{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.08,
  "eval_steps": 1000,
  "global_step": 4000,
  "is_hyper_param_search": false,
  "is_local_process_zero": false,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.002,
      "grad_norm": 19.660429000854492,
      "learning_rate": 7.920000000000001e-05,
      "loss": 67.8543,
      "step": 100
    },
    {
      "epoch": 0.004,
      "grad_norm": 24.802082061767578,
      "learning_rate": 0.00015920000000000002,
      "loss": 54.1444,
      "step": 200
    },
    {
      "epoch": 0.006,
      "grad_norm": 16.059066772460938,
      "learning_rate": 0.00023920000000000001,
      "loss": 51.5921,
      "step": 300
    },
    {
      "epoch": 0.008,
      "grad_norm": 27.705520629882812,
      "learning_rate": 0.0003192,
      "loss": 50.0536,
      "step": 400
    },
    {
      "epoch": 0.01,
      "grad_norm": 25.79052734375,
      "learning_rate": 0.0003992,
      "loss": 48.6987,
      "step": 500
    },
    {
      "epoch": 0.012,
      "grad_norm": 14.82126522064209,
      "learning_rate": 0.00047920000000000005,
      "loss": 47.3182,
      "step": 600
    },
    {
      "epoch": 0.014,
      "grad_norm": 12.00988483428955,
      "learning_rate": 0.0005592,
      "loss": 46.6574,
      "step": 700
    },
    {
      "epoch": 0.016,
      "grad_norm": 32.48664474487305,
      "learning_rate": 0.0006392,
      "loss": 46.0747,
      "step": 800
    },
    {
      "epoch": 0.018,
      "grad_norm": 25.646282196044922,
      "learning_rate": 0.0007191999999999999,
      "loss": 45.6098,
      "step": 900
    },
    {
      "epoch": 0.02,
      "grad_norm": 16.071203231811523,
      "learning_rate": 0.0007992,
      "loss": 44.8537,
      "step": 1000
    },
    {
      "epoch": 0.02,
      "eval_accuracy": 0.18606457925636008,
      "eval_loss": 44.18140411376953,
      "eval_runtime": 14.581,
      "eval_samples_per_second": 8.573,
      "eval_steps_per_second": 0.137,
      "step": 1000
    },
    {
      "epoch": 0.022,
      "grad_norm": 14.277446746826172,
      "learning_rate": 0.0008792,
      "loss": 44.223,
      "step": 1100
    },
    {
      "epoch": 0.024,
      "grad_norm": 9.809870719909668,
      "learning_rate": 0.0009592000000000001,
      "loss": 43.5719,
      "step": 1200
    },
    {
      "epoch": 0.026,
      "grad_norm": 16.994014739990234,
      "learning_rate": 0.0010391999999999999,
      "loss": 43.0323,
      "step": 1300
    },
    {
      "epoch": 0.028,
      "grad_norm": 8.082005500793457,
      "learning_rate": 0.0011192,
      "loss": 42.3631,
      "step": 1400
    },
    {
      "epoch": 0.03,
      "grad_norm": 7.080636024475098,
      "learning_rate": 0.0011992,
      "loss": 42.0285,
      "step": 1500
    },
    {
      "epoch": 0.032,
      "grad_norm": 8.547815322875977,
      "learning_rate": 0.0012791999999999999,
      "loss": 42.6829,
      "step": 1600
    },
    {
      "epoch": 0.034,
      "grad_norm": 7.541939735412598,
      "learning_rate": 0.0013592,
      "loss": 41.9796,
      "step": 1700
    },
    {
      "epoch": 0.036,
      "grad_norm": 6.902923583984375,
      "learning_rate": 0.0014392,
      "loss": 41.5016,
      "step": 1800
    },
    {
      "epoch": 0.038,
      "grad_norm": 7.68479061126709,
      "learning_rate": 0.0015192,
      "loss": 41.2607,
      "step": 1900
    },
    {
      "epoch": 0.04,
      "grad_norm": 8.936702728271484,
      "learning_rate": 0.0015992,
      "loss": 40.7545,
      "step": 2000
    },
    {
      "epoch": 0.04,
      "eval_accuracy": 0.21657142857142858,
      "eval_loss": 40.417686462402344,
      "eval_runtime": 3.051,
      "eval_samples_per_second": 40.97,
      "eval_steps_per_second": 0.656,
      "step": 2000
    },
    {
      "epoch": 0.042,
