{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.7836990595611285,
  "eval_steps": 500,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 10.742109775543213,
      "epoch": 0.003918495297805642,
      "grad_norm": 4.9375,
      "learning_rate": 2e-06,
      "loss": 10.7674,
      "mean_token_accuracy": 9.596928721293807e-05,
      "num_tokens": 9415.0,
      "step": 5
    },
    {
      "entropy": 10.742099189758301,
      "epoch": 0.007836990595611285,
      "grad_norm": 5.65625,
      "learning_rate": 4.5e-06,
      "loss": 10.7184,
      "mean_token_accuracy": 0.0,
      "num_tokens": 17669.0,
      "step": 10
    },
    {
      "entropy": 10.742108917236328,
      "epoch": 0.011755485893416929,
      "grad_norm": 5.0625,
      "learning_rate": 7e-06,
      "loss": 10.7241,
      "mean_token_accuracy": 0.00021240997593849898,
      "num_tokens": 26558.0,
      "step": 15
    },
    {
      "entropy": 10.742095565795898,
      "epoch": 0.01567398119122257,
      "grad_norm": 5.40625,
      "learning_rate": 9.5e-06,
      "loss": 10.6562,
      "mean_token_accuracy": 0.0,
      "num_tokens": 34940.0,
      "step": 20
    },
    {
      "entropy": 10.742070865631103,
      "epoch": 0.019592476489028215,
      "grad_norm": 4.40625,
      "learning_rate": 1.2e-05,
      "loss": 10.5966,
      "mean_token_accuracy": 0.0032189773628488183,
      "num_tokens": 44835.0,
      "step": 25
    },
    {
      "entropy": 10.741683006286621,
      "epoch": 0.023510971786833857,
      "grad_norm": 4.03125,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 10.4748,
      "mean_token_accuracy": 0.02363332137465477,
      "num_tokens": 53623.0,
      "step": 30
    },
    {
      "entropy": 10.740711116790772,
      "epoch": 0.0274294670846395,
      "grad_norm": 3.5,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 10.3625,
      "mean_token_accuracy": 0.03914350755512715,
      "num_tokens": 63198.0,
      "step": 35
    },
    {
      "entropy": 10.738054847717285,
      "epoch": 0.03134796238244514,
      "grad_norm": 2.828125,
      "learning_rate": 1.95e-05,
      "loss": 10.2445,
      "mean_token_accuracy": 0.043806169740855695,
      "num_tokens": 72379.0,
      "step": 40
    },
    {
      "entropy": 10.73265199661255,
      "epoch": 0.03526645768025078,
      "grad_norm": 2.453125,
      "learning_rate": 2.2e-05,
      "loss": 10.1124,
      "mean_token_accuracy": 0.04499626830220223,
      "num_tokens": 81098.0,
      "step": 45
    },
    {
      "entropy": 10.725791835784912,
      "epoch": 0.03918495297805643,
      "grad_norm": 2.109375,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 10.0549,
      "mean_token_accuracy": 0.04616287313401699,
      "num_tokens": 90637.0,
      "step": 50
    },
    {
      "entropy": 10.720808124542236,
      "epoch": 0.04310344827586207,
      "grad_norm": 2.03125,
      "learning_rate": 2.7e-05,
      "loss": 9.9466,
      "mean_token_accuracy": 0.04513598307967186,
      "num_tokens": 98505.0,
      "step": 55
    },
    {
      "entropy": 10.717309474945068,
      "epoch": 0.047021943573667714,
      "grad_norm": 1.90625,
      "learning_rate": 2.95e-05,
      "loss": 9.8942,
      "mean_token_accuracy": 0.048630540072917935,
      "num_tokens": 107295.0,
      "step": 60
    },
    {
      "entropy": 10.71382761001587,
      "epoch": 0.050940438871473356,
      "grad_norm": 2.0,
      "learning_rate": 3.2e-05,
      "loss": 9.8181,
      "mean_token_accuracy": 0.04877537116408348,
      "num_tokens": 116381.0,
      "step": 65
    },
    {
      "entropy": 10.70920639038086,
      "epoch": 0.054858934169279,
      "grad_norm": 1.984375,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.7896,
      "mean_token_accuracy": 0.04598020613193512,
      "num_tokens": 124826.0,
      "step": 70
    },
    {
      "entropy": 10.703414630889892,
      "epoch": 0.05877742946708464,
      "grad_norm": 1.9140625,
      "learning_rate": 3.7e-05,
      "loss": 9.7231,
      "mean_token_accuracy": 0.04887317828834057,
      "num_tokens": 133255.0,
      "step": 75
    },
    {
      "entropy": 10.69604263305664,
      "epoch": 0.06269592476489028,
      "grad_norm": 1.9609375,
      "learning_rate": 3.95e-05,
      "loss": 9.6411,
      "mean_token_accuracy": 0.05134939663112163,
      "num_tokens": 141626.0,
      "step": 80
    },
    {
      "entropy": 10.6855975151062,
      "epoch": 0.06661442006269593,
      "grad_norm": 1.9375,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.5459,
      "mean_token_accuracy": 0.06545204371213913,
      "num_tokens": 150051.0,
      "step": 85
    },
    {
      "entropy": 10.666928005218505,
      "epoch": 0.07053291536050156,
      "grad_norm": 1.953125,
      "learning_rate": 4.45e-05,
      "loss": 9.5417,
      "mean_token_accuracy": 0.061743459478020665,
      "num_tokens": 159325.0,
      "step": 90
    },
    {
      "entropy": 10.645073413848877,
      "epoch": 0.07445141065830721,
      "grad_norm": 1.859375,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 9.4416,
      "mean_token_accuracy": 0.06291964910924434,
      "num_tokens": 168142.0,
      "step": 95
    },
    {
      "entropy": 10.608468723297118,
      "epoch": 0.07836990595611286,
      "grad_norm": 1.765625,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 9.3001,
      "mean_token_accuracy": 0.06304978355765342,
      "num_tokens": 177033.0,
      "step": 100
    },
    {
      "entropy": 10.553625392913819,
      "epoch": 0.0822884012539185,
      "grad_norm": 1.8046875,
      "learning_rate": 5.2e-05,
      "loss": 9.229,
      "mean_token_accuracy": 0.056359085440635684,
      "num_tokens": 185605.0,
      "step": 105
    },
    {
      "entropy": 10.505783081054688,
      "epoch": 0.08620689655172414,
      "grad_norm": 1.703125,
      "learning_rate": 5.45e-05,
      "loss": 9.0952,
      "mean_token_accuracy": 0.0655987836420536,
      "num_tokens": 193863.0,
      "step": 110
    },
    {
      "entropy": 10.448099517822266,
      "epoch": 0.09012539184952978,
      "grad_norm": 1.671875,
      "learning_rate": 5.7e-05,
      "loss": 8.9999,
      "mean_token_accuracy": 0.0653056938201189,
      "num_tokens": 202504.0,
      "step": 115
    },
    {
      "entropy": 10.432163047790528,
      "epoch": 0.09404388714733543,
      "grad_norm": 1.5390625,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.9214,
      "mean_token_accuracy": 0.06190854795277119,
      "num_tokens": 211172.0,
      "step": 120
    },
    {
      "entropy": 10.393641376495362,
      "epoch": 0.09796238244514106,
      "grad_norm": 1.46875,
      "learning_rate": 6.2e-05,
      "loss": 8.8333,
      "mean_token_accuracy": 0.060395382717251776,
