{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.4126984126984126,
  "eval_steps": 500,
  "global_step": 152,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 0.25798711776733396,
      "epoch": 0.07936507936507936,
      "grad_norm": 0.6395628452301025,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.33170533180236816,
      "mean_token_accuracy": 0.9083019256591797,
      "num_tokens": 2249744.0,
      "step": 5
    },
    {
      "entropy": 0.1518770933151245,
      "epoch": 0.15873015873015872,
      "grad_norm": 0.27742329239845276,
      "learning_rate": 4.9966852247120764e-05,
      "loss": 0.14515713453292847,
      "mean_token_accuracy": 0.9476617932319641,
      "num_tokens": 4502357.0,
      "step": 10
    },
    {
      "entropy": 0.1205204889178276,
      "epoch": 0.23809523809523808,
      "grad_norm": 0.15280286967754364,
      "learning_rate": 4.976460088644493e-05,
      "loss": 0.11978392601013184,
      "mean_token_accuracy": 0.9552689790725708,
      "num_tokens": 6750630.0,
      "step": 15
    },
    {
      "entropy": 0.1094709500670433,
      "epoch": 0.31746031746031744,
      "grad_norm": 0.13621626794338226,
      "learning_rate": 4.938000100611456e-05,
      "loss": 0.10746543407440186,
      "mean_token_accuracy": 0.9598726153373718,
      "num_tokens": 8991471.0,
      "step": 20
    },
    {
      "entropy": 0.10419090241193771,
      "epoch": 0.3968253968253968,
      "grad_norm": 0.16320250928401947,
      "learning_rate": 4.881588452008456e-05,
      "loss": 0.10661067962646484,
      "mean_token_accuracy": 0.9599695205688477,
      "num_tokens": 11228489.0,
      "step": 25
    },
    {
      "entropy": 0.09569432139396668,
      "epoch": 0.47619047619047616,
      "grad_norm": 0.15873780846595764,
      "learning_rate": 4.80764051721044e-05,
      "loss": 0.09594419002532958,
      "mean_token_accuracy": 0.9637031435966492,
      "num_tokens": 13464387.0,
      "step": 30
    },
    {
      "entropy": 0.09625398516654968,
      "epoch": 0.5555555555555556,
      "grad_norm": 0.15297284722328186,
      "learning_rate": 4.7167007950568505e-05,
      "loss": 0.09495124220848083,
      "mean_token_accuracy": 0.964220929145813,
      "num_tokens": 15709681.0,
      "step": 35
    },
    {
      "entropy": 0.09195081293582916,
      "epoch": 0.6349206349206349,
      "grad_norm": 0.15219081938266754,
      "learning_rate": 4.609438899557964e-05,
      "loss": 0.09150891304016114,
      "mean_token_accuracy": 0.9652478575706482,
      "num_tokens": 17987403.0,
      "step": 40
    },
    {
      "entropy": 0.08980809897184372,
      "epoch": 0.7142857142857143,
      "grad_norm": 0.1412411779165268,
      "learning_rate": 4.4866446293440626e-05,
      "loss": 0.0898817539215088,
      "mean_token_accuracy": 0.965602707862854,
      "num_tokens": 20231746.0,
      "step": 45
    },
    {
      "entropy": 0.08743101209402085,
      "epoch": 0.7936507936507936,
      "grad_norm": 0.16695629060268402,
      "learning_rate": 4.3492221521624174e-05,
      "loss": 0.08715485334396363,
      "mean_token_accuracy": 0.9668790459632873,
      "num_tokens": 22491762.0,
      "step": 50
    },
    {
      "entropy": 0.08765677958726883,
      "epoch": 0.873015873015873,
      "grad_norm": 0.21143023669719696,
      "learning_rate": 4.198183347243233e-05,
      "loss": 0.08766236305236816,
      "mean_token_accuracy": 0.9668452143669128,
      "num_tokens": 24752715.0,
      "step": 55
    },
    {
      "entropy": 0.08127444982528687,
      "epoch": 0.9523809523809523,
      "grad_norm": 0.18779045343399048,
      "learning_rate": 4.03464035455653e-05,
      "loss": 0.08236484527587891,
      "mean_token_accuracy": 0.9689654350280762,
      "num_tokens": 26982538.0,
      "step": 60
    },
    {
      "entropy": 0.07894417196512223,
      "epoch": 1.0317460317460316,
      "grad_norm": 0.1385810822248459,
      "learning_rate": 3.859797385821812e-05,
      "loss": 0.0778856873512268,
      "mean_token_accuracy": 0.9699163198471069,
      "num_tokens": 29221704.0,
      "step": 65
    },
    {
      "entropy": 0.07985353171825409,
      "epoch": 1.1111111111111112,
      "grad_norm": 0.16439372301101685,
      "learning_rate": 3.6749418575683e-05,
      "loss": 0.07885851860046386,
      "mean_token_accuracy": 0.969633674621582,
      "num_tokens": 31470338.0,
      "step": 70
    },
    {
      "entropy": 0.07194689065217971,
      "epoch": 1.1904761904761905,
      "grad_norm": 0.16862133145332336,
      "learning_rate": 3.48143491153543e-05,
      "loss": 0.074225252866745,
      "mean_token_accuracy": 0.9716025114059448,
      "num_tokens": 33719303.0,
      "step": 75
    },
    {
      "entropy": 0.07757648378610611,
