{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 2011,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 0.8093053457140923,
      "epoch": 0.02486325211337643,
      "grad_norm": 3.875,
      "learning_rate": 7.277227722772277e-06,
      "loss": 1.7423,
      "mean_token_accuracy": 0.671543929874897,
      "num_tokens": 12997394.0,
      "step": 50
    },
    {
      "entropy": 1.2584100252389907,
      "epoch": 0.04972650422675286,
      "grad_norm": 0.62109375,
      "learning_rate": 1.4702970297029704e-05,
      "loss": 1.3055,
      "mean_token_accuracy": 0.7016283966600895,
      "num_tokens": 25982466.0,
      "step": 100
    },
    {
      "entropy": 1.1815398697555066,
      "epoch": 0.07458975634012929,
      "grad_norm": 0.5625,
      "learning_rate": 2.212871287128713e-05,
      "loss": 1.1874,
      "mean_token_accuracy": 0.7216299413144589,
      "num_tokens": 38965230.0,
      "step": 150
    },
    {
      "entropy": 1.1402550649642944,
      "epoch": 0.09945300845350571,
      "grad_norm": 0.5390625,
      "learning_rate": 2.9554455445544555e-05,
      "loss": 1.142,
      "mean_token_accuracy": 0.7295443145930767,
      "num_tokens": 51948935.0,
      "step": 200
    },
    {
      "entropy": 1.110574235022068,
      "epoch": 0.12431626056688215,
      "grad_norm": 0.5546875,
      "learning_rate": 2.9950061158395005e-05,
      "loss": 1.1098,
      "mean_token_accuracy": 0.7351739694178104,
      "num_tokens": 64951043.0,
      "step": 250
    },
    {
      "entropy": 1.0846257837116717,
      "epoch": 0.14917951268025859,
      "grad_norm": 0.5390625,
      "learning_rate": 2.9787675503104344e-05,
      "loss": 1.0836,
      "mean_token_accuracy": 0.7398869113624096,
      "num_tokens": 77936279.0,
      "step": 300
    },
    {
      "entropy": 1.0673179541528226,
      "epoch": 0.174042764793635,
      "grad_norm": 0.54296875,
      "learning_rate": 2.9513863052820796e-05,
      "loss": 1.0648,
      "mean_token_accuracy": 0.7432914447784423,
      "num_tokens": 90932336.0,
      "step": 350
    },
    {
      "entropy": 1.052229733467102,
      "epoch": 0.19890601690701143,
      "grad_norm": 0.5703125,
      "learning_rate": 2.913068701509568e-05,
      "loss": 1.049,
      "mean_token_accuracy": 0.746162725687027,
      "num_tokens": 103921567.0,
      "step": 400
    },
    {
      "entropy": 1.0345054648816585,
      "epoch": 0.22376926902038788,
      "grad_norm": 0.53515625,
      "learning_rate": 2.864103466438891e-05,
      "loss": 1.0312,
      "mean_token_accuracy": 0.7494943365454674,
      "num_tokens": 116912063.0,
      "step": 450
    },
    {
      "entropy": 1.0316937543451785,
      "epoch": 0.2486325211337643,
      "grad_norm": 0.53125,
      "learning_rate": 2.8048595586131855e-05,
      "loss": 1.0284,
      "mean_token_accuracy": 0.7500163938105107,
      "num_tokens": 129905218.0,
      "step": 500
    },
    {
      "entropy": 1.0241920906305313,
      "epoch": 0.27349577324714075,
      "grad_norm": 0.55078125,
      "learning_rate": 2.7357833875286894e-05,
      "loss": 1.0203,
      "mean_token_accuracy": 0.7514490978419781,
      "num_tokens": 142885565.0,
      "step": 550
    },
    {
      "entropy": 1.0306694743037224,
      "epoch": 0.29835902536051717,
      "grad_norm": 0.54296875,
      "learning_rate": 2.6573954498890738e-05,
      "loss": 1.0269,
      "mean_token_accuracy": 0.750104363411665,
      "num_tokens": 155872064.0,
      "step": 600
    },
    {
      "entropy": 1.0152570338547229,
      "epoch": 0.3232222774738936,
      "grad_norm": 0.5234375,
      "learning_rate": 2.5702864076043493e-05,
      "loss": 1.0108,
      "mean_token_accuracy": 0.7531800110638142,
      "num_tokens": 168867280.0,
      "step": 650
    },
    {
      "entropy": 1.0094216690957547,
      "epoch": 0.34808552958727,
      "grad_norm": 0.53125,
