{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 345,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 1.6703570872545241,
      "epoch": 0.087527352297593,
      "grad_norm": 0.65625,
      "learning_rate": 0.00019478260869565218,
      "loss": 2.0278825759887695,
      "mean_token_accuracy": 0.612476097419858,
      "num_tokens": 18728.0,
      "step": 10
    },
    {
      "entropy": 1.2492346689105034,
      "epoch": 0.175054704595186,
      "grad_norm": 0.75,
      "learning_rate": 0.0001889855072463768,
      "loss": 1.2333297729492188,
      "mean_token_accuracy": 0.7266808465123177,
      "num_tokens": 37328.0,
      "step": 20
    },
    {
      "entropy": 1.1282426938414574,
      "epoch": 0.26258205689277897,
      "grad_norm": 0.6796875,
      "learning_rate": 0.00018318840579710147,
      "loss": 1.1297640800476074,
      "mean_token_accuracy": 0.7533167637884617,
      "num_tokens": 56036.0,
      "step": 30
    },
    {
      "entropy": 0.9922276919707655,
      "epoch": 0.350109409190372,
      "grad_norm": 0.640625,
      "learning_rate": 0.0001773913043478261,
      "loss": 1.033622646331787,
      "mean_token_accuracy": 0.7703150801360608,
      "num_tokens": 75072.0,
      "step": 40
    },
    {
      "entropy": 1.0213249990716577,
      "epoch": 0.437636761487965,
      "grad_norm": 0.53125,
      "learning_rate": 0.00017159420289855073,
      "loss": 1.042306900024414,
      "mean_token_accuracy": 0.7690192483365536,
      "num_tokens": 93505.0,
      "step": 50
    },
    {
      "entropy": 0.9650674186646938,
      "epoch": 0.5251641137855579,
      "grad_norm": 0.56640625,
      "learning_rate": 0.00016579710144927536,
      "loss": 1.0220633506774903,
      "mean_token_accuracy": 0.7742902040481567,
      "num_tokens": 112039.0,
      "step": 60
    },
    {
      "entropy": 1.0567748602479696,
      "epoch": 0.612691466083151,
      "grad_norm": 0.59765625,
      "learning_rate": 0.00016,
      "loss": 1.0543122291564941,
      "mean_token_accuracy": 0.7587149851024151,
      "num_tokens": 131322.0,
      "step": 70
    },
    {
      "entropy": 0.9168455514125526,
      "epoch": 0.700218818380744,
      "grad_norm": 0.6328125,
      "learning_rate": 0.00015420289855072464,
      "loss": 0.95663423538208,
      "mean_token_accuracy": 0.78614791482687,
      "num_tokens": 149646.0,
      "step": 80
    },
    {
      "entropy": 0.9514417068101466,
      "epoch": 0.787746170678337,
      "grad_norm": 0.68359375,
      "learning_rate": 0.00014840579710144927,
      "loss": 0.9753911972045899,
      "mean_token_accuracy": 0.781362334638834,
      "num_tokens": 168182.0,
      "step": 90
    },
    {
      "entropy": 0.9971431667916477,
      "epoch": 0.87527352297593,
      "grad_norm": 0.6015625,
      "learning_rate": 0.00014260869565217393,
      "loss": 1.0196686744689942,
      "mean_token_accuracy": 0.7682865738868714,
      "num_tokens": 187221.0,
      "step": 100
    },
    {
      "entropy": 0.9537320947274566,
      "epoch": 0.962800875273523,
      "grad_norm": 0.515625,
      "learning_rate": 0.00013681159420289856,
      "loss": 0.9510069847106933,
      "mean_token_accuracy": 0.782170583307743,
      "num_tokens": 205633.0,
      "step": 110
    },
    {
      "entropy": 0.907145096643551,
      "epoch": 1.0437636761487965,
      "grad_norm": 0.484375,
