{
  "best_metric": 0.9943507313728333,
  "best_model_checkpoint": "outputs/checkpoint-230",
  "epoch": 4.966261808367071,
  "eval_steps": 500,
  "global_step": 230,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.4318488529014845,
      "grad_norm": 1.4057484865188599,
      "learning_rate": 2e-05,
      "loss": 2.8828,
      "step": 20
    },
    {
      "epoch": 0.863697705802969,
      "grad_norm": 0.5148554444313049,
      "learning_rate": 4e-05,
      "loss": 2.4498,
      "step": 40
    },
    {
      "epoch": 0.9932523616734144,
      "eval_loss": 1.945718765258789,
      "eval_runtime": 35.3592,
      "eval_samples_per_second": 31.562,
      "eval_steps_per_second": 3.959,
      "step": 46
    },
    {
      "epoch": 1.2955465587044535,
      "grad_norm": 0.46272239089012146,
      "learning_rate": 6e-05,
      "loss": 2.0821,
      "step": 60
    },
    {
      "epoch": 1.7273954116059378,
      "grad_norm": 0.5183445811271667,
      "learning_rate": 8e-05,
      "loss": 1.8908,
      "step": 80
    },
    {
      "epoch": 1.9865047233468287,
      "eval_loss": 1.5600017309188843,
      "eval_runtime": 35.3207,
      "eval_samples_per_second": 31.596,
      "eval_steps_per_second": 3.964,
      "step": 92
    },
    {
      "epoch": 2.1592442645074224,
      "grad_norm": 0.7157186269760132,
      "learning_rate": 0.0001,
      "loss": 1.8117,
      "step": 100
    },
    {
      "epoch": 2.591093117408907,
      "grad_norm": 0.7727251052856445,
      "learning_rate": 9.42728012826605e-05,
      "loss": 1.652,
      "step": 120
    },
    {
      "epoch": 2.979757085020243,
      "eval_loss": 1.2167656421661377,
      "eval_runtime": 35.3143,
      "eval_samples_per_second": 31.602,
      "eval_steps_per_second": 3.964,
      "step": 138
    },
    {
      "epoch": 3.0229419703103915,
      "grad_norm": 1.1024483442306519,
      "learning_rate": 7.840323733655778e-05,
      "loss": 1.5965,
      "step": 140
    },
    {
      "epoch": 3.454790823211876,
      "grad_norm": 1.2325457334518433,
      "learning_rate": 5.602683401276615e-05,
      "loss": 1.4536,
      "step": 160
    },
    {
      "epoch": 3.8866396761133606,
      "grad_norm": 1.153786063194275,
      "learning_rate": 3.226975564787322e-05,
      "loss": 1.423,
      "step": 180
    },
    {
      "epoch": 3.9946018893387314,
      "eval_loss": 1.030889868736267,
      "eval_runtime": 35.3197,
      "eval_samples_per_second": 31.597,
      "eval_steps_per_second": 3.964,
      "step": 185
    },
    {
      "epoch": 4.318488529014845,
      "grad_norm": 1.3771637678146362,
      "learning_rate": 1.257446259144494e-05,
      "loss": 1.3297,
      "step": 200
    },
    {
      "epoch": 4.75033738191633,
      "grad_norm": 1.443164587020874,
      "learning_rate": 1.4529091286973995e-06,
      "loss": 1.3588,
      "step": 220
    },
    {
      "epoch": 4.966261808367071,
      "eval_loss": 0.9943507313728333,
      "eval_runtime": 35.323,
      "eval_samples_per_second": 31.594,
      "eval_steps_per_second": 3.963,
      "step": 230
    }
  ],
  "logging_steps": 20,
  "max_steps": 230,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 1.990131695389901e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}