      "num_tokens": 220345.0,
      "step": 125
    },
    {
      "entropy": 10.306912040710449,
      "epoch": 0.10188087774294671,
      "grad_norm": 1.421875,
      "learning_rate": 6.450000000000001e-05,
      "loss": 8.6192,
      "mean_token_accuracy": 0.0663460198789835,
      "num_tokens": 228396.0,
      "step": 130
    },
    {
      "entropy": 10.185036849975585,
      "epoch": 0.10579937304075235,
      "grad_norm": 1.390625,
      "learning_rate": 6.7e-05,
      "loss": 8.5194,
      "mean_token_accuracy": 0.06784758679568767,
      "num_tokens": 236773.0,
      "step": 135
    },
    {
      "entropy": 10.10704460144043,
      "epoch": 0.109717868338558,
      "grad_norm": 1.4140625,
      "learning_rate": 6.950000000000001e-05,
      "loss": 8.4114,
      "mean_token_accuracy": 0.06528282575309277,
      "num_tokens": 245279.0,
      "step": 140
    },
    {
      "entropy": 9.997784900665284,
      "epoch": 0.11363636363636363,
      "grad_norm": 1.203125,
      "learning_rate": 7.2e-05,
      "loss": 8.2873,
      "mean_token_accuracy": 0.06404702439904213,
      "num_tokens": 253965.0,
      "step": 145
    },
    {
      "entropy": 9.862957191467284,
      "epoch": 0.11755485893416928,
      "grad_norm": 1.21875,
      "learning_rate": 7.45e-05,
      "loss": 8.2398,
      "mean_token_accuracy": 0.061683330684900284,
      "num_tokens": 262859.0,
      "step": 150
    },
    {
      "entropy": 9.73002061843872,
      "epoch": 0.12147335423197492,
      "grad_norm": 1.171875,
      "learning_rate": 7.7e-05,
      "loss": 8.1396,
      "mean_token_accuracy": 0.06386828869581222,
      "num_tokens": 270757.0,
      "step": 155
    },
    {
      "entropy": 9.495413970947265,
      "epoch": 0.12539184952978055,
      "grad_norm": 1.1015625,
      "learning_rate": 7.950000000000001e-05,
      "loss": 8.0706,
      "mean_token_accuracy": 0.060359453037381175,
      "num_tokens": 279737.0,
      "step": 160
    },
    {
      "entropy": 9.4040997505188,
      "epoch": 0.12931034482758622,
      "grad_norm": 1.0546875,
      "learning_rate": 8.2e-05,
      "loss": 7.895,
      "mean_token_accuracy": 0.061074954271316526,
      "num_tokens": 287788.0,
      "step": 165
    },
    {
      "entropy": 9.116162109375,
      "epoch": 0.13322884012539185,
      "grad_norm": 0.90625,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.8633,
      "mean_token_accuracy": 0.06705749183893203,
      "num_tokens": 296481.0,
      "step": 170
    },
    {
      "entropy": 8.948561000823975,
      "epoch": 0.1371473354231975,
      "grad_norm": 0.8671875,
      "learning_rate": 8.7e-05,
      "loss": 7.76,
      "mean_token_accuracy": 0.06502468548715115,
      "num_tokens": 305412.0,
      "step": 175
    },
    {
      "entropy": 8.750289058685302,
      "epoch": 0.14106583072100312,
      "grad_norm": 0.8828125,
      "learning_rate": 8.95e-05,
      "loss": 7.7497,
      "mean_token_accuracy": 0.06412209160625934,
      "num_tokens": 314412.0,
      "step": 180
    },
    {
      "entropy": 8.61333589553833,
      "epoch": 0.14498432601880878,
      "grad_norm": 0.80859375,
      "learning_rate": 9.2e-05,
      "loss": 7.6727,
      "mean_token_accuracy": 0.06552823781967163,
      "num_tokens": 322928.0,
      "step": 185
    },
    {
      "entropy": 8.46501522064209,
      "epoch": 0.14890282131661442,
      "grad_norm": 0.875,
      "learning_rate": 9.45e-05,
      "loss": 7.693,
      "mean_token_accuracy": 0.0656155750155449,
      "num_tokens": 331605.0,
      "step": 190
    },
    {
      "entropy": 8.362890625,
      "epoch": 0.15282131661442006,
      "grad_norm": 0.86328125,
      "learning_rate": 9.7e-05,
      "loss": 7.6608,
      "mean_token_accuracy": 0.06488309130072593,
      "num_tokens": 340654.0,
      "step": 195
    },
    {
      "entropy": 8.327593994140624,
      "epoch": 0.15673981191222572,
      "grad_norm": 0.88671875,
      "learning_rate": 9.95e-05,
      "loss": 7.6721,
      "mean_token_accuracy": 0.06592198349535465,
      "num_tokens": 348905.0,
      "step": 200
    },
    {
      "entropy": 8.273851680755616,
      "epoch": 0.16065830721003135,
      "grad_norm": 0.921875,
      "learning_rate": 0.000102,
      "loss": 7.5357,
      "mean_token_accuracy": 0.06864950023591518,
      "num_tokens": 357276.0,
      "step": 205
    },
    {
      "entropy": 8.201261711120605,
      "epoch": 0.164576802507837,
      "grad_norm": 0.79296875,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.578,
      "mean_token_accuracy": 0.06853164285421372,
      "num_tokens": 365707.0,
      "step": 210
    },
    {
      "entropy": 8.147554111480712,
      "epoch": 0.16849529780564262,
      "grad_norm": 1.265625,
      "learning_rate": 0.000107,
      "loss": 7.5605,
      "mean_token_accuracy": 0.06806492805480957,
      "num_tokens": 374595.0,
      "step": 215
    },
    {
      "entropy": 8.051411581039428,
      "epoch": 0.1724137931034483,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0001095,
      "loss": 7.5988,
      "mean_token_accuracy": 0.06651411205530167,
      "num_tokens": 383953.0,
      "step": 220
    },
    {
      "entropy": 8.147121715545655,
      "epoch": 0.17633228840125392,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000112,
      "loss": 7.5603,
      "mean_token_accuracy": 0.06881510317325593,
      "num_tokens": 393524.0,
      "step": 225
    },
    {
      "entropy": 8.113260126113891,
      "epoch": 0.18025078369905956,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0001145,
      "loss": 7.439,
      "mean_token_accuracy": 0.0729409821331501,
      "num_tokens": 402079.0,
      "step": 230
    },
    {
      "entropy": 8.0263240814209,
      "epoch": 0.1841692789968652,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.4926,
      "mean_token_accuracy": 0.06627442128956318,
      "num_tokens": 410306.0,
      "step": 235
    },
    {
      "entropy": 8.06832423210144,
      "epoch": 0.18808777429467086,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.508,
      "mean_token_accuracy": 0.07089690454304218,
      "num_tokens": 419157.0,
      "step": 240
    },
    {
      "entropy": 8.074864530563355,
      "epoch": 0.1920062695924765,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000122,
      "loss": 7.463,
      "mean_token_accuracy": 0.06991211287677288,
      "num_tokens": 428216.0,
      "step": 245
    },
    {
      "entropy": 7.9781323909759525,
      "epoch": 0.19592476489028213,
      "grad_norm": 1.03125,
      "learning_rate": 0.0001245,
      "loss": 7.4362,
      "mean_token_accuracy": 0.0743593417108059,
      "num_tokens": 436695.0,
      "step": 250
    },
    {