      "learning_rate": 0.00013101449275362319,
      "loss": 0.9373687744140625,
      "mean_token_accuracy": 0.7886275141625791,
      "num_tokens": 222954.0,
      "step": 120
    },
    {
      "entropy": 0.8457014967687428,
      "epoch": 1.1312910284463895,
      "grad_norm": 0.6328125,
      "learning_rate": 0.00012521739130434784,
      "loss": 0.8454959869384766,
      "mean_token_accuracy": 0.7963103860616684,
      "num_tokens": 241678.0,
      "step": 130
    },
    {
      "entropy": 0.8462455681525171,
      "epoch": 1.2188183807439825,
      "grad_norm": 0.703125,
      "learning_rate": 0.00011942028985507247,
      "loss": 0.8574352264404297,
      "mean_token_accuracy": 0.8001780763268471,
      "num_tokens": 260144.0,
      "step": 140
    },
    {
      "entropy": 0.8343841714784503,
      "epoch": 1.3063457330415755,
      "grad_norm": 0.703125,
      "learning_rate": 0.00011362318840579711,
      "loss": 0.8432804107666015,
      "mean_token_accuracy": 0.7991413235664367,
      "num_tokens": 279020.0,
      "step": 150
    },
    {
      "entropy": 0.7861125588417053,
      "epoch": 1.3938730853391685,
      "grad_norm": 0.7578125,
      "learning_rate": 0.00010782608695652174,
      "loss": 0.8140290260314942,
      "mean_token_accuracy": 0.8104772590100765,
      "num_tokens": 297696.0,
      "step": 160
    },
    {
      "entropy": 0.8265796534717083,
      "epoch": 1.4814004376367615,
      "grad_norm": 0.7421875,
      "learning_rate": 0.00010202898550724637,
      "loss": 0.8296893119812012,
      "mean_token_accuracy": 0.8040477402508259,
      "num_tokens": 315741.0,
      "step": 170
    },
    {
      "entropy": 0.7689724578522146,
      "epoch": 1.5689277899343543,
      "grad_norm": 0.80859375,
      "learning_rate": 9.623188405797103e-05,
      "loss": 0.7881460189819336,
      "mean_token_accuracy": 0.808928444236517,
      "num_tokens": 334515.0,
      "step": 180
    },
    {
      "entropy": 0.7925019094720482,
      "epoch": 1.6564551422319473,
      "grad_norm": 0.7890625,
      "learning_rate": 9.043478260869566e-05,
      "loss": 0.8049691200256348,
      "mean_token_accuracy": 0.8096396245062352,
      "num_tokens": 352933.0,
      "step": 190
    },
    {
      "entropy": 0.7968289134092629,
      "epoch": 1.7439824945295404,
      "grad_norm": 0.859375,
      "learning_rate": 8.463768115942029e-05,
      "loss": 0.8357898712158203,
      "mean_token_accuracy": 0.8038821592926979,
      "num_tokens": 371642.0,
      "step": 200
    },
    {
      "entropy": 0.8211181129328906,
      "epoch": 1.8315098468271334,
      "grad_norm": 0.80078125,
      "learning_rate": 7.884057971014493e-05,
      "loss": 0.8084160804748535,
      "mean_token_accuracy": 0.8103035606443882,
      "num_tokens": 390297.0,
      "step": 210
    },
    {
      "entropy": 0.8177960739005357,
      "epoch": 1.9190371991247264,
      "grad_norm": 0.73828125,
      "learning_rate": 7.304347826086957e-05,
      "loss": 0.8264873504638672,
      "mean_token_accuracy": 0.8067296646535397,
      "num_tokens": 409295.0,
      "step": 220
    },
    {
      "entropy": 0.8154785256530788,
      "epoch": 2.0,
      "grad_norm": 1.6171875,
      "learning_rate": 6.72463768115942e-05,
      "loss": 0.82596435546875,
      "mean_token_accuracy": 0.8030677026993519,
      "num_tokens": 426750.0,
      "step": 230