      "grad_norm": 5.834764003753662,
      "learning_rate": 0.0016792,
      "loss": 40.2024,
      "step": 2100
    },
    {
      "epoch": 0.044,
      "grad_norm": 3.9864397048950195,
      "learning_rate": 0.0017592,
      "loss": 39.4534,
      "step": 2200
    },
    {
      "epoch": 0.046,
      "grad_norm": 6.052990436553955,
      "learning_rate": 0.0018392,
      "loss": 40.0088,
      "step": 2300
    },
    {
      "epoch": 0.048,
      "grad_norm": 10.583518981933594,
      "learning_rate": 0.0019192,
      "loss": 40.099,
      "step": 2400
    },
    {
      "epoch": 0.05,
      "grad_norm": 3.686554193496704,
      "learning_rate": 0.0019992,
      "loss": 39.5664,
      "step": 2500
    },
    {
      "epoch": 0.052,
      "grad_norm": 2.981701612472534,
      "learning_rate": 0.001999978563623903,
      "loss": 39.5087,
      "step": 2600
    },
    {
      "epoch": 0.054,
      "grad_norm": 3.382185459136963,
      "learning_rate": 0.0019999133871331223,
      "loss": 38.9185,
      "step": 2700
    },
    {
      "epoch": 0.056,
      "grad_norm": 2.8874025344848633,
      "learning_rate": 0.0019998044711915177,
      "loss": 38.1472,
      "step": 2800
    },
    {
      "epoch": 0.058,
      "grad_norm": 3.083789825439453,
      "learning_rate": 0.0019996518205634257,
      "loss": 37.5833,
      "step": 2900
    },
    {
      "epoch": 0.06,
      "grad_norm": 2.315279483795166,
      "learning_rate": 0.0019994554419262797,
      "loss": 37.927,
      "step": 3000
    },
    {
      "epoch": 0.06,
      "eval_accuracy": 0.23828180039138944,
      "eval_loss": 37.86294174194336,
      "eval_runtime": 3.1437,
      "eval_samples_per_second": 39.763,
      "eval_steps_per_second": 0.636,
      "step": 3000
    },
    {
      "epoch": 0.062,
      "grad_norm": 4.6366777420043945,
      "learning_rate": 0.001999215343870317,
      "loss": 38.432,
      "step": 3100
    },
    {
      "epoch": 0.064,
      "grad_norm": 2.8195507526397705,
      "learning_rate": 0.0019989315368982054,
      "loss": 37.8144,
      "step": 3200
    },
    {
      "epoch": 0.066,
      "grad_norm": 2.320726156234741,
      "learning_rate": 0.0019986040334245797,
      "loss": 38.1739,
      "step": 3300
    },
    {
      "epoch": 0.068,
      "grad_norm": 2.575469732284546,
      "learning_rate": 0.001998232847775504,
      "loss": 37.1996,
      "step": 3400
    },
    {
      "epoch": 0.07,
      "grad_norm": 2.4692764282226562,
      "learning_rate": 0.0019978179961878404,
      "loss": 36.8725,
      "step": 3500
    },
    {
      "epoch": 0.072,
      "grad_norm": 2.555267810821533,
      "learning_rate": 0.001997359496808541,
      "loss": 36.2038,
      "step": 3600
    },
    {
      "epoch": 0.074,
      "grad_norm": 1.9989144802093506,
      "learning_rate": 0.001996857369693855,
      "loss": 36.3164,
      "step": 3700
    },
    {
      "epoch": 0.076,
      "grad_norm": 2.223381519317627,
      "learning_rate": 0.0019963116368084486,
      "loss": 37.1086,
      "step": 3800
    },
    {
      "epoch": 0.078,
      "grad_norm": 1.996813416481018,
      "learning_rate": 0.001995722322024446,
      "loss": 36.5327,
      "step": 3900
    },
    {
      "epoch": 0.08,
      "grad_norm": 1.8378113508224487,
      "learning_rate": 0.001995089451120385,
      "loss": 36.314,
      "step": 4000
    },
    {
      "epoch": 0.08,
      "eval_accuracy": 0.2553561643835616,
      "eval_loss": 36.08281707763672,
      "eval_runtime": 3.1615,
      "eval_samples_per_second": 39.538,
      "eval_steps_per_second": 0.633,
      "step": 4000
    }
  ],
  "logging_steps": 100,
  "max_steps": 50000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 2000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}