{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.04,
  "eval_steps": 1000,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": false,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.002,
      "grad_norm": 22.030515670776367,
      "learning_rate": 7.920000000000001e-05,
      "loss": 71.6835,
      "step": 100
    },
    {
      "epoch": 0.004,
      "grad_norm": 21.3626651763916,
      "learning_rate": 0.00015920000000000002,
      "loss": 56.4135,
      "step": 200
    },
    {
      "epoch": 0.006,
      "grad_norm": 21.728944778442383,
      "learning_rate": 0.00023920000000000001,
      "loss": 53.2586,
      "step": 300
    },
    {
      "epoch": 0.008,
      "grad_norm": 16.323501586914062,
      "learning_rate": 0.0003192,
      "loss": 51.3419,
      "step": 400
    },
    {
      "epoch": 0.01,
      "grad_norm": 11.02919864654541,
      "learning_rate": 0.0003992,
      "loss": 49.8526,
      "step": 500
    },
    {
      "epoch": 0.012,
      "grad_norm": 18.37352180480957,
      "learning_rate": 0.00047920000000000005,
      "loss": 48.3665,
      "step": 600
    },
    {
      "epoch": 0.014,
      "grad_norm": 13.519577026367188,
      "learning_rate": 0.0005592,
      "loss": 47.4206,
      "step": 700
    },
    {
      "epoch": 0.016,
      "grad_norm": 20.760446548461914,
      "learning_rate": 0.0006392,
      "loss": 46.6753,
      "step": 800
    },
    {
      "epoch": 0.018,
      "grad_norm": 25.183908462524414,
      "learning_rate": 0.0007191999999999999,
      "loss": 46.4672,
      "step": 900
    },
    {
      "epoch": 0.02,
      "grad_norm": 8.60799789428711,
      "learning_rate": 0.0007992,
      "loss": 45.7788,
      "step": 1000
    },
    {
      "epoch": 0.02,
      "eval_accuracy": 0.17874951076320938,
      "eval_loss": 45.424434661865234,
      "eval_runtime": 18.125,
      "eval_samples_per_second": 6.897,
      "eval_steps_per_second": 0.11,
      "step": 1000
    },
    {
      "epoch": 0.022,
      "grad_norm": 7.3993916511535645,
      "learning_rate": 0.0008792,
      "loss": 44.8989,
      "step": 1100
    },
    {
      "epoch": 0.024,
      "grad_norm": 11.284933090209961,
      "learning_rate": 0.0009592000000000001,
      "loss": 44.3576,
      "step": 1200
    },
    {
      "epoch": 0.026,
      "grad_norm": 4.662988185882568,
      "learning_rate": 0.0010391999999999999,
      "loss": 43.4894,
      "step": 1300
    },
    {
      "epoch": 0.028,
      "grad_norm": 7.507369041442871,
      "learning_rate": 0.0011192,
      "loss": 42.9519,
      "step": 1400
    },
    {
      "epoch": 0.03,
      "grad_norm": 6.868042469024658,
      "learning_rate": 0.0011992,
      "loss": 42.3116,
      "step": 1500
    },
    {
      "epoch": 0.032,
      "grad_norm": 10.319790840148926,
      "learning_rate": 0.0012791999999999999,
      "loss": 43.2361,
      "step": 1600
    },
    {
      "epoch": 0.034,
      "grad_norm": 5.94151496887207,
      "learning_rate": 0.0013592,
      "loss": 42.3295,
      "step": 1700
    },
    {
      "epoch": 0.036,
      "grad_norm": 6.9707865715026855,
      "learning_rate": 0.0014392,
      "loss": 42.1999,
      "step": 1800
    },
    {
      "epoch": 0.038,
      "grad_norm": 17.769460678100586,
      "learning_rate": 0.0015192,
      "loss": 41.7166,
      "step": 1900
    },
    {
      "epoch": 0.04,
      "grad_norm": 3.8550658226013184,
      "learning_rate": 0.0015992,
      "loss": 41.1934,
      "step": 2000
    },
    {
      "epoch": 0.04,
      "eval_accuracy": 0.20908414872798434,
      "eval_loss": 41.37644958496094,
      "eval_runtime": 3.4901,
      "eval_samples_per_second": 35.816,
      "eval_steps_per_second": 0.573,
      "step": 2000
    }
  ],
  "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
}