{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.761904761904762,
  "eval_steps": 10,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.23809523809523808,
      "grad_norm": 107.2943344116211,
      "learning_rate": 7e-06,
      "loss": 4.5568,
      "step": 10
    },
    {
      "epoch": 0.23809523809523808,
      "eval_loss": 4.329397678375244,
      "eval_runtime": 2.6051,
      "eval_samples_per_second": 982.322,
      "eval_steps_per_second": 15.355,
      "step": 10
    },
    {
      "epoch": 0.47619047619047616,
      "grad_norm": 82.67948150634766,
      "learning_rate": 1.7e-05,
      "loss": 3.937,
      "step": 20
    },
    {
      "epoch": 0.47619047619047616,
      "eval_loss": 3.166707754135132,
      "eval_runtime": 2.6689,
      "eval_samples_per_second": 958.832,
      "eval_steps_per_second": 14.988,
      "step": 20
    },
    {
      "epoch": 0.7142857142857143,
      "grad_norm": 81.49203491210938,
      "learning_rate": 1.9222222222222225e-05,
      "loss": 2.7135,
      "step": 30
    },
    {
      "epoch": 0.7142857142857143,
      "eval_loss": 1.7821017503738403,
      "eval_runtime": 3.0515,
      "eval_samples_per_second": 838.597,
      "eval_steps_per_second": 13.108,
      "step": 30
    },
    {
      "epoch": 0.9523809523809523,
      "grad_norm": 67.98638153076172,
      "learning_rate": 1.8111111111111112e-05,
      "loss": 1.6075,
      "step": 40
    },
    {
      "epoch": 0.9523809523809523,
      "eval_loss": 0.9218789339065552,
      "eval_runtime": 2.6581,
      "eval_samples_per_second": 962.702,
      "eval_steps_per_second": 15.048,
      "step": 40
    },
    {
      "epoch": 1.1904761904761905,
      "grad_norm": 41.94712448120117,
      "learning_rate": 1.7e-05,
      "loss": 0.7666,
      "step": 50
    },
    {
      "epoch": 1.1904761904761905,
      "eval_loss": 0.5395554304122925,
      "eval_runtime": 2.7233,
      "eval_samples_per_second": 939.664,
      "eval_steps_per_second": 14.688,
      "step": 50
    },
    {
      "epoch": 1.4285714285714286,
      "grad_norm": 45.76442337036133,
      "learning_rate": 1.588888888888889e-05,
      "loss": 0.4077,
      "step": 60
    },
    {
      "epoch": 1.4285714285714286,
      "eval_loss": 0.395812064409256,
      "eval_runtime": 2.6568,
      "eval_samples_per_second": 963.196,
      "eval_steps_per_second": 15.056,
      "step": 60
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 36.85358428955078,
      "learning_rate": 1.477777777777778e-05,
      "loss": 0.2843,
      "step": 70
    },
    {
      "epoch": 1.6666666666666665,
      "eval_loss": 0.3538413345813751,
      "eval_runtime": 2.958,
      "eval_samples_per_second": 865.099,
      "eval_steps_per_second": 13.522,
      "step": 70
    },
    {
      "epoch": 1.9047619047619047,
      "grad_norm": 17.167036056518555,
      "learning_rate": 1.3666666666666667e-05,
      "loss": 0.2165,
      "step": 80
    },
    {
      "epoch": 1.9047619047619047,
      "eval_loss": 0.3396788537502289,
      "eval_runtime": 3.0374,
      "eval_samples_per_second": 842.51,
      "eval_steps_per_second": 13.169,
      "step": 80
    },
    {
      "epoch": 2.142857142857143,
      "grad_norm": 28.207691192626953,
      "learning_rate": 1.2555555555555557e-05,
      "loss": 0.1979,
      "step": 90
    },
    {
      "epoch": 2.142857142857143,
      "eval_loss": 0.32209891080856323,
      "eval_runtime": 2.6763,
      "eval_samples_per_second": 956.167,
      "eval_steps_per_second": 14.946,
      "step": 90
    },
    {
      "epoch": 2.380952380952381,
      "grad_norm": 20.979476928710938,
      "learning_rate": 1.1444444444444444e-05,
      "loss": 0.1476,
      "step": 100
    },
    {
      "epoch": 2.380952380952381,
      "eval_loss": 0.30111950635910034,
      "eval_runtime": 2.7597,
      "eval_samples_per_second": 927.283,
      "eval_steps_per_second": 14.494,
      "step": 100
    },
    {
      "epoch": 2.619047619047619,
