{
  "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": 0.8529030084609985,
      "epoch": 0.02531645569620253,
      "grad_norm": 78.52010345458984,
      "learning_rate": 0.0,
      "loss": 15.181689262390137,
      "mean_token_accuracy": 0.28961846977472305,
      "num_tokens": 7286.0,
      "step": 1
    },
    {
      "entropy": 0.8989771902561188,
      "epoch": 0.05063291139240506,
      "grad_norm": 75.26119232177734,
      "learning_rate": 5e-05,
      "loss": 14.95038890838623,
      "mean_token_accuracy": 0.27897340804338455,
      "num_tokens": 14954.0,
      "step": 2
    },
    {
      "entropy": 0.9821003973484039,
      "epoch": 0.0759493670886076,
      "grad_norm": 74.54902648925781,
      "learning_rate": 0.0001,
      "loss": 13.742369651794434,
      "mean_token_accuracy": 0.2827275022864342,
      "num_tokens": 21366.0,
      "step": 3
    },
    {
      "entropy": 1.0873801112174988,
      "epoch": 0.10126582278481013,
      "grad_norm": 34.19524383544922,
      "learning_rate": 0.00015000000000000001,
      "loss": 11.25832462310791,
      "mean_token_accuracy": 0.44820327311754227,
      "num_tokens": 28250.0,
      "step": 4
    },
    {
      "entropy": 1.0790044069290161,
      "epoch": 0.12658227848101267,
      "grad_norm": 28.061630249023438,
      "learning_rate": 0.0002,
      "loss": 9.99978256225586,
      "mean_token_accuracy": 0.5489031970500946,
      "num_tokens": 34914.0,
      "step": 5
    },
    {
      "entropy": 1.1546767055988312,
      "epoch": 0.1518987341772152,
      "grad_norm": 24.789119720458984,
      "learning_rate": 0.0001999145758387301,
      "loss": 8.632344245910645,
      "mean_token_accuracy": 0.554060235619545,
      "num_tokens": 42532.0,
      "step": 6
    },
    {
      "entropy": 1.0596064329147339,
      "epoch": 0.17721518987341772,
      "grad_norm": 12.310446739196777,
      "learning_rate": 0.000199658449300667,
      "loss": 7.437180995941162,
      "mean_token_accuracy": 0.5873144865036011,
      "num_tokens": 49872.0,
      "step": 7
    },
    {
      "entropy": 0.7889958769083023,
      "epoch": 0.20253164556962025,
      "grad_norm": 10.70737361907959,
      "learning_rate": 0.0001992320579737045,
      "loss": 7.076646327972412,
      "mean_token_accuracy": 0.5939096212387085,
      "num_tokens": 55639.0,
      "step": 8
    },
    {
      "entropy": 0.9904351979494095,
      "epoch": 0.22784810126582278,
      "grad_norm": 39.2529296875,
      "learning_rate": 0.00019863613034027224,
      "loss": 7.060567855834961,
      "mean_token_accuracy": 0.6403393149375916,
      "num_tokens": 61649.0,
      "step": 9
    },
    {
      "entropy": 1.3839358687400818,
      "epoch": 0.25316455696202533,
      "grad_norm": 15.882004737854004,
      "learning_rate": 0.00019787168453273544,
      "loss": 6.710319519042969,
      "mean_token_accuracy": 0.603007972240448,
      "num_tokens": 69689.0,
      "step": 10
    },
    {
      "entropy": 1.1738347113132477,
      "epoch": 0.27848101265822783,
      "grad_norm": 8.31102180480957,
      "learning_rate": 0.00019694002659393305,
      "loss": 6.315446853637695,
      "mean_token_accuracy": 0.627836138010025,
      "num_tokens": 77528.0,
      "step": 11
    },
    {
      "entropy": 1.2701810002326965,
      "epoch": 0.3037974683544304,
      "grad_norm": 6.591884136199951,
      "learning_rate": 0.0001958427482458253,
      "loss": 6.042618751525879,
      "mean_token_accuracy": 0.6690841615200043,
      "num_tokens": 85064.0,
      "step": 12
    },
    {
      "entropy": 1.2295758128166199,
      "epoch": 0.3291139240506329,
      "grad_norm": 15.478667259216309,
      "learning_rate": 0.00019458172417006347,
      "loss": 6.213565826416016,
      "mean_token_accuracy": 0.6405076384544373,
      "num_tokens": 91009.0,
      "step": 13
    },
    {
      "entropy": 1.268456757068634,
      "epoch": 0.35443037974683544,
      "grad_norm": 7.625810146331787,
      "learning_rate": 0.0001931591088051279,
      "loss": 6.13183069229126,