      "entropy": 8.013546228408813,
      "epoch": 0.19984326018808776,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000127,
      "loss": 7.4641,
      "mean_token_accuracy": 0.07037879787385463,
      "num_tokens": 446801.0,
      "step": 255
    },
    {
      "entropy": 7.9741381168365475,
      "epoch": 0.20376175548589343,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001295,
      "loss": 7.4692,
      "mean_token_accuracy": 0.0736465446650982,
      "num_tokens": 454980.0,
      "step": 260
    },
    {
      "entropy": 8.027658605575562,
      "epoch": 0.20768025078369906,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000132,
      "loss": 7.4402,
      "mean_token_accuracy": 0.07250457666814328,
      "num_tokens": 463699.0,
      "step": 265
    },
    {
      "entropy": 8.018376350402832,
      "epoch": 0.2115987460815047,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.334,
      "mean_token_accuracy": 0.07564162723720073,
      "num_tokens": 473005.0,
      "step": 270
    },
    {
      "entropy": 7.891370153427124,
      "epoch": 0.21551724137931033,
      "grad_norm": 1.0625,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.3218,
      "mean_token_accuracy": 0.07659829519689083,
      "num_tokens": 481104.0,
      "step": 275
    },
    {
      "entropy": 7.955986738204956,
      "epoch": 0.219435736677116,
      "grad_norm": 0.890625,
      "learning_rate": 0.0001395,
      "loss": 7.4639,
      "mean_token_accuracy": 0.07242306545376778,
      "num_tokens": 490028.0,
      "step": 280
    },
    {
      "entropy": 7.935779190063476,
      "epoch": 0.22335423197492163,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.46,
      "mean_token_accuracy": 0.07522198893129825,
      "num_tokens": 499093.0,
      "step": 285
    },
    {
      "entropy": 8.021220302581787,
      "epoch": 0.22727272727272727,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0001445,
      "loss": 7.4596,
      "mean_token_accuracy": 0.07582265920937062,
      "num_tokens": 508049.0,
      "step": 290
    },
    {
      "entropy": 7.857068347930908,
      "epoch": 0.23119122257053293,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000147,
      "loss": 7.276,
      "mean_token_accuracy": 0.080800349265337,
      "num_tokens": 516173.0,
      "step": 295
    },
    {
      "entropy": 7.936904716491699,
      "epoch": 0.23510971786833856,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001495,
      "loss": 7.3857,
      "mean_token_accuracy": 0.07901114635169507,
      "num_tokens": 524990.0,
      "step": 300
    },
    {
      "entropy": 7.885950946807862,
      "epoch": 0.2390282131661442,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000152,
      "loss": 7.4693,
      "mean_token_accuracy": 0.07808374315500259,
      "num_tokens": 533676.0,
      "step": 305
    },
    {
      "entropy": 7.922548103332519,
      "epoch": 0.24294670846394983,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.4113,
      "mean_token_accuracy": 0.0751194790005684,
      "num_tokens": 542757.0,
      "step": 310
    },
    {
      "entropy": 7.911523008346558,
      "epoch": 0.2468652037617555,
      "grad_norm": 1.3125,
      "learning_rate": 0.000157,
      "loss": 7.3816,
      "mean_token_accuracy": 0.07483164295554161,
      "num_tokens": 551958.0,
      "step": 315
    },
    {
      "entropy": 7.87831482887268,
      "epoch": 0.2507836990595611,
      "grad_norm": 1.09375,
      "learning_rate": 0.0001595,
      "loss": 7.297,
      "mean_token_accuracy": 0.07816217541694641,
      "num_tokens": 560850.0,
      "step": 320
    },
    {
      "entropy": 7.742319774627686,
      "epoch": 0.2547021943573668,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000162,
      "loss": 7.2442,
      "mean_token_accuracy": 0.07955843061208726,
      "num_tokens": 569226.0,
      "step": 325
    },
    {
      "entropy": 7.862335586547852,
      "epoch": 0.25862068965517243,
      "grad_norm": 1.125,
      "learning_rate": 0.00016450000000000001,
      "loss": 7.2487,
      "mean_token_accuracy": 0.08338543474674225,
      "num_tokens": 577674.0,
      "step": 330
    },
    {
      "entropy": 7.8038520336151125,
      "epoch": 0.26253918495297807,
      "grad_norm": 1.078125,
      "learning_rate": 0.00016700000000000002,
      "loss": 7.3003,
      "mean_token_accuracy": 0.08087649494409561,
      "num_tokens": 587095.0,
      "step": 335
    },
    {
      "entropy": 7.815746021270752,
      "epoch": 0.2664576802507837,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.2334,
      "mean_token_accuracy": 0.08010689094662667,
      "num_tokens": 595690.0,
      "step": 340
    },
    {
      "entropy": 7.813080167770385,
      "epoch": 0.27037617554858934,
      "grad_norm": 1.0,
      "learning_rate": 0.00017199999999999998,
      "loss": 7.2638,
      "mean_token_accuracy": 0.08633363470435143,
      "num_tokens": 603962.0,
      "step": 345
    },
    {
      "entropy": 7.854545068740845,
      "epoch": 0.274294670846395,
      "grad_norm": 1.1875,
      "learning_rate": 0.00017449999999999999,
      "loss": 7.2481,
      "mean_token_accuracy": 0.08236029744148254,
      "num_tokens": 612407.0,
      "step": 350
    },
    {
      "entropy": 7.811877584457397,
      "epoch": 0.2782131661442006,
      "grad_norm": 1.203125,
      "learning_rate": 0.000177,
      "loss": 7.3141,
      "mean_token_accuracy": 0.08180430978536606,
      "num_tokens": 620970.0,
      "step": 355
    },
    {
      "entropy": 7.737411594390869,
      "epoch": 0.28213166144200624,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0001795,
      "loss": 7.2908,
      "mean_token_accuracy": 0.07814904972910881,
      "num_tokens": 630701.0,
      "step": 360
    },
    {
      "entropy": 7.802158451080322,
      "epoch": 0.28605015673981193,
      "grad_norm": 1.09375,
      "learning_rate": 0.000182,
      "loss": 7.2974,
      "mean_token_accuracy": 0.08439014628529548,
      "num_tokens": 640246.0,
      "step": 365
    },
    {
      "entropy": 7.708374309539795,
      "epoch": 0.28996865203761757,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0001845,
      "loss": 7.2173,
      "mean_token_accuracy": 0.08501755222678184,
      "num_tokens": 648698.0,
      "step": 370
    },
    {
      "entropy": 7.808612918853759,
      "epoch": 0.2938871473354232,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000187,
      "loss": 7.2139,
      "mean_token_accuracy": 0.09233642369508743,
      "num_tokens": 658087.0,
      "step": 375
    },
    {
      "entropy": 7.6729991912841795,
      "epoch": 0.29780564263322884,
      "grad_norm": 1.046875,
      "learning_rate": 0.0001895,
