{
  "best_global_step": null,
  "best_metric": Infinity,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 40,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 0.8637090027332306,
      "epoch": 0.02531645569620253,
      "grad_norm": 99.94273376464844,
      "learning_rate": 0.0,
      "loss": 15.196868896484375,
      "mean_token_accuracy": 0.28961846977472305,
      "num_tokens": 7286.0,
      "step": 1
    },
    {
      "entropy": 0.9094592481851578,
      "epoch": 0.05063291139240506,
      "grad_norm": 95.93968200683594,
      "learning_rate": 5e-05,
      "loss": 14.926185607910156,
      "mean_token_accuracy": 0.27897340804338455,
      "num_tokens": 14954.0,
      "step": 2
    },
    {
      "entropy": 1.1258608400821686,
      "epoch": 0.0759493670886076,
      "grad_norm": 64.25172424316406,
      "learning_rate": 0.0001,
      "loss": 12.606666564941406,
      "mean_token_accuracy": 0.35700511932373047,
      "num_tokens": 21366.0,
      "step": 3
    },
    {
      "entropy": 1.0687015354633331,
      "epoch": 0.10126582278481013,
      "grad_norm": 36.90766906738281,
      "learning_rate": 0.00015000000000000001,
      "loss": 9.619781494140625,
      "mean_token_accuracy": 0.540277749300003,
      "num_tokens": 28250.0,
      "step": 4
    },
    {
      "entropy": 1.1662852466106415,
      "epoch": 0.12658227848101267,
      "grad_norm": 29.526512145996094,
      "learning_rate": 0.0002,
      "loss": 8.147899627685547,
      "mean_token_accuracy": 0.567672997713089,
      "num_tokens": 34914.0,
      "step": 5
    },
    {
      "entropy": 0.727982223033905,
      "epoch": 0.1518987341772152,
      "grad_norm": 13.896689414978027,
      "learning_rate": 0.0001999145758387301,
      "loss": 7.228446006774902,
      "mean_token_accuracy": 0.6021915972232819,
      "num_tokens": 42532.0,
      "step": 6
    },
    {
      "entropy": 1.246531367301941,
      "epoch": 0.17721518987341772,
      "grad_norm": 19.760786056518555,
      "learning_rate": 0.000199658449300667,
      "loss": 6.584984302520752,
      "mean_token_accuracy": 0.6384725868701935,
      "num_tokens": 49872.0,
      "step": 7
    },
    {
      "entropy": 1.175276130437851,
      "epoch": 0.20253164556962025,
      "grad_norm": 12.119710922241211,
      "learning_rate": 0.0001992320579737045,
      "loss": 6.310988426208496,
      "mean_token_accuracy": 0.6375854164361954,
      "num_tokens": 55639.0,
      "step": 8
    },
    {
      "entropy": 1.0149184465408325,
      "epoch": 0.22784810126582278,
      "grad_norm": 7.977464199066162,
      "learning_rate": 0.00019863613034027224,
      "loss": 6.036639213562012,
      "mean_token_accuracy": 0.666093721985817,
      "num_tokens": 61649.0,
      "step": 9
    },
    {
      "entropy": 1.022455170750618,
      "epoch": 0.25316455696202533,
      "grad_norm": 7.014733791351318,
      "learning_rate": 0.00019787168453273544,
      "loss": 5.827225685119629,
      "mean_token_accuracy": 0.6803832501173019,
      "num_tokens": 69689.0,
      "step": 10
    },
    {
      "entropy": 1.1049264669418335,
      "epoch": 0.27848101265822783,
      "grad_norm": 5.545648574829102,
      "learning_rate": 0.00019694002659393305,
      "loss": 5.525129795074463,
      "mean_token_accuracy": 0.704516813158989,
      "num_tokens": 77528.0,
      "step": 11
    },
    {
      "entropy": 0.8492173552513123,
      "epoch": 0.3037974683544304,
      "grad_norm": 6.59067964553833,
      "learning_rate": 0.0001958427482458253,
      "loss": 5.324199199676514,
      "mean_token_accuracy": 0.8436015695333481,
      "num_tokens": 85064.0,
      "step": 12
    },
    {
      "entropy": 0.6838414371013641,
      "epoch": 0.3291139240506329,
      "grad_norm": 3.6837315559387207,
      "learning_rate": 0.00019458172417006347,
      "loss": 5.420000076293945,
      "mean_token_accuracy": 0.816602423787117,
      "num_tokens": 91009.0,
      "step": 13
    },
    {
      "entropy": 0.6191327422857285,
      "epoch": 0.35443037974683544,
      "grad_norm": 3.5020408630371094,
      "learning_rate": 0.0001931591088051279,
      "loss": 5.348326683044434,
