checkpoint-73632/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 73632,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 11.612536430358887,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 0.5193,
      "step": 24544
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8372898624554254,
      "eval_loss": 0.4270987808704376,
      "eval_runtime": 41.8372,
      "eval_samples_per_second": 234.6,
      "eval_steps_per_second": 29.328,
      "step": 24544
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.1507811546325684,
      "learning_rate": 6.666666666666667e-06,
      "loss