{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 4455,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06734006734006734,
      "grad_norm": 1277.759765625,
      "learning_rate": 0.00014849999999999998,
      "loss": 306.280546875,
      "step": 100
    },
    {
      "epoch": 0.13468013468013468,
      "grad_norm": 45.35179138183594,
      "learning_rate": 0.0002985,
      "loss": 116.630830078125,
      "step": 200
    },
    {
      "epoch": 0.20202020202020202,
      "grad_norm": 38.45261764526367,
      "learning_rate": 0.00029301997649823735,
      "loss": 81.141572265625,
      "step": 300
    },
    {
      "epoch": 0.26936026936026936,
      "grad_norm": 30.12755584716797,
      "learning_rate": 0.0002859694477085781,
      "loss": 71.679775390625,
      "step": 400
    },
    {
      "epoch": 0.3367003367003367,
      "grad_norm": 26.099262237548828,
      "learning_rate": 0.0002789189189189189,
      "loss": 67.6156201171875,
      "step": 500
    },
    {
      "epoch": 0.3367003367003367,
      "eval_loss": 6.426710605621338,
      "eval_runtime": 71.9526,
      "eval_samples_per_second": 13.342,
      "eval_steps_per_second": 3.336,
      "step": 500
    },
    {
      "epoch": 0.40404040404040403,
      "grad_norm": 25.531190872192383,
      "learning_rate": 0.00027186839012925966,
      "loss": 65.3996484375,
      "step": 600
    },
    {
      "epoch": 0.4713804713804714,
      "grad_norm": 25.456987380981445,
      "learning_rate": 0.00026481786133960043,
      "loss": 63.7711328125,
      "step": 700
    },
    {
      "epoch": 0.5387205387205387,
      "grad_norm": 24.871572494506836,
      "learning_rate": 0.00025776733254994125,
      "loss": 62.25517578125,
      "step": 800
    },
    {
      "epoch": 0.6060606060606061,
      "grad_norm": 29.645320892333984,
      "learning_rate": 0.000250716803760282,
      "loss": 60.900390625,
      "step": 900
    },
    {
      "epoch": 0.6734006734006734,
      "grad_norm": 25.431486129760742,
      "learning_rate": 0.00024366627497062276,
      "loss": 60.542490234375,
      "step": 1000
    },
    {
      "epoch": 0.6734006734006734,
      "eval_loss": 5.928141117095947,
      "eval_runtime": 71.83,
      "eval_samples_per_second": 13.365,
      "eval_steps_per_second": 3.341,
      "step": 1000
    },
    {
      "epoch": 0.7407407407407407,
      "grad_norm": 23.62899398803711,
      "learning_rate": 0.00023661574618096356,
      "loss": 59.750263671875,
      "step": 1100
    },
    {
      "epoch": 0.8080808080808081,
      "grad_norm": 18.71799087524414,
      "learning_rate": 0.00022956521739130433,
      "loss": 58.8070458984375,
      "step": 1200
    },
    {
      "epoch": 0.8754208754208754,
      "grad_norm": 21.216167449951172,
      "learning_rate": 0.00022251468860164512,
      "loss": 58.904775390625,
      "step": 1300
    },
    {
      "epoch": 0.9427609427609428,
      "grad_norm": 25.631328582763672,
      "learning_rate": 0.00021546415981198587,
      "loss": 57.6883837890625,
      "step": 1400
    },
    {
      "epoch": 1.0101010101010102,
      "grad_norm": 24.354522705078125,
      "learning_rate": 0.00020841363102232666,
      "loss": 57.590849609375,
      "step": 1500
    },
    {
      "epoch": 1.0101010101010102,
      "eval_loss": 5.705260276794434,
      "eval_runtime": 72.7006,
      "eval_samples_per_second": 13.205,
      "eval_steps_per_second": 3.301,
      "step": 1500
    },
    {
      "epoch": 1.0774410774410774,
      "grad_norm": 28.926172256469727,
      "learning_rate": 0.00020136310223266743,
      "loss": 54.9418896484375,
      "step": 1600
    },
    {
      "epoch": 1.144781144781145,
      "grad_norm": 23.21477508544922,
      "learning_rate": 0.00019431257344300823,
      "loss": 55.0524072265625,
      "step": 1700
    },
    {
      "epoch": 1.2121212121212122,
      "grad_norm": 22.93968391418457,
      "learning_rate": 0.00018726204465334897,
      "loss": 54.60087890625,
      "step": 1800
    },
    {
      "epoch": 1.2794612794612794,
      "grad_norm": 21.029342651367188,
      "learning_rate": 0.00018021151586368976,
      "loss": 53.961689453125,
      "step": 1900
    },
    {
      "epoch": 1.3468013468013469,
      "grad_norm": 22.02585220336914,
      "learning_rate": 0.00017316098707403053,
      "loss": 54.352119140625,
      "step": 2000
    },
    {
      "epoch": 1.3468013468013469,
      "eval_loss": 5.576821804046631,
      "eval_runtime": 70.7614,
      "eval_samples_per_second": 13.567,
      "eval_steps_per_second": 3.392,
      "step": 2000
    },
    {
      "epoch": 1.4141414141414141,
      "grad_norm": 22.914306640625,
      "learning_rate": 0.00016611045828437133,
      "loss": 53.9832177734375,
      "step": 2100
    },
    {
      "epoch": 1.4814814814814814,
      "grad_norm": 22.693891525268555,
      "learning_rate": 0.00015905992949471207,
      "loss": 53.846845703125,
      "step": 2200
    },
    {
      "epoch": 1.5488215488215489,