    },
    {
      "entropy": 0.6059559936635196,
      "epoch": 2.087527352297593,
      "grad_norm": 1.6796875,
      "learning_rate": 6.144927536231884e-05,
      "loss": 0.6000402450561524,
      "mean_token_accuracy": 0.8544376268982887,
      "num_tokens": 445112.0,
      "step": 240
    },
    {
      "entropy": 0.6742822146276012,
      "epoch": 2.175054704595186,
      "grad_norm": 0.9453125,
      "learning_rate": 5.565217391304348e-05,
      "loss": 0.6495694160461426,
      "mean_token_accuracy": 0.8368013821542263,
      "num_tokens": 464055.0,
      "step": 250
    },
    {
      "entropy": 0.6670764476526528,
      "epoch": 2.262582056892779,
      "grad_norm": 1.21875,
      "learning_rate": 4.985507246376812e-05,
      "loss": 0.6383417129516602,
      "mean_token_accuracy": 0.8340263485908508,
      "num_tokens": 483243.0,
      "step": 260
    },
    {
      "entropy": 0.6482093236874789,
      "epoch": 2.350109409190372,
      "grad_norm": 1.0390625,
      "learning_rate": 4.405797101449275e-05,
      "loss": 0.6531811237335206,
      "mean_token_accuracy": 0.8335508003830909,
      "num_tokens": 501911.0,
      "step": 270
    },
    {
      "entropy": 0.6858879348263145,
      "epoch": 2.437636761487965,
      "grad_norm": 1.0546875,
      "learning_rate": 3.8260869565217395e-05,
      "loss": 0.6682425498962402,
      "mean_token_accuracy": 0.8321790546178818,
      "num_tokens": 520579.0,
      "step": 280
    },
    {
      "entropy": 0.6361387457232922,
      "epoch": 2.525164113785558,
      "grad_norm": 1.1171875,
      "learning_rate": 3.246376811594203e-05,
      "loss": 0.628416633605957,
      "mean_token_accuracy": 0.8387672163546085,
      "num_tokens": 539205.0,
      "step": 290
    },
    {
      "entropy": 0.6545329387299716,
      "epoch": 2.612691466083151,
      "grad_norm": 1.015625,
      "learning_rate": 2.6666666666666667e-05,
      "loss": 0.6523668766021729,
      "mean_token_accuracy": 0.8395993508398533,
      "num_tokens": 557726.0,
      "step": 300
    },
    {
      "entropy": 0.5724413711111993,
      "epoch": 2.700218818380744,
      "grad_norm": 1.09375,
      "learning_rate": 2.0869565217391303e-05,
      "loss": 0.5660404682159423,
      "mean_token_accuracy": 0.860739403963089,
      "num_tokens": 575763.0,
      "step": 310
    },
    {
      "entropy": 0.6278647107072175,
      "epoch": 2.787746170678337,
      "grad_norm": 1.0859375,
      "learning_rate": 1.5072463768115944e-05,
      "loss": 0.6288604736328125,
      "mean_token_accuracy": 0.8464278854429722,
      "num_tokens": 594054.0,
      "step": 320
    },
    {
      "entropy": 0.6267774282023311,
      "epoch": 2.87527352297593,
      "grad_norm": 1.03125,
      "learning_rate": 9.27536231884058e-06,
      "loss": 0.6052563190460205,
      "mean_token_accuracy": 0.8475193127989769,
      "num_tokens": 613053.0,
      "step": 330
    },
    {
      "entropy": 0.6937473515979946,
      "epoch": 2.962800875273523,
      "grad_norm": 1.0625,
      "learning_rate": 3.4782608695652175e-06,
      "loss": 0.6603917121887207,
      "mean_token_accuracy": 0.8315372362732887,
      "num_tokens": 632039.0,
      "step": 340
    }
  ],
  "logging_steps": 10,
  "max_steps": 345,
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.7311532200064e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}