trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9961360544217687,
  "global_step": 143,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.28,
      "eval_loss": 1.8657242059707642,
      "eval_runtime": 240.0653,
      "eval_samples_per_second": 18.37,
      "eval_steps_per_second": 9.185,
      "step": 40
    },
    {
      "epoch": 0.56,
      "eval_loss": 1.6290868520736694,
      "eval_runtime": 240.1735,
      "eval_samples_per_second": 18.362,
      "eval_steps_per_second": 9.181,
      "step": 80
    },
    {
      "epoch": 0.84,
      "eval_loss": 1.4208073616027832,
      "eval_runtime": 237.5896,
      "eval_samples_per_second": 18.561,
      "eval_steps_per_second": 9.281,
      "step": 120
    },
    {
      "epoch": 1.0,
      "step": 143,
      "total_flos": 1.1430717843131597e+18,
      "train_loss": 1.8730214792531688,
      "train_runtime": 31924.7845,
      "train_samples_per_second": 4.605,
      "train_steps_per_second": 0.004
    }
  ],
  "max_steps": 143,
  "num_train_epochs": 1,
  "total_flos": 1.1430717843131597e+18,
  "trial_name": null,
  "trial_params": null
}