{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 43960,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 2.494273384451866,
      "epoch": 0.022747952684258416,
      "grad_norm": 18.375,
      "learning_rate": 6.991084023229653e-05,
      "loss": 2.56,
      "mean_token_accuracy": 0.47454266212880614,
      "num_tokens": 137746.0,
      "step": 1000
    },
    {
      "entropy": 2.1963223298788073,
      "epoch": 0.04549590536851683,
      "grad_norm": 36.75,
      "learning_rate": 6.964345916042913e-05,
      "loss": 2.2203,
      "mean_token_accuracy": 0.5251638530790805,
      "num_tokens": 274474.0,
      "step": 2000
    },
    {
      "entropy": 2.071790741562843,
      "epoch": 0.06824385805277525,
      "grad_norm": 18.625,
      "learning_rate": 6.919922168729661e-05,
      "loss": 2.0846,
      "mean_token_accuracy": 0.5476762084960938,
      "num_tokens": 410511.0,
      "step": 3000
    },
    {
      "entropy": 1.9665795323848725,
      "epoch": 0.09099181073703366,
      "grad_norm": 19.375,
      "learning_rate": 6.858039566497577e-05,
      "loss": 1.9827,
      "mean_token_accuracy": 0.5608653167784214,
      "num_tokens": 546387.0,
      "step": 4000
    },
    {
      "entropy": 1.8746658574938775,
      "epoch": 0.11373976342129208,
      "grad_norm": 14.5,
      "learning_rate": 6.779014022785937e-05,
      "loss": 1.878,
      "mean_token_accuracy": 0.5797940475940704,
      "num_tokens": 685188.0,
      "step": 5000
    },
    {
      "entropy": 1.8111666428446769,
      "epoch": 0.1364877161055505,
      "grad_norm": 25.375,
      "learning_rate": 6.683248966513431e-05,
      "loss": 1.8199,
      "mean_token_accuracy": 0.5904675594568253,
      "num_tokens": 823629.0,
      "step": 6000
    },
    {
      "entropy": 1.7517779313921928,
      "epoch": 0.1592356687898089,
      "grad_norm": 15.0625,
      "learning_rate": 6.571233282555582e-05,
      "loss": 1.7533,
      "mean_token_accuracy": 0.6037232976555824,
      "num_tokens": 962974.0,
      "step": 7000
    },
    {
      "entropy": 1.7067503632903098,
      "epoch": 0.18198362147406733,
      "grad_norm": 15.25,
      "learning_rate": 6.443538815965688e-05,
      "loss": 1.7095,
      "mean_token_accuracy": 0.6099362835586071,
      "num_tokens": 1097086.0,
      "step": 8000
    },
    {
      "entropy": 1.6539168149232863,
      "epoch": 0.20473157415832574,
      "grad_norm": 21.25,
      "learning_rate": 6.300817452680371e-05,
      "loss": 1.6574,
      "mean_token_accuracy": 0.6192078518867493,
      "num_tokens": 1241889.0,
      "step": 9000
    },
    {
      "entropy": 1.6205334926843644,
      "epoch": 0.22747952684258416,
      "grad_norm": 11.5625,
      "learning_rate": 6.143797791612843e-05,
      "loss": 1.623,
      "mean_token_accuracy": 0.6248027366697788,
      "num_tokens": 1378131.0,
      "step": 10000
    },
    {
      "entropy": 1.5953178288936616,
      "epoch": 0.2502274795268426,
      "grad_norm": 16.25,
      "learning_rate": 5.9732814251230294e-05,
      "loss": 1.5932,
      "mean_token_accuracy": 0.6300182574093341,
      "num_tokens": 1513496.0,
      "step": 11000
    },
    {
      "entropy": 1.5745678083896637,
      "epoch": 0.272975432211101,
      "grad_norm": 16.5,
      "learning_rate": 5.7901388468528755e-05,
      "loss": 1.5751,
      "mean_token_accuracy": 0.6331984028220177,
      "num_tokens": 1651603.0,
      "step": 12000
    },
    {
      "entropy": 1.576361115962267,
      "epoch": 0.29572338489535943,
      "grad_norm": 19.5,
      "learning_rate": 5.595305007817556e-05,
      "loss": 1.5795,
      "mean_token_accuracy": 0.6308481710553169,
      "num_tokens": 1790851.0,
      "step": 13000
    },
    {
      "entropy": 1.506616612225771,
      "epoch": 0.3184713375796178,
      "grad_norm": 15.8125,
      "learning_rate": 5.3897745434389274e-05,
      "loss": 1.504,
      "mean_token_accuracy": 0.6438729543685913,
      "num_tokens": 1926179.0,
      "step": 14000
    },
    {
      "entropy": 1.4861243069171906,
      "epoch": 0.34121929026387626,
      "grad_norm": 17.75,
