{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.3529411764705883,
  "eval_steps": 10,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.11764705882352941,
      "grad_norm": 139.67864990234375,
      "learning_rate": 7e-06,
      "loss": 4.668,
      "step": 10
    },
    {
      "epoch": 0.11764705882352941,
      "eval_loss": 4.418745517730713,
      "eval_runtime": 12.6399,
      "eval_samples_per_second": 430.778,
      "eval_steps_per_second": 6.804,
      "step": 10
    },
    {
      "epoch": 0.23529411764705882,
      "grad_norm": 82.81034088134766,
      "learning_rate": 1.7e-05,
      "loss": 4.1806,
      "step": 20
    },
    {
      "epoch": 0.23529411764705882,
      "eval_loss": 3.803370237350464,
      "eval_runtime": 12.5875,
      "eval_samples_per_second": 432.573,
      "eval_steps_per_second": 6.832,
      "step": 20
    },
    {
      "epoch": 0.35294117647058826,
      "grad_norm": 84.353271484375,
      "learning_rate": 1.9222222222222225e-05,
      "loss": 3.4003,
      "step": 30
    },
    {
      "epoch": 0.35294117647058826,
      "eval_loss": 3.040264844894409,
      "eval_runtime": 12.8707,
      "eval_samples_per_second": 423.054,
      "eval_steps_per_second": 6.682,
      "step": 30
    },
    {
      "epoch": 0.47058823529411764,
      "grad_norm": 103.20001220703125,
      "learning_rate": 1.8111111111111112e-05,
      "loss": 2.2696,
      "step": 40
    },
    {
      "epoch": 0.47058823529411764,
      "eval_loss": 2.149442195892334,
      "eval_runtime": 13.2329,
      "eval_samples_per_second": 411.473,
      "eval_steps_per_second": 6.499,
      "step": 40
    },
    {
      "epoch": 0.5882352941176471,
      "grad_norm": 76.80607604980469,
      "learning_rate": 1.7e-05,
      "loss": 1.274,
      "step": 50
    },
    {
      "epoch": 0.5882352941176471,
      "eval_loss": 1.369861125946045,
      "eval_runtime": 13.2745,
      "eval_samples_per_second": 410.185,
      "eval_steps_per_second": 6.479,
      "step": 50
    },
    {
      "epoch": 0.7058823529411765,
      "grad_norm": 59.09462356567383,
      "learning_rate": 1.588888888888889e-05,
      "loss": 0.6758,
      "step": 60
    },
    {
      "epoch": 0.7058823529411765,
      "eval_loss": 1.0373408794403076,
      "eval_runtime": 13.2464,
      "eval_samples_per_second": 411.056,
      "eval_steps_per_second": 6.492,
      "step": 60
    },
    {
      "epoch": 0.8235294117647058,
      "grad_norm": 45.71690368652344,
      "learning_rate": 1.477777777777778e-05,
      "loss": 0.3574,
      "step": 70
    },
    {
      "epoch": 0.8235294117647058,
      "eval_loss": 0.9310317039489746,
      "eval_runtime": 13.0465,
      "eval_samples_per_second": 417.354,
      "eval_steps_per_second": 6.592,
      "step": 70
    },
    {
      "epoch": 0.9411764705882353,
      "grad_norm": 44.961952209472656,
      "learning_rate": 1.3666666666666667e-05,
      "loss": 0.3313,
      "step": 80
    },
    {
      "epoch": 0.9411764705882353,
      "eval_loss": 0.9087818264961243,
      "eval_runtime": 13.0319,
      "eval_samples_per_second": 417.82,
      "eval_steps_per_second": 6.599,
      "step": 80
    },
    {
      "epoch": 1.0588235294117647,
      "grad_norm": 49.15745544433594,
      "learning_rate": 1.2555555555555557e-05,
      "loss": 0.3025,
      "step": 90
    },
    {
      "epoch": 1.0588235294117647,
      "eval_loss": 0.8983781337738037,
      "eval_runtime": 13.3434,
      "eval_samples_per_second": 408.067,
      "eval_steps_per_second": 6.445,
      "step": 90
    },
    {
      "epoch": 1.1764705882352942,
      "grad_norm": 79.36802673339844,
      "learning_rate": 1.1444444444444444e-05,
      "loss": 0.2912,
      "step": 100
    },
    {
      "epoch": 1.1764705882352942,
      "eval_loss": 0.9225805401802063,
      "eval_runtime": 13.1562,
      "eval_samples_per_second": 413.873,
      "eval_steps_per_second": 6.537,
      "step": 100
    },
    {
