{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.764705882352941,
  "eval_steps": 25,
  "global_step": 8,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.47058823529411764,
      "grad_norm": 1.2521203756332397,
      "learning_rate": 0.000175,
      "loss": 9.5734,
      "step": 1
    },
    {
      "epoch": 0.9411764705882353,
      "grad_norm": 1.106857419013977,
      "learning_rate": 0.00015000000000000001,
      "loss": 9.4203,
      "step": 2
    },
    {
      "epoch": 1.4117647058823528,
      "grad_norm": 1.2563574314117432,
      "learning_rate": 0.000125,
      "loss": 9.2271,
      "step": 3
    },
    {
      "epoch": 1.8823529411764706,
      "grad_norm": 1.5731617212295532,
      "learning_rate": 0.0001,
      "loss": 9.0003,
      "step": 4
    },
    {
      "epoch": 2.3529411764705883,
      "grad_norm": 1.5413283109664917,
      "learning_rate": 7.500000000000001e-05,
      "loss": 8.8589,
      "step": 5
    },
    {
      "epoch": 2.8235294117647056,
      "grad_norm": 1.7880181074142456,
      "learning_rate": 5e-05,
      "loss": 8.6594,
      "step": 6
    },
    {
      "epoch": 3.2941176470588234,
      "grad_norm": 2.0251083374023438,
      "learning_rate": 2.5e-05,
      "loss": 8.5469,
      "step": 7
    },
    {
      "epoch": 3.764705882352941,
      "grad_norm": 2.190796136856079,
      "learning_rate": 0.0,
      "loss": 8.4155,
      "step": 8
    },
    {
      "epoch": 3.764705882352941,
      "step": 8,
      "total_flos": 35226970482492.0,
      "train_loss": 8.96271562576294,
      "train_runtime": 144.1141,
      "train_samples_per_second": 0.916,
      "train_steps_per_second": 0.056
    }
  ],
  "logging_steps": 1,
  "max_steps": 8,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 35226970482492.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}