{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 625,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 1.5365671901141895,
      "epoch": 0.0016,
      "grad_norm": 4.21875,
      "learning_rate": 0.0,
      "loss": 5.494313716888428,
      "mean_token_accuracy": 0.6573571685482474,
      "num_tokens": 348701.0,
      "step": 1
    },
    {
      "entropy": 1.348903712299135,
      "epoch": 0.016,
      "grad_norm": 2.59375,
      "learning_rate": 3.157894736842105e-05,
      "loss": 5.544558631049262,
      "mean_token_accuracy": 0.6651431231035126,
      "num_tokens": 371496.0,
      "step": 10
    },
    {
      "entropy": 1.360134594142437,
      "epoch": 0.032,
      "grad_norm": 3.296875,
      "learning_rate": 6.666666666666667e-05,
      "loss": 5.742576599121094,
      "mean_token_accuracy": 0.6662841200828552,
      "num_tokens": 400839.0,
      "step": 20
    },
    {
      "entropy": 1.375771728157997,
      "epoch": 0.048,
      "grad_norm": 3.140625,
      "learning_rate": 0.0001017543859649123,
      "loss": 5.714028167724609,
      "mean_token_accuracy": 0.6566085889935493,
      "num_tokens": 431025.0,
      "step": 30
    },
    {
      "entropy": 1.2748979568481444,
      "epoch": 0.064,
      "grad_norm": 2.828125,
      "learning_rate": 0.0001368421052631579,
      "loss": 5.11612548828125,
      "mean_token_accuracy": 0.6816287323832512,
      "num_tokens": 459089.0,
      "step": 40
    },
    {
      "entropy": 1.3155348658561707,
      "epoch": 0.08,
      "grad_norm": 3.15625,
      "learning_rate": 0.00017192982456140353,
      "loss": 5.35040512084961,
      "mean_token_accuracy": 0.6721501812338829,
      "num_tokens": 484648.0,
      "step": 50
    },
    {
      "entropy": 1.1791150525212288,
      "epoch": 0.096,
      "grad_norm": 2.375,
      "learning_rate": 0.00019999940277008808,
      "loss": 4.642951583862304,
      "mean_token_accuracy": 0.7028968930244446,
      "num_tokens": 511463.0,
      "step": 60
    },
    {
      "entropy": 1.1177749201655387,
      "epoch": 0.112,
      "grad_norm": 2.421875,
      "learning_rate": 0.0001999785004721968,
      "loss": 4.523546981811523,
      "mean_token_accuracy": 0.7057401552796364,
      "num_tokens": 538438.0,
      "step": 70
    },
    {
      "entropy": 1.1784485399723053,
      "epoch": 0.128,
      "grad_norm": 2.25,
      "learning_rate": 0.00019992774381199778,
      "loss": 4.712202072143555,
      "mean_token_accuracy": 0.7004493281245232,
      "num_tokens": 569421.0,
      "step": 80
    },
    {
      "entropy": 1.0666756927967072,
      "epoch": 0.144,
      "grad_norm": 4.25,
      "learning_rate": 0.00019984714794582683,
      "loss": 4.181539154052734,
      "mean_token_accuracy": 0.7218294009566307,
      "num_tokens": 595806.0,
      "step": 90
    },
    {
      "entropy": 1.1410984337329864,
      "epoch": 0.16,
      "grad_norm": 2.59375,
      "learning_rate": 0.00019973673694024,
      "loss": 4.578577041625977,
      "mean_token_accuracy": 0.70914496332407,
      "num_tokens": 626238.0,
      "step": 100
    },
    {
      "entropy": 1.058976437151432,
      "epoch": 0.176,
      "grad_norm": 4.34375,
      "learning_rate": 0.0001995965437648273,
      "loss": 4.2924964904785154,
      "mean_token_accuracy": 0.7180401891469955,
      "num_tokens": 653441.0,
      "step": 110
    },
    {
      "entropy": 1.2249037250876427,
      "epoch": 0.192,
      "grad_norm": 3.96875,
      "learning_rate": 0.00019942661028236745,
      "loss": 5.084912490844727,
      "mean_token_accuracy": 0.6847136527299881,
      "num_tokens": 682582.0,
