{
  "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": 1.9540774886310102,
      "epoch": 0.07407407407407407,
      "grad_norm": 0.8241696953773499,
      "learning_rate": 8.032786885245902e-05,
      "loss": 1.7451,
      "mean_token_accuracy": 0.6189013833552599,
      "num_tokens": 379953.0,
      "step": 50
    },
    {
      "entropy": 0.5046265083923935,
      "epoch": 0.14814814814814814,
      "grad_norm": 4.424446105957031,
      "learning_rate": 9.990765991730485e-05,
      "loss": 0.4302,
      "mean_token_accuracy": 0.8861582314968109,
      "num_tokens": 757047.0,
      "step": 100
    },
    {
      "entropy": 0.24775007627904416,
      "epoch": 0.2222222222222222,
      "grad_norm": 0.30575165152549744,
      "learning_rate": 9.950545603782162e-05,
      "loss": 0.2323,
      "mean_token_accuracy": 0.940535937026143,
      "num_tokens": 1134378.0,
      "step": 150
    },
    {
      "entropy": 0.1925063591822982,
      "epoch": 0.2962962962962963,
      "grad_norm": 0.24072137475013733,
      "learning_rate": 9.878674879048427e-05,
      "loss": 0.1856,
      "mean_token_accuracy": 0.9529062640666962,
      "num_tokens": 1516152.0,
      "step": 200
    },
    {
      "entropy": 0.18322961997240783,
      "epoch": 0.37037037037037035,
      "grad_norm": 0.18854060769081116,
      "learning_rate": 9.775613308830824e-05,
      "loss": 0.1775,
      "mean_token_accuracy": 0.9546041788160801,
      "num_tokens": 1896494.0,
      "step": 250
    },
    {
      "entropy": 0.17187482433393597,
      "epoch": 0.4444444444444444,
      "grad_norm": 0.18019668757915497,
      "learning_rate": 9.642019796948866e-05,
      "loss": 0.164,
      "mean_token_accuracy": 0.9579361644387245,
      "num_tokens": 2277194.0,
      "step": 300
    },
    {
      "entropy": 0.16193925650790333,
      "epoch": 0.5185185185185185,
      "grad_norm": 0.23759332299232483,
      "learning_rate": 9.478748447168449e-05,
      "loss": 0.1539,
      "mean_token_accuracy": 0.9614081564545631,
      "num_tokens": 2658243.0,
      "step": 350
    },
    {
      "entropy": 0.15039873549714686,
      "epoch": 0.5925925925925926,
      "grad_norm": 0.2590758204460144,
      "learning_rate": 9.28684310265789e-05,
      "loss": 0.1443,
      "mean_token_accuracy": 0.9650989197194576,
      "num_tokens": 3041430.0,
      "step": 400
    },
    {
      "entropy": 0.14228019634261727,
      "epoch": 0.6666666666666666,
      "grad_norm": 0.18062612414360046,
      "learning_rate": 9.067530672382544e-05,
      "loss": 0.1366,
      "mean_token_accuracy": 0.9667340110242367,
      "num_tokens": 3422368.0,
      "step": 450
    },
    {
      "entropy": 0.1391275341436267,
      "epoch": 0.7407407407407407,
      "grad_norm": 0.1984509527683258,
      "learning_rate": 8.822213287104348e-05,
      "loss": 0.1352,
      "mean_token_accuracy": 0.9674024738371372,
      "num_tokens": 3801968.0,
      "step": 500
    },
    {
      "entropy": 0.13673329239711166,
      "epoch": 0.8148148148148148,
      "grad_norm": 0.13088534772396088,
      "learning_rate": 8.552459335135381e-05,
      "loss": 0.1354,
      "mean_token_accuracy": 0.9673544447124004,
      "num_tokens": 4182102.0,
      "step": 550
    },
    {
      "entropy": 0.13533455861732363,
      "epoch": 0.8888888888888888,
      "grad_norm": 0.33587542176246643,
      "learning_rate": 8.259993435156559e-05,
      "loss": 0.1325,
      "mean_token_accuracy": 0.9676820485293866,
      "num_tokens": 4561998.0,
      "step": 600
    },
    {
      "entropy": 0.13281562993302942,
      "epoch": 0.9629629629629629,
      "grad_norm": 0.11626797914505005,
      "learning_rate": 7.946685410208296e-05,
      "loss": 0.1313,
      "mean_token_accuracy": 0.9678142921626568,
      "num_tokens": 4936347.0,
      "step": 650
    },
    {
      "entropy": 0.1338025047816336,
      "epoch": 1.037037037037037,
      "grad_norm": 0.10883153975009918,
