last-checkpoint/trainer_state.json

{
  "best_metric": 2.5578970909118652,
  "best_model_checkpoint": "./outputs/checkpoint-300",
  "epoch": 0.2185792349726776,
  "eval_steps": 100,
  "global_step": 300,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 2.7401,
      "step": 100
    },
    {
      "epoch": 0.07,
      "eval_loss": 2.6424810886383057,
      "eval_runtime": 206.4108,
      "eval_samples_per_second": 30.396,
      "eval_steps_per_second": 3.803,
      "step": 100
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002,
      "loss": 2.6061,
      "step": 200
    },
    {
      "epoch": 0.15,
      "eval_loss": 2.5929574966430664,
      "eval_runtime": 206.2177,
      "eval_samples_per_second": 30.424,
      "eval_steps_per_second": 3.807,
      "step": 200
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002,
      "loss": 2.5643,
      "step": 300
    },
    {
      "epoch": 0.22,
      "eval_loss": 2.5578970909118652,
      "eval_runtime": 206.509,
      "eval_samples_per_second": 30.381,
      "eval_steps_per_second": 3.801,
      "step": 300
    }
  ],
  "logging_steps": 100,
  "max_steps": 4116,
  "num_train_epochs": 3,
  "save_steps": 100,
  "total_flos": 8750270976491520.0,
  "trial_name": null,
  "trial_params": null
}