{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.962962962962963,
  "eval_steps": 500,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 1.6574764619767666,
      "epoch": 0.07407407407407407,
      "grad_norm": 0.6482250690460205,
      "learning_rate": 8.032786885245902e-05,
      "loss": 1.4512,
      "mean_token_accuracy": 0.6979378089308739,
      "num_tokens": 408597.0,
      "step": 50
    },
    {
      "entropy": 0.38226877618581057,
      "epoch": 0.14814814814814814,
      "grad_norm": 0.2513386011123657,
      "learning_rate": 9.990765991730485e-05,
      "loss": 0.3105,
      "mean_token_accuracy": 0.92415526881814,
      "num_tokens": 817567.0,
      "step": 100
    },
    {
      "entropy": 0.16286218419671059,
      "epoch": 0.2222222222222222,
      "grad_norm": 0.2887406051158905,
      "learning_rate": 9.950545603782162e-05,
      "loss": 0.1493,
      "mean_token_accuracy": 0.9622085580229759,
      "num_tokens": 1226265.0,
      "step": 150
    },
    {
      "entropy": 0.12771075297147036,
      "epoch": 0.2962962962962963,
      "grad_norm": 0.15249371528625488,
      "learning_rate": 9.878674879048427e-05,
      "loss": 0.1177,
      "mean_token_accuracy": 0.9713320073485374,
      "num_tokens": 1635020.0,
      "step": 200
    },
    {
      "entropy": 0.1111781226657331,
      "epoch": 0.37037037037037035,
      "grad_norm": 0.18299971520900726,
      "learning_rate": 9.775613308830824e-05,
      "loss": 0.1014,
      "mean_token_accuracy": 0.9749903282523156,
      "num_tokens": 2044395.0,
      "step": 250
    },
    {
      "entropy": 0.09641788197681307,
      "epoch": 0.4444444444444444,
      "grad_norm": 0.13050603866577148,
      "learning_rate": 9.642019796948866e-05,
      "loss": 0.0915,
      "mean_token_accuracy": 0.977490707486868,
      "num_tokens": 2453123.0,
      "step": 300
    },
    {
      "entropy": 0.09188288618810475,
      "epoch": 0.5185185185185185,
      "grad_norm": 0.11683323234319687,
      "learning_rate": 9.478748447168449e-05,
      "loss": 0.0883,
      "mean_token_accuracy": 0.9783088724315167,
      "num_tokens": 2862139.0,
      "step": 350
    },
    {
      "entropy": 0.09006990000605583,
      "epoch": 0.5925925925925926,
      "grad_norm": 0.16509227454662323,
      "learning_rate": 9.28684310265789e-05,
      "loss": 0.0874,
      "mean_token_accuracy": 0.9784132435917854,
      "num_tokens": 3270859.0,
      "step": 400
    },
    {
      "entropy": 0.08776539071463048,
      "epoch": 0.6666666666666666,
      "grad_norm": 0.13524411618709564,
      "learning_rate": 9.067530672382544e-05,
      "loss": 0.0866,
      "mean_token_accuracy": 0.9787498603761197,
      "num_tokens": 3680243.0,
      "step": 450
    },
    {
      "entropy": 0.08806963111273944,
      "epoch": 0.7407407407407407,
      "grad_norm": 0.09137360751628876,
      "learning_rate": 8.822213287104348e-05,
      "loss": 0.086,
      "mean_token_accuracy": 0.9788302226364612,
      "num_tokens": 4089386.0,
      "step": 500
    },
    {
      "entropy": 0.08409048081375659,
      "epoch": 0.8148148148148148,
      "grad_norm": 0.18463748693466187,
      "learning_rate": 8.552459335135381e-05,
      "loss": 0.0831,
      "mean_token_accuracy": 0.9793832650780678,
      "num_tokens": 4498160.0,
      "step": 550
    },
    {
      "entropy": 0.08435689969919621,
      "epoch": 0.8888888888888888,
      "grad_norm": 0.09467488527297974,
      "learning_rate": 8.259993435156559e-05,
      "loss": 0.0831,
      "mean_token_accuracy": 0.9795491580665111,
      "num_tokens": 4907148.0,
      "step": 600
    },
    {
      "entropy": 0.08570322282612323,
      "epoch": 0.9629629629629629,
      "grad_norm": 0.11093126982450485,
      "learning_rate": 7.946685410208296e-05,
      "loss": 0.0844,
      "mean_token_accuracy": 0.9790063916146755,
      "num_tokens": 5315908.0,
      "step": 650
    },
    {
      "entropy": 0.08327227901667357,
      "epoch": 1.037037037037037,
