checkpoint-60/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 60,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.85,
      "eval_loss": 0.3328920304775238,
      "eval_runtime": 19.3098,
      "eval_samples_per_second": 2.071,
      "eval_steps_per_second": 0.259,
      "step": 20
    },
    {
      "epoch": 1.5,
      "grad_norm": 1.1108067035675049,
      "learning_rate": 2.5e-05,
      "loss": 0.5935,
      "step": 30
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9,
      "eval_loss": 0.36818569898605347,
      "eval_runtime": 19.4153,
      "eval_samples_per_second": 2.06,
      "eval_steps_per_second": 0.258,
      "step": 40
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.13847129046916962,
      "learning_rate": 0.0,
      "loss": 0.1087,
      "step": 60
    }
  ],
  "logging_steps": 30,
  "max_steps": 60,
  "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": 63584351354880.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}