      "step": 120
    },
    {
      "entropy": 1.1726351112127305,
      "epoch": 0.208,
      "grad_norm": 3.296875,
      "learning_rate": 0.00019922698723632767,
      "loss": 5.018433380126953,
      "mean_token_accuracy": 0.7055773109197616,
      "num_tokens": 710468.0,
      "step": 130
    },
    {
      "entropy": 1.3469643473625184,
      "epoch": 0.224,
      "grad_norm": 2.875,
      "learning_rate": 0.000198997734235711,
      "loss": 5.7476848602294925,
      "mean_token_accuracy": 0.6790369123220443,
      "num_tokens": 740979.0,
      "step": 140
    },
    {
      "entropy": 1.187085197865963,
      "epoch": 0.24,
      "grad_norm": 2.359375,
      "learning_rate": 0.0001987389197372567,
      "loss": 5.085608673095703,
      "mean_token_accuracy": 0.7057169727981091,
      "num_tokens": 768427.0,
      "step": 150
    },
    {
      "entropy": 1.2938067495822907,
      "epoch": 0.256,
      "grad_norm": 2.359375,
      "learning_rate": 0.0001984506210249986,
      "loss": 5.570721054077149,
      "mean_token_accuracy": 0.6840770006179809,
      "num_tokens": 796744.0,
      "step": 160
    },
    {
      "entropy": 1.1161220327019692,
      "epoch": 0.272,
      "grad_norm": 2.265625,
      "learning_rate": 0.00019813292418718732,
      "loss": 4.821302795410157,
      "mean_token_accuracy": 0.7168306574225426,
      "num_tokens": 822225.0,
      "step": 170
    },
    {
      "entropy": 1.2217196941375732,
      "epoch": 0.288,
      "grad_norm": 2.484375,
      "learning_rate": 0.00019778592409058378,
      "loss": 5.171672821044922,
      "mean_token_accuracy": 0.6998881995677948,
      "num_tokens": 848049.0,
      "step": 180
    },
    {
      "entropy": 1.1091715931892394,
      "epoch": 0.304,
      "grad_norm": 3.109375,
      "learning_rate": 0.00019740972435213115,
      "loss": 4.743664169311524,
      "mean_token_accuracy": 0.7203769966959953,
      "num_tokens": 873172.0,
      "step": 190
    },
    {
      "entropy": 1.2154748886823654,
      "epoch": 0.32,
      "grad_norm": 2.21875,
      "learning_rate": 0.00019700443730801413,
      "loss": 5.212123107910156,
      "mean_token_accuracy": 0.7076091229915619,
      "num_tokens": 901716.0,
      "step": 200
    },
    {
      "entropy": 1.161145742237568,
      "epoch": 0.336,
      "grad_norm": 2.078125,
      "learning_rate": 0.00019657018398011434,
      "loss": 4.889891052246094,
      "mean_token_accuracy": 0.7145821407437325,
      "num_tokens": 928506.0,
      "step": 210
    },
    {
      "entropy": 1.1209837168455123,
      "epoch": 0.352,
      "grad_norm": 2.09375,
      "learning_rate": 0.00019610709403987246,
      "loss": 4.862021636962891,
      "mean_token_accuracy": 0.7152845054864884,
      "num_tokens": 958723.0,
      "step": 220
    },
    {
      "entropy": 1.1322738587856294,
      "epoch": 0.368,
      "grad_norm": 2.203125,
      "learning_rate": 0.00019561530576956703,
      "loss": 4.911846923828125,
      "mean_token_accuracy": 0.7134342208504677,
      "num_tokens": 987751.0,
      "step": 230
    },
    {
      "entropy": 1.1780440196394921,
      "epoch": 0.384,
      "grad_norm": 2.421875,
      "learning_rate": 0.00019509496602102252,
      "loss": 5.0671440124511715,
      "mean_token_accuracy": 0.7104542285203934,
      "num_tokens": 1018315.0,
      "step": 240
    },
    {
      "entropy": 1.1776200592517854,
      "epoch": 0.4,
      "grad_norm": 2.359375,
      "learning_rate": 0.00019454623017175812,
      "loss": 5.145317459106446,
      "mean_token_accuracy": 0.7086253471672534,
      "num_tokens": 1046627.0,
