{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 252,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.11904761904761904,
      "grad_norm": 12.131371063020744,
      "learning_rate": 1.3157894736842106e-06,
      "loss": 0.8467,
      "step": 10
    },
    {
      "epoch": 0.23809523809523808,
      "grad_norm": 2.8702347939026684,
      "learning_rate": 2.631578947368421e-06,
      "loss": 0.7188,
      "step": 20
    },
    {
      "epoch": 0.35714285714285715,
      "grad_norm": 2.5230518976542506,
      "learning_rate": 3.947368421052632e-06,
      "loss": 0.635,
      "step": 30
    },
    {
      "epoch": 0.47619047619047616,
      "grad_norm": 1.6425584319246187,
      "learning_rate": 4.998922515567496e-06,
      "loss": 0.5914,
      "step": 40
    },
    {
      "epoch": 0.5952380952380952,
      "grad_norm": 3.9561567299131952,
      "learning_rate": 4.9613079925074435e-06,
      "loss": 0.5662,
      "step": 50
    },
    {
      "epoch": 0.7142857142857143,
      "grad_norm": 2.8506571881322786,
      "learning_rate": 4.870744433740688e-06,
      "loss": 0.5495,
      "step": 60
    },
    {
      "epoch": 0.8333333333333334,
      "grad_norm": 2.9284294355648997,
      "learning_rate": 4.729180093253106e-06,
      "loss": 0.5354,
      "step": 70
    },
    {
      "epoch": 0.9523809523809523,
      "grad_norm": 1.608687573355936,
      "learning_rate": 4.53966038240406e-06,
      "loss": 0.523,
      "step": 80
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.06461798399686813,
      "eval_runtime": 89.5731,
      "eval_samples_per_second": 202.103,
      "eval_steps_per_second": 0.402,
      "step": 84
    },
    {
      "epoch": 1.0714285714285714,
      "grad_norm": 3.5277524106115705,
      "learning_rate": 4.306262355332006e-06,
      "loss": 0.5029,
      "step": 90
    },
    {
      "epoch": 1.1904761904761905,
      "grad_norm": 2.9157076219013502,
      "learning_rate": 4.034007001082985e-06,
      "loss": 0.4873,
      "step": 100
    },
    {
      "epoch": 1.3095238095238095,
      "grad_norm": 1.9515685257710755,
      "learning_rate": 3.7287512292828364e-06,
      "loss": 0.4835,
      "step": 110
    },
    {
      "epoch": 1.4285714285714286,
      "grad_norm": 3.0996510453406376,
      "learning_rate": 3.39706187301784e-06,
      "loss": 0.4795,
      "step": 120
    },
    {
      "epoch": 1.5476190476190477,
      "grad_norm": 1.6487179646275694,
      "learning_rate": 3.0460744194443658e-06,
      "loss": 0.4738,
      "step": 130
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 1.6412743986342637,
      "learning_rate": 2.68333950719376e-06,
      "loss": 0.4706,
      "step": 140
    },
    {
      "epoch": 1.7857142857142856,
      "grad_norm": 1.756612697830082,
      "learning_rate": 2.3166604928062407e-06,
      "loss": 0.4675,
      "step": 150
    },
    {
      "epoch": 1.9047619047619047,
      "grad_norm": 1.0971644453746074,
      "learning_rate": 1.9539255805556346e-06,
      "loss": 0.4633,
      "step": 160
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.06034243851900101,
      "eval_runtime": 89.3139,
      "eval_samples_per_second": 202.69,
      "eval_steps_per_second": 0.403,
      "step": 168
    },
    {
      "epoch": 2.0238095238095237,
      "grad_norm": 1.174978789267966,
      "learning_rate": 1.6029381269821607e-06,
      "loss": 0.4564,
      "step": 170
    },
    {
      "epoch": 2.142857142857143,
      "grad_norm": 0.8542389334158027,
      "learning_rate": 1.2712487707171645e-06,
      "loss": 0.4356,
      "step": 180
    },
    {
      "epoch": 2.261904761904762,
      "grad_norm": 0.7154367298759456,
      "learning_rate": 9.659929989170156e-07,
      "loss": 0.4335,
      "step": 190
    },
    {
      "epoch": 2.380952380952381,
      "grad_norm": 0.5531641628675295,
      "learning_rate": 6.93737644667995e-07,
      "loss": 0.43,
      "step": 200
    },
    {
      "epoch": 2.5,
      "grad_norm": 0.505015673695498,
      "learning_rate": 4.6033961759594045e-07,
      "loss": 0.429,
      "step": 210
    },
    {
      "epoch": 2.619047619047619,
      "grad_norm": 0.48866524545920653,
      "learning_rate": 2.708199067468939e-07,
      "loss": 0.4311,
      "step": 220
    },
    {
      "epoch": 2.738095238095238,
      "grad_norm": 0.46556587701838564,
      "learning_rate": 1.2925556625931174e-07,
      "loss": 0.4303,
      "step": 230
    },
    {
      "epoch": 2.857142857142857,
      "grad_norm": 0.45472603633578945,
      "learning_rate": 3.869200749255703e-08,
      "loss": 0.4296,
      "step": 240
    },
    {
      "epoch": 2.9761904761904763,
      "grad_norm": 0.43987009685051254,
      "learning_rate": 1.0774844325039946e-09,
      "loss": 0.4277,
      "step": 250
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.05977506935596466,
      "eval_runtime": 88.5977,
      "eval_samples_per_second": 204.328,
      "eval_steps_per_second": 0.406,
      "step": 252
    },
    {
      "epoch": 3.0,
      "step": 252,
      "total_flos": 3375200049561600.0,
      "train_loss": 0.5070906683566079,
      "train_runtime": 15180.8601,
      "train_samples_per_second": 67.971,
      "train_steps_per_second": 0.017
    }
  ],
  "logging_steps": 10,
  "max_steps": 252,
  "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": 3375200049561600.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}