      "learning_rate": 5.1745966958874055e-05,
      "loss": 1.4832,
      "mean_token_accuracy": 0.6503438740372658,
      "num_tokens": 2062340.0,
      "step": 15000
    },
    {
      "entropy": 1.4704010844826698,
      "epoch": 0.36396724294813465,
      "grad_norm": 21.0,
      "learning_rate": 4.9508699576539914e-05,
      "loss": 1.4694,
      "mean_token_accuracy": 0.6534121400117874,
      "num_tokens": 2207639.0,
      "step": 16000
    },
    {
      "entropy": 1.4741187323331832,
      "epoch": 0.3867151956323931,
      "grad_norm": 19.375,
      "learning_rate": 4.7197364636971925e-05,
      "loss": 1.4749,
      "mean_token_accuracy": 0.6518012735545635,
      "num_tokens": 2348887.0,
      "step": 17000
    },
    {
      "entropy": 1.4345429268479348,
      "epoch": 0.4094631483166515,
      "grad_norm": 19.25,
      "learning_rate": 4.482376160793216e-05,
      "loss": 1.4354,
      "mean_token_accuracy": 0.6596047645211219,
      "num_tokens": 2483718.0,
      "step": 18000
    },
    {
      "entropy": 1.4003090425133704,
      "epoch": 0.4322111010009099,
      "grad_norm": 20.625,
      "learning_rate": 4.240000783855147e-05,
      "loss": 1.3979,
      "mean_token_accuracy": 0.6681542407870292,
      "num_tokens": 2622656.0,
      "step": 19000
    },
    {
      "entropy": 1.4163592108488083,
      "epoch": 0.4549590536851683,
      "grad_norm": 16.5,
      "learning_rate": 3.993847669972281e-05,
      "loss": 1.4083,
      "mean_token_accuracy": 0.664802198857069,
      "num_tokens": 2760831.0,
      "step": 20000
    },
    {
      "entropy": 1.4052879491746426,
      "epoch": 0.47770700636942676,
      "grad_norm": 23.875,
      "learning_rate": 3.745173441749185e-05,
      "loss": 1.4074,
      "mean_token_accuracy": 0.6652796367108822,
      "num_tokens": 2900387.0,
      "step": 21000
    },
    {
      "entropy": 1.3997035399377347,
      "epoch": 0.5004549590536852,
      "grad_norm": 25.875,
      "learning_rate": 3.495247592191375e-05,
      "loss": 1.3972,
      "mean_token_accuracy": 0.6675955319404602,
      "num_tokens": 3037204.0,
      "step": 22000
    },
    {
      "entropy": 1.415595789283514,
      "epoch": 0.5232029117379435,
      "grad_norm": 26.75,
      "learning_rate": 3.245346003886994e-05,
      "loss": 1.4129,
      "mean_token_accuracy": 0.6636776500046253,
      "num_tokens": 3179823.0,
      "step": 23000
    },
    {
      "entropy": 1.379241144567728,
      "epoch": 0.545950864422202,
      "grad_norm": 20.0,
      "learning_rate": 2.996744435569409e-05,
      "loss": 1.3739,
      "mean_token_accuracy": 0.6730512301325798,
      "num_tokens": 3318434.0,
      "step": 24000
    },
    {
      "entropy": 1.3975665314793586,
      "epoch": 0.5686988171064604,
      "grad_norm": 20.875,
      "learning_rate": 2.7507120093120825e-05,
      "loss": 1.3976,
      "mean_token_accuracy": 0.6678333807885647,
      "num_tokens": 3461132.0,
      "step": 25000
    },
    {
      "entropy": 1.380986104875803,
      "epoch": 0.5914467697907189,
      "grad_norm": 14.6875,
      "learning_rate": 2.5085047316038814e-05,
      "loss": 1.3817,
      "mean_token_accuracy": 0.6719385531246662,
      "num_tokens": 3601560.0,
      "step": 26000
    },
    {
      "entropy": 1.385195587992668,
      "epoch": 0.6141947224749773,
      "grad_norm": 21.25,
      "learning_rate": 2.271359081380046e-05,
      "loss": 1.3815,
      "mean_token_accuracy": 0.6699076734781265,
      "num_tokens": 3745377.0,
      "step": 27000
    },
    {
      "entropy": 1.3656318633258342,
      "epoch": 0.6369426751592356,
      "grad_norm": 22.5,
      "learning_rate": 2.040485697742177e-05,
      "loss": 1.357,
      "mean_token_accuracy": 0.6759414212107658,
      "num_tokens": 3879126.0,
      "step": 28000
    },
    {
      "entropy": 1.3772443866729736,
      "epoch": 0.6596906278434941,
      "grad_norm": 26.0,
      "learning_rate": 1.8170631995917233e-05,
      "loss": 1.3745,
      "mean_token_accuracy": 0.6744600256979465,
      "num_tokens": 4017925.0,
      "step": 29000