      "learning_rate": 2.4751126370870668e-05,
      "loss": 1.005,
      "mean_token_accuracy": 0.7541881857812405,
      "num_tokens": 181860562.0,
      "step": 700
    },
    {
      "entropy": 1.0075355672836304,
      "epoch": 0.37294878170064644,
      "grad_norm": 0.5703125,
      "learning_rate": 2.3725912833823623e-05,
      "loss": 1.0029,
      "mean_token_accuracy": 0.7544908618927002,
      "num_tokens": 194857393.0,
      "step": 750
    },
    {
      "entropy": 1.0026545779407023,
      "epoch": 0.39781203381402286,
      "grad_norm": 0.54296875,
      "learning_rate": 2.263494856399534e-05,
      "loss": 0.9983,
      "mean_token_accuracy": 0.7555115695297718,
      "num_tokens": 207830748.0,
      "step": 800
    },
    {
      "entropy": 0.9955900900065899,
      "epoch": 0.4226752859273993,
      "grad_norm": 0.55859375,
      "learning_rate": 2.148645409963155e-05,
      "loss": 0.991,
      "mean_token_accuracy": 0.7569138373434544,
      "num_tokens": 220811854.0,
      "step": 850
    },
    {
      "entropy": 0.9966490264236927,
      "epoch": 0.44753853804077576,
      "grad_norm": 0.5234375,
      "learning_rate": 2.0289083475452206e-05,
      "loss": 0.9917,
      "mean_token_accuracy": 0.7566816800832749,
      "num_tokens": 233809152.0,
      "step": 900
    },
    {
      "entropy": 0.9946993951499462,
      "epoch": 0.4724017901541522,
      "grad_norm": 0.5703125,
      "learning_rate": 1.9051859013528333e-05,
      "loss": 0.9902,
      "mean_token_accuracy": 0.7567966990172863,
      "num_tokens": 246795198.0,
      "step": 950
    },
    {
      "entropy": 0.9998035056889057,
      "epoch": 0.4972650422675286,
      "grad_norm": 0.50390625,
      "learning_rate": 1.7784103339072398e-05,
      "loss": 0.9943,
      "mean_token_accuracy": 0.7560001449286937,
      "num_tokens": 259785941.0,
      "step": 1000
    },
    {
      "entropy": 0.991586543917656,
      "epoch": 0.522128294380905,
      "grad_norm": 0.515625,
      "learning_rate": 1.649536913341075e-05,
      "loss": 0.9866,
      "mean_token_accuracy": 0.7574710394442081,
      "num_tokens": 272778072.0,
      "step": 1050
    },
    {
      "entropy": 0.9825779674947261,
      "epoch": 0.5469915464942815,
      "grad_norm": 0.5234375,
      "learning_rate": 1.5195367153457442e-05,
      "loss": 0.9782,
      "mean_token_accuracy": 0.7593892233073711,
      "num_tokens": 285763198.0,
      "step": 1100
    },
    {
      "entropy": 0.9868022166192532,
      "epoch": 0.5718547986076579,
      "grad_norm": 0.52734375,
      "learning_rate": 1.3893893060070753e-05,
      "loss": 0.982,
      "mean_token_accuracy": 0.7586962369084358,
      "num_tokens": 298755470.0,
      "step": 1150
    },
    {
      "entropy": 0.9881981492042542,
      "epoch": 0.5967180507210343,
      "grad_norm": 0.51953125,
      "learning_rate": 1.260075360664893e-05,
      "loss": 0.983,
      "mean_token_accuracy": 0.7583484600484371,
      "num_tokens": 311732964.0,
      "step": 1200
    },
    {
      "entropy": 0.990956412255764,
      "epoch": 0.6215813028344107,
      "grad_norm": 0.54296875,
      "learning_rate": 1.1325692744142444e-05,
      "loss": 0.9855,
      "mean_token_accuracy": 0.7576845416426659,
      "num_tokens": 324722969.0,
      "step": 1250
    },
    {
      "entropy": 0.9845810621976853,
      "epoch": 0.6464445549477872,
      "grad_norm": 0.5234375,
      "learning_rate": 1.0078318199289694e-05,
      "loss": 0.98,
      "mean_token_accuracy": 0.7590868937969207,
      "num_tokens": 337708515.0,
      "step": 1300
    },
    {
      "entropy": 0.9848950608074665,
      "epoch": 0.6713078070611636,
      "grad_norm": 0.51171875,
      "learning_rate": 8.868029079317466e-06,
      "loss": 0.9794,
      "mean_token_accuracy": 0.7590427026152611,
      "num_tokens": 350694304.0,
      "step": 1350
    },
    {
      "entropy": 0.9880955889821053,
