": 0.16810715943574905,
      "num_tokens": 9007885.0,
      "step": 5190
    },
    {
      "entropy": 5.655215787887573,
      "epoch": 0.4042015172145497,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004988000418274736,
      "loss": 5.5149,
      "mean_token_accuracy": 0.16204234063625336,
      "num_tokens": 9016752.0,
      "step": 5195
    },
    {
      "entropy": 5.696285581588745,
      "epoch": 0.40459054658626725,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004987971815448341,
      "loss": 5.5101,
      "mean_token_accuracy": 0.16552756428718568,
      "num_tokens": 9025547.0,
      "step": 5200
    },
    {
      "entropy": 5.683474922180176,
      "epoch": 0.40497957595798484,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004987943178664285,
      "loss": 5.552,
      "mean_token_accuracy": 0.15770690143108368,
      "num_tokens": 9034711.0,
      "step": 5205
    },
    {
      "entropy": 5.697414779663086,
      "epoch": 0.40536860532970237,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004987914507923004,
      "loss": 5.4505,
      "mean_token_accuracy": 0.16524471193552018,
      "num_tokens": 9042565.0,
      "step": 5210
    },
    {
      "entropy": 5.592935276031494,
      "epoch": 0.40575763470141996,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004987885803224931,
      "loss": 5.5219,
      "mean_token_accuracy": 0.15521206110715866,
      "num_tokens": 9051590.0,
      "step": 5215
    },
    {
      "entropy": 5.702045011520386,
      "epoch": 0.40614666407313754,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004987857064570505,
      "loss": 5.6005,
      "mean_token_accuracy": 0.16120531857013704,
      "num_tokens": 9060415.0,
      "step": 5220
    },
    {
      "entropy": 5.619015216827393,
      "epoch": 0.4065356934448551,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004987828291960158,
      "loss": 5.3811,
      "mean_token_accuracy": 0.16975099742412567,
      "num_tokens": 9069492.0,
      "step": 5225
    },
    {
      "entropy": 5.593218278884888,
      "epoch": 0.40692472281657266,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000498779948539433,
      "loss": 5.443,
      "mean_token_accuracy": 0.16820174008607863,
      "num_tokens": 9078183.0,
      "step": 5230
    },
    {
      "entropy": 5.754888248443604,
      "epoch": 0.4073137521882902,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004987770644873455,
      "loss": 5.6682,
      "mean_token_accuracy": 0.15733300000429154,
      "num_tokens": 9086661.0,
      "step": 5235
    },
    {
      "entropy": 5.627206754684448,
      "epoch": 0.4077027815600078,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004987741770397974,
      "loss": 5.3865,
      "mean_token_accuracy": 0.16997277289628981,
      "num_tokens": 9095449.0,
      "step": 5240
    },
    {
      "entropy": 5.628784608840943,
      "epoch": 0.40809181093172536,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004987712861968323,
      "loss": 5.5283,
      "mean_token_accuracy": 0.15532390177249908,
      "num_tokens": 9103347.0,
      "step": 5245
    },
    {
      "entropy": 5.539442920684815,
      "epoch": 0.4084808403034429,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000498768391958494,
      "loss": 5.4372,
      "mean_token_accuracy": 0.1731258362531662,
      "num_tokens": 9112382.0,
      "step": 5250
    },
    {
      "entropy": 5.647328233718872,
      "epoch": 0.4088698696751605,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004987654943248266,
      "loss": 5.4053,
      "mean_token_accuracy": 0.16700365394353867,
      "num_tokens": 9120628.0,
      "step": 5255
    },
    {
      "entropy": 5.694366407394409,
      "epoch": 0.40925889904687807,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004987625932958739,
      "loss": 5.545,
      "mean_token_accuracy": 0.15544501841068267,
      "num_tokens": 9130280.0,
      "step": 5260
    },
    {
      "entropy": 5.819931268692017,
      "epoch": 0.4096479284185956,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004987596888716801,
      "loss": 5.6471,
      "mean_token_accuracy": 0.15586191713809966,
      "num_tokens": 9139720.0,
      "step": 5265
    },
    {
      "entropy": 5.632251214981079,
      "epoch": 0.4100369577903132,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004987567810522892,
      "loss": 5.5686,
      "mean_token_accuracy": 0.15921868681907653,
      "num_tokens": 9148537.0,
      "step": 5270
    },
    {
      "entropy": 5.702337598800659,
      "epoch": 0.4104259871620307,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004987538698377451,
      "loss": 5.6348,
      "mean_token_accuracy": 0.16866708248853685,
      "num_tokens": 9156965.0,
      "step": 5275
    },
    {
      "entropy": 5.637188529968261,
      "epoch": 0.4108150165337483,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004987509552280924,
      "loss": 5.3663,
      "mean_token_accuracy": 0.17250857055187224,
      "num_tokens": 9164982.0,
      "step": 5280
    },
    {
      "entropy": 5.670972013473511,
      "epoch": 0.4112040459054659,
      "grad_norm": 1.046875,
      "learning_rate": 0.000498748037223375,
      "loss": 5.5863,
      "mean_token_accuracy": 0.1595808021724224,
      "num_tokens": 9173383.0,
      "step": 5285
    },
    {
      "entropy": 5.714626264572144,
      "epoch": 0.4115930752771834,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004987451158236372,
      "loss": 5.5843,
      "mean_token_accuracy": 0.17200423628091813,
      "num_tokens": 9182743.0,
      "step": 5290
    },
    {
      "entropy": 5.722370147705078,