    },
    {
      "entropy": 1.365410826742649,
      "epoch": 0.6824385805277525,
      "grad_norm": 22.5,
      "learning_rate": 1.602232168728024e-05,
      "loss": 1.3584,
      "mean_token_accuracy": 0.675611907929182,
      "num_tokens": 4156599.0,
      "step": 30000
    },
    {
      "entropy": 1.3873618737459184,
      "epoch": 0.705186533212011,
      "grad_norm": 13.25,
      "learning_rate": 1.3970893271274471e-05,
      "loss": 1.3803,
      "mean_token_accuracy": 0.6701690441966057,
      "num_tokens": 4295734.0,
      "step": 31000
    },
    {
      "entropy": 1.359022274851799,
      "epoch": 0.7279344858962693,
      "grad_norm": 16.75,
      "learning_rate": 1.202681938128876e-05,
      "loss": 1.3556,
      "mean_token_accuracy": 0.6761138562858104,
      "num_tokens": 4431490.0,
      "step": 32000
    },
    {
      "entropy": 1.3703680724203586,
      "epoch": 0.7506824385805277,
      "grad_norm": 19.0,
      "learning_rate": 1.0200024601077386e-05,
      "loss": 1.362,
      "mean_token_accuracy": 0.6743936349153519,
      "num_tokens": 4571660.0,
      "step": 33000
    },
    {
      "entropy": 1.3698465181291104,
      "epoch": 0.7734303912647862,
      "grad_norm": 24.25,
      "learning_rate": 8.49983479931827e-06,
      "loss": 1.3757,
      "mean_token_accuracy": 0.6719635992050171,
      "num_tokens": 4707985.0,
      "step": 34000
    },
    {
      "entropy": 1.3624781457483768,
      "epoch": 0.7961783439490446,
      "grad_norm": 16.875,
      "learning_rate": 6.9349295206380985e-06,
      "loss": 1.3518,
      "mean_token_accuracy": 0.6769196209311485,
      "num_tokens": 4844765.0,
      "step": 35000
    },
    {
      "entropy": 1.3442125609219073,
      "epoch": 0.818926296633303,
      "grad_norm": 15.25,
      "learning_rate": 5.513297676150713e-06,
      "loss": 1.3335,
      "mean_token_accuracy": 0.6805432761013508,
      "num_tokens": 4978615.0,
      "step": 36000
    },
    {
      "entropy": 1.355837997198105,
      "epoch": 0.8416742493175614,
      "grad_norm": 16.0,
      "learning_rate": 4.242196759710179e-06,
      "loss": 1.3377,
      "mean_token_accuracy": 0.6823569060564041,
      "num_tokens": 5117536.0,
      "step": 37000
    },
    {
      "entropy": 1.3855277094841003,
      "epoch": 0.8644222020018199,
      "grad_norm": 17.625,
      "learning_rate": 3.1281157980815473e-06,
      "loss": 1.3731,
      "mean_token_accuracy": 0.6734900210201741,
      "num_tokens": 5258021.0,
      "step": 38000
    },
    {
      "entropy": 1.3616491684913634,
      "epoch": 0.8871701546860783,
      "grad_norm": 33.25,
      "learning_rate": 2.1767422241703795e-06,
      "loss": 1.3462,
      "mean_token_accuracy": 0.6789434304237366,
      "num_tokens": 5395368.0,
      "step": 39000
    },
    {
      "entropy": 1.3761241393685342,
      "epoch": 0.9099181073703366,
      "grad_norm": 16.0,
      "learning_rate": 1.392932842424574e-06,
      "loss": 1.3787,
      "mean_token_accuracy": 0.6720339168906212,
      "num_tokens": 5533958.0,
      "step": 40000
    },
    {
      "entropy": 1.3613062560856342,
      "epoch": 0.9326660600545951,
      "grad_norm": 31.375,
      "learning_rate": 7.806890346314221e-07,
      "loss": 1.3621,
      "mean_token_accuracy": 0.6752942685186863,
      "num_tokens": 5668914.0,
      "step": 41000
    },
    {
      "entropy": 1.3506658849716187,
      "epoch": 0.9554140127388535,
      "grad_norm": 13.8125,
      "learning_rate": 3.431363326854719e-07,
      "loss": 1.3478,
      "mean_token_accuracy": 0.6773408466875553,
      "num_tokens": 5806407.0,
      "step": 42000
    },
    {
      "entropy": 1.3391023662090302,
      "epoch": 0.978161965423112,
      "grad_norm": 13.125,
      "learning_rate": 8.250846260903998e-08,
      "loss": 1.3275,
      "mean_token_accuracy": 0.68198375582695,
      "num_tokens": 5942532.0,
      "step": 43000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 43960,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.3040032742440704e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}