{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.35362014690451204,
  "eval_steps": 500,
  "global_step": 337,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.005246589716684155,
      "grad_norm": 8.869540214538574,
      "learning_rate": 8.401166180758017e-06,
      "loss": 0.9257,
      "mean_token_accuracy": 0.8380203127861023,
      "num_tokens": 289686.0,
      "step": 5
    },
    {
      "epoch": 0.01049317943336831,
      "grad_norm": 1.9420901536941528,
      "learning_rate": 1.8902623906705538e-05,
      "loss": 0.5266,
      "mean_token_accuracy": 0.8593480110168457,
      "num_tokens": 584447.0,
      "step": 10
    },
    {
      "epoch": 0.015739769150052464,
      "grad_norm": 1.2173879146575928,
      "learning_rate": 2.9404081632653062e-05,
      "loss": 0.3377,
      "mean_token_accuracy": 0.894003939628601,
      "num_tokens": 865283.0,
      "step": 15
    },
    {
      "epoch": 0.02098635886673662,
      "grad_norm": 0.9192031621932983,
      "learning_rate": 3.990553935860058e-05,
      "loss": 0.3091,
      "mean_token_accuracy": 0.9075875878334045,
      "num_tokens": 1150898.0,
      "step": 20
    },
    {
      "epoch": 0.026232948583420776,
      "grad_norm": 0.8734805583953857,
      "learning_rate": 5.0406997084548104e-05,
      "loss": 0.2471,
      "mean_token_accuracy": 0.9180729031562805,
      "num_tokens": 1456620.0,
      "step": 25
    },
    {
      "epoch": 0.03147953830010493,
      "grad_norm": 0.7508789300918579,
      "learning_rate": 6.090845481049563e-05,
      "loss": 0.2155,
      "mean_token_accuracy": 0.9300779223442077,
      "num_tokens": 1751457.0,
      "step": 30
    },
    {
      "epoch": 0.03672612801678909,
      "grad_norm": 0.8536575436592102,
      "learning_rate": 7.140991253644314e-05,
      "loss": 0.2623,
      "mean_token_accuracy": 0.9183142781257629,
      "num_tokens": 2039245.0,
      "step": 35
    },
    {
      "epoch": 0.04197271773347324,
      "grad_norm": 0.868010938167572,
      "learning_rate": 7.350762136663497e-05,
      "loss": 0.2239,
      "mean_token_accuracy": 0.9314786314964294,
      "num_tokens": 2330149.0,
      "step": 40
    },
    {
      "epoch": 0.0472193074501574,
      "grad_norm": 0.7415319681167603,
      "learning_rate": 7.349712991638426e-05,
      "loss": 0.2124,
      "mean_token_accuracy": 0.935946810245514,
      "num_tokens": 2627934.0,
      "step": 45
    },
    {
      "epoch": 0.05246589716684155,
      "grad_norm": 0.6589397192001343,
      "learning_rate": 7.347857138053374e-05,
      "loss": 0.2258,
      "mean_token_accuracy": 0.9339556813240051,
      "num_tokens": 2909298.0,
      "step": 50
    },
    {
      "epoch": 0.05771248688352571,
      "grad_norm": 0.8945339322090149,
      "learning_rate": 7.345195119268144e-05,
      "loss": 0.2135,
      "mean_token_accuracy": 0.9362031579017639,
      "num_tokens": 3215160.0,
      "step": 55
    },
    {
      "epoch": 0.06295907660020986,
      "grad_norm": 0.6015563607215881,
      "learning_rate": 7.341727714672879e-05,
      "loss": 0.223,
      "mean_token_accuracy": 0.9350167393684388,
      "num_tokens": 3512700.0,
      "step": 60
    },
    {
      "epoch": 0.06820566631689402,
      "grad_norm": 0.9376577138900757,
      "learning_rate": 7.337455939459886e-05,
      "loss": 0.1949,
      "mean_token_accuracy": 0.9446848630905151,
      "num_tokens": 3798243.0,
      "step": 65
    },
    {
      "epoch": 0.07345225603357818,