      "mean_token_accuracy": 0.8181768357753754,
      "num_tokens": 97909.0,
      "step": 14
    },
    {
      "entropy": 0.5245884731411934,
      "epoch": 0.379746835443038,
      "grad_norm": 3.6154603958129883,
      "learning_rate": 0.00019157733266550575,
      "loss": 5.308533191680908,
      "mean_token_accuracy": 0.8219696879386902,
      "num_tokens": 104187.0,
      "step": 15
    },
    {
      "entropy": 0.4121474325656891,
      "epoch": 0.4050632911392405,
      "grad_norm": 2.822946071624756,
      "learning_rate": 0.0001898390981891979,
      "loss": 5.135751247406006,
      "mean_token_accuracy": 0.8321253657341003,
      "num_tokens": 110300.0,
      "step": 16
    },
    {
      "entropy": 0.392690971493721,
      "epoch": 0.43037974683544306,
      "grad_norm": 5.22876501083374,
      "learning_rate": 0.0001879473751206489,
      "loss": 5.34346866607666,
      "mean_token_accuracy": 0.8164242208003998,
      "num_tokens": 117157.0,
      "step": 17
    },
    {
      "entropy": 0.3356513977050781,
      "epoch": 0.45569620253164556,
      "grad_norm": 3.489384412765503,
      "learning_rate": 0.00018590539543698854,
      "loss": 4.941544532775879,
      "mean_token_accuracy": 0.8337943404912949,
      "num_tokens": 123378.0,
      "step": 18
    },
    {
      "entropy": 0.32025614380836487,
      "epoch": 0.4810126582278481,
      "grad_norm": 2.2713074684143066,
      "learning_rate": 0.00018371664782625287,
      "loss": 5.1571879386901855,
      "mean_token_accuracy": 0.8307176530361176,
      "num_tokens": 129042.0,
      "step": 19
    },
    {
      "entropy": 0.3084551766514778,
      "epoch": 0.5063291139240507,
      "grad_norm": 1.9176894426345825,
      "learning_rate": 0.0001813848717270195,
      "loss": 5.014953136444092,
      "mean_token_accuracy": 0.8466087132692337,
      "num_tokens": 135747.0,
      "step": 20
    },
    {
      "entropy": 0.26904235407710075,
      "epoch": 0.5316455696202531,
      "grad_norm": 1.4043225049972534,
      "learning_rate": 0.00017891405093963938,
      "loss": 4.949811935424805,
      "mean_token_accuracy": 0.8474528640508652,
      "num_tokens": 141848.0,
      "step": 21
    },
    {
      "entropy": 0.2922343239188194,
      "epoch": 0.5569620253164557,
      "grad_norm": 1.2163947820663452,
      "learning_rate": 0.00017630840681998066,
      "loss": 5.169545650482178,
      "mean_token_accuracy": 0.8475807756185532,
      "num_tokens": 147566.0,
      "step": 22
    },
    {
      "entropy": 0.2782471589744091,
      "epoch": 0.5822784810126582,
      "grad_norm": 1.0059943199157715,
      "learning_rate": 0.00017357239106731317,
      "loss": 5.164263725280762,
      "mean_token_accuracy": 0.8462631553411484,
      "num_tokens": 154620.0,
      "step": 23
    },
    {
      "entropy": 0.2586139217019081,
      "epoch": 0.6075949367088608,
      "grad_norm": 1.22340726852417,
      "learning_rate": 0.00017071067811865476,
      "loss": 5.076096534729004,
      "mean_token_accuracy": 0.8611445128917694,
      "num_tokens": 161701.0,
      "step": 24
    },
    {
      "entropy": 0.28559913113713264,
      "epoch": 0.6329113924050633,
      "grad_norm": 2.1334292888641357,
      "learning_rate": 0.00016772815716257412,
      "loss": 5.121291160583496,
      "mean_token_accuracy": 0.8396630734205246,
      "num_tokens": 168435.0,
      "step": 25
    },
    {
      "entropy": 0.26606928184628487,
      "epoch": 0.6582278481012658,
      "grad_norm": 0.8918235301971436,
      "learning_rate": 0.00016462992378609407,
      "loss": 5.148040771484375,
      "mean_token_accuracy": 0.8437058180570602,
      "num_tokens": 175650.0,
      "step": 26
    },
    {
      "entropy": 0.23234782740473747,
      "epoch": 0.6835443037974683,
      "grad_norm": 1.8136335611343384,
      "learning_rate": 0.0001614212712689668,
      "loss": 5.150457382202148,
      "mean_token_accuracy": 0.8433783650398254,
      "num_tokens": 181490.0,
      "step": 27
    },
    {
      "entropy": 0.24661532044410706,
      "epoch": 0.7088607594936709,
      "grad_norm": 1.321523904800415,
      "learning_rate": 0.00015810768154019385,
