{
  "best_global_step": 3375,
  "best_metric": 0.25075,
  "best_model_checkpoint": "./output/checkpoint-3375",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 3375,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.014814814814814815,
      "grad_norm": 15.2734375,
      "learning_rate": 2.4197530864197535e-07,
      "loss": 2.7766494750976562,
      "step": 50
    },
    {
      "epoch": 0.02962962962962963,
      "grad_norm": 14.0625,
      "learning_rate": 4.88888888888889e-07,
      "loss": 2.6392501831054687,
      "step": 100
    },
    {
      "epoch": 0.044444444444444446,
      "grad_norm": 7.6171875,
      "learning_rate": 7.358024691358024e-07,
      "loss": 2.662928161621094,
      "step": 150
    },
    {
      "epoch": 0.05925925925925926,
      "grad_norm": 13.4296875,
      "learning_rate": 9.827160493827162e-07,
      "loss": 3.1913311767578123,
      "step": 200
    },
    {
      "epoch": 0.07407407407407407,
      "grad_norm": 1.96484375,
      "learning_rate": 1.2296296296296297e-06,
      "loss": 2.97621826171875,
      "step": 250
    },
    {
      "epoch": 0.08888888888888889,
      "grad_norm": 2.583984375,
      "learning_rate": 1.4765432098765435e-06,
      "loss": 2.795979309082031,
      "step": 300
    },
    {
      "epoch": 0.1037037037037037,
      "grad_norm": 3.095703125,
      "learning_rate": 1.723456790123457e-06,
      "loss": 2.7952374267578124,
      "step": 350
    },
    {
      "epoch": 0.11851851851851852,
      "grad_norm": 2.453125,
      "learning_rate": 1.9703703703703703e-06,
      "loss": 2.7849395751953123,
      "step": 400
    },
    {
      "epoch": 0.13333333333333333,
      "grad_norm": 6.875,
      "learning_rate": 2.217283950617284e-06,
      "loss": 2.7812811279296876,
      "step": 450
    },
    {
      "epoch": 0.14814814814814814,
      "grad_norm": 2.33203125,
      "learning_rate": 2.464197530864198e-06,
      "loss": 2.7828436279296875,
      "step": 500
    },
    {
      "epoch": 0.16296296296296298,
      "grad_norm": 2.53125,
      "learning_rate": 2.7111111111111116e-06,
      "loss": 2.7977508544921874,
      "step": 550
    },
    {
      "epoch": 0.17777777777777778,
      "grad_norm": 3.587890625,
      "learning_rate": 2.9580246913580254e-06,
      "loss": 2.783578186035156,
      "step": 600
    },
    {
      "epoch": 0.1925925925925926,
      "grad_norm": 1.9501953125,
      "learning_rate": 3.2049382716049387e-06,
      "loss": 2.7933184814453127,
      "step": 650
    },
    {
      "epoch": 0.2074074074074074,
      "grad_norm": 3.435546875,
      "learning_rate": 3.4518518518518524e-06,
      "loss": 2.80977783203125,
      "step": 700
    },
    {
      "epoch": 0.2222222222222222,
      "grad_norm": 1.8955078125,
      "learning_rate": 3.6987654320987658e-06,
      "loss": 2.7978311157226563,
      "step": 750
    },
    {
      "epoch": 0.23703703703703705,
      "grad_norm": 2.455078125,
      "learning_rate": 3.94567901234568e-06,
      "loss": 2.7951055908203126,
      "step": 800
    },
    {
      "epoch": 0.2518518518518518,
      "grad_norm": 4.1171875,
      "learning_rate": 4.192592592592593e-06,
      "loss": 2.785648193359375,
      "step": 850
    },
    {
      "epoch": 0.26666666666666666,
      "grad_norm": 1.84375,
      "learning_rate": 4.439506172839507e-06,
      "loss": 2.7951544189453124,
      "step": 900
    },
    {
      "epoch": 0.2814814814814815,
      "grad_norm": 2.359375,
      "learning_rate": 4.68641975308642e-06,
      "loss": 2.796750793457031,
      "step": 950
    },
    {
      "epoch": 0.2962962962962963,
      "grad_norm": 2.39453125,
      "learning_rate": 4.933333333333334e-06,
      "loss": 2.7995571899414062,
      "step": 1000
    },
    {
      "epoch": 0.3111111111111111,
      "grad_norm": 2.55078125,
      "learning_rate": 5.180246913580247e-06,
      "loss": 2.7804888916015624,
      "step": 1050
    },
    {
      "epoch": 0.32592592592592595,
      "grad_norm": 3.67578125,
      "learning_rate": 5.427160493827161e-06,