      "grad_norm": 0.8647239804267883,
      "learning_rate": 7.332381044326387e-05,
      "loss": 0.192,
      "mean_token_accuracy": 0.9473262071609497,
      "num_tokens": 4088918.0,
      "step": 70
    },
    {
      "epoch": 0.07869884575026233,
      "grad_norm": 0.6758660078048706,
      "learning_rate": 7.326504515108353e-05,
      "loss": 0.1608,
      "mean_token_accuracy": 0.9533717513084412,
      "num_tokens": 4373400.0,
      "step": 75
    },
    {
      "epoch": 0.08394543546694648,
      "grad_norm": 0.6870194673538208,
      "learning_rate": 7.319828072345472e-05,
      "loss": 0.1719,
      "mean_token_accuracy": 0.947171425819397,
      "num_tokens": 4658256.0,
      "step": 80
    },
    {
      "epoch": 0.08919202518363065,
      "grad_norm": 0.677169680595398,
      "learning_rate": 7.312353670777409e-05,
      "loss": 0.2406,
      "mean_token_accuracy": 0.9317703485488892,
      "num_tokens": 4942712.0,
      "step": 85
    },
    {
      "epoch": 0.0944386149003148,
      "grad_norm": 0.7355192303657532,
      "learning_rate": 7.304083498771491e-05,
      "loss": 0.1559,
      "mean_token_accuracy": 0.9547448277473449,
      "num_tokens": 5236965.0,
      "step": 90
    },
    {
      "epoch": 0.09968520461699895,
      "grad_norm": 0.8258501291275024,
      "learning_rate": 7.295019977681995e-05,
      "loss": 0.1484,
      "mean_token_accuracy": 0.9531127452850342,
      "num_tokens": 5531311.0,
      "step": 95
    },
    {
      "epoch": 0.1049317943336831,
      "grad_norm": 0.6302676200866699,
      "learning_rate": 7.285165761141225e-05,
      "loss": 0.1302,
      "mean_token_accuracy": 0.9609502911567688,
      "num_tokens": 5824530.0,
      "step": 100
    },
    {
      "epoch": 0.11017838405036726,
      "grad_norm": 0.7048377990722656,
      "learning_rate": 7.274523734282567e-05,
      "loss": 0.1655,
      "mean_token_accuracy": 0.9510802745819091,
      "num_tokens": 6126073.0,
      "step": 105
    },
    {
      "epoch": 0.11542497376705142,
      "grad_norm": 0.708620011806488,
      "learning_rate": 7.263097012895783e-05,
      "loss": 0.1919,
      "mean_token_accuracy": 0.9440585255622864,
      "num_tokens": 6428088.0,
      "step": 110
    },
    {
      "epoch": 0.12067156348373557,
      "grad_norm": 0.7772646546363831,
      "learning_rate": 7.250888942514763e-05,
      "loss": 0.1791,
      "mean_token_accuracy": 0.9519138216972352,
      "num_tokens": 6709759.0,
      "step": 115
    },
    {
      "epoch": 0.1259181532004197,
      "grad_norm": 0.728533148765564,
      "learning_rate": 7.237903097438013e-05,
      "loss": 0.1794,
      "mean_token_accuracy": 0.9493731379508972,
      "num_tokens": 7009928.0,
      "step": 120
    },
    {
      "epoch": 0.1311647429171039,
      "grad_norm": 0.5031924843788147,
      "learning_rate": 7.224143279682167e-05,
      "loss": 0.1966,
      "mean_token_accuracy": 0.9432877659797668,
      "num_tokens": 7299912.0,
      "step": 125
    },
    {
      "epoch": 0.13641133263378805,
      "grad_norm": 0.7311460375785828,
      "learning_rate": 7.209613517868828e-05,
      "loss": 0.1623,
      "mean_token_accuracy": 0.9553608179092408,
      "num_tokens": 7587766.0,
      "step": 130
    },
    {
      "epoch": 0.1416579223504722,
      "grad_norm": 0.4527396261692047,
      "learning_rate": 7.194318066045057e-05,
      "loss": 0.1552,
      "mean_token_accuracy": 0.9542324781417847,
      "num_tokens": 7883332.0,