      "epoch": 0.411982104648901,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004987421910289234,
      "loss": 5.4729,
      "mean_token_accuracy": 0.16518608927726747,
      "num_tokens": 9191770.0,
      "step": 5295
    },
    {
      "entropy": 5.507548522949219,
      "epoch": 0.41237113402061853,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004987392628392781,
      "loss": 5.4645,
      "mean_token_accuracy": 0.17184146940708162,
      "num_tokens": 9200885.0,
      "step": 5300
    },
    {
      "entropy": 5.75942063331604,
      "epoch": 0.4127601633923361,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004987363312547456,
      "loss": 5.5127,
      "mean_token_accuracy": 0.16049581915140151,
      "num_tokens": 9210114.0,
      "step": 5305
    },
    {
      "entropy": 5.726402950286865,
      "epoch": 0.4131491927640537,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004987333962753703,
      "loss": 5.4866,
      "mean_token_accuracy": 0.16796217560768129,
      "num_tokens": 9218608.0,
      "step": 5310
    },
    {
      "entropy": 5.61348934173584,
      "epoch": 0.41353822213577124,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004987304579011967,
      "loss": 5.5281,
      "mean_token_accuracy": 0.16807485967874528,
      "num_tokens": 9227737.0,
      "step": 5315
    },
    {
      "entropy": 5.612955093383789,
      "epoch": 0.4139272515074888,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004987275161322697,
      "loss": 5.4455,
      "mean_token_accuracy": 0.16860099285840988,
      "num_tokens": 9236589.0,
      "step": 5320
    },
    {
      "entropy": 5.645352697372436,
      "epoch": 0.41431628087920636,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004987245709686337,
      "loss": 5.4712,
      "mean_token_accuracy": 0.1744957759976387,
      "num_tokens": 9245313.0,
      "step": 5325
    },
    {
      "entropy": 5.609607744216919,
      "epoch": 0.41470531025092394,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004987216224103334,
      "loss": 5.5259,
      "mean_token_accuracy": 0.15794259160757065,
      "num_tokens": 9254001.0,
      "step": 5330
    },
    {
      "entropy": 5.725889730453491,
      "epoch": 0.41509433962264153,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004987186704574136,
      "loss": 5.5794,
      "mean_token_accuracy": 0.15721883326768876,
      "num_tokens": 9262485.0,
      "step": 5335
    },
    {
      "entropy": 5.671980142593384,
      "epoch": 0.41548336899435906,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004987157151099189,
      "loss": 5.4654,
      "mean_token_accuracy": 0.16698906123638152,
      "num_tokens": 9271490.0,
      "step": 5340
    },
    {
      "entropy": 5.617227935791016,
      "epoch": 0.41587239836607665,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004987127563678945,
      "loss": 5.4886,
      "mean_token_accuracy": 0.16965751051902772,
      "num_tokens": 9280595.0,
      "step": 5345
    },
    {
      "entropy": 5.653589868545533,
      "epoch": 0.4162614277377942,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004987097942313852,
      "loss": 5.4076,
      "mean_token_accuracy": 0.172671402990818,
      "num_tokens": 9289260.0,
      "step": 5350
    },
    {
      "entropy": 5.703232002258301,
      "epoch": 0.41665045710951176,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004987068287004355,
      "loss": 5.5825,
      "mean_token_accuracy": 0.16232939064502716,
      "num_tokens": 9297842.0,
      "step": 5355
    },
    {
      "entropy": 5.719019365310669,
      "epoch": 0.41703948648122935,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004987038597750909,
      "loss": 5.5775,
      "mean_token_accuracy": 0.16258386820554732,
      "num_tokens": 9306533.0,
      "step": 5360
    },
    {
      "entropy": 5.695370769500732,
      "epoch": 0.4174285158529469,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004987008874553964,
      "loss": 5.4541,
      "mean_token_accuracy": 0.16366709172725677,
      "num_tokens": 9314534.0,
      "step": 5365
    },
    {
      "entropy": 5.738441705703735,
      "epoch": 0.41781754522466447,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004986979117413969,
      "loss": 5.6188,
      "mean_token_accuracy": 0.1668700933456421,
      "num_tokens": 9323179.0,
      "step": 5370
    },
    {
      "entropy": 5.623295402526855,
      "epoch": 0.41820657459638205,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004986949326331376,
      "loss": 5.4067,
      "mean_token_accuracy": 0.1757480889558792,
      "num_tokens": 9331265.0,
      "step": 5375
    },
    {
      "entropy": 5.637596559524536,
      "epoch": 0.4185956039680996,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004986919501306638,
      "loss": 5.4783,
      "mean_token_accuracy": 0.15887642353773118,
      "num_tokens": 9340856.0,
      "step": 5380
    },
    {
      "entropy": 5.651981019973755,
      "epoch": 0.41898463333981717,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004986889642340207,
      "loss": 5.4972,
      "mean_token_accuracy": 0.15731943473219873,
      "num_tokens": 9350728.0,
      "step": 5385
    },
    {
      "entropy": 5.718332433700562,
      "epoch": 0.4193736627115347,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004986859749432536,
      "loss": 5.6181,
      "mean_token_accuracy": 0.15903780460357667,
      "num_tokens": 9360109.0,
      "step": 5390
    },
    {
      "entropy": 5.748874664306641,
      "epoch": 0.4197626920832523,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004986829822584078,
      "loss": 5.585,
      "mean_token_accuracy": 0.1624787151813507,