      "step": 250
    },
    {
      "entropy": 1.1513659209012985,
      "epoch": 0.416,
      "grad_norm": 2.140625,
      "learning_rate": 0.00019396926207859084,
      "loss": 4.8750255584716795,
      "mean_token_accuracy": 0.7203947469592095,
      "num_tokens": 1075027.0,
      "step": 260
    },
    {
      "entropy": 1.1364729449152946,
      "epoch": 0.432,
      "grad_norm": 2.265625,
      "learning_rate": 0.00019336423402870653,
      "loss": 4.866281890869141,
      "mean_token_accuracy": 0.7150357633829116,
      "num_tokens": 1102334.0,
      "step": 270
    },
    {
      "entropy": 1.108440762758255,
      "epoch": 0.448,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00019273132668821364,
      "loss": 4.697020721435547,
      "mean_token_accuracy": 0.7265092357993126,
      "num_tokens": 1131415.0,
      "step": 280
    },
    {
      "entropy": 1.1001321360468865,
      "epoch": 0.464,
      "grad_norm": 2.125,
      "learning_rate": 0.00019207072904819486,
      "loss": 4.747102355957031,
      "mean_token_accuracy": 0.7231316924095154,
      "num_tokens": 1158280.0,
      "step": 290
    },
    {
      "entropy": 1.1048941344022751,
      "epoch": 0.48,
      "grad_norm": 2.234375,
      "learning_rate": 0.00019138263836827288,
      "loss": 4.644550323486328,
      "mean_token_accuracy": 0.7270571634173393,
      "num_tokens": 1188651.0,
      "step": 300
    },
    {
      "entropy": 1.0945911705493927,
      "epoch": 0.496,
      "grad_norm": 1.984375,
      "learning_rate": 0.00019066726011770726,
      "loss": 4.7493640899658205,
      "mean_token_accuracy": 0.7258608743548394,
      "num_tokens": 1216504.0,
      "step": 310
    },
    {
      "entropy": 1.0356923416256905,
      "epoch": 0.512,
      "grad_norm": 2.1875,
      "learning_rate": 0.00018992480791403958,
      "loss": 4.3843944549560545,
      "mean_token_accuracy": 0.7372462660074234,
      "num_tokens": 1243732.0,
      "step": 320
    },
    {
      "entropy": 1.0773125097155571,
      "epoch": 0.528,
      "grad_norm": 2.328125,
      "learning_rate": 0.0001891555034593055,
      "loss": 4.6180767059326175,
      "mean_token_accuracy": 0.7283408164978027,
      "num_tokens": 1269272.0,
      "step": 330
    },
    {
      "entropy": 1.050987295806408,
      "epoch": 0.544,
      "grad_norm": 1.640625,
      "learning_rate": 0.00018835957647383303,
      "loss": 4.4863533020019535,
      "mean_token_accuracy": 0.7359548002481461,
      "num_tokens": 1296001.0,
      "step": 340
    },
    {
      "entropy": 1.0461898490786552,
      "epoch": 0.56,
      "grad_norm": 2.453125,
      "learning_rate": 0.000187537264627646,
      "loss": 4.434906005859375,
      "mean_token_accuracy": 0.7349108412861824,
      "num_tokens": 1322180.0,
      "step": 350
    },
    {
      "entropy": 1.0046504065394402,
      "epoch": 0.576,
      "grad_norm": 2.125,
      "learning_rate": 0.00018668881346949417,
      "loss": 4.309226989746094,
      "mean_token_accuracy": 0.73991359770298,
      "num_tokens": 1347532.0,
      "step": 360
    },
    {
      "entropy": 1.0861794739961623,
      "epoch": 0.592,
      "grad_norm": 1.890625,
      "learning_rate": 0.0001858144763535302,
      "loss": 4.599692535400391,
      "mean_token_accuracy": 0.7285229310393333,
      "num_tokens": 1373584.0,
      "step": 370
    },
    {
      "entropy": 1.0600647926330566,
      "epoch": 0.608,
      "grad_norm": 2.203125,
      "learning_rate": 0.00018491451436365627,
      "loss": 4.5491493225097654,
      "mean_token_accuracy": 0.7357145607471466,