      "epoch": 1.2698412698412698,
      "grad_norm": 0.14261098206043243,
      "learning_rate": 3.280701392214487e-05,
      "loss": 0.07595503330230713,
      "mean_token_accuracy": 0.9707554697990417,
      "num_tokens": 35977913.0,
      "step": 80
    },
    {
      "entropy": 0.07388863414525985,
      "epoch": 1.3492063492063493,
      "grad_norm": 0.16554316878318787,
      "learning_rate": 3.07421935532949e-05,
      "loss": 0.07574511766433716,
      "mean_token_accuracy": 0.9704798936843873,
      "num_tokens": 38220353.0,
      "step": 85
    },
    {
      "entropy": 0.07249651849269867,
      "epoch": 1.4285714285714286,
      "grad_norm": 0.18186147511005402,
      "learning_rate": 2.8635091845092247e-05,
      "loss": 0.07062947750091553,
      "mean_token_accuracy": 0.9724894285202026,
      "num_tokens": 40476134.0,
      "step": 90
    },
    {
      "entropy": 0.06985943019390106,
      "epoch": 1.507936507936508,
      "grad_norm": 0.13512398302555084,
      "learning_rate": 2.650122396287371e-05,
      "loss": 0.06756666898727418,
      "mean_token_accuracy": 0.9738536238670349,
      "num_tokens": 42711931.0,
      "step": 95
    },
    {
      "entropy": 0.06709291264414788,
      "epoch": 1.5873015873015874,
      "grad_norm": 0.13330888748168945,
      "learning_rate": 2.4356302158625288e-05,
      "loss": 0.07059981226921082,
      "mean_token_accuracy": 0.9723513960838318,
      "num_tokens": 44948562.0,
      "step": 100
    },
    {
      "entropy": 0.06978753134608269,
      "epoch": 1.6666666666666665,
      "grad_norm": 0.16254326701164246,
      "learning_rate": 2.221612007737978e-05,
      "loss": 0.06924379467964173,
      "mean_token_accuracy": 0.9731807231903076,
      "num_tokens": 47194289.0,
      "step": 105
    },
    {
      "entropy": 0.06934378147125245,
      "epoch": 1.746031746031746,
      "grad_norm": 0.13457117974758148,
      "learning_rate": 2.009643646429508e-05,
      "loss": 0.06761763095855713,
      "mean_token_accuracy": 0.9738588094711303,
      "num_tokens": 49437033.0,
      "step": 110
    },
    {
      "entropy": 0.06849471032619477,
      "epoch": 1.8253968253968254,
      "grad_norm": 0.14807309210300446,
      "learning_rate": 1.8012859128709766e-05,
      "loss": 0.06851363182067871,
      "mean_token_accuracy": 0.9734799265861511,
      "num_tokens": 51685754.0,
      "step": 115
    },
    {
      "entropy": 0.06743545532226562,
      "epoch": 1.9047619047619047,
      "grad_norm": 0.14509981870651245,
      "learning_rate": 1.598073001958042e-05,
      "loss": 0.06817712187767029,
      "mean_token_accuracy": 0.9735557198524475,
      "num_tokens": 53931825.0,
      "step": 120
    },
    {
      "entropy": 0.06640436649322509,
      "epoch": 1.9841269841269842,
      "grad_norm": 0.13425911962985992,
      "learning_rate": 1.4015012258521454e-05,
      "loss": 0.0663202166557312,
      "mean_token_accuracy": 0.9740958571434021,
      "num_tokens": 56203400.0,
      "step": 125
    },
    {
      "entropy": 0.06114112734794617,
      "epoch": 2.0634920634920633,
      "grad_norm": 0.14675261080265045,
      "learning_rate": 1.213017996225424e-05,
      "loss": 0.0597736656665802,
      "mean_token_accuracy": 0.9770057320594787,
      "num_tokens": 58447264.0,
      "step": 130
    },
    {
      "entropy": 0.06077109202742577,
      "epoch": 2.142857142857143,
      "grad_norm": 0.14722956717014313,
      "learning_rate": 1.034011166573261e-05,
      "loss": 0.0608077347278595,
      "mean_token_accuracy": 0.9764052987098694,
      "num_tokens": 60680203.0,
      "step": 135
    },
    {
      "entropy": 0.06085996180772781,
      "epoch": 2.2222222222222223,
      "grad_norm": 0.13919851183891296,
      "learning_rate": 8.657988130699848e-06,
      "loss": 0.06093168258666992,
      "mean_token_accuracy": 0.975907814502716,
      "num_tokens": 62928491.0,
      "step": 140
    },
    {
      "entropy": 0.06154095679521561,
      "epoch": 2.3015873015873014,
      "grad_norm": 0.1471860557794571,
      "learning_rate": 7.096195292140173e-06,
      "loss": 0.06225389242172241,
      "mean_token_accuracy": 0.9755991458892822,
      "num_tokens": 65183865.0,
      "step": 145
    },
    {
      "entropy": 0.06028798446059227,
      "epoch": 2.380952380952381,
      "grad_norm": 0.1381225436925888,
      "learning_rate": 5.666233057256756e-06,
      "loss": 0.05946478247642517,
      "mean_token_accuracy": 0.9772057175636292,
      "num_tokens": 67455033.0,
      "step": 150
    }
  ],
  "logging_steps": 5,
  "max_steps": 189,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 8,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.8107234557965107e+18,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}