      "learning_rate": 7.614538333345735e-05,
      "loss": 0.1328,
      "mean_token_accuracy": 0.9673172944784164,
      "num_tokens": 5313396.0,
      "step": 700
    },
    {
      "entropy": 0.1330988533422351,
      "epoch": 1.1111111111111112,
      "grad_norm": 0.14333182573318481,
      "learning_rate": 7.265675721386285e-05,
      "loss": 0.1317,
      "mean_token_accuracy": 0.9678454534709453,
      "num_tokens": 5694821.0,
      "step": 750
    },
    {
      "entropy": 0.1315150342695415,
      "epoch": 1.1851851851851851,
      "grad_norm": 0.07893866300582886,
      "learning_rate": 6.902327958623736e-05,
      "loss": 0.131,
      "mean_token_accuracy": 0.9678919970989227,
      "num_tokens": 6071147.0,
      "step": 800
    },
    {
      "entropy": 0.1348141137883067,
      "epoch": 1.2592592592592593,
      "grad_norm": 0.10548827797174454,
      "learning_rate": 6.526818037306228e-05,
      "loss": 0.1337,
      "mean_token_accuracy": 0.9671075843274594,
      "num_tokens": 6454454.0,
      "step": 850
    },
    {
      "entropy": 0.12924523117020725,
      "epoch": 1.3333333333333333,
      "grad_norm": 0.13734276592731476,
      "learning_rate": 6.14154670604355e-05,
      "loss": 0.1294,
      "mean_token_accuracy": 0.9686036820709706,
      "num_tokens": 6831407.0,
      "step": 900
    },
    {
      "entropy": 0.13089272173121572,
      "epoch": 1.4074074074074074,
      "grad_norm": 0.062484513968229294,
      "learning_rate": 5.7489771210944564e-05,
      "loss": 0.1304,
      "mean_token_accuracy": 0.9679886139929295,
      "num_tokens": 7211253.0,
      "step": 950
    },
    {
      "entropy": 0.1295284123532474,
      "epoch": 1.4814814814814814,
      "grad_norm": 0.07434429228305817,
      "learning_rate": 5.351619098663021e-05,
      "loss": 0.1289,
      "mean_token_accuracy": 0.9682336232066154,
      "num_tokens": 7591172.0,
      "step": 1000
    },
    {
      "entropy": 0.13286457041278482,
      "epoch": 1.5555555555555556,
      "grad_norm": 0.1009301245212555,
      "learning_rate": 4.952013068883795e-05,
      "loss": 0.1326,
      "mean_token_accuracy": 0.9677493931353092,
      "num_tokens": 7976360.0,
      "step": 1050
    },
    {
      "entropy": 0.13181614426895977,
      "epoch": 1.6296296296296298,
      "grad_norm": 0.09136403352022171,
      "learning_rate": 4.5527138340828776e-05,
      "loss": 0.1317,
      "mean_token_accuracy": 0.9678143452107907,
      "num_tokens": 8355146.0,
      "step": 1100
    },
    {
      "entropy": 0.12889527762308717,
      "epoch": 1.7037037037037037,
      "grad_norm": 0.11126290261745453,
      "learning_rate": 4.156274235153189e-05,
      "loss": 0.1289,
      "mean_token_accuracy": 0.9681750671565532,
      "num_tokens": 8736037.0,
      "step": 1150
    },
    {
      "entropy": 0.1293605554662645,
      "epoch": 1.7777777777777777,
      "grad_norm": 0.10993292182683945,
      "learning_rate": 3.765228830469794e-05,
      "loss": 0.1293,
      "mean_token_accuracy": 0.9684934197366237,
      "num_tokens": 9113158.0,
      "step": 1200
    },
    {
      "entropy": 0.12954492604359985,
      "epoch": 1.8518518518518519,
      "grad_norm": 0.1266999989748001,
      "learning_rate": 3.3820776916908857e-05,
      "loss": 0.1296,
      "mean_token_accuracy": 0.9680785122513771,
      "num_tokens": 9492245.0,
      "step": 1250
    },
    {
      "entropy": 0.13069430900737644,
      "epoch": 1.925925925925926,
      "grad_norm": 0.07351736724376678,
      "learning_rate": 3.0092704200428058e-05,
      "loss": 0.131,
      "mean_token_accuracy": 0.9679618345201015,
      "num_tokens": 9874560.0,
      "step": 1300
    },
    {
      "entropy": 0.12986605327576398,
      "epoch": 2.0,
      "grad_norm": 0.07602707296609879,
      "learning_rate": 2.649190485277792e-05,
      "loss": 0.1301,
      "mean_token_accuracy": 0.9679882827401162,
      "num_tokens": 10251446.0,