      "loss": 5.084752082824707,
      "mean_token_accuracy": 0.8424803614616394,
      "num_tokens": 188582.0,
      "step": 28
    },
    {
      "entropy": 0.19287967681884766,
      "epoch": 0.7341772151898734,
      "grad_norm": 0.9798034429550171,
      "learning_rate": 0.00015469481581224272,
      "loss": 5.147141456604004,
      "mean_token_accuracy": 0.866110622882843,
      "num_tokens": 193783.0,
      "step": 29
    },
    {
      "entropy": 0.23315827921032906,
      "epoch": 0.759493670886076,
      "grad_norm": 3.1756558418273926,
      "learning_rate": 0.00015118850490896012,
      "loss": 5.3330183029174805,
      "mean_token_accuracy": 0.8204634040594101,
      "num_tokens": 200686.0,
      "step": 30
    },
    {
      "entropy": 0.1938694752752781,
      "epoch": 0.7848101265822784,
      "grad_norm": 1.9134807586669922,
      "learning_rate": 0.00014759473930370736,
      "loss": 5.052967071533203,
      "mean_token_accuracy": 0.8582398444414139,
      "num_tokens": 208440.0,
      "step": 31
    },
    {
      "entropy": 0.22125443816184998,
      "epoch": 0.810126582278481,
      "grad_norm": 1.5088441371917725,
      "learning_rate": 0.00014391965888473703,
      "loss": 5.049187660217285,
      "mean_token_accuracy": 0.8358375877141953,
      "num_tokens": 215131.0,
      "step": 32
    },
    {
      "entropy": 0.20639389753341675,
      "epoch": 0.8354430379746836,
      "grad_norm": 1.491873860359192,
      "learning_rate": 0.00014016954246529696,
      "loss": 5.055438041687012,
      "mean_token_accuracy": 0.8552371561527252,
      "num_tokens": 222195.0,
      "step": 33
    },
    {
      "entropy": 0.20158018916845322,
      "epoch": 0.8607594936708861,
      "grad_norm": 1.46501624584198,
      "learning_rate": 0.00013635079705638298,
      "loss": 5.16477108001709,
      "mean_token_accuracy": 0.8602638840675354,
      "num_tokens": 229926.0,
      "step": 34
    },
    {
      "entropy": 0.21367954835295677,
      "epoch": 0.8860759493670886,
      "grad_norm": 1.7227668762207031,
      "learning_rate": 0.00013246994692046836,
      "loss": 5.240202903747559,
      "mean_token_accuracy": 0.8288360238075256,
      "num_tokens": 235766.0,
      "step": 35
    },
    {
      "entropy": 0.2060464285314083,
      "epoch": 0.9113924050632911,
      "grad_norm": 1.615714192390442,
      "learning_rate": 0.00012853362242491053,
      "loss": 5.13973331451416,
      "mean_token_accuracy": 0.8528883457183838,
      "num_tokens": 242731.0,
      "step": 36
    },
    {
      "entropy": 0.2112499214708805,
      "epoch": 0.9367088607594937,
      "grad_norm": 1.70606529712677,
      "learning_rate": 0.00012454854871407994,
      "loss": 5.105003833770752,
      "mean_token_accuracy": 0.8559168428182602,
      "num_tokens": 249818.0,
      "step": 37
    },
    {
      "entropy": 0.1963396780192852,
      "epoch": 0.9620253164556962,
      "grad_norm": 1.8689205646514893,
      "learning_rate": 0.00012052153421956342,
      "loss": 5.083700180053711,
      "mean_token_accuracy": 0.8450468629598618,
      "num_tokens": 256043.0,
      "step": 38
    },
    {
      "entropy": 0.17459847778081894,
      "epoch": 0.9873417721518988,
      "grad_norm": 1.3199875354766846,
      "learning_rate": 0.00011645945902807341,
      "loss": 5.136896133422852,
      "mean_token_accuracy": 0.8544726371765137,
      "num_tokens": 263628.0,
      "step": 39
    },
    {
      "entropy": 0.19893981516361237,
      "epoch": 1.0,
      "grad_norm": 2.03135347366333,
      "learning_rate": 0.00011236926312693479,
      "loss": 5.0191426277160645,
      "mean_token_accuracy": 0.8476890921592712,
      "num_tokens": 266935.0,
      "step": 40
    },
    {
      "epoch": 1.0,
      "eval_entropy": 0.0,
      "eval_loss": NaN,
      "eval_mean_token_accuracy": 0.0,
      "eval_num_tokens": 266935.0,
      "eval_runtime": 5.2068,
      "eval_samples_per_second": 42.637,
      "eval_steps_per_second": 10.755,
      "step": 40
    }
  ],
  "logging_steps": 1,
  "max_steps": 80,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 558442705059840.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}