      "loss": 7.1842,
      "mean_token_accuracy": 0.08876543864607811,
      "num_tokens": 666664.0,
      "step": 380
    },
    {
      "entropy": 7.760433673858643,
      "epoch": 0.3017241379310345,
      "grad_norm": 1.0625,
      "learning_rate": 0.000192,
      "loss": 7.2527,
      "mean_token_accuracy": 0.08252104483544827,
      "num_tokens": 675738.0,
      "step": 385
    },
    {
      "entropy": 7.659246635437012,
      "epoch": 0.3056426332288401,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0001945,
      "loss": 7.1341,
      "mean_token_accuracy": 0.0883593812584877,
      "num_tokens": 684751.0,
      "step": 390
    },
    {
      "entropy": 7.6855284690856935,
      "epoch": 0.30956112852664575,
      "grad_norm": 1.03125,
      "learning_rate": 0.00019700000000000002,
      "loss": 7.1392,
      "mean_token_accuracy": 0.09001046866178512,
      "num_tokens": 693819.0,
      "step": 395
    },
    {
      "entropy": 7.675692749023438,
      "epoch": 0.31347962382445144,
      "grad_norm": 1.046875,
      "learning_rate": 0.00019950000000000002,
      "loss": 7.2002,
      "mean_token_accuracy": 0.08597821295261383,
      "num_tokens": 703082.0,
      "step": 400
    },
    {
      "entropy": 7.6546083927154545,
      "epoch": 0.31739811912225707,
      "grad_norm": 1.203125,
      "learning_rate": 0.000202,
      "loss": 7.1151,
      "mean_token_accuracy": 0.08707506135106087,
      "num_tokens": 711721.0,
      "step": 405
    },
    {
      "entropy": 7.646405744552612,
      "epoch": 0.3213166144200627,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00020449999999999998,
      "loss": 7.1256,
      "mean_token_accuracy": 0.08994124978780746,
      "num_tokens": 719825.0,
      "step": 410
    },
    {
      "entropy": 7.639235925674439,
      "epoch": 0.32523510971786834,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000207,
      "loss": 7.053,
      "mean_token_accuracy": 0.08337244391441345,
      "num_tokens": 728575.0,
      "step": 415
    },
    {
      "entropy": 7.728874778747558,
      "epoch": 0.329153605015674,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002095,
      "loss": 7.2405,
      "mean_token_accuracy": 0.08313387259840965,
      "num_tokens": 738379.0,
      "step": 420
    },
    {
      "entropy": 7.568649673461914,
      "epoch": 0.3330721003134796,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000212,
      "loss": 7.0653,
      "mean_token_accuracy": 0.0943116620182991,
      "num_tokens": 747484.0,
      "step": 425
    },
    {
      "entropy": 7.662053632736206,
      "epoch": 0.33699059561128525,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002145,
      "loss": 7.0976,
      "mean_token_accuracy": 0.09167452827095986,
      "num_tokens": 756250.0,
      "step": 430
    },
    {
      "entropy": 7.631493139266968,
      "epoch": 0.3409090909090909,
      "grad_norm": 1.1875,
      "learning_rate": 0.00021700000000000002,
      "loss": 7.1553,
      "mean_token_accuracy": 0.09070844799280167,
      "num_tokens": 765393.0,
      "step": 435
    },
    {
      "entropy": 7.533075618743896,
      "epoch": 0.3448275862068966,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002195,
      "loss": 7.1282,
      "mean_token_accuracy": 0.09012485444545745,
      "num_tokens": 773726.0,
      "step": 440
    },
    {
      "entropy": 7.584697198867798,
      "epoch": 0.3487460815047022,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000222,
      "loss": 7.0359,
      "mean_token_accuracy": 0.09066722467541695,
      "num_tokens": 782335.0,
      "step": 445
    },
    {
      "entropy": 7.568591690063476,
      "epoch": 0.35266457680250785,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002245,
      "loss": 6.9887,
      "mean_token_accuracy": 0.09229951500892639,
      "num_tokens": 791546.0,
      "step": 450
    },
    {
      "entropy": 7.518783092498779,
      "epoch": 0.3565830721003135,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.964,
      "mean_token_accuracy": 0.09283199757337571,
      "num_tokens": 800650.0,
      "step": 455
    },
    {
      "entropy": 7.47516098022461,
      "epoch": 0.3605015673981191,
      "grad_norm": 1.109375,
      "learning_rate": 0.00022950000000000002,
      "loss": 7.0328,
      "mean_token_accuracy": 0.09423457458615303,
      "num_tokens": 809968.0,
      "step": 460
    },
    {
      "entropy": 7.544811105728149,
      "epoch": 0.36442006269592475,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00023200000000000003,
      "loss": 7.0201,
      "mean_token_accuracy": 0.09420157298445701,
      "num_tokens": 818800.0,
      "step": 465
    },
    {
      "entropy": 7.581027984619141,
      "epoch": 0.3683385579937304,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.9864,
      "mean_token_accuracy": 0.09675867483019829,
      "num_tokens": 827586.0,
      "step": 470
    },
    {
      "entropy": 7.303757953643799,
      "epoch": 0.3722570532915361,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000237,
      "loss": 6.9384,
      "mean_token_accuracy": 0.09580465108156204,
      "num_tokens": 836730.0,
      "step": 475
    },
    {
      "entropy": 7.696502780914306,
      "epoch": 0.3761755485893417,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002395,
      "loss": 7.1422,
      "mean_token_accuracy": 0.09025072157382966,
      "num_tokens": 845105.0,
      "step": 480
    },
    {
      "entropy": 7.521749973297119,
      "epoch": 0.38009404388714735,
      "grad_norm": 1.328125,
      "learning_rate": 0.000242,
      "loss": 7.0783,
      "mean_token_accuracy": 0.09295645728707314,
      "num_tokens": 853584.0,
      "step": 485
    },
    {
      "entropy": 7.556596755981445,
      "epoch": 0.384012539184953,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002445,
      "loss": 7.0668,
      "mean_token_accuracy": 0.08822977468371392,
      "num_tokens": 862442.0,
      "step": 490
    },
    {
      "entropy": 7.4543486595153805,
      "epoch": 0.3879310344827586,
      "grad_norm": 1.234375,
      "learning_rate": 0.000247,
      "loss": 6.9643,
      "mean_token_accuracy": 0.0951703853905201,
      "num_tokens": 870833.0,
      "step": 495
    },
    {
      "entropy": 7.596196413040161,
      "epoch": 0.39184952978056425,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002495,
      "loss": 7.1087,
      "mean_token_accuracy": 0.09554079249501228,
      "num_tokens": 879461.0,
      "step": 500
    },
    {
      "epoch": 0.39184952978056425,
      "eval_entropy": 7.3791314095489735,
      "eval_loss": 7.098165035247803,
      "eval_mean_token_accuracy": 0.09636555407051892,