      "num_tokens": 9369101.0,
      "step": 5395
    },
    {
      "entropy": 5.699647426605225,
      "epoch": 0.4201517214549699,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004986799861795289,
      "loss": 5.5483,
      "mean_token_accuracy": 0.16545858085155488,
      "num_tokens": 9377686.0,
      "step": 5400
    },
    {
      "entropy": 5.719456434249878,
      "epoch": 0.4205407508266874,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004986769867066622,
      "loss": 5.5296,
      "mean_token_accuracy": 0.16334747076034545,
      "num_tokens": 9386439.0,
      "step": 5405
    },
    {
      "entropy": 5.654419565200806,
      "epoch": 0.420929780198405,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004986739838398532,
      "loss": 5.5743,
      "mean_token_accuracy": 0.15611925572156907,
      "num_tokens": 9395086.0,
      "step": 5410
    },
    {
      "entropy": 5.651913070678711,
      "epoch": 0.4213188095701225,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004986709775791475,
      "loss": 5.4338,
      "mean_token_accuracy": 0.16861581802368164,
      "num_tokens": 9403762.0,
      "step": 5415
    },
    {
      "entropy": 5.638100814819336,
      "epoch": 0.4217078389418401,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004986679679245907,
      "loss": 5.4745,
      "mean_token_accuracy": 0.16845559030771257,
      "num_tokens": 9412386.0,
      "step": 5420
    },
    {
      "entropy": 5.654151153564453,
      "epoch": 0.4220968683135577,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004986649548762286,
      "loss": 5.4395,
      "mean_token_accuracy": 0.1766769051551819,
      "num_tokens": 9420538.0,
      "step": 5425
    },
    {
      "entropy": 5.578556871414184,
      "epoch": 0.4224858976852752,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004986619384341066,
      "loss": 5.4181,
      "mean_token_accuracy": 0.16903934627771378,
      "num_tokens": 9429083.0,
      "step": 5430
    },
    {
      "entropy": 5.645332145690918,
      "epoch": 0.4228749270569928,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004986589185982708,
      "loss": 5.5277,
      "mean_token_accuracy": 0.16904726922512053,
      "num_tokens": 9436892.0,
      "step": 5435
    },
    {
      "entropy": 5.62878098487854,
      "epoch": 0.42326395642871034,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004986558953687671,
      "loss": 5.5471,
      "mean_token_accuracy": 0.1671071842312813,
      "num_tokens": 9445267.0,
      "step": 5440
    },
    {
      "entropy": 5.692759227752686,
      "epoch": 0.42365298580042793,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004986528687456409,
      "loss": 5.5473,
      "mean_token_accuracy": 0.1619514286518097,
      "num_tokens": 9454292.0,
      "step": 5445
    },
    {
      "entropy": 5.692660617828369,
      "epoch": 0.4240420151721455,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004986498387289384,
      "loss": 5.434,
      "mean_token_accuracy": 0.16928142309188843,
      "num_tokens": 9463126.0,
      "step": 5450
    },
    {
      "entropy": 5.680186367034912,
      "epoch": 0.42443104454386305,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004986468053187058,
      "loss": 5.526,
      "mean_token_accuracy": 0.16752272248268127,
      "num_tokens": 9471703.0,
      "step": 5455
    },
    {
      "entropy": 5.583491611480713,
      "epoch": 0.42482007391558063,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004986437685149887,
      "loss": 5.4078,
      "mean_token_accuracy": 0.17276425659656525,
      "num_tokens": 9480029.0,
      "step": 5460
    },
    {
      "entropy": 5.669733619689941,
      "epoch": 0.4252091032872982,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004986407283178334,
      "loss": 5.5275,
      "mean_token_accuracy": 0.1660941243171692,
      "num_tokens": 9488277.0,
      "step": 5465
    },
    {
      "entropy": 5.674656867980957,
      "epoch": 0.42559813265901575,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004986376847272861,
      "loss": 5.5736,
      "mean_token_accuracy": 0.16575695127248763,
      "num_tokens": 9496096.0,
      "step": 5470
    },
    {
      "entropy": 5.657446384429932,
      "epoch": 0.42598716203073334,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004986346377433929,
      "loss": 5.4813,
      "mean_token_accuracy": 0.16967480331659318,
      "num_tokens": 9504653.0,
      "step": 5475
    },
    {
      "entropy": 5.6826355934143065,
      "epoch": 0.42637619140245087,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004986315873661999,
      "loss": 5.4467,
      "mean_token_accuracy": 0.16702346205711366,
      "num_tokens": 9513088.0,
      "step": 5480
    },
    {
      "entropy": 5.585157108306885,
      "epoch": 0.42676522077416845,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004986285335957536,
      "loss": 5.3617,
      "mean_token_accuracy": 0.17300552427768706,
      "num_tokens": 9521714.0,
      "step": 5485
    },
    {
      "entropy": 5.5741489887237545,
      "epoch": 0.42715425014588604,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004986254764321002,
      "loss": 5.467,
      "mean_token_accuracy": 0.17281568050384521,
      "num_tokens": 9530629.0,
      "step": 5490
    },
    {
      "entropy": 5.612785673141479,
      "epoch": 0.42754327951760357,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004986224158752861,
      "loss": 5.4437,
      "mean_token_accuracy": 0.17414536476135253,
      "num_tokens": 9539293.0,
      "step": 5495
    },
    {
      "entropy": 5.575432538986206,