      "grad_norm": 22.133243560791016,
      "learning_rate": 0.00015200940070505287,
      "loss": 54.00986328125,
      "step": 2300
    },
    {
      "epoch": 1.6161616161616161,
      "grad_norm": 23.607688903808594,
      "learning_rate": 0.00014495887191539364,
      "loss": 53.8654638671875,
      "step": 2400
    },
    {
      "epoch": 1.6835016835016834,
      "grad_norm": 22.71261215209961,
      "learning_rate": 0.0001379083431257344,
      "loss": 53.5878173828125,
      "step": 2500
    },
    {
      "epoch": 1.6835016835016834,
      "eval_loss": 5.490618705749512,
      "eval_runtime": 72.5632,
      "eval_samples_per_second": 13.23,
      "eval_steps_per_second": 3.307,
      "step": 2500
    },
    {
      "epoch": 1.7508417508417509,
      "grad_norm": 19.942703247070312,
      "learning_rate": 0.0001308578143360752,
      "loss": 52.879384765625,
      "step": 2600
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 20.36492919921875,
      "learning_rate": 0.00012380728554641597,
      "loss": 52.678017578125,
      "step": 2700
    },
    {
      "epoch": 1.8855218855218854,
      "grad_norm": 20.429353713989258,
      "learning_rate": 0.00011675675675675675,
      "loss": 53.1516796875,
      "step": 2800
    },
    {
      "epoch": 1.9528619528619529,
      "grad_norm": 20.908979415893555,
      "learning_rate": 0.00010970622796709752,
      "loss": 52.3822607421875,
      "step": 2900
    },
    {
      "epoch": 2.0202020202020203,
      "grad_norm": 24.01288604736328,
      "learning_rate": 0.0001026556991774383,
      "loss": 51.8819384765625,
      "step": 3000
    },
    {
      "epoch": 2.0202020202020203,
      "eval_loss": 5.448428153991699,
      "eval_runtime": 70.4982,
      "eval_samples_per_second": 13.617,
      "eval_steps_per_second": 3.404,
      "step": 3000
    },
    {
      "epoch": 2.0875420875420874,
      "grad_norm": 19.735408782958984,
      "learning_rate": 9.560517038777907e-05,
      "loss": 51.3095654296875,
      "step": 3100
    },
    {
      "epoch": 2.154882154882155,
      "grad_norm": 23.40346336364746,
      "learning_rate": 8.855464159811985e-05,
      "loss": 51.257080078125,
      "step": 3200
    },
    {
      "epoch": 2.2222222222222223,
      "grad_norm": 19.5986385345459,
      "learning_rate": 8.150411280846062e-05,
      "loss": 50.9421484375,
      "step": 3300
    },
    {
      "epoch": 2.28956228956229,
      "grad_norm": 22.559755325317383,
      "learning_rate": 7.44535840188014e-05,
      "loss": 50.7065576171875,
      "step": 3400
    },
    {
      "epoch": 2.356902356902357,
      "grad_norm": 21.577171325683594,
      "learning_rate": 6.740305522914217e-05,
      "loss": 51.1507177734375,
      "step": 3500
    },
    {
      "epoch": 2.356902356902357,
      "eval_loss": 5.395638942718506,
      "eval_runtime": 73.4284,
      "eval_samples_per_second": 13.074,
      "eval_steps_per_second": 3.268,
      "step": 3500
    },
    {
      "epoch": 2.4242424242424243,
      "grad_norm": 29.17755889892578,
      "learning_rate": 6.0352526439482956e-05,
      "loss": 50.8910009765625,
      "step": 3600
    },
    {
      "epoch": 2.4915824915824913,
      "grad_norm": 20.923023223876953,
      "learning_rate": 5.330199764982373e-05,
      "loss": 50.6894189453125,
      "step": 3700
    },
    {
      "epoch": 2.558922558922559,
      "grad_norm": 18.506322860717773,
      "learning_rate": 4.625146886016451e-05,
      "loss": 51.0062451171875,
      "step": 3800
    },
    {
      "epoch": 2.6262626262626263,
      "grad_norm": 22.548017501831055,
      "learning_rate": 3.920094007050528e-05,
      "loss": 50.7424365234375,
      "step": 3900
    },
    {
      "epoch": 2.6936026936026938,
      "grad_norm": 22.250715255737305,
      "learning_rate": 3.2150411280846066e-05,
      "loss": 50.8699951171875,
      "step": 4000
    },
    {
      "epoch": 2.6936026936026938,
      "eval_loss": 5.3642354011535645,
      "eval_runtime": 71.4784,
      "eval_samples_per_second": 13.431,
      "eval_steps_per_second": 3.358,
      "step": 4000
    },
    {
      "epoch": 2.760942760942761,
      "grad_norm": 21.673240661621094,
      "learning_rate": 2.5099882491186838e-05,
      "loss": 50.46708984375,
      "step": 4100
    },
    {
      "epoch": 2.8282828282828283,
      "grad_norm": 21.410390853881836,
      "learning_rate": 1.8049353701527613e-05,
      "loss": 50.4184521484375,
      "step": 4200
    },
    {
      "epoch": 2.8956228956228958,
      "grad_norm": 21.75411033630371,
      "learning_rate": 1.0998824911868389e-05,
      "loss": 50.68787109375,
      "step": 4300
    },
    {
      "epoch": 2.962962962962963,
      "grad_norm": 21.319902420043945,
      "learning_rate": 3.948296122209165e-06,
      "loss": 50.4838671875,
      "step": 4400
    }
  ],
  "logging_steps": 100,
  "max_steps": 4455,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 1.714151970521088e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}