      "grad_norm": 14.82142448425293,
      "learning_rate": 1.0333333333333335e-05,
      "loss": 0.1061,
      "step": 110
    },
    {
      "epoch": 2.619047619047619,
      "eval_loss": 0.30764198303222656,
      "eval_runtime": 2.701,
      "eval_samples_per_second": 947.422,
      "eval_steps_per_second": 14.809,
      "step": 110
    },
    {
      "epoch": 2.857142857142857,
      "grad_norm": 19.38449478149414,
      "learning_rate": 9.222222222222224e-06,
      "loss": 0.0907,
      "step": 120
    },
    {
      "epoch": 2.857142857142857,
      "eval_loss": 0.31123870611190796,
      "eval_runtime": 2.7275,
      "eval_samples_per_second": 938.22,
      "eval_steps_per_second": 14.665,
      "step": 120
    },
    {
      "epoch": 3.0952380952380953,
      "grad_norm": 15.523360252380371,
      "learning_rate": 8.111111111111112e-06,
      "loss": 0.0573,
      "step": 130
    },
    {
      "epoch": 3.0952380952380953,
      "eval_loss": 0.306456595659256,
      "eval_runtime": 3.037,
      "eval_samples_per_second": 842.596,
      "eval_steps_per_second": 13.171,
      "step": 130
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 9.912215232849121,
      "learning_rate": 7e-06,
      "loss": 0.0638,
      "step": 140
    },
    {
      "epoch": 3.3333333333333335,
      "eval_loss": 0.3030768036842346,
      "eval_runtime": 2.784,
      "eval_samples_per_second": 919.181,
      "eval_steps_per_second": 14.368,
      "step": 140
    },
    {
      "epoch": 3.571428571428571,
      "grad_norm": 32.101985931396484,
      "learning_rate": 5.88888888888889e-06,
      "loss": 0.0716,
      "step": 150
    },
    {
      "epoch": 3.571428571428571,
      "eval_loss": 0.2990337312221527,
      "eval_runtime": 2.8128,
      "eval_samples_per_second": 909.761,
      "eval_steps_per_second": 14.221,
      "step": 150
    },
    {
      "epoch": 3.8095238095238093,
      "grad_norm": 17.976055145263672,
      "learning_rate": 4.777777777777778e-06,
      "loss": 0.0841,
      "step": 160
    },
    {
      "epoch": 3.8095238095238093,
      "eval_loss": 0.3003748655319214,
      "eval_runtime": 2.7098,
      "eval_samples_per_second": 944.346,
      "eval_steps_per_second": 14.761,
      "step": 160
    },
    {
      "epoch": 4.0476190476190474,
      "grad_norm": 16.045225143432617,
      "learning_rate": 3.6666666666666666e-06,
      "loss": 0.096,
      "step": 170
    },
    {
      "epoch": 4.0476190476190474,
      "eval_loss": 0.3065829873085022,
      "eval_runtime": 2.7428,
      "eval_samples_per_second": 933.004,
      "eval_steps_per_second": 14.584,
      "step": 170
    },
    {
      "epoch": 4.285714285714286,
      "grad_norm": 24.23316764831543,
      "learning_rate": 2.5555555555555557e-06,
      "loss": 0.0728,
      "step": 180
    },
    {
      "epoch": 4.285714285714286,
      "eval_loss": 0.3128698170185089,
      "eval_runtime": 3.0252,
      "eval_samples_per_second": 845.905,
      "eval_steps_per_second": 13.222,
      "step": 180
    },
    {
      "epoch": 4.523809523809524,
      "grad_norm": 10.853073120117188,
      "learning_rate": 1.4444444444444445e-06,
      "loss": 0.0594,
      "step": 190
    },
    {
      "epoch": 4.523809523809524,
      "eval_loss": 0.3163248300552368,
      "eval_runtime": 2.763,
      "eval_samples_per_second": 926.153,
      "eval_steps_per_second": 14.477,
      "step": 190
    },
    {
      "epoch": 4.761904761904762,
      "grad_norm": 18.616168975830078,
      "learning_rate": 3.3333333333333335e-07,
      "loss": 0.071,
      "step": 200
    },
    {
      "epoch": 4.761904761904762,
      "eval_loss": 0.3181891143321991,
      "eval_runtime": 3.0801,
      "eval_samples_per_second": 830.819,
      "eval_steps_per_second": 12.987,
      "step": 200
    }
  ],
  "logging_steps": 10,
  "max_steps": 200,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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
}