      "epoch": 0.42793230888932116,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004986193519253578,
      "loss": 5.4067,
      "mean_token_accuracy": 0.16896552741527557,
      "num_tokens": 9547628.0,
      "step": 5500
    },
    {
      "entropy": 5.604017972946167,
      "epoch": 0.4283213382610387,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004986162845823618,
      "loss": 5.4267,
      "mean_token_accuracy": 0.17168065458536147,
      "num_tokens": 9557260.0,
      "step": 5505
    },
    {
      "entropy": 5.7184590816497805,
      "epoch": 0.4287103676327563,
      "grad_norm": 1.0,
      "learning_rate": 0.0004986132138463446,
      "loss": 5.4712,
      "mean_token_accuracy": 0.15693661868572234,
      "num_tokens": 9565923.0,
      "step": 5510
    },
    {
      "entropy": 5.600460720062256,
      "epoch": 0.42909939700447386,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004986101397173528,
      "loss": 5.4645,
      "mean_token_accuracy": 0.16751811951398848,
      "num_tokens": 9574373.0,
      "step": 5515
    },
    {
      "entropy": 5.683349561691284,
      "epoch": 0.4294884263761914,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000498607062195433,
      "loss": 5.5504,
      "mean_token_accuracy": 0.1631235733628273,
      "num_tokens": 9583652.0,
      "step": 5520
    },
    {
      "entropy": 5.680144786834717,
      "epoch": 0.429877455747909,
      "grad_norm": 0.984375,
      "learning_rate": 0.000498603981280632,
      "loss": 5.5133,
      "mean_token_accuracy": 0.16775375306606294,
      "num_tokens": 9592625.0,
      "step": 5525
    },
    {
      "entropy": 5.579809713363647,
      "epoch": 0.4302664851196265,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004986008969729964,
      "loss": 5.4202,
      "mean_token_accuracy": 0.16815870851278306,
      "num_tokens": 9600934.0,
      "step": 5530
    },
    {
      "entropy": 5.538936901092529,
      "epoch": 0.4306555144913441,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004985978092725731,
      "loss": 5.4112,
      "mean_token_accuracy": 0.16719519346952438,
      "num_tokens": 9609481.0,
      "step": 5535
    },
    {
      "entropy": 5.59554648399353,
      "epoch": 0.4310445438630617,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000498594718179409,
      "loss": 5.3769,
      "mean_token_accuracy": 0.17795415222644806,
      "num_tokens": 9618063.0,
      "step": 5540
    },
    {
      "entropy": 5.682264947891236,
      "epoch": 0.4314335732347792,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004985916236935508,
      "loss": 5.4672,
      "mean_token_accuracy": 0.16837858706712722,
      "num_tokens": 9626144.0,
      "step": 5545
    },
    {
      "entropy": 5.451812887191773,
      "epoch": 0.4318226026064968,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004985885258150458,
      "loss": 5.4215,
      "mean_token_accuracy": 0.167401722073555,
      "num_tokens": 9635328.0,
      "step": 5550
    },
    {
      "entropy": 5.6543515682220455,
      "epoch": 0.43221163197821433,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004985854245439408,
      "loss": 5.5584,
      "mean_token_accuracy": 0.15709087401628494,
      "num_tokens": 9644529.0,
      "step": 5555
    },
    {
      "entropy": 5.713449096679687,
      "epoch": 0.4326006613499319,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004985823198802827,
      "loss": 5.5353,
      "mean_token_accuracy": 0.16077228337526323,
      "num_tokens": 9653763.0,
      "step": 5560
    },
    {
      "entropy": 5.6186117172241214,
      "epoch": 0.4329896907216495,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004985792118241188,
      "loss": 5.3736,
      "mean_token_accuracy": 0.171256360411644,
      "num_tokens": 9661433.0,
      "step": 5565
    },
    {
      "entropy": 5.568835210800171,
      "epoch": 0.43337872009336703,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004985761003754961,
      "loss": 5.3526,
      "mean_token_accuracy": 0.17259980142116546,
      "num_tokens": 9669688.0,
      "step": 5570
    },
    {
      "entropy": 5.657050752639771,
      "epoch": 0.4337677494650846,
      "grad_norm": 1.0,
      "learning_rate": 0.0004985729855344622,
      "loss": 5.474,
      "mean_token_accuracy": 0.1650309681892395,
      "num_tokens": 9678403.0,
      "step": 5575
    },
    {
      "entropy": 5.683776950836181,
      "epoch": 0.4341567788368022,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000498569867301064,
      "loss": 5.5725,
      "mean_token_accuracy": 0.15945140421390533,
      "num_tokens": 9687504.0,
      "step": 5580
    },
    {
      "entropy": 5.635408067703247,
      "epoch": 0.43454580820851973,
      "grad_norm": 1.0,
      "learning_rate": 0.0004985667456753489,
      "loss": 5.3825,
      "mean_token_accuracy": 0.17245262265205383,
      "num_tokens": 9695772.0,
      "step": 5585
    },
    {
      "entropy": 5.624304962158203,
      "epoch": 0.4349348375802373,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004985636206573642,
      "loss": 5.4634,
      "mean_token_accuracy": 0.16656662225723268,
      "num_tokens": 9704150.0,
      "step": 5590
    },
    {
      "entropy": 5.524925708770752,
      "epoch": 0.43532386695195485,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004985604922471575,
      "loss": 5.3729,
      "mean_token_accuracy": 0.17111615985631942,
      "num_tokens": 9712245.0,
      "step": 5595
    },
    {
      "entropy": 5.619245958328247,
      "epoch": 0.43571289632367244,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004985573604447761,
      "loss": 5.5051,
      "mean_token_accuracy": 0.16338210701942443,
      "num_tokens": 9721027.0,