      "num_tokens": 1399594.0,
      "step": 380
    },
    {
      "entropy": 1.098224511742592,
      "epoch": 0.624,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00018398919623556238,
      "loss": 4.8250572204589846,
      "mean_token_accuracy": 0.7251888766884804,
      "num_tokens": 1428792.0,
      "step": 390
    },
    {
      "entropy": 1.1232757449150086,
      "epoch": 0.64,
      "grad_norm": 2.03125,
      "learning_rate": 0.00018303879827647975,
      "loss": 4.712010192871094,
      "mean_token_accuracy": 0.7287421196699142,
      "num_tokens": 1457889.0,
      "step": 400
    },
    {
      "entropy": 1.0411028936505318,
      "epoch": 0.656,
      "grad_norm": 1.859375,
      "learning_rate": 0.00018206360428267332,
      "loss": 4.405958938598633,
      "mean_token_accuracy": 0.7350577011704444,
      "num_tokens": 1484479.0,
      "step": 410
    },
    {
      "entropy": 1.1146624743938447,
      "epoch": 0.672,
      "grad_norm": 1.734375,
      "learning_rate": 0.00018106390545469795,
      "loss": 4.742184448242187,
      "mean_token_accuracy": 0.7282734125852585,
      "num_tokens": 1514268.0,
      "step": 420
    },
    {
      "entropy": 1.0480840012431145,
      "epoch": 0.688,
      "grad_norm": 1.78125,
      "learning_rate": 0.0001800400003104436,
      "loss": 4.4207916259765625,
      "mean_token_accuracy": 0.7365467861294747,
      "num_tokens": 1540474.0,
      "step": 430
    },
    {
      "entropy": 1.0067586719989776,
      "epoch": 0.704,
      "grad_norm": 1.921875,
      "learning_rate": 0.0001789921945959958,
      "loss": 4.31513671875,
      "mean_token_accuracy": 0.7437789484858512,
      "num_tokens": 1568557.0,
      "step": 440
    },
    {
      "entropy": 1.0166094586253167,
      "epoch": 0.72,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0001779208011943371,
      "loss": 4.292739486694336,
      "mean_token_accuracy": 0.7471857726573944,
      "num_tokens": 1594403.0,
      "step": 450
    },
    {
      "entropy": 1.0385540708899499,
      "epoch": 0.736,
      "grad_norm": 2.75,
      "learning_rate": 0.00017682614003191807,
      "loss": 4.415458679199219,
      "mean_token_accuracy": 0.7400641202926636,
      "num_tokens": 1624111.0,
      "step": 460
    },
    {
      "entropy": 1.0706985503435136,
      "epoch": 0.752,
      "grad_norm": 2.265625,
      "learning_rate": 0.0001757085379831246,
      "loss": 4.554729080200195,
      "mean_token_accuracy": 0.7342943042516709,
      "num_tokens": 1652609.0,
      "step": 470
    },
    {
      "entropy": 0.9689434483647347,
      "epoch": 0.768,
      "grad_norm": 2.03125,
      "learning_rate": 0.00017456832877267084,
      "loss": 4.2303211212158205,
      "mean_token_accuracy": 0.7474748462438583,
      "num_tokens": 1678589.0,
      "step": 480
    },
    {
      "entropy": 1.1837652444839477,
      "epoch": 0.784,
      "grad_norm": 2.078125,
      "learning_rate": 0.00017340585287594604,
      "loss": 4.968061447143555,
      "mean_token_accuracy": 0.7166377156972885,
      "num_tokens": 1708513.0,
      "step": 490
    },
    {
      "entropy": 1.042286379635334,
      "epoch": 0.8,
      "grad_norm": 2.515625,
      "learning_rate": 0.00017222145741734626,
      "loss": 4.422880172729492,
      "mean_token_accuracy": 0.7405225187540054,
      "num_tokens": 1736283.0,
      "step": 500
    },
    {
      "entropy": 0.9832519263029098,
      "epoch": 0.816,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00017101549606662024,
      "loss": 4.119276428222657,
      "mean_token_accuracy": 0.7495438739657402,
