last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.1428571428571428,
  "eval_steps": 1,
  "global_step": 3,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.38095238095238093,
      "eval_loss": 10.373360633850098,
      "eval_runtime": 0.0348,
      "eval_samples_per_second": 517.209,
      "eval_steps_per_second": 86.201,
      "step": 1
    },
    {
      "epoch": 0.7619047619047619,
      "eval_loss": 10.373323440551758,
      "eval_runtime": 0.0366,
      "eval_samples_per_second": 491.159,
      "eval_steps_per_second": 81.86,
      "step": 2
    },
    {
      "epoch": 1.1428571428571428,
      "eval_loss": 10.373238563537598,
      "eval_runtime": 0.0371,
      "eval_samples_per_second": 485.034,
      "eval_steps_per_second": 80.839,
      "step": 3
    }
  ],
  "logging_steps": 10,
  "max_steps": 3,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 1,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 614218137600.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}