      "step": 5600
    },
    {
      "entropy": 5.671468448638916,
      "epoch": 0.43610192569539,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004985542252502676,
      "loss": 5.4159,
      "mean_token_accuracy": 0.1692680224776268,
      "num_tokens": 9728939.0,
      "step": 5605
    },
    {
      "entropy": 5.716129922866822,
      "epoch": 0.43649095506710756,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004985510866636796,
      "loss": 5.5658,
      "mean_token_accuracy": 0.16524840891361237,
      "num_tokens": 9737863.0,
      "step": 5610
    },
    {
      "entropy": 5.680402708053589,
      "epoch": 0.43687998443882514,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004985479446850596,
      "loss": 5.5584,
      "mean_token_accuracy": 0.1601756900548935,
      "num_tokens": 9747666.0,
      "step": 5615
    },
    {
      "entropy": 5.643349838256836,
      "epoch": 0.4372690138105427,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004985447993144554,
      "loss": 5.4342,
      "mean_token_accuracy": 0.1627214342355728,
      "num_tokens": 9756595.0,
      "step": 5620
    },
    {
      "entropy": 5.62148151397705,
      "epoch": 0.43765804318226026,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004985416505519147,
      "loss": 5.4696,
      "mean_token_accuracy": 0.16752104312181473,
      "num_tokens": 9764747.0,
      "step": 5625
    },
    {
      "entropy": 5.599878311157227,
      "epoch": 0.43804707255397785,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004985384983974853,
      "loss": 5.4112,
      "mean_token_accuracy": 0.16520307958126068,
      "num_tokens": 9772992.0,
      "step": 5630
    },
    {
      "entropy": 5.655435037612915,
      "epoch": 0.4384361019256954,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004985353428512148,
      "loss": 5.3982,
      "mean_token_accuracy": 0.169257228076458,
      "num_tokens": 9781107.0,
      "step": 5635
    },
    {
      "entropy": 5.58502049446106,
      "epoch": 0.43882513129741296,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004985321839131514,
      "loss": 5.4798,
      "mean_token_accuracy": 0.16393501609563826,
      "num_tokens": 9790618.0,
      "step": 5640
    },
    {
      "entropy": 5.657913780212402,
      "epoch": 0.4392141606691305,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004985290215833428,
      "loss": 5.4852,
      "mean_token_accuracy": 0.16647476851940154,
      "num_tokens": 9800147.0,
      "step": 5645
    },
    {
      "entropy": 5.7092164039611815,
      "epoch": 0.4396031900408481,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004985258558618372,
      "loss": 5.5373,
      "mean_token_accuracy": 0.16442038416862487,
      "num_tokens": 9809056.0,
      "step": 5650
    },
    {
      "entropy": 5.572237920761109,
      "epoch": 0.43999221941256567,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004985226867486825,
      "loss": 5.4018,
      "mean_token_accuracy": 0.16231081634759903,
      "num_tokens": 9817188.0,
      "step": 5655
    },
    {
      "entropy": 5.643007230758667,
      "epoch": 0.4403812487842832,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004985195142439267,
      "loss": 5.5829,
      "mean_token_accuracy": 0.16034058555960656,
      "num_tokens": 9826904.0,
      "step": 5660
    },
    {
      "entropy": 5.700199747085572,
      "epoch": 0.4407702781560008,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004985163383476181,
      "loss": 5.5507,
      "mean_token_accuracy": 0.16313511431217192,
      "num_tokens": 9835937.0,
      "step": 5665
    },
    {
      "entropy": 5.65016770362854,
      "epoch": 0.44115930752771837,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004985131590598048,
      "loss": 5.3874,
      "mean_token_accuracy": 0.17442827373743058,
      "num_tokens": 9843695.0,
      "step": 5670
    },
    {
      "entropy": 5.626602268218994,
      "epoch": 0.4415483368994359,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004985099763805352,
      "loss": 5.4665,
      "mean_token_accuracy": 0.16676409542560577,
      "num_tokens": 9852693.0,
      "step": 5675
    },
    {
      "entropy": 5.594214963912964,
      "epoch": 0.4419373662711535,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004985067903098574,
      "loss": 5.4733,
      "mean_token_accuracy": 0.1682109758257866,
      "num_tokens": 9860251.0,
      "step": 5680
    },
    {
      "entropy": 5.5587108612060545,
      "epoch": 0.442326395642871,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004985036008478197,
      "loss": 5.391,
      "mean_token_accuracy": 0.16241925805807114,
      "num_tokens": 9868319.0,
      "step": 5685
    },
    {
      "entropy": 5.572983074188232,
      "epoch": 0.4427154250145886,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004985004079944707,
      "loss": 5.4393,
      "mean_token_accuracy": 0.16255133897066115,
      "num_tokens": 9877119.0,
      "step": 5690
    },
    {
      "entropy": 5.727364730834961,
      "epoch": 0.4431044543863062,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004984972117498587,
      "loss": 5.5136,
      "mean_token_accuracy": 0.16677001416683196,
      "num_tokens": 9886015.0,
      "step": 5695
    },
    {
      "entropy": 5.618843507766724,
      "epoch": 0.4434934837580237,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004984940121140323,
      "loss": 5.4561,
      "mean_token_accuracy": 0.1649298220872879,
      "num_tokens": 9894829.0,
      "step": 5700
    },
    {
      "entropy": 5.781460523605347,
      "epoch": 0.4438825131297413,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00049849080908704,