      "grad_norm": 0.07517968118190765,
      "learning_rate": 7.614538333345735e-05,
      "loss": 0.0824,
      "mean_token_accuracy": 0.979544368237257,
      "num_tokens": 5724704.0,
      "step": 700
    },
    {
      "entropy": 0.08300686337985098,
      "epoch": 1.1111111111111112,
      "grad_norm": 0.0655263215303421,
      "learning_rate": 7.265675721386285e-05,
      "loss": 0.0818,
      "mean_token_accuracy": 0.9796155346930027,
      "num_tokens": 6133646.0,
      "step": 750
    },
    {
      "entropy": 0.08471525728702545,
      "epoch": 1.1851851851851851,
      "grad_norm": 0.063040591776371,
      "learning_rate": 6.902327958623736e-05,
      "loss": 0.0841,
      "mean_token_accuracy": 0.9790562556684017,
      "num_tokens": 6542365.0,
      "step": 800
    },
    {
      "entropy": 0.08242542044259608,
      "epoch": 1.2592592592592593,
      "grad_norm": 0.16829481720924377,
      "learning_rate": 6.526818037306228e-05,
      "loss": 0.0818,
      "mean_token_accuracy": 0.9798041236400604,
      "num_tokens": 6951319.0,
      "step": 850
    },
    {
      "entropy": 0.08193825788795948,
      "epoch": 1.3333333333333333,
      "grad_norm": 0.08218955248594284,
      "learning_rate": 6.14154670604355e-05,
      "loss": 0.0812,
      "mean_token_accuracy": 0.9799344435334205,
      "num_tokens": 7360266.0,
      "step": 900
    },
    {
      "entropy": 0.0820022343751043,
      "epoch": 1.4074074074074074,
      "grad_norm": 0.07167431712150574,
      "learning_rate": 5.7489771210944564e-05,
      "loss": 0.0814,
      "mean_token_accuracy": 0.979758190214634,
      "num_tokens": 7769643.0,
      "step": 950
    },
    {
      "entropy": 0.08256705107167363,
      "epoch": 1.4814814814814814,
      "grad_norm": 0.1688249707221985,
      "learning_rate": 5.351619098663021e-05,
      "loss": 0.0818,
      "mean_token_accuracy": 0.9795773278176785,
      "num_tokens": 8178399.0,
      "step": 1000
    },
    {
      "entropy": 0.08179685108363628,
      "epoch": 1.5555555555555556,
      "grad_norm": 0.06633222848176956,
      "learning_rate": 4.952013068883795e-05,
      "loss": 0.0814,
      "mean_token_accuracy": 0.979800873696804,
      "num_tokens": 8587121.0,
      "step": 1050
    },
    {
      "entropy": 0.08153917868621648,
      "epoch": 1.6296296296296298,
      "grad_norm": 0.07971940189599991,
      "learning_rate": 4.5527138340828776e-05,
      "loss": 0.0812,
      "mean_token_accuracy": 0.9798491597175598,
      "num_tokens": 8996096.0,
      "step": 1100
    },
    {
      "entropy": 0.0821349940635264,
      "epoch": 1.7037037037037037,
      "grad_norm": 0.0924430713057518,
      "learning_rate": 4.156274235153189e-05,
      "loss": 0.0815,
      "mean_token_accuracy": 0.9795667895674706,
      "num_tokens": 9404772.0,
      "step": 1150
    },
    {
      "entropy": 0.08265135439112782,
      "epoch": 1.7777777777777777,
      "grad_norm": 0.05620681121945381,
      "learning_rate": 3.765228830469794e-05,
      "loss": 0.082,
      "mean_token_accuracy": 0.9795762729644776,
      "num_tokens": 9813274.0,
      "step": 1200
    },
    {
      "entropy": 0.08118220455944539,
      "epoch": 1.8518518518518519,
      "grad_norm": 0.07570701092481613,
      "learning_rate": 3.3820776916908857e-05,
      "loss": 0.0809,
      "mean_token_accuracy": 0.9796575351059437,
      "num_tokens": 10222375.0,
      "step": 1250
    },
    {
      "entropy": 0.08083926324732602,
      "epoch": 1.925925925925926,
      "grad_norm": 0.060712188482284546,
      "learning_rate": 3.0092704200428058e-05,
      "loss": 0.0802,
      "mean_token_accuracy": 0.9798884338140488,
      "num_tokens": 10631168.0,
      "step": 1300
    },
    {
      "entropy": 0.080772600248456,
      "epoch": 2.0,
      "grad_norm": 0.07269751280546188,
      "learning_rate": 2.649190485277792e-05,
      "loss": 0.0796,
      "mean_token_accuracy": 0.9800221265852451,
      "num_tokens": 11040558.0,
      "step": 1350
    },