      "eval_num_tokens": 879461.0,
      "eval_runtime": 2.2153,
      "eval_samples_per_second": 1860.68,
      "eval_steps_per_second": 232.924,
      "step": 500
    },
    {
      "entropy": 7.323290538787842,
      "epoch": 0.3957680250783699,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000252,
      "loss": 6.8468,
      "mean_token_accuracy": 0.10375453308224677,
      "num_tokens": 887893.0,
      "step": 505
    },
    {
      "entropy": 7.584144163131714,
      "epoch": 0.3996865203761755,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002545,
      "loss": 7.078,
      "mean_token_accuracy": 0.09517347961664199,
      "num_tokens": 896129.0,
      "step": 510
    },
    {
      "entropy": 7.514800357818603,
      "epoch": 0.4036050156739812,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000257,
      "loss": 7.0467,
      "mean_token_accuracy": 0.0908431887626648,
      "num_tokens": 904448.0,
      "step": 515
    },
    {
      "entropy": 7.4618706703186035,
      "epoch": 0.40752351097178685,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002595,
      "loss": 6.9808,
      "mean_token_accuracy": 0.09821045622229577,
      "num_tokens": 912486.0,
      "step": 520
    },
    {
      "entropy": 7.48437385559082,
      "epoch": 0.4114420062695925,
      "grad_norm": 1.0,
      "learning_rate": 0.000262,
      "loss": 6.9342,
      "mean_token_accuracy": 0.0975078672170639,
      "num_tokens": 920913.0,
      "step": 525
    },
    {
      "entropy": 7.505733489990234,
      "epoch": 0.4153605015673981,
      "grad_norm": 1.140625,
      "learning_rate": 0.00026450000000000003,
      "loss": 7.0757,
      "mean_token_accuracy": 0.08893783465027809,
      "num_tokens": 930227.0,
      "step": 530
    },
    {
      "entropy": 7.452349138259888,
      "epoch": 0.41927899686520376,
      "grad_norm": 1.140625,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.9462,
      "mean_token_accuracy": 0.09549824744462967,
      "num_tokens": 938835.0,
      "step": 535
    },
    {
      "entropy": 7.436382722854614,
      "epoch": 0.4231974921630094,
      "grad_norm": 1.046875,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.9256,
      "mean_token_accuracy": 0.10100051760673523,
      "num_tokens": 946930.0,
      "step": 540
    },
    {
      "entropy": 7.43581748008728,
      "epoch": 0.427115987460815,
      "grad_norm": 1.0625,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.9024,
      "mean_token_accuracy": 0.09818659126758575,
      "num_tokens": 955450.0,
      "step": 545
    },
    {
      "entropy": 7.412800025939942,
      "epoch": 0.43103448275862066,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002745,
      "loss": 6.9804,
      "mean_token_accuracy": 0.09347106665372848,
      "num_tokens": 964456.0,
      "step": 550
    },
    {
      "entropy": 7.413184976577758,
      "epoch": 0.43495297805642635,
      "grad_norm": 1.0625,
      "learning_rate": 0.000277,
      "loss": 6.8786,
      "mean_token_accuracy": 0.0999380074441433,
      "num_tokens": 973088.0,
      "step": 555
    },
    {
      "entropy": 7.320905685424805,
      "epoch": 0.438871473354232,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002795,
      "loss": 6.8844,
      "mean_token_accuracy": 0.09666496440768242,
      "num_tokens": 982182.0,
      "step": 560
    },
    {
      "entropy": 7.376274633407593,
      "epoch": 0.4427899686520376,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.8457,
      "mean_token_accuracy": 0.10074941292405129,
      "num_tokens": 990513.0,
      "step": 565
    },
    {
      "entropy": 7.300549602508545,
      "epoch": 0.44670846394984326,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002845,
      "loss": 6.857,
      "mean_token_accuracy": 0.09506906270980835,
      "num_tokens": 999394.0,
      "step": 570
    },
    {
      "entropy": 7.375359916687012,
      "epoch": 0.4506269592476489,
      "grad_norm": 1.09375,
      "learning_rate": 0.000287,
      "loss": 6.8453,
      "mean_token_accuracy": 0.1013949416577816,
      "num_tokens": 1007793.0,
      "step": 575
    },
    {
      "entropy": 7.34481520652771,
      "epoch": 0.45454545454545453,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002895,
      "loss": 6.9084,
      "mean_token_accuracy": 0.09777404814958572,
      "num_tokens": 1017198.0,
      "step": 580
    },
    {
      "entropy": 7.3425328731536865,
      "epoch": 0.45846394984326017,
      "grad_norm": 1.25,
      "learning_rate": 0.000292,
      "loss": 6.8047,
      "mean_token_accuracy": 0.10280357003211975,
      "num_tokens": 1025896.0,
      "step": 585
    },
    {
      "entropy": 7.445605945587158,
      "epoch": 0.46238244514106586,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002945,
      "loss": 7.0133,
      "mean_token_accuracy": 0.09696159362792969,
      "num_tokens": 1034540.0,
      "step": 590
    },
    {
      "entropy": 7.213116216659546,
      "epoch": 0.4663009404388715,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000297,
      "loss": 6.8321,
      "mean_token_accuracy": 0.10862131938338279,
      "num_tokens": 1043006.0,
      "step": 595
    },
    {
      "entropy": 7.316764163970947,
      "epoch": 0.4702194357366771,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002995,
      "loss": 6.8099,
      "mean_token_accuracy": 0.10350289791822434,
      "num_tokens": 1052530.0,
      "step": 600
    },
    {
      "entropy": 7.323724317550659,
      "epoch": 0.47413793103448276,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000302,
      "loss": 6.8137,
      "mean_token_accuracy": 0.10041582360863685,
      "num_tokens": 1060822.0,
      "step": 605
    },
    {
      "entropy": 7.30889310836792,
      "epoch": 0.4780564263322884,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003045,
      "loss": 6.8802,
      "mean_token_accuracy": 0.10094830319285393,
      "num_tokens": 1069962.0,
      "step": 610
    },
    {
      "entropy": 7.313723564147949,
      "epoch": 0.48197492163009403,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000307,
      "loss": 6.806,
      "mean_token_accuracy": 0.10286147370934487,
      "num_tokens": 1078903.0,
      "step": 615
    },
    {
      "entropy": 7.307069587707519,
      "epoch": 0.48589341692789967,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003095,
      "loss": 6.8794,
      "mean_token_accuracy": 0.09798636510968209,
      "num_tokens": 1087921.0,
      "step": 620
    },
    {
      "entropy": 7.2093301773071286,
      "epoch": 0.4898119122257053,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000312,
      "loss": 6.7742,
      "mean_token_accuracy": 0.10318923816084861,