      "step": 135
    },
    {
      "epoch": 0.14690451206715635,
      "grad_norm": 0.5695252418518066,
      "learning_rate": 7.178261402437872e-05,
      "loss": 0.1409,
      "mean_token_accuracy": 0.9568296670913696,
      "num_tokens": 8184963.0,
      "step": 140
    },
    {
      "epoch": 0.1521511017838405,
      "grad_norm": 0.47259852290153503,
      "learning_rate": 7.161448228143099e-05,
      "loss": 0.1465,
      "mean_token_accuracy": 0.9609991312026978,
      "num_tokens": 8481153.0,
      "step": 145
    },
    {
      "epoch": 0.15739769150052466,
      "grad_norm": 0.5303276181221008,
      "learning_rate": 7.143883465748988e-05,
      "loss": 0.183,
      "mean_token_accuracy": 0.9521325826644897,
      "num_tokens": 8775155.0,
      "step": 150
    },
    {
      "epoch": 0.1626442812172088,
      "grad_norm": 0.46558475494384766,
      "learning_rate": 7.125572257894959e-05,
      "loss": 0.205,
      "mean_token_accuracy": 0.9444481015205384,
      "num_tokens": 9071761.0,
      "step": 155
    },
    {
      "epoch": 0.16789087093389296,
      "grad_norm": 0.6111699342727661,
      "learning_rate": 7.106519965765934e-05,
      "loss": 0.1717,
      "mean_token_accuracy": 0.9499832153320312,
      "num_tokens": 9363710.0,
      "step": 160
    },
    {
      "epoch": 0.1731374606505771,
      "grad_norm": 0.7169287204742432,
      "learning_rate": 7.086732167522687e-05,
      "loss": 0.2031,
      "mean_token_accuracy": 0.9435471892356873,
      "num_tokens": 9650095.0,
      "step": 165
    },
    {
      "epoch": 0.1783840503672613,
      "grad_norm": 0.41854023933410645,
      "learning_rate": 7.066214656668653e-05,
      "loss": 0.128,
      "mean_token_accuracy": 0.9642794132232666,
      "num_tokens": 9943925.0,
      "step": 170
    },
    {
      "epoch": 0.18363064008394545,
      "grad_norm": 0.5462023019790649,
      "learning_rate": 7.044973440353702e-05,
      "loss": 0.1692,
      "mean_token_accuracy": 0.9510787844657898,
      "num_tokens": 10239935.0,
      "step": 175
    },
    {
      "epoch": 0.1888772298006296,
      "grad_norm": 0.682062566280365,
      "learning_rate": 7.023014737615354e-05,
      "loss": 0.0925,
      "mean_token_accuracy": 0.9718472838401795,
      "num_tokens": 10542283.0,
      "step": 180
    },
    {
      "epoch": 0.19412381951731375,
      "grad_norm": 0.4446072578430176,
      "learning_rate": 7.000344977557959e-05,
      "loss": 0.1831,
      "mean_token_accuracy": 0.9482220768928528,
      "num_tokens": 10833066.0,
      "step": 185
    },
    {
      "epoch": 0.1993704092339979,
      "grad_norm": 0.35908645391464233,
      "learning_rate": 6.976970797470374e-05,
      "loss": 0.1553,
      "mean_token_accuracy": 0.9547260999679565,
      "num_tokens": 11138524.0,
      "step": 190
    },
    {
      "epoch": 0.20461699895068206,
      "grad_norm": 0.5753610134124756,
      "learning_rate": 6.95289904088269e-05,
      "loss": 0.1783,
      "mean_token_accuracy": 0.9493900299072265,
      "num_tokens": 11434453.0,
      "step": 195
    },
    {
      "epoch": 0.2098635886673662,
      "grad_norm": 0.5665861964225769,
      "learning_rate": 6.928136755562561e-05,
      "loss": 0.1653,
      "mean_token_accuracy": 0.9550609827041626,
      "num_tokens": 11730328.0,
      "step": 200
    },
    {
      "epoch": 0.21511017838405036,
      "grad_norm": 0.5666924118995667,
      "learning_rate": 6.902691191451765e-05,
      "loss": 0.1927,