      "epoch": 1.2941176470588236,
      "grad_norm": 35.45604705810547,
      "learning_rate": 1.0333333333333335e-05,
      "loss": 0.3154,
      "step": 110
    },
    {
      "epoch": 1.2941176470588236,
      "eval_loss": 0.9188823699951172,
      "eval_runtime": 13.2204,
      "eval_samples_per_second": 411.863,
      "eval_steps_per_second": 6.505,
      "step": 110
    },
    {
      "epoch": 1.4117647058823528,
      "grad_norm": 40.45594787597656,
      "learning_rate": 9.222222222222224e-06,
      "loss": 0.2466,
      "step": 120
    },
    {
      "epoch": 1.4117647058823528,
      "eval_loss": 0.8772405982017517,
      "eval_runtime": 13.0716,
      "eval_samples_per_second": 416.552,
      "eval_steps_per_second": 6.579,
      "step": 120
    },
    {
      "epoch": 1.5294117647058822,
      "grad_norm": 29.228164672851562,
      "learning_rate": 8.111111111111112e-06,
      "loss": 0.2124,
      "step": 130
    },
    {
      "epoch": 1.5294117647058822,
      "eval_loss": 0.8616596460342407,
      "eval_runtime": 13.1898,
      "eval_samples_per_second": 412.817,
      "eval_steps_per_second": 6.52,
      "step": 130
    },
    {
      "epoch": 1.6470588235294117,
      "grad_norm": 50.698936462402344,
      "learning_rate": 7e-06,
      "loss": 0.2917,
      "step": 140
    },
    {
      "epoch": 1.6470588235294117,
      "eval_loss": 0.855468213558197,
      "eval_runtime": 13.3221,
      "eval_samples_per_second": 408.718,
      "eval_steps_per_second": 6.455,
      "step": 140
    },
    {
      "epoch": 1.7647058823529411,
      "grad_norm": 35.14857482910156,
      "learning_rate": 5.88888888888889e-06,
      "loss": 0.1959,
      "step": 150
    },
    {
      "epoch": 1.7647058823529411,
      "eval_loss": 0.8391666412353516,
      "eval_runtime": 13.0706,
      "eval_samples_per_second": 416.582,
      "eval_steps_per_second": 6.58,
      "step": 150
    },
    {
      "epoch": 1.8823529411764706,
      "grad_norm": 36.21611785888672,
      "learning_rate": 4.777777777777778e-06,
      "loss": 0.1968,
      "step": 160
    },
    {
      "epoch": 1.8823529411764706,
      "eval_loss": 0.8412739038467407,
      "eval_runtime": 13.1041,
      "eval_samples_per_second": 415.52,
      "eval_steps_per_second": 6.563,
      "step": 160
    },
    {
      "epoch": 2.0,
      "grad_norm": 44.26481246948242,
      "learning_rate": 3.6666666666666666e-06,
      "loss": 0.1807,
      "step": 170
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.8440327644348145,
      "eval_runtime": 13.2994,
      "eval_samples_per_second": 409.418,
      "eval_steps_per_second": 6.466,
      "step": 170
    },
    {
      "epoch": 2.1176470588235294,
      "grad_norm": 42.70452117919922,
      "learning_rate": 2.5555555555555557e-06,
      "loss": 0.1773,
      "step": 180
    },
    {
      "epoch": 2.1176470588235294,
      "eval_loss": 0.8340872526168823,
      "eval_runtime": 13.2643,
      "eval_samples_per_second": 410.5,
      "eval_steps_per_second": 6.484,
      "step": 180
    },
    {
      "epoch": 2.235294117647059,
      "grad_norm": 11.37656307220459,
      "learning_rate": 1.4444444444444445e-06,
      "loss": 0.1533,
      "step": 190
    },
    {
      "epoch": 2.235294117647059,
      "eval_loss": 0.8336626291275024,
      "eval_runtime": 13.39,
      "eval_samples_per_second": 406.648,
      "eval_steps_per_second": 6.423,
      "step": 190
    },
    {
      "epoch": 2.3529411764705883,
      "grad_norm": 26.42169761657715,
      "learning_rate": 3.3333333333333335e-07,
      "loss": 0.1541,
      "step": 200
    },
    {
      "epoch": 2.3529411764705883,
      "eval_loss": 0.8328158855438232,
      "eval_runtime": 13.1345,
      "eval_samples_per_second": 414.558,
      "eval_steps_per_second": 6.548,
      "step": 200
    }
  ],
  "logging_steps": 10,
  "max_steps": 200,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 2500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}