{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 5268,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.28,
      "learning_rate": 1.810174639331815e-05,
      "loss": 0.2994,
      "step": 500
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6203492786636296e-05,
      "loss": 0.1291,
      "step": 1000
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4305239179954442e-05,
      "loss": 0.0928,
      "step": 1500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9757461588273385,
      "eval_f1": 0.8900808714309293,
      "eval_loss": 0.08755526691675186,
      "eval_precision": 0.8732189119170984,
      "eval_recall": 0.9076068663749579,
      "eval_runtime": 6.4367,
      "eval_samples_per_second": 504.92,
      "eval_steps_per_second": 63.232,
      "step": 1756
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.240698557327259e-05,
      "loss": 0.076,
      "step": 2000
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0508731966590738e-05,
      "loss": 0.0546,
      "step": 2500
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.610478359908885e-06,
      "loss": 0.0571,
      "step": 3000
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.712224753227031e-06,
      "loss": 0.0456,
      "step": 3500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.981235650791782,
      "eval_f1": 0.919607190412783,
      "eval_loss": 0.0778830349445343,
      "eval_precision": 0.9096147513994073,
      "eval_recall": 0.929821608885897,
      "eval_runtime": 6.5577,
      "eval_samples_per_second": 495.604,
      "eval_steps_per_second": 62.065,
      "step": 3512
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.8139711465451785e-06,
      "loss": 0.0321,
      "step": 4000
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.9157175398633257e-06,
      "loss": 0.027,
      "step": 4500
    },
    {
      "epoch": 2.85,
      "learning_rate": 1.0174639331814731e-06,
      "loss": 0.0316,
      "step": 5000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9829722729145818,
      "eval_f1": 0.9245094778849351,
      "eval_loss": 0.07030738890171051,
      "eval_precision": 0.9135721327637201,
      "eval_recall": 0.9357118815213733,
      "eval_runtime": 6.7636,
      "eval_samples_per_second": 480.516,
      "eval_steps_per_second": 60.175,
      "step": 5268
    }
  ],
  "logging_steps": 500,
  "max_steps": 5268,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 460336113849150.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}