      "loss": 5.7107,
      "mean_token_accuracy": 0.15197371542453766,
      "num_tokens": 9904840.0,
      "step": 5705
    },
    {
      "entropy": 5.779392242431641,
      "epoch": 0.44427154250145884,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004984876026689302,
      "loss": 5.5529,
      "mean_token_accuracy": 0.1596390500664711,
      "num_tokens": 9913535.0,
      "step": 5710
    },
    {
      "entropy": 5.697728252410888,
      "epoch": 0.4446605718731764,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000498484392859752,
      "loss": 5.456,
      "mean_token_accuracy": 0.16690233200788498,
      "num_tokens": 9921676.0,
      "step": 5715
    },
    {
      "entropy": 5.586839818954468,
      "epoch": 0.445049601244894,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004984811796595538,
      "loss": 5.5227,
      "mean_token_accuracy": 0.1633313000202179,
      "num_tokens": 9930321.0,
      "step": 5720
    },
    {
      "entropy": 5.598290538787841,
      "epoch": 0.44543863061661154,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004984779630683843,
      "loss": 5.4535,
      "mean_token_accuracy": 0.16856093257665633,
      "num_tokens": 9939324.0,
      "step": 5725
    },
    {
      "entropy": 5.642680406570435,
      "epoch": 0.44582765998832913,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004984747430862924,
      "loss": 5.474,
      "mean_token_accuracy": 0.16791820228099824,
      "num_tokens": 9948603.0,
      "step": 5730
    },
    {
      "entropy": 5.5791689395904545,
      "epoch": 0.44621668936004666,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004984715197133271,
      "loss": 5.3856,
      "mean_token_accuracy": 0.16801765114068984,
      "num_tokens": 9957165.0,
      "step": 5735
    },
    {
      "entropy": 5.667300176620484,
      "epoch": 0.44660571873176425,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004984682929495371,
      "loss": 5.5544,
      "mean_token_accuracy": 0.16268503814935684,
      "num_tokens": 9966064.0,
      "step": 5740
    },
    {
      "entropy": 5.567746019363403,
      "epoch": 0.44699474810348183,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004984650627949715,
      "loss": 5.3963,
      "mean_token_accuracy": 0.163088820874691,
      "num_tokens": 9975146.0,
      "step": 5745
    },
    {
      "entropy": 5.624360799789429,
      "epoch": 0.44738377747519936,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004984618292496792,
      "loss": 5.4927,
      "mean_token_accuracy": 0.16582848578691484,
      "num_tokens": 9983725.0,
      "step": 5750
    },
    {
      "entropy": 5.682122611999512,
      "epoch": 0.44777280684691695,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004984585923137093,
      "loss": 5.5265,
      "mean_token_accuracy": 0.16172722578048707,
      "num_tokens": 9992627.0,
      "step": 5755
    },
    {
      "entropy": 5.592145872116089,
      "epoch": 0.4481618362186345,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000498455351987111,
      "loss": 5.3514,
      "mean_token_accuracy": 0.17467245012521743,
      "num_tokens": 10000801.0,
      "step": 5760
    },
    {
      "entropy": 5.673874187469482,
      "epoch": 0.44855086559035207,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004984521082699333,
      "loss": 5.5033,
      "mean_token_accuracy": 0.16748406738042831,
      "num_tokens": 10009421.0,
      "step": 5765
    },
    {
      "entropy": 5.611412572860718,
      "epoch": 0.44893989496206965,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004984488611622256,
      "loss": 5.4978,
      "mean_token_accuracy": 0.16486949399113654,
      "num_tokens": 10018527.0,
      "step": 5770
    },
    {
      "entropy": 5.619416761398315,
      "epoch": 0.4493289243337872,
      "grad_norm": 1.015625,
      "learning_rate": 0.000498445610664037,
      "loss": 5.4203,
      "mean_token_accuracy": 0.17014821916818618,
      "num_tokens": 10026906.0,
      "step": 5775
    },
    {
      "entropy": 5.60166482925415,
      "epoch": 0.44971795370550477,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000498442356775417,
      "loss": 5.4081,
      "mean_token_accuracy": 0.17366300374269486,
      "num_tokens": 10035972.0,
      "step": 5780
    },
    {
      "entropy": 5.606394481658936,
      "epoch": 0.45010698307722236,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004984390994964148,
      "loss": 5.5028,
      "mean_token_accuracy": 0.16163644939661026,
      "num_tokens": 10044544.0,
      "step": 5785
    },
    {
      "entropy": 5.731960916519165,
      "epoch": 0.4504960124489399,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00049843583882708,
      "loss": 5.4952,
      "mean_token_accuracy": 0.16808821707963945,
      "num_tokens": 10052880.0,
      "step": 5790
    },
    {
      "entropy": 5.686711406707763,
      "epoch": 0.4508850418206575,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000498432574767462,
      "loss": 5.5556,
      "mean_token_accuracy": 0.15909216105937957,
      "num_tokens": 10062373.0,
      "step": 5795
    },
    {
      "entropy": 5.576024532318115,
      "epoch": 0.451274071192375,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004984293073176103,
      "loss": 5.3637,
      "mean_token_accuracy": 0.17339247167110444,
      "num_tokens": 10070473.0,
      "step": 5800
    },
    {
      "entropy": 5.663080501556396,
      "epoch": 0.4516631005640926,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004984260364775744,
      "loss": 5.4869,
      "mean_token_accuracy": 0.16736487299203873,
      "num_tokens": 10079857.0,