      "epoch": 0.69617105917454,
      "grad_norm": 0.5234375,
      "learning_rate": 7.703945048612838e-06,
      "loss": 0.9829,
      "mean_token_accuracy": 0.758459353595972,
      "num_tokens": 363687831.0,
      "step": 1400
    },
    {
      "entropy": 0.9795867702364922,
      "epoch": 0.7210343112879165,
      "grad_norm": 0.5,
      "learning_rate": 6.594837611028224e-06,
      "loss": 0.9748,
      "mean_token_accuracy": 0.760101655125618,
      "num_tokens": 376685856.0,
      "step": 1450
    },
    {
      "entropy": 0.9814675351977349,
      "epoch": 0.7458975634012929,
      "grad_norm": 0.51171875,
      "learning_rate": 5.549064015615166e-06,
      "loss": 0.9759,
      "mean_token_accuracy": 0.7595324893295765,
      "num_tokens": 389669946.0,
      "step": 1500
    },
    {
      "entropy": 0.9784578867256641,
      "epoch": 0.7707608155146694,
      "grad_norm": 0.515625,
      "learning_rate": 4.574504283814536e-06,
      "loss": 0.9735,
      "mean_token_accuracy": 0.7601935516297817,
      "num_tokens": 402664698.0,
      "step": 1550
    },
    {
      "entropy": 0.9806207512319088,
      "epoch": 0.7956240676280457,
      "grad_norm": 0.52734375,
      "learning_rate": 3.67850183261035e-06,
      "loss": 0.9754,
      "mean_token_accuracy": 0.7597657778859138,
      "num_tokens": 415650519.0,
      "step": 1600
    },
    {
      "entropy": 0.981166479587555,
      "epoch": 0.8204873197414222,
      "grad_norm": 0.498046875,
      "learning_rate": 2.8678081410584267e-06,
      "loss": 0.9759,
      "mean_token_accuracy": 0.7597530463337898,
      "num_tokens": 428641539.0,
      "step": 1650
    },
    {
      "entropy": 0.9809097257256508,
      "epoch": 0.8453505718547986,
      "grad_norm": 0.50390625,
      "learning_rate": 2.1485318771337776e-06,
      "loss": 0.9757,
      "mean_token_accuracy": 0.7594838063418865,
      "num_tokens": 441636832.0,
      "step": 1700
    },
    {
      "entropy": 0.9838246862590313,
      "epoch": 0.870213823968175,
      "grad_norm": 0.4921875,
      "learning_rate": 1.5260928682316267e-06,
      "loss": 0.9788,
      "mean_token_accuracy": 0.7591392694413662,
      "num_tokens": 454630338.0,
      "step": 1750
    },
    {
      "entropy": 0.9816241288185119,
      "epoch": 0.8950770760815515,
      "grad_norm": 0.51171875,
      "learning_rate": 1.0051812621595896e-06,
      "loss": 0.9763,
      "mean_token_accuracy": 0.7593851044774056,
      "num_tokens": 467615642.0,
      "step": 1800
    },
    {
      "entropy": 0.9775546994805336,
      "epoch": 0.9199403281949279,
      "grad_norm": 0.51953125,
      "learning_rate": 5.89722186347399e-07,
      "loss": 0.9727,
      "mean_token_accuracy": 0.7600829027593136,
      "num_tokens": 480607840.0,
      "step": 1850
    },
    {
      "entropy": 0.9843708150088787,
      "epoch": 0.9448035803083044,
      "grad_norm": 0.498046875,
      "learning_rate": 2.828461715710806e-07,
      "loss": 0.9798,
      "mean_token_accuracy": 0.7588261432945729,
      "num_tokens": 493591064.0,
      "step": 1900
    },
    {
      "entropy": 0.9855945162475109,
      "epoch": 0.9696668324216807,
      "grad_norm": 0.5234375,
      "learning_rate": 8.686556305214144e-08,
      "loss": 0.9812,
      "mean_token_accuracy": 0.7586447390913963,
      "num_tokens": 506570590.0,
      "step": 1950
    },
    {
      "entropy": 0.9880490954220295,
      "epoch": 0.9945300845350572,
      "grad_norm": 0.51171875,
      "learning_rate": 3.2570966767209166e-09,
      "loss": 0.9837,
      "mean_token_accuracy": 0.7583207984268665,
      "num_tokens": 519553398.0,
      "step": 2000
    }
  ],
  "logging_steps": 50,
  "max_steps": 2011,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 50000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.417734959016444e+18,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}