      "mean_token_accuracy": 0.6365684270858765,
      "num_tokens": 97909.0,
      "step": 14
    },
    {
      "entropy": 1.2001140415668488,
      "epoch": 0.379746835443038,
      "grad_norm": 7.8025221824646,
      "learning_rate": 0.00019157733266550575,
      "loss": 6.064075469970703,
      "mean_token_accuracy": 0.6330698281526566,
      "num_tokens": 104187.0,
      "step": 15
    },
    {
      "entropy": 1.066677764058113,
      "epoch": 0.4050632911392405,
      "grad_norm": 8.150992393493652,
      "learning_rate": 0.0001898390981891979,
      "loss": 5.798349380493164,
      "mean_token_accuracy": 0.646238312125206,
      "num_tokens": 110300.0,
      "step": 16
    },
    {
      "entropy": 1.0704185664653778,
      "epoch": 0.43037974683544306,
      "grad_norm": 3.722452402114868,
      "learning_rate": 0.0001879473751206489,
      "loss": 5.87784481048584,
      "mean_token_accuracy": 0.6626823246479034,
      "num_tokens": 117157.0,
      "step": 17
    },
    {
      "entropy": 1.1619559824466705,
      "epoch": 0.45569620253164556,
      "grad_norm": 4.019715785980225,
      "learning_rate": 0.00018590539543698854,
      "loss": 5.398146152496338,
      "mean_token_accuracy": 0.714724525809288,
      "num_tokens": 123378.0,
      "step": 18
    },
    {
      "entropy": 1.072197139263153,
      "epoch": 0.4810126582278481,
      "grad_norm": 3.3014461994171143,
      "learning_rate": 0.00018371664782625287,
      "loss": 5.488443851470947,
      "mean_token_accuracy": 0.8233065903186798,
      "num_tokens": 129042.0,
      "step": 19
    },
    {
      "entropy": 0.9806628823280334,
      "epoch": 0.5063291139240507,
      "grad_norm": 2.70117449760437,
      "learning_rate": 0.0001813848717270195,
      "loss": 5.3289875984191895,
      "mean_token_accuracy": 0.8143462389707565,
      "num_tokens": 135747.0,
      "step": 20
    },
    {
      "entropy": 0.8040178865194321,
      "epoch": 0.5316455696202531,
      "grad_norm": 3.4564270973205566,
      "learning_rate": 0.00017891405093963938,
      "loss": 5.2513909339904785,
      "mean_token_accuracy": 0.8225801885128021,
      "num_tokens": 141848.0,
      "step": 21
    },
    {
      "entropy": 0.7444496601819992,
      "epoch": 0.5569620253164557,
      "grad_norm": 2.3326470851898193,
      "learning_rate": 0.00017630840681998066,
      "loss": 5.4000067710876465,
      "mean_token_accuracy": 0.8329752385616302,
      "num_tokens": 147566.0,
      "step": 22
    },
    {
      "entropy": 0.6936303377151489,
      "epoch": 0.5822784810126582,
      "grad_norm": 2.478424072265625,
      "learning_rate": 0.00017357239106731317,
      "loss": 5.346803665161133,
      "mean_token_accuracy": 0.8282431960105896,
      "num_tokens": 154620.0,
      "step": 23
    },
    {
      "entropy": 0.6478566229343414,
      "epoch": 0.6075949367088608,
      "grad_norm": 2.6104896068573,
      "learning_rate": 0.00017071067811865476,
      "loss": 5.191404342651367,
      "mean_token_accuracy": 0.8425344824790955,
      "num_tokens": 161701.0,
      "step": 24
    },
    {
      "entropy": 0.6078712940216064,
      "epoch": 0.6329113924050633,
      "grad_norm": 4.111148357391357,
      "learning_rate": 0.00016772815716257412,
      "loss": 5.243197917938232,
      "mean_token_accuracy": 0.8070437759160995,
      "num_tokens": 168435.0,
      "step": 25
    },
    {
      "entropy": 0.5901128649711609,
      "epoch": 0.6582278481012658,
      "grad_norm": 2.244276285171509,
      "learning_rate": 0.00016462992378609407,
      "loss": 5.272450923919678,
      "mean_token_accuracy": 0.8324518501758575,
      "num_tokens": 175650.0,
      "step": 26
    },
    {
      "entropy": 0.5017133206129074,
      "epoch": 0.6835443037974683,
      "grad_norm": 2.1174073219299316,
      "learning_rate": 0.0001614212712689668,
      "loss": 5.215684413909912,
      "mean_token_accuracy": 0.8321826756000519,
      "num_tokens": 181490.0,
      "step": 27
    },
    {
      "entropy": 0.4748859405517578,