      "step": 5805
    },
    {
      "entropy": 5.643650674819947,
      "epoch": 0.4520521299358102,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000498422762247404,
      "loss": 5.445,
      "mean_token_accuracy": 0.16632652133703232,
      "num_tokens": 10088326.0,
      "step": 5810
    },
    {
      "entropy": 5.631912469863892,
      "epoch": 0.4524411593075277,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004984194846271489,
      "loss": 5.3867,
      "mean_token_accuracy": 0.16788820028305054,
      "num_tokens": 10097208.0,
      "step": 5815
    },
    {
      "entropy": 5.653847026824951,
      "epoch": 0.4528301886792453,
      "grad_norm": 1.0,
      "learning_rate": 0.0004984162036168586,
      "loss": 5.4881,
      "mean_token_accuracy": 0.1633862353861332,
      "num_tokens": 10106057.0,
      "step": 5820
    },
    {
      "entropy": 5.569620895385742,
      "epoch": 0.4532192180509628,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004984129192165831,
      "loss": 5.4138,
      "mean_token_accuracy": 0.1673375114798546,
      "num_tokens": 10115384.0,
      "step": 5825
    },
    {
      "entropy": 5.686632299423218,
      "epoch": 0.4536082474226804,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004984096314263722,
      "loss": 5.5095,
      "mean_token_accuracy": 0.16176718175411225,
      "num_tokens": 10123958.0,
      "step": 5830
    },
    {
      "entropy": 5.65466160774231,
      "epoch": 0.453997276794398,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004984063402462757,
      "loss": 5.4413,
      "mean_token_accuracy": 0.17231336385011672,
      "num_tokens": 10133199.0,
      "step": 5835
    },
    {
      "entropy": 5.604842758178711,
      "epoch": 0.45438630616611553,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004984030456763435,
      "loss": 5.408,
      "mean_token_accuracy": 0.17446824312210082,
      "num_tokens": 10141796.0,
      "step": 5840
    },
    {
      "entropy": 5.606010675430298,
      "epoch": 0.4547753355378331,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004983997477166257,
      "loss": 5.424,
      "mean_token_accuracy": 0.1695821091532707,
      "num_tokens": 10149941.0,
      "step": 5845
    },
    {
      "entropy": 5.619622421264649,
      "epoch": 0.45516436490955064,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004983964463671723,
      "loss": 5.4935,
      "mean_token_accuracy": 0.1641951784491539,
      "num_tokens": 10158796.0,
      "step": 5850
    },
    {
      "entropy": 5.665774250030518,
      "epoch": 0.45555339428126823,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004983931416280334,
      "loss": 5.4422,
      "mean_token_accuracy": 0.17011956125497818,
      "num_tokens": 10167328.0,
      "step": 5855
    },
    {
      "entropy": 5.644601631164551,
      "epoch": 0.4559424236529858,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004983898334992591,
      "loss": 5.4961,
      "mean_token_accuracy": 0.16651429533958434,
      "num_tokens": 10175937.0,
      "step": 5860
    },
    {
      "entropy": 5.647733211517334,
      "epoch": 0.45633145302470335,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004983865219808998,
      "loss": 5.4556,
      "mean_token_accuracy": 0.1606528416275978,
      "num_tokens": 10184828.0,
      "step": 5865
    },
    {
      "entropy": 5.616971063613891,
      "epoch": 0.45672048239642093,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004983832070730055,
      "loss": 5.4497,
      "mean_token_accuracy": 0.16973941326141356,
      "num_tokens": 10193871.0,
      "step": 5870
    },
    {
      "entropy": 5.587744855880738,
      "epoch": 0.4571095117681385,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004983798887756265,
      "loss": 5.4516,
      "mean_token_accuracy": 0.16800768971443175,
      "num_tokens": 10202740.0,
      "step": 5875
    },
    {
      "entropy": 5.597984743118286,
      "epoch": 0.45749854113985605,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004983765670888133,
      "loss": 5.4087,
      "mean_token_accuracy": 0.17071811556816102,
      "num_tokens": 10210903.0,
      "step": 5880
    },
    {
      "entropy": 5.551643514633179,
      "epoch": 0.45788757051157364,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004983732420126163,
      "loss": 5.3589,
      "mean_token_accuracy": 0.1736896276473999,
      "num_tokens": 10219305.0,
      "step": 5885
    },
    {
      "entropy": 5.681756496429443,
      "epoch": 0.45827659988329117,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004983699135470858,
      "loss": 5.5142,
      "mean_token_accuracy": 0.15853092521429063,
      "num_tokens": 10228575.0,
      "step": 5890
    },
    {
      "entropy": 5.646949529647827,
      "epoch": 0.45866562925500876,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004983665816922723,
      "loss": 5.5034,
      "mean_token_accuracy": 0.16626327931880952,
      "num_tokens": 10237813.0,
      "step": 5895
    },
    {
      "entropy": 5.689908361434936,
      "epoch": 0.45905465862672634,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004983632464482267,
      "loss": 5.5037,
      "mean_token_accuracy": 0.16461766958236695,
      "num_tokens": 10246606.0,
      "step": 5900
    },
    {
      "entropy": 5.560118007659912,
      "epoch": 0.4594436879984439,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004983599078149991,
      "loss": 5.4526,
      "mean_token_accuracy": 0.17055174559354783,
      "num_tokens": 10254996.0,
      "step": 5905
    },
    {
      "entropy": 5.599699831008911,
      "epoch": 0.45983271737016146,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004983565657926405,