      "mean_token_accuracy": 0.9518372297286988,
      "num_tokens": 12015188.0,
      "step": 205
    },
    {
      "epoch": 0.2203567681007345,
      "grad_norm": 0.41499361395835876,
      "learning_rate": 6.876569798543542e-05,
      "loss": 0.1149,
      "mean_token_accuracy": 0.9646952271461486,
      "num_tokens": 12310232.0,
      "step": 210
    },
    {
      "epoch": 0.22560335781741866,
      "grad_norm": 0.44563302397727966,
      "learning_rate": 6.84978022470137e-05,
      "loss": 0.1195,
      "mean_token_accuracy": 0.9647434711456299,
      "num_tokens": 12591060.0,
      "step": 215
    },
    {
      "epoch": 0.23084994753410285,
      "grad_norm": 0.9801908731460571,
      "learning_rate": 6.822330313419822e-05,
      "loss": 0.1501,
      "mean_token_accuracy": 0.9570110678672791,
      "num_tokens": 12879679.0,
      "step": 220
    },
    {
      "epoch": 0.236096537250787,
      "grad_norm": 0.590543806552887,
      "learning_rate": 6.79422810152813e-05,
      "loss": 0.1879,
      "mean_token_accuracy": 0.949954628944397,
      "num_tokens": 13169419.0,
      "step": 225
    },
    {
      "epoch": 0.24134312696747115,
      "grad_norm": 0.5752514004707336,
      "learning_rate": 6.765481816837146e-05,
      "loss": 0.1507,
      "mean_token_accuracy": 0.9556933283805847,
      "num_tokens": 13471416.0,
      "step": 230
    },
    {
      "epoch": 0.2465897166841553,
      "grad_norm": 0.47486424446105957,
      "learning_rate": 6.736099875730397e-05,
      "loss": 0.1675,
      "mean_token_accuracy": 0.956990647315979,
      "num_tokens": 13767085.0,
      "step": 235
    },
    {
      "epoch": 0.2518363064008394,
      "grad_norm": 0.7542303800582886,
      "learning_rate": 6.706090880699916e-05,
      "loss": 0.1798,
      "mean_token_accuracy": 0.9498309254646301,
      "num_tokens": 14071045.0,
      "step": 240
    },
    {
      "epoch": 0.2570828961175236,
      "grad_norm": 0.5474414825439453,
      "learning_rate": 6.675463617827598e-05,
      "loss": 0.1698,
      "mean_token_accuracy": 0.9535765051841736,
      "num_tokens": 14362641.0,
      "step": 245
    },
    {
      "epoch": 0.2623294858342078,
      "grad_norm": 0.5496485829353333,
      "learning_rate": 6.644227054212787e-05,
      "loss": 0.1457,
      "mean_token_accuracy": 0.9581978678703308,
      "num_tokens": 14664981.0,
      "step": 250
    },
    {
      "epoch": 0.2675760755508919,
      "grad_norm": 0.49473029375076294,
      "learning_rate": 6.612390335346883e-05,
      "loss": 0.155,
      "mean_token_accuracy": 0.9565532088279725,
      "num_tokens": 14959011.0,
      "step": 255
    },
    {
      "epoch": 0.2728226652675761,
      "grad_norm": 0.303487628698349,
      "learning_rate": 6.579962782435706e-05,
      "loss": 0.1509,
      "mean_token_accuracy": 0.9600463390350342,
      "num_tokens": 15245374.0,
      "step": 260
    },
    {
      "epoch": 0.2780692549842602,
      "grad_norm": 0.5282312631607056,
      "learning_rate": 6.546953889670418e-05,
      "loss": 0.1702,
      "mean_token_accuracy": 0.9531951785087586,
      "num_tokens": 15525453.0,
      "step": 265
    },
    {
      "epoch": 0.2833158447009444,
      "grad_norm": 0.28557926416397095,
      "learning_rate": 6.513373321447803e-05,
      "loss": 0.1937,
      "mean_token_accuracy": 0.9494660019874572,
      "num_tokens": 15809199.0,
      "step": 270
    },
    {
      "epoch": 0.2885624344176285,
      "grad_norm": 0.4175744652748108,
