last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.1399972000559989,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07,
      "learning_rate": 1.982500349993e-05,
      "loss": 3.4571,
      "step": 500
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9650006999860005e-05,
      "loss": 3.3857,
      "step": 1000
    },
    {
      "epoch": 0.14,
      "eval_loss": 3.312843084335327,
      "eval_runtime": 161.8869,
      "eval_samples_per_second": 7.647,
      "eval_steps_per_second": 7.647,
      "step": 1000
    }
  ],
  "max_steps": 57144,
  "num_train_epochs": 8,
  "total_flos": 1019939328000000.0,
  "trial_name": null,
  "trial_params": null
}