      "num_tokens": 1096778.0,
      "step": 625
    },
    {
      "entropy": 7.321264219284058,
      "epoch": 0.493730407523511,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003145,
      "loss": 6.8046,
      "mean_token_accuracy": 0.10321516692638397,
      "num_tokens": 1105000.0,
      "step": 630
    },
    {
      "entropy": 7.175268077850342,
      "epoch": 0.49764890282131663,
      "grad_norm": 1.296875,
      "learning_rate": 0.000317,
      "loss": 6.6925,
      "mean_token_accuracy": 0.10435841083526612,
      "num_tokens": 1113475.0,
      "step": 635
    },
    {
      "entropy": 7.1677467823028564,
      "epoch": 0.5015673981191222,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003195,
      "loss": 6.7436,
      "mean_token_accuracy": 0.10689813792705535,
      "num_tokens": 1122188.0,
      "step": 640
    },
    {
      "entropy": 7.284512710571289,
      "epoch": 0.5054858934169278,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000322,
      "loss": 6.8446,
      "mean_token_accuracy": 0.10488534867763519,
      "num_tokens": 1130781.0,
      "step": 645
    },
    {
      "entropy": 7.213377618789673,
      "epoch": 0.5094043887147336,
      "grad_norm": 1.125,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.771,
      "mean_token_accuracy": 0.09970749393105507,
      "num_tokens": 1139889.0,
      "step": 650
    },
    {
      "entropy": 7.230893850326538,
      "epoch": 0.5133228840125392,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.7344,
      "mean_token_accuracy": 0.10886386185884475,
      "num_tokens": 1147904.0,
      "step": 655
    },
    {
      "entropy": 7.088760614395142,
      "epoch": 0.5172413793103449,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.696,
      "mean_token_accuracy": 0.1124557226896286,
      "num_tokens": 1157413.0,
      "step": 660
    },
    {
      "entropy": 7.306563091278076,
      "epoch": 0.5211598746081505,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.8164,
      "mean_token_accuracy": 0.10038655102252961,
      "num_tokens": 1165635.0,
      "step": 665
    },
    {
      "entropy": 7.134089422225952,
      "epoch": 0.5250783699059561,
      "grad_norm": 1.125,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.8298,
      "mean_token_accuracy": 0.09688482508063316,
      "num_tokens": 1174681.0,
      "step": 670
    },
    {
      "entropy": 7.235197401046753,
      "epoch": 0.5289968652037618,
      "grad_norm": 1.078125,
      "learning_rate": 0.000337,
      "loss": 6.8225,
      "mean_token_accuracy": 0.10358083248138428,
      "num_tokens": 1183454.0,
      "step": 675
    },
    {
      "entropy": 7.254007053375244,
      "epoch": 0.5329153605015674,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003395,
      "loss": 6.7803,
      "mean_token_accuracy": 0.10250866636633874,
      "num_tokens": 1191836.0,
      "step": 680
    },
    {
      "entropy": 7.161578273773193,
      "epoch": 0.536833855799373,
      "grad_norm": 1.21875,
      "learning_rate": 0.000342,
      "loss": 6.7006,
      "mean_token_accuracy": 0.10597401708364487,
      "num_tokens": 1200280.0,
      "step": 685
    },
    {
      "entropy": 7.203849601745605,
      "epoch": 0.5407523510971787,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.8066,
      "mean_token_accuracy": 0.09864948689937592,
      "num_tokens": 1208773.0,
      "step": 690
    },
    {
      "entropy": 7.236851644515991,
      "epoch": 0.5446708463949843,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000347,
      "loss": 6.795,
      "mean_token_accuracy": 0.10489028915762902,
      "num_tokens": 1218171.0,
      "step": 695
    },
    {
      "entropy": 7.15051794052124,
      "epoch": 0.54858934169279,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003495,
      "loss": 6.7532,
      "mean_token_accuracy": 0.10080546215176582,
      "num_tokens": 1226549.0,
      "step": 700
    },
    {
      "entropy": 7.200709390640259,
      "epoch": 0.5525078369905956,
      "grad_norm": 1.203125,
      "learning_rate": 0.000352,
      "loss": 6.7439,
      "mean_token_accuracy": 0.11170587316155434,
      "num_tokens": 1234995.0,
      "step": 705
    },
    {
      "entropy": 7.17590970993042,
      "epoch": 0.5564263322884012,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003545,
      "loss": 6.787,
      "mean_token_accuracy": 0.10626475065946579,
      "num_tokens": 1243807.0,
      "step": 710
    },
    {
      "entropy": 7.112601041793823,
      "epoch": 0.5603448275862069,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000357,
      "loss": 6.7665,
      "mean_token_accuracy": 0.10710262060165406,
      "num_tokens": 1252773.0,
      "step": 715
    },
    {
      "entropy": 7.164514064788818,
      "epoch": 0.5642633228840125,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003595,
      "loss": 6.7316,
      "mean_token_accuracy": 0.09972823187708854,
      "num_tokens": 1262514.0,
      "step": 720
    },
    {
      "entropy": 7.2263566017150875,
      "epoch": 0.5681818181818182,
      "grad_norm": 1.09375,
      "learning_rate": 0.000362,
      "loss": 6.7304,
      "mean_token_accuracy": 0.10351964756846428,
      "num_tokens": 1271329.0,
      "step": 725
    },
    {
      "entropy": 7.020602416992188,
      "epoch": 0.5721003134796239,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003645,
      "loss": 6.6153,
      "mean_token_accuracy": 0.1127767838537693,
      "num_tokens": 1279563.0,
      "step": 730
    },
    {
      "entropy": 7.153738784790039,
      "epoch": 0.5760188087774295,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000367,
      "loss": 6.7494,
      "mean_token_accuracy": 0.10431532934308052,
      "num_tokens": 1288906.0,
      "step": 735
    },
    {
      "entropy": 7.086837673187256,
      "epoch": 0.5799373040752351,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003695,
      "loss": 6.6582,
      "mean_token_accuracy": 0.10919351503252983,
      "num_tokens": 1297880.0,
      "step": 740
    },
    {
      "entropy": 7.083261871337891,
      "epoch": 0.5838557993730408,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000372,
      "loss": 6.6923,
      "mean_token_accuracy": 0.10958038419485092,
      "num_tokens": 1306740.0,
      "step": 745
    },
    {
      "entropy": 7.1014440059661865,
      "epoch": 0.5877742946708464,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003745,
      "loss": 6.6471,
      "mean_token_accuracy": 0.10642243772745133,
      "num_tokens": 1315997.0,
      "step": 750
    },
    {
      "entropy": 7.118290996551513,
      "epoch": 0.591692789968652,
      "grad_norm": 1.15625,