      "learning_rate": 6.479230909540701e-05,
      "loss": 0.125,
      "mean_token_accuracy": 0.9632657885551452,
      "num_tokens": 16101826.0,
      "step": 275
    },
    {
      "epoch": 0.2938090241343127,
      "grad_norm": 0.46477630734443665,
      "learning_rate": 6.444536650219457e-05,
      "loss": 0.1477,
      "mean_token_accuracy": 0.9573351144790649,
      "num_tokens": 16390495.0,
      "step": 280
    },
    {
      "epoch": 0.2990556138509968,
      "grad_norm": 0.40762028098106384,
      "learning_rate": 6.409300701325186e-05,
      "loss": 0.1447,
      "mean_token_accuracy": 0.9613166928291321,
      "num_tokens": 16685503.0,
      "step": 285
    },
    {
      "epoch": 0.304302203567681,
      "grad_norm": 0.24077750742435455,
      "learning_rate": 6.37353337929575e-05,
      "loss": 0.172,
      "mean_token_accuracy": 0.9532236933708191,
      "num_tokens": 16964030.0,
      "step": 290
    },
    {
      "epoch": 0.3095487932843652,
      "grad_norm": 0.826016366481781,
      "learning_rate": 6.337245156145292e-05,
      "loss": 0.1803,
      "mean_token_accuracy": 0.955608320236206,
      "num_tokens": 17262831.0,
      "step": 295
    },
    {
      "epoch": 0.3147953830010493,
      "grad_norm": 0.4392148554325104,
      "learning_rate": 6.300446656398228e-05,
      "loss": 0.1769,
      "mean_token_accuracy": 0.950160825252533,
      "num_tokens": 17542500.0,
      "step": 300
    },
    {
      "epoch": 0.3200419727177335,
      "grad_norm": 0.5865193009376526,
      "learning_rate": 6.263148653978572e-05,
      "loss": 0.1245,
      "mean_token_accuracy": 0.9670758128166199,
      "num_tokens": 17833796.0,
      "step": 305
    },
    {
      "epoch": 0.3252885624344176,
      "grad_norm": 0.5191839337348938,
      "learning_rate": 6.225362069055539e-05,
      "loss": 0.192,
      "mean_token_accuracy": 0.947128975391388,
      "num_tokens": 18131801.0,
      "step": 310
    },
    {
      "epoch": 0.3305351521511018,
      "grad_norm": 0.41704511642456055,
      "learning_rate": 6.187097964846317e-05,
      "loss": 0.1353,
      "mean_token_accuracy": 0.9619265556335449,
      "num_tokens": 18429696.0,
      "step": 315
    },
    {
      "epoch": 0.3357817418677859,
      "grad_norm": 0.5766462087631226,
      "learning_rate": 6.148367544376953e-05,
      "loss": 0.1647,
      "mean_token_accuracy": 0.9554041147232055,
      "num_tokens": 18726485.0,
      "step": 320
    },
    {
      "epoch": 0.3410283315844701,
      "grad_norm": 0.4433589577674866,
      "learning_rate": 6.109182147202329e-05,
      "loss": 0.1643,
      "mean_token_accuracy": 0.9574645280838012,
      "num_tokens": 19017396.0,
      "step": 325
    },
    {
      "epoch": 0.3462749213011542,
      "grad_norm": 0.4165564179420471,
      "learning_rate": 6.069553246086131e-05,
      "loss": 0.1669,
      "mean_token_accuracy": 0.9528910636901855,
      "num_tokens": 19320603.0,
      "step": 330
    },
    {
      "epoch": 0.3515215110178384,
      "grad_norm": 0.6303921341896057,
      "learning_rate": 6.0294924436418496e-05,
      "loss": 0.1796,
      "mean_token_accuracy": 0.9498880624771118,
      "num_tokens": 19599038.0,
      "step": 335
    }
  ],
  "logging_steps": 5,
  "max_steps": 953,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "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": 4.871895722016768e+17,
  "train_batch_size": 24,
  "trial_name": null,
  "trial_params": null
}