last-checkpoint/trainer_state.json

{
  "best_metric": 0.5486997676789124,
  "best_model_checkpoint": "training_dir/checkpoint-1000",
  "epoch": 0.1059658789869662,
  "eval_steps": 1000,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1059658789869662,
      "grad_norm": 15.513019561767578,
      "learning_rate": 9.946734846063707e-06,
      "loss": 0.8105,
      "step": 1000
    },
    {
      "epoch": 0.1059658789869662,
      "eval_accuracy": 0.6267482517482518,
      "eval_f1": 0.5486997676789124,
      "eval_loss": 0.8523119688034058,
      "eval_precision": 0.6663256334906192,
      "eval_recall": 0.6267482517482518,
      "eval_runtime": 26.6488,
      "eval_samples_per_second": 85.857,
      "eval_steps_per_second": 10.732,
      "step": 1000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 94370,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 970057431418608.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}