    {
      "entropy": 0.08113629410974682,
      "epoch": 2.074074074074074,
      "grad_norm": 0.062111906707286835,
      "learning_rate": 2.3041399874302905e-05,
      "loss": 0.081,
      "mean_token_accuracy": 0.979648227095604,
      "num_tokens": 11449495.0,
      "step": 1400
    },
    {
      "entropy": 0.08001561598852276,
      "epoch": 2.148148148148148,
      "grad_norm": 0.05966750532388687,
      "learning_rate": 1.976324938794482e-05,
      "loss": 0.0797,
      "mean_token_accuracy": 0.9802096697688103,
      "num_tokens": 11858437.0,
      "step": 1450
    },
    {
      "entropy": 0.08092246975749731,
      "epoch": 2.2222222222222223,
      "grad_norm": 0.18608996272087097,
      "learning_rate": 1.667841160219835e-05,
      "loss": 0.0801,
      "mean_token_accuracy": 0.980140280276537,
      "num_tokens": 12267300.0,
      "step": 1500
    },
    {
      "entropy": 0.08059992772992701,
      "epoch": 2.2962962962962963,
      "grad_norm": 0.11162838339805603,
      "learning_rate": 1.3806608818939203e-05,
      "loss": 0.0804,
      "mean_token_accuracy": 0.9798828169703484,
      "num_tokens": 12676684.0,
      "step": 1550
    },
    {
      "entropy": 0.07961415925063192,
      "epoch": 2.3703703703703702,
      "grad_norm": 0.07311662286520004,
      "learning_rate": 1.1166201342777438e-05,
      "loss": 0.0794,
      "mean_token_accuracy": 0.9800658442080021,
      "num_tokens": 13085168.0,
      "step": 1600
    },
    {
      "entropy": 0.08146837581880391,
      "epoch": 2.4444444444444446,
      "grad_norm": 0.06994790583848953,
      "learning_rate": 8.774070098071668e-06,
      "loss": 0.0809,
      "mean_token_accuracy": 0.9797783401608468,
      "num_tokens": 13493932.0,
      "step": 1650
    },
    {
      "entropy": 0.08123130862601101,
      "epoch": 2.5185185185185186,
      "grad_norm": 0.05672051012516022,
      "learning_rate": 6.645508704069003e-06,
      "loss": 0.0803,
      "mean_token_accuracy": 0.9800009173154831,
      "num_tokens": 13902471.0,
      "step": 1700
    },
    {
      "entropy": 0.0805498613230884,
      "epoch": 2.5925925925925926,
      "grad_norm": 0.16635239124298096,
      "learning_rate": 4.794125698167262e-06,
      "loss": 0.0794,
      "mean_token_accuracy": 0.9803060111403465,
      "num_tokens": 14311608.0,
      "step": 1750
    },
    {
      "entropy": 0.08146414666436613,
      "epoch": 2.6666666666666665,
      "grad_norm": 0.06827282160520554,
      "learning_rate": 3.231757532415458e-06,
      "loss": 0.081,
      "mean_token_accuracy": 0.9797043937444687,
      "num_tokens": 14720611.0,
      "step": 1800
    },
    {
      "entropy": 0.07917289328761398,
      "epoch": 2.7407407407407405,
      "grad_norm": 0.059548377990722656,
      "learning_rate": 1.9683928994924385e-06,
      "loss": 0.0793,
      "mean_token_accuracy": 0.9801959751546383,
      "num_tokens": 15129409.0,
      "step": 1850
    },
    {
      "entropy": 0.08008888635784388,
      "epoch": 2.814814814814815,
      "grad_norm": 0.09395159035921097,
      "learning_rate": 1.0121088719706296e-06,
      "loss": 0.0799,
      "mean_token_accuracy": 0.9799904689192772,
      "num_tokens": 15537648.0,
      "step": 1900
    },
    {
      "entropy": 0.0811097707413137,
      "epoch": 2.888888888888889,
      "grad_norm": 0.06502337753772736,
      "learning_rate": 3.6901926314575894e-07,
      "loss": 0.0806,
      "mean_token_accuracy": 0.9798469330370426,
      "num_tokens": 15947039.0,
      "step": 1950
    },
    {
      "entropy": 0.0801746395137161,
      "epoch": 2.962962962962963,
      "grad_norm": 0.06600493937730789,
      "learning_rate": 4.323553957759629e-08,
      "loss": 0.0798,
      "mean_token_accuracy": 0.9801910315454007,
      "num_tokens": 16356235.0,
      "step": 2000
    }
  ],
  "logging_steps": 50,
  "max_steps": 2025,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 6.832435895635661e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}