      "learning_rate": 0.000377,
      "loss": 6.634,
      "mean_token_accuracy": 0.11355772763490676,
      "num_tokens": 1324064.0,
      "step": 755
    },
    {
      "entropy": 7.116359090805053,
      "epoch": 0.5956112852664577,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003795,
      "loss": 6.7624,
      "mean_token_accuracy": 0.10623705238103867,
      "num_tokens": 1332560.0,
      "step": 760
    },
    {
      "entropy": 7.0753131866455075,
      "epoch": 0.5995297805642633,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000382,
      "loss": 6.6853,
      "mean_token_accuracy": 0.10491749420762062,
      "num_tokens": 1341367.0,
      "step": 765
    },
    {
      "entropy": 7.208959865570068,
      "epoch": 0.603448275862069,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003845,
      "loss": 6.8643,
      "mean_token_accuracy": 0.09926253855228424,
      "num_tokens": 1350602.0,
      "step": 770
    },
    {
      "entropy": 7.111758089065551,
      "epoch": 0.6073667711598746,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.5959,
      "mean_token_accuracy": 0.11473986878991127,
      "num_tokens": 1358538.0,
      "step": 775
    },
    {
      "entropy": 6.9510876655578615,
      "epoch": 0.6112852664576802,
      "grad_norm": 1.125,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.5966,
      "mean_token_accuracy": 0.11128686666488648,
      "num_tokens": 1367089.0,
      "step": 780
    },
    {
      "entropy": 7.095315790176391,
      "epoch": 0.6152037617554859,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.6987,
      "mean_token_accuracy": 0.107613255828619,
      "num_tokens": 1376095.0,
      "step": 785
    },
    {
      "entropy": 7.0291649341583256,
      "epoch": 0.6191222570532915,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.7278,
      "mean_token_accuracy": 0.10492047220468521,
      "num_tokens": 1385132.0,
      "step": 790
    },
    {
      "entropy": 7.174719142913818,
      "epoch": 0.6230407523510971,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.8042,
      "mean_token_accuracy": 0.10808081552386284,
      "num_tokens": 1394187.0,
      "step": 795
    },
    {
      "entropy": 7.009328317642212,
      "epoch": 0.6269592476489029,
      "grad_norm": 1.125,
      "learning_rate": 0.0003995,
      "loss": 6.6543,
      "mean_token_accuracy": 0.10319057255983352,
      "num_tokens": 1403250.0,
      "step": 800
    },
    {
      "entropy": 7.114693927764892,
      "epoch": 0.6308777429467085,
      "grad_norm": 1.15625,
      "learning_rate": 0.000402,
      "loss": 6.7523,
      "mean_token_accuracy": 0.10417917445302009,
      "num_tokens": 1412768.0,
      "step": 805
    },
    {
      "entropy": 7.1693981170654295,
      "epoch": 0.6347962382445141,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004045,
      "loss": 6.7831,
      "mean_token_accuracy": 0.10479006171226501,
      "num_tokens": 1422437.0,
      "step": 810
    },
    {
      "entropy": 7.221331596374512,
      "epoch": 0.6387147335423198,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.8352,
      "mean_token_accuracy": 0.10820698514580726,
      "num_tokens": 1432540.0,
      "step": 815
    },
    {
      "entropy": 6.996674919128418,
      "epoch": 0.6426332288401254,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004095,
      "loss": 6.6518,
      "mean_token_accuracy": 0.10877189710736275,
      "num_tokens": 1441403.0,
      "step": 820
    },
    {
      "entropy": 6.932708883285523,
      "epoch": 0.646551724137931,
      "grad_norm": 1.171875,
      "learning_rate": 0.000412,
      "loss": 6.4891,
      "mean_token_accuracy": 0.10969182550907135,
      "num_tokens": 1450059.0,
      "step": 825
    },
    {
      "entropy": 7.004702997207642,
      "epoch": 0.6504702194357367,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004145,
      "loss": 6.6424,
      "mean_token_accuracy": 0.11137095913290977,
      "num_tokens": 1458413.0,
      "step": 830
    },
    {
      "entropy": 6.982443046569824,
      "epoch": 0.6543887147335423,
      "grad_norm": 1.15625,
      "learning_rate": 0.000417,
      "loss": 6.622,
      "mean_token_accuracy": 0.11147240921854973,
      "num_tokens": 1467343.0,
      "step": 835
    },
    {
      "entropy": 6.956045198440552,
      "epoch": 0.658307210031348,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004195,
      "loss": 6.5607,
      "mean_token_accuracy": 0.11432239189743995,
      "num_tokens": 1476250.0,
      "step": 840
    },
    {
      "entropy": 7.017556047439575,
      "epoch": 0.6622257053291536,
      "grad_norm": 1.265625,
      "learning_rate": 0.000422,
      "loss": 6.5268,
      "mean_token_accuracy": 0.11572841182351112,
      "num_tokens": 1484663.0,
      "step": 845
    },
    {
      "entropy": 7.062141704559326,
      "epoch": 0.6661442006269592,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004245,
      "loss": 6.6401,
      "mean_token_accuracy": 0.10651951879262925,
      "num_tokens": 1493898.0,
      "step": 850
    },
    {
      "entropy": 6.941916561126709,
      "epoch": 0.6700626959247649,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000427,
      "loss": 6.6059,
      "mean_token_accuracy": 0.1140934869647026,
      "num_tokens": 1502429.0,
      "step": 855
    },
    {
      "entropy": 6.927605772018433,
      "epoch": 0.6739811912225705,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004295,
      "loss": 6.6293,
      "mean_token_accuracy": 0.11264230906963349,
      "num_tokens": 1511671.0,
      "step": 860
    },
    {
      "entropy": 6.936274194717408,
      "epoch": 0.6778996865203761,
      "grad_norm": 1.078125,
      "learning_rate": 0.000432,
      "loss": 6.5421,
      "mean_token_accuracy": 0.11239867955446244,
      "num_tokens": 1520481.0,
      "step": 865
    },
    {
      "entropy": 6.910920095443726,
      "epoch": 0.6818181818181818,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004345,
      "loss": 6.4874,
      "mean_token_accuracy": 0.11565134301781654,
      "num_tokens": 1528589.0,
      "step": 870
    },
    {
      "entropy": 6.936212968826294,
      "epoch": 0.6857366771159875,
      "grad_norm": 1.1875,
      "learning_rate": 0.000437,
      "loss": 6.6584,
      "mean_token_accuracy": 0.108811154961586,
      "num_tokens": 1537555.0,
      "step": 875
    },
    {
      "entropy": 6.945295000076294,
      "epoch": 0.6896551724137931,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004395,
      "loss": 6.6032,
      "mean_token_accuracy": 0.11123565956950188,
      "num_tokens": 1546397.0,
      "step": 880
    },
    {
      "entropy": 7.034606647491455,
