checkpoint-87/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 87,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9,
      "eval_loss": 0.2707670331001282,
      "eval_runtime": 19.5658,
      "eval_samples_per_second": 2.044,
      "eval_steps_per_second": 0.256,
      "step": 29
    },
    {
      "epoch": 1.0344827586206897,
      "grad_norm": 4.015429496765137,
      "learning_rate": 3.275862068965517e-05,
      "loss": 0.7211,
      "step": 30
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9,
      "eval_loss": 0.34923750162124634,
      "eval_runtime": 19.9116,
      "eval_samples_per_second": 2.009,
      "eval_steps_per_second": 0.251,
      "step": 58
    },
    {
      "epoch": 2.0689655172413794,
      "grad_norm": 0.996155321598053,
      "learning_rate": 1.5517241379310346e-05,
      "loss": 0.1894,
      "step": 60
    }
  ],
  "logging_steps": 30,
  "max_steps": 87,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 90210298484736.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}