{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.857142857142857,
  "eval_steps": 500,
  "global_step": 15,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.19,
      "grad_norm": 1.2310200929641724,
      "learning_rate": 0.0001,
      "loss": 4.5852,
      "step": 1
    },
    {
      "epoch": 0.38,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 4.2091,
      "step": 2
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.9326909184455872,
      "learning_rate": 0.0002,
      "loss": 4.3676,
      "step": 3
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.9319186806678772,
      "learning_rate": 0.00018461538461538463,
      "loss": 4.7135,
      "step": 4
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.8805396556854248,
      "learning_rate": 0.00016923076923076923,
      "loss": 4.6183,
      "step": 5
    },
    {
      "epoch": 1.14,
      "grad_norm": 1.1888014078140259,
      "learning_rate": 0.00015384615384615385,
      "loss": 4.1549,
      "step": 6
    },
    {
      "epoch": 1.33,
      "grad_norm": 1.1059948205947876,
      "learning_rate": 0.00013846153846153847,
      "loss": 4.1283,
      "step": 7
    },
    {
      "epoch": 1.52,
      "grad_norm": 1.4926124811172485,
      "learning_rate": 0.0001230769230769231,
      "loss": 4.4317,
      "step": 8
    },
    {
      "epoch": 1.71,
      "grad_norm": 1.425428032875061,
      "learning_rate": 0.0001076923076923077,
      "loss": 4.1823,
      "step": 9
    },
    {
      "epoch": 1.9,
      "grad_norm": 1.4075274467468262,
      "learning_rate": 9.230769230769232e-05,
      "loss": 4.0179,
      "step": 10
    },
    {
      "epoch": 2.1,
      "grad_norm": NaN,
      "learning_rate": 9.230769230769232e-05,
      "loss": 3.9234,
      "step": 11
    },
    {
      "epoch": 2.29,
      "grad_norm": 1.9703055620193481,
      "learning_rate": 7.692307692307693e-05,
      "loss": 3.8184,
      "step": 12
    },
    {
      "epoch": 2.48,
      "grad_norm": 1.3431462049484253,
      "learning_rate": 6.153846153846155e-05,
      "loss": 3.7346,
      "step": 13
    },
    {
      "epoch": 2.67,
      "grad_norm": 1.342360019683838,
      "learning_rate": 4.615384615384616e-05,
      "loss": 4.0967,
      "step": 14
    },
    {
      "epoch": 2.86,
      "grad_norm": 1.568617343902588,
      "learning_rate": 3.0769230769230774e-05,
      "loss": 3.9014,
      "step": 15
    }
  ],
  "logging_steps": 1,
  "max_steps": 15,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 186722426290176.0,
  "train_batch_size": 5,
  "trial_name": null,
  "trial_params": null
}