      "epoch": 0.6935736677115988,
      "grad_norm": 1.140625,
      "learning_rate": 0.000442,
      "loss": 6.6331,
      "mean_token_accuracy": 0.11215913444757461,
      "num_tokens": 1555603.0,
      "step": 885
    },
    {
      "entropy": 6.92329592704773,
      "epoch": 0.6974921630094044,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004445,
      "loss": 6.5938,
      "mean_token_accuracy": 0.11114713251590728,
      "num_tokens": 1564688.0,
      "step": 890
    },
    {
      "entropy": 6.946131467819214,
      "epoch": 0.70141065830721,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000447,
      "loss": 6.5727,
      "mean_token_accuracy": 0.1131402924656868,
      "num_tokens": 1573490.0,
      "step": 895
    },
    {
      "entropy": 6.9099291324615475,
      "epoch": 0.7053291536050157,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.5425,
      "mean_token_accuracy": 0.11272019520401955,
      "num_tokens": 1582668.0,
      "step": 900
    },
    {
      "entropy": 6.931637620925903,
      "epoch": 0.7092476489028213,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.6098,
      "mean_token_accuracy": 0.11104563847184182,
      "num_tokens": 1591667.0,
      "step": 905
    },
    {
      "entropy": 6.885412693023682,
      "epoch": 0.713166144200627,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.5154,
      "mean_token_accuracy": 0.11723392531275749,
      "num_tokens": 1600248.0,
      "step": 910
    },
    {
      "entropy": 6.9636177062988285,
      "epoch": 0.7170846394984326,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.5222,
      "mean_token_accuracy": 0.11748122572898864,
      "num_tokens": 1609094.0,
      "step": 915
    },
    {
      "entropy": 6.80591516494751,
      "epoch": 0.7210031347962382,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.5523,
      "mean_token_accuracy": 0.11386217251420021,
      "num_tokens": 1617845.0,
      "step": 920
    },
    {
      "entropy": 7.011699295043945,
      "epoch": 0.7249216300940439,
      "grad_norm": 1.171875,
      "learning_rate": 0.000462,
      "loss": 6.5865,
      "mean_token_accuracy": 0.11231657788157463,
      "num_tokens": 1626596.0,
      "step": 925
    },
    {
      "entropy": 6.840612173080444,
      "epoch": 0.7288401253918495,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004645,
      "loss": 6.5,
      "mean_token_accuracy": 0.11078331917524338,
      "num_tokens": 1635246.0,
      "step": 930
    },
    {
      "entropy": 6.838626718521118,
      "epoch": 0.7327586206896551,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000467,
      "loss": 6.556,
      "mean_token_accuracy": 0.11381039470434189,
      "num_tokens": 1645041.0,
      "step": 935
    },
    {
      "entropy": 6.995179224014282,
      "epoch": 0.7366771159874608,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004695,
      "loss": 6.6265,
      "mean_token_accuracy": 0.10761500224471092,
      "num_tokens": 1653636.0,
      "step": 940
    },
    {
      "entropy": 6.835239219665527,
      "epoch": 0.7405956112852664,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000472,
      "loss": 6.5227,
      "mean_token_accuracy": 0.11463285759091377,
      "num_tokens": 1662485.0,
      "step": 945
    },
    {
      "entropy": 6.926279592514038,
      "epoch": 0.7445141065830722,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004745,
      "loss": 6.5461,
      "mean_token_accuracy": 0.12085802108049393,
      "num_tokens": 1670414.0,
      "step": 950
    },
    {
      "entropy": 6.897070741653442,
      "epoch": 0.7484326018808778,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000477,
      "loss": 6.5507,
      "mean_token_accuracy": 0.11035623922944068,
      "num_tokens": 1678853.0,
      "step": 955
    },
    {
      "entropy": 6.948398160934448,
      "epoch": 0.7523510971786834,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004795,
      "loss": 6.4881,
      "mean_token_accuracy": 0.11652906686067581,
      "num_tokens": 1687286.0,
      "step": 960
    },
    {
      "entropy": 6.789101934432983,
      "epoch": 0.7562695924764891,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000482,
      "loss": 6.5127,
      "mean_token_accuracy": 0.1135128065943718,
      "num_tokens": 1696302.0,
      "step": 965
    },
    {
      "entropy": 6.88850679397583,
      "epoch": 0.7601880877742947,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004845,
      "loss": 6.6328,
      "mean_token_accuracy": 0.10952614322304725,
      "num_tokens": 1704622.0,
      "step": 970
    },
    {
      "entropy": 6.836090850830078,
      "epoch": 0.7641065830721003,
      "grad_norm": 1.140625,
      "learning_rate": 0.000487,
      "loss": 6.4682,
      "mean_token_accuracy": 0.11953761354088784,
      "num_tokens": 1713060.0,
      "step": 975
    },
    {
      "entropy": 6.769405508041382,
      "epoch": 0.768025078369906,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004895,
      "loss": 6.4437,
      "mean_token_accuracy": 0.1201899990439415,
      "num_tokens": 1721654.0,
      "step": 980
    },
    {
      "entropy": 6.872002363204956,
      "epoch": 0.7719435736677116,
      "grad_norm": 1.203125,
      "learning_rate": 0.000492,
      "loss": 6.4967,
      "mean_token_accuracy": 0.11558816209435463,
      "num_tokens": 1730243.0,
      "step": 985
    },
    {
      "entropy": 6.746385478973389,
      "epoch": 0.7758620689655172,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004945,
      "loss": 6.4224,
      "mean_token_accuracy": 0.11981726884841919,
      "num_tokens": 1739347.0,
      "step": 990
    },
    {
      "entropy": 6.771752119064331,
      "epoch": 0.7797805642633229,
      "grad_norm": 1.125,
      "learning_rate": 0.000497,
      "loss": 6.4607,
      "mean_token_accuracy": 0.12106571421027183,
      "num_tokens": 1747802.0,
      "step": 995
    },
    {
      "entropy": 6.778915500640869,
      "epoch": 0.7836990595611285,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004995,
      "loss": 6.4581,
      "mean_token_accuracy": 0.11269956901669502,
      "num_tokens": 1757407.0,
      "step": 1000
    },
    {
      "epoch": 0.7836990595611285,
      "eval_entropy": 6.515588334364485,
      "eval_loss": 6.588331699371338,
      "eval_mean_token_accuracy": 0.1149799013077173,
      "eval_num_tokens": 1757407.0,
      "eval_runtime": 2.3882,
      "eval_samples_per_second": 1725.989,
      "eval_steps_per_second": 216.063,
      "step": 1000
    }
  ],
  "logging_steps": 5,
  "max_steps": 12750,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 349979963719680.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}