      "epoch": 0.7088607594936709,
      "grad_norm": 2.087001323699951,
      "learning_rate": 0.00015810768154019385,
      "loss": 5.15059757232666,
      "mean_token_accuracy": 0.8388571739196777,
      "num_tokens": 188582.0,
      "step": 28
    },
    {
      "entropy": 0.3888755962252617,
      "epoch": 0.7341772151898734,
      "grad_norm": 1.487061858177185,
      "learning_rate": 0.00015469481581224272,
      "loss": 5.174810409545898,
      "mean_token_accuracy": 0.8587576746940613,
      "num_tokens": 193783.0,
      "step": 29
    },
    {
      "entropy": 0.4288812130689621,
      "epoch": 0.759493670886076,
      "grad_norm": 2.868377685546875,
      "learning_rate": 0.00015118850490896012,
      "loss": 5.343541145324707,
      "mean_token_accuracy": 0.8163273632526398,
      "num_tokens": 200686.0,
      "step": 30
    },
    {
      "entropy": 0.39349576085805893,
      "epoch": 0.7848101265822784,
      "grad_norm": 1.7554408311843872,
      "learning_rate": 0.00014759473930370736,
      "loss": 5.089621543884277,
      "mean_token_accuracy": 0.8510452210903168,
      "num_tokens": 208440.0,
      "step": 31
    },
    {
      "entropy": 0.37267740815877914,
      "epoch": 0.810126582278481,
      "grad_norm": 1.210075855255127,
      "learning_rate": 0.00014391965888473703,
      "loss": 5.0665974617004395,
      "mean_token_accuracy": 0.8576832711696625,
      "num_tokens": 215131.0,
      "step": 32
    },
    {
      "entropy": 0.3740547075867653,
      "epoch": 0.8354430379746836,
      "grad_norm": 2.1259520053863525,
      "learning_rate": 0.00014016954246529696,
      "loss": 5.144742012023926,
      "mean_token_accuracy": 0.8442028760910034,
      "num_tokens": 222195.0,
      "step": 33
    },
    {
      "entropy": 0.33895356953144073,
      "epoch": 0.8607594936708861,
      "grad_norm": 2.1364142894744873,
      "learning_rate": 0.00013635079705638298,
      "loss": 5.227801322937012,
      "mean_token_accuracy": 0.8455548137426376,
      "num_tokens": 229926.0,
      "step": 34
    },
    {
      "entropy": 0.3552927002310753,
      "epoch": 0.8860759493670886,
      "grad_norm": 2.4019367694854736,
      "learning_rate": 0.00013246994692046836,
      "loss": 5.275204181671143,
      "mean_token_accuracy": 0.8322960436344147,
      "num_tokens": 235766.0,
      "step": 35
    },
    {
      "entropy": 0.3192586973309517,
      "epoch": 0.9113924050632911,
      "grad_norm": 1.8943588733673096,
      "learning_rate": 0.00012853362242491053,
      "loss": 5.1949639320373535,
      "mean_token_accuracy": 0.8491553515195847,
      "num_tokens": 242731.0,
      "step": 36
    },
    {
      "entropy": 0.31374089419841766,
      "epoch": 0.9367088607594937,
      "grad_norm": 1.5797899961471558,
      "learning_rate": 0.00012454854871407994,
      "loss": 5.137860298156738,
      "mean_token_accuracy": 0.8447228074073792,
      "num_tokens": 249818.0,
      "step": 37
    },
    {
      "entropy": 0.3000646084547043,
      "epoch": 0.9620253164556962,
      "grad_norm": 1.9470438957214355,
      "learning_rate": 0.00012052153421956342,
      "loss": 5.12597131729126,
      "mean_token_accuracy": 0.8450468629598618,
      "num_tokens": 256043.0,
      "step": 38
    },
    {
      "entropy": 0.2730342224240303,
      "epoch": 0.9873417721518988,
      "grad_norm": 1.3998013734817505,
      "learning_rate": 0.00011645945902807341,
      "loss": 5.182764530181885,
      "mean_token_accuracy": 0.8505663871765137,
      "num_tokens": 263628.0,
      "step": 39
    },
    {
      "entropy": 0.31550464034080505,
      "epoch": 1.0,
      "grad_norm": 1.9481233358383179,
      "learning_rate": 0.00011236926312693479,
      "loss": 5.046781063079834,
      "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": 9.0564,
      "eval_samples_per_second": 24.513,
      "eval_steps_per_second": 6.184,
      "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": 542852652204032.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}