      "loss": 5.4982,
      "mean_token_accuracy": 0.1632937014102936,
      "num_tokens": 10263570.0,
      "step": 5910
    },
    {
      "entropy": 5.5971846103668215,
      "epoch": 0.460221746741879,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004983532203812017,
      "loss": 5.3542,
      "mean_token_accuracy": 0.1691060781478882,
      "num_tokens": 10272237.0,
      "step": 5915
    },
    {
      "entropy": 5.601367378234864,
      "epoch": 0.4606107761135966,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004983498715807331,
      "loss": 5.4901,
      "mean_token_accuracy": 0.16562240570783615,
      "num_tokens": 10281215.0,
      "step": 5920
    },
    {
      "entropy": 5.646333360671997,
      "epoch": 0.46099980548531416,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004983465193912857,
      "loss": 5.5213,
      "mean_token_accuracy": 0.16282440721988678,
      "num_tokens": 10291054.0,
      "step": 5925
    },
    {
      "entropy": 5.700567531585693,
      "epoch": 0.4613888348570317,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004983431638129104,
      "loss": 5.5128,
      "mean_token_accuracy": 0.16725291311740875,
      "num_tokens": 10299847.0,
      "step": 5930
    },
    {
      "entropy": 5.592577791213989,
      "epoch": 0.4617778642287493,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004983398048456581,
      "loss": 5.4352,
      "mean_token_accuracy": 0.17061688899993896,
      "num_tokens": 10309195.0,
      "step": 5935
    },
    {
      "entropy": 5.607589292526245,
      "epoch": 0.4621668936004668,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004983364424895796,
      "loss": 5.4602,
      "mean_token_accuracy": 0.16673509031534195,
      "num_tokens": 10318386.0,
      "step": 5940
    },
    {
      "entropy": 5.568243932723999,
      "epoch": 0.4625559229721844,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004983330767447262,
      "loss": 5.3737,
      "mean_token_accuracy": 0.17284452468156813,
      "num_tokens": 10327230.0,
      "step": 5945
    },
    {
      "entropy": 5.588061189651489,
      "epoch": 0.462944952343902,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004983297076111489,
      "loss": 5.3662,
      "mean_token_accuracy": 0.17034709453582764,
      "num_tokens": 10335617.0,
      "step": 5950
    },
    {
      "entropy": 5.618640327453614,
      "epoch": 0.4633339817156195,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004983263350888987,
      "loss": 5.4454,
      "mean_token_accuracy": 0.1700463205575943,
      "num_tokens": 10344302.0,
      "step": 5955
    },
    {
      "entropy": 5.6334014415740965,
      "epoch": 0.4637230110873371,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004983229591780268,
      "loss": 5.4823,
      "mean_token_accuracy": 0.16464230567216873,
      "num_tokens": 10352490.0,
      "step": 5960
    },
    {
      "entropy": 5.602584028244019,
      "epoch": 0.46411204045905463,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004983195798785844,
      "loss": 5.4313,
      "mean_token_accuracy": 0.17336985617876052,
      "num_tokens": 10361249.0,
      "step": 5965
    },
    {
      "entropy": 5.642180490493774,
      "epoch": 0.4645010698307722,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004983161971906228,
      "loss": 5.5149,
      "mean_token_accuracy": 0.15877831131219863,
      "num_tokens": 10370367.0,
      "step": 5970
    },
    {
      "entropy": 5.640953063964844,
      "epoch": 0.4648900992024898,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004983128111141935,
      "loss": 5.4891,
      "mean_token_accuracy": 0.16095846146345139,
      "num_tokens": 10378429.0,
      "step": 5975
    },
    {
      "entropy": 5.6266193866729735,
      "epoch": 0.46527912857420733,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004983094216493475,
      "loss": 5.3981,
      "mean_token_accuracy": 0.17122958600521088,
      "num_tokens": 10386762.0,
      "step": 5980
    },
    {
      "entropy": 5.591618394851684,
      "epoch": 0.4656681579459249,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004983060287961367,
      "loss": 5.4247,
      "mean_token_accuracy": 0.17049588710069657,
      "num_tokens": 10395150.0,
      "step": 5985
    },
    {
      "entropy": 5.609709787368774,
      "epoch": 0.4660571873176425,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004983026325546123,
      "loss": 5.4429,
      "mean_token_accuracy": 0.16723257601261138,
      "num_tokens": 10403941.0,
      "step": 5990
    },
    {
      "entropy": 5.741720724105835,
      "epoch": 0.46644621668936004,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004982992329248257,
      "loss": 5.5161,
      "mean_token_accuracy": 0.1589074283838272,
      "num_tokens": 10413955.0,
      "step": 5995
    },
    {
      "entropy": 5.628343296051026,
      "epoch": 0.4668352460610776,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004982958299068289,
      "loss": 5.4341,
      "mean_token_accuracy": 0.170184688270092,
      "num_tokens": 10421676.0,
      "step": 6000
    },
    {
      "epoch": 0.4668352460610776,
      "eval_entropy": 5.3785627919510075,
      "eval_loss": 5.484496593475342,
      "eval_mean_token_accuracy": 0.1744253120305334,
      "eval_num_tokens": 10421676.0,
      "eval_runtime": 21.695,
      "eval_samples_per_second": 1915.553,
      "eval_steps_per_second": 239.456,
      "step": 6000
    }
  ],
  "logging_steps": 5,
  "max_steps": 128520,
  "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": 2097246697390080.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}