      "step": 1350
    },
    {
      "entropy": 0.13015099691227078,
      "epoch": 2.074074074074074,
      "grad_norm": 0.07422789186239243,
      "learning_rate": 2.3041399874302905e-05,
      "loss": 0.13,
      "mean_token_accuracy": 0.968013653755188,
      "num_tokens": 10637110.0,
      "step": 1400
    },
    {
      "entropy": 0.12833521047607063,
      "epoch": 2.148148148148148,
      "grad_norm": 0.06928899884223938,
      "learning_rate": 1.976324938794482e-05,
      "loss": 0.1287,
      "mean_token_accuracy": 0.9684618780016899,
      "num_tokens": 11014301.0,
      "step": 1450
    },
    {
      "entropy": 0.12765121564269066,
      "epoch": 2.2222222222222223,
      "grad_norm": 0.07737194001674652,
      "learning_rate": 1.667841160219835e-05,
      "loss": 0.1279,
      "mean_token_accuracy": 0.9685468013584614,
      "num_tokens": 11390895.0,
      "step": 1500
    },
    {
      "entropy": 0.13076237022876738,
      "epoch": 2.2962962962962963,
      "grad_norm": 0.0690469890832901,
      "learning_rate": 1.3806608818939203e-05,
      "loss": 0.1308,
      "mean_token_accuracy": 0.9680638153851032,
      "num_tokens": 11773360.0,
      "step": 1550
    },
    {
      "entropy": 0.1271959487348795,
      "epoch": 2.3703703703703702,
      "grad_norm": 0.08058126270771027,
      "learning_rate": 1.1166201342777438e-05,
      "loss": 0.1275,
      "mean_token_accuracy": 0.9684904217720032,
      "num_tokens": 12152692.0,
      "step": 1600
    },
    {
      "entropy": 0.12783636916428803,
      "epoch": 2.4444444444444446,
      "grad_norm": 0.08844149112701416,
      "learning_rate": 8.774070098071668e-06,
      "loss": 0.1278,
      "mean_token_accuracy": 0.9686374716460705,
      "num_tokens": 12532074.0,
      "step": 1650
    },
    {
      "entropy": 0.12949045987799765,
      "epoch": 2.5185185185185186,
      "grad_norm": 0.0735594779253006,
      "learning_rate": 6.645508704069003e-06,
      "loss": 0.13,
      "mean_token_accuracy": 0.9678716999292374,
      "num_tokens": 12914270.0,
      "step": 1700
    },
    {
      "entropy": 0.12908031923696398,
      "epoch": 2.5925925925925926,
      "grad_norm": 0.0713183730840683,
      "learning_rate": 4.794125698167262e-06,
      "loss": 0.1291,
      "mean_token_accuracy": 0.9681964771449566,
      "num_tokens": 13293683.0,
      "step": 1750
    },
    {
      "entropy": 0.12857461655512453,
      "epoch": 2.6666666666666665,
      "grad_norm": 0.06526947021484375,
      "learning_rate": 3.231757532415458e-06,
      "loss": 0.1287,
      "mean_token_accuracy": 0.9684041538834571,
      "num_tokens": 13671470.0,
      "step": 1800
    },
    {
      "entropy": 0.12981407037004827,
      "epoch": 2.7407407407407405,
      "grad_norm": 0.06781283766031265,
      "learning_rate": 1.9683928994924385e-06,
      "loss": 0.1304,
      "mean_token_accuracy": 0.9678549686074257,
      "num_tokens": 14051413.0,
      "step": 1850
    },
    {
      "entropy": 0.1281242691539228,
      "epoch": 2.814814814814815,
      "grad_norm": 0.06520246714353561,
      "learning_rate": 1.0121088719706296e-06,
      "loss": 0.1291,
      "mean_token_accuracy": 0.968293984234333,
      "num_tokens": 14429791.0,
      "step": 1900
    },
    {
      "entropy": 0.12915834257379175,
      "epoch": 2.888888888888889,
      "grad_norm": 0.07606221735477448,
      "learning_rate": 3.6901926314575894e-07,
      "loss": 0.1292,
      "mean_token_accuracy": 0.9681269869208335,
      "num_tokens": 14808357.0,
      "step": 1950
    },
    {
      "entropy": 0.1287903120368719,
      "epoch": 2.962962962962963,
      "grad_norm": 0.07113504409790039,
      "learning_rate": 4.323553957759629e-08,
      "loss": 0.1286,
      "mean_token_accuracy": 0.9684909208118916,
      "num_tokens": 15187872.0,
      "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": 6.344379487768412e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}