      "num_tokens": 1765017.0,
      "step": 510
    },
    {
      "entropy": 0.9872105062007904,
      "epoch": 0.832,
      "grad_norm": 1.71875,
      "learning_rate": 0.00016978832893326074,
      "loss": 4.234106826782226,
      "mean_token_accuracy": 0.7473902180790901,
      "num_tokens": 1790682.0,
      "step": 520
    },
    {
      "entropy": 0.9845283895730972,
      "epoch": 0.848,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00016854032245897308,
      "loss": 4.146430969238281,
      "mean_token_accuracy": 0.7520387843251228,
      "num_tokens": 1820050.0,
      "step": 530
    },
    {
      "entropy": 1.0613365799188614,
      "epoch": 0.864,
      "grad_norm": 2.046875,
      "learning_rate": 0.00016727184930825288,
      "loss": 4.48931655883789,
      "mean_token_accuracy": 0.7396015107631684,
      "num_tokens": 1847079.0,
      "step": 540
    },
    {
      "entropy": 0.9684726029634476,
      "epoch": 0.88,
      "grad_norm": 2.03125,
      "learning_rate": 0.00016598328825710533,
      "loss": 4.12475357055664,
      "mean_token_accuracy": 0.7526269048452378,
      "num_tokens": 1872383.0,
      "step": 550
    },
    {
      "entropy": 0.9990086019039154,
      "epoch": 0.896,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00016467502407993992,
      "loss": 4.360863494873047,
      "mean_token_accuracy": 0.7387492001056671,
      "num_tokens": 1903630.0,
      "step": 560
    },
    {
      "entropy": 0.9708257809281349,
      "epoch": 0.912,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00016334744743467364,
      "loss": 4.054442596435547,
      "mean_token_accuracy": 0.7545965671539306,
      "num_tokens": 1930592.0,
      "step": 570
    },
    {
      "entropy": 1.0035204842686654,
      "epoch": 0.928,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00016200095474607753,
      "loss": 4.25194206237793,
      "mean_token_accuracy": 0.7464351058006287,
      "num_tokens": 1960212.0,
      "step": 580
    },
    {
      "entropy": 0.9635734960436821,
      "epoch": 0.944,
      "grad_norm": 1.53125,
      "learning_rate": 0.00016063594808740113,
      "loss": 4.0587310791015625,
      "mean_token_accuracy": 0.7547481089830399,
      "num_tokens": 1986049.0,
      "step": 590
    },
    {
      "entropy": 0.9986386775970459,
      "epoch": 0.96,
      "grad_norm": 1.703125,
      "learning_rate": 0.0001592528350603103,
      "loss": 4.331460952758789,
      "mean_token_accuracy": 0.7503352165222168,
      "num_tokens": 2013266.0,
      "step": 600
    },
    {
      "entropy": 1.006168755888939,
      "epoch": 0.976,
      "grad_norm": 1.734375,
      "learning_rate": 0.00015785202867317407,
      "loss": 4.238505554199219,
      "mean_token_accuracy": 0.7441465124487877,
      "num_tokens": 2041824.0,
      "step": 610
    },
    {
      "entropy": 1.0482723653316497,
      "epoch": 0.992,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0001564339472177373,
      "loss": 4.394336700439453,
      "mean_token_accuracy": 0.7416493371129036,
      "num_tokens": 2071693.0,
      "step": 620
    },
    {
      "epoch": 1.0,
      "eval_entropy": 1.0188530464172363,
      "eval_loss": 1.0276753902435303,
      "eval_mean_token_accuracy": 0.7517141411304474,
      "eval_num_tokens": 2084150.0,
      "eval_runtime": 72.9791,
      "eval_samples_per_second": 13.703,
      "eval_steps_per_second": 3.426,
      "step": 625
    }
  ],
  "logging_steps": 10,
  "max_steps": 1875,
  "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": 5.400684024323021e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}