      "loss": 2.792048034667969,
      "step": 1100
    },
    {
      "epoch": 0.34074074074074073,
      "grad_norm": 2.130859375,
      "learning_rate": 5.674074074074074e-06,
      "loss": 2.789242858886719,
      "step": 1150
    },
    {
      "epoch": 0.35555555555555557,
      "grad_norm": 2.412109375,
      "learning_rate": 5.920987654320987e-06,
      "loss": 2.788144836425781,
      "step": 1200
    },
    {
      "epoch": 0.37037037037037035,
      "grad_norm": 2.72265625,
      "learning_rate": 6.167901234567902e-06,
      "loss": 2.8027859497070313,
      "step": 1250
    },
    {
      "epoch": 0.3851851851851852,
      "grad_norm": 2.6484375,
      "learning_rate": 6.414814814814815e-06,
      "loss": 2.7877798461914063,
      "step": 1300
    },
    {
      "epoch": 0.4,
      "grad_norm": 2.3359375,
      "learning_rate": 6.661728395061728e-06,
      "loss": 2.7876470947265624,
      "step": 1350
    },
    {
      "epoch": 0.4148148148148148,
      "grad_norm": 1.9970703125,
      "learning_rate": 6.908641975308642e-06,
      "loss": 2.77328857421875,
      "step": 1400
    },
    {
      "epoch": 0.42962962962962964,
      "grad_norm": 1.875,
      "learning_rate": 7.155555555555556e-06,
      "loss": 2.79718017578125,
      "step": 1450
    },
    {
      "epoch": 0.4444444444444444,
      "grad_norm": 2.248046875,
      "learning_rate": 7.40246913580247e-06,
      "loss": 2.7796466064453127,
      "step": 1500
    },
    {
      "epoch": 0.45925925925925926,
      "grad_norm": 1.8623046875,
      "learning_rate": 7.649382716049384e-06,
      "loss": 2.7859207153320313,
      "step": 1550
    },
    {
      "epoch": 0.4740740740740741,
      "grad_norm": 1.5302734375,
      "learning_rate": 7.896296296296297e-06,
      "loss": 2.781698913574219,
      "step": 1600
    },
    {
      "epoch": 0.4888888888888889,
      "grad_norm": 1.3701171875,
      "learning_rate": 8.14320987654321e-06,
      "loss": 2.7813421630859376,
      "step": 1650
    },
    {
      "epoch": 0.5037037037037037,
      "grad_norm": 1.40625,
      "learning_rate": 8.390123456790124e-06,
      "loss": 2.781036071777344,
      "step": 1700
    },
    {
      "epoch": 0.5185185185185185,
      "grad_norm": 2.236328125,
      "learning_rate": 8.637037037037037e-06,
      "loss": 2.783014221191406,
      "step": 1750
    },
    {
      "epoch": 0.5333333333333333,
      "grad_norm": 1.826171875,
      "learning_rate": 8.883950617283952e-06,
      "loss": 2.7817703247070313,
      "step": 1800
    },
    {
      "epoch": 0.5481481481481482,
      "grad_norm": 1.82421875,
      "learning_rate": 9.130864197530866e-06,
      "loss": 2.7779510498046873,
      "step": 1850
    },
    {
      "epoch": 0.562962962962963,
      "grad_norm": 2.1953125,
      "learning_rate": 9.377777777777779e-06,
      "loss": 2.7827484130859377,
      "step": 1900
    },
    {
      "epoch": 0.5777777777777777,
      "grad_norm": 1.353515625,
      "learning_rate": 9.624691358024692e-06,
      "loss": 2.7817633056640627,
      "step": 1950
    },
    {
      "epoch": 0.5925925925925926,
      "grad_norm": 2.69921875,
      "learning_rate": 9.871604938271606e-06,
      "loss": 2.7813009643554687,
      "step": 2000
    },
    {
      "epoch": 0.6074074074074074,
      "grad_norm": 1.2294921875,
      "learning_rate": 9.970370370370372e-06,
      "loss": 2.77718994140625,
      "step": 2050
    },
    {
      "epoch": 0.6222222222222222,
      "grad_norm": 1.6513671875,
      "learning_rate": 9.908641975308642e-06,
      "loss": 2.778031921386719,
      "step": 2100
    },
    {
      "epoch": 0.6370370370370371,
      "grad_norm": 1.0498046875,
      "learning_rate": 9.846913580246915e-06,
      "loss": 2.778972473144531,
      "step": 2150
    },
    {
      "epoch": 0.6518518518518519,
      "grad_norm": 1.7890625,
      "learning_rate": 9.785185185185187e-06,
      "loss": 2.7754339599609374,
      "step": 2200
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 0.3349609375,
      "learning_rate": 9.723456790123457e-06,
      "loss": 2.7735556030273436,
      "step": 2250
    },
    {
      "epoch": 0.6814814814814815,
      "grad_norm": 0.53271484375,
      "learning_rate": 9.66172839506173e-06,
      "loss": 2.7741458129882814,
      "step": 2300
    },
    {
      "epoch": 0.6962962962962963,
      "grad_norm": 1.763671875,
      "learning_rate": 9.600000000000001e-06,
      "loss": 2.773301086425781,
      "step": 2350
    },
    {
      "epoch": 0.7111111111111111,
      "grad_norm": 1.1982421875,
      "learning_rate": 9.538271604938273e-06,
      "loss": 2.7742254638671877,
      "step": 2400
    },
    {
      "epoch": 0.725925925925926,
      "grad_norm": 2.26953125,
      "learning_rate": 9.476543209876544e-06,
      "loss": 2.7740939331054686,
      "step": 2450
    },
    {
      "epoch": 0.7407407407407407,
      "grad_norm": NaN,
      "learning_rate": 9.414814814814816e-06,
      "loss": 3.09239013671875,
      "step": 2500
    },
    {
      "epoch": 0.7555555555555555,
      "grad_norm": NaN,
      "learning_rate": 9.353086419753086e-06,
      "loss": 0.0,
      "step": 2550
    },
    {
      "epoch": 0.7703703703703704,
      "grad_norm": NaN,
      "learning_rate": 9.29135802469136e-06,
      "loss": 0.0,
      "step": 2600
    },
    {
      "epoch": 0.7851851851851852,
      "grad_norm": NaN,
      "learning_rate": 9.22962962962963e-06,
      "loss": 0.0,
      "step": 2650
    },
    {
      "epoch": 0.8,
      "grad_norm": NaN,
      "learning_rate": 9.167901234567903e-06,
      "loss": 0.0,
      "step": 2700
    },
    {
      "epoch": 0.8148148148148148,
      "grad_norm": NaN,
      "learning_rate": 9.106172839506173e-06,
      "loss": 0.0,
      "step": 2750
    },
    {
      "epoch": 0.8296296296296296,
      "grad_norm": NaN,
      "learning_rate": 9.044444444444445e-06,
      "loss": 0.0,
      "step": 2800
    },
    {
      "epoch": 0.8444444444444444,
      "grad_norm": NaN,
      "learning_rate": 8.982716049382717e-06,
      "loss": 0.0,
      "step": 2850
    },
    {
      "epoch": 0.8592592592592593,
      "grad_norm": NaN,
      "learning_rate": 8.92098765432099e-06,
      "loss": 0.0,
      "step": 2900
    },
    {
      "epoch": 0.8740740740740741,
      "grad_norm": NaN,
      "learning_rate": 8.85925925925926e-06,
      "loss": 0.0,
      "step": 2950
    },
    {
      "epoch": 0.8888888888888888,
      "grad_norm": NaN,
      "learning_rate": 8.797530864197532e-06,
      "loss": 0.0,
      "step": 3000
    },
    {
      "epoch": 0.9037037037037037,
      "grad_norm": NaN,
      "learning_rate": 8.735802469135804e-06,
      "loss": 0.0,
      "step": 3050
    },
    {
      "epoch": 0.9185185185185185,
      "grad_norm": NaN,
      "learning_rate": 8.674074074074074e-06,
      "loss": 0.0,
      "step": 3100
    },
    {
      "epoch": 0.9333333333333333,
      "grad_norm": NaN,
      "learning_rate": 8.612345679012346e-06,
      "loss": 0.0,
      "step": 3150
    },
    {
      "epoch": 0.9481481481481482,
      "grad_norm": NaN,
      "learning_rate": 8.550617283950618e-06,
      "loss": 0.0,
      "step": 3200
    },
    {
      "epoch": 0.9629629629629629,
      "grad_norm": NaN,
      "learning_rate": 8.48888888888889e-06,
      "loss": 0.0,
      "step": 3250
    },
    {
      "epoch": 0.9777777777777777,
      "grad_norm": NaN,
      "learning_rate": 8.42716049382716e-06,
      "loss": 0.0,
      "step": 3300
    },
    {
      "epoch": 0.9925925925925926,
      "grad_norm": NaN,
      "learning_rate": 8.365432098765433e-06,
      "loss": 0.0,
      "step": 3350
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.25075,
      "eval_loss": NaN,
      "eval_runtime": 17.5086,
      "eval_samples_per_second": 685.376,
      "eval_steps_per_second": 5.369,
      "step": 3375
    }
  ],
  "logging_steps": 50,
  "max_steps": 10125,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 2,
        "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": 7104253464576000.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}