trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 1144,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 2e-05,
      "loss": 1.2868,
      "step": 1
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.2366484158895118e-05,
      "loss": 0.9791,
      "step": 500
    },
    {
      "epoch": 0.87,
      "learning_rate": 9.382120760135128e-07,
      "loss": 0.9384,
      "step": 1000
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.9400383830070496,
      "eval_runtime": 173.4479,
      "eval_samples_per_second": 93.434,
      "eval_steps_per_second": 0.369,
      "step": 1144
    },
    {
      "epoch": 1.0,
      "step": 1144,
      "total_flos": 472360503214080.0,
      "train_loss": 0.9564839349343226,
      "train_runtime": 7467.4621,
      "train_samples_per_second": 19.608,
      "train_steps_per_second": 0.153
    }
  ],
  "logging_steps": 500,
  "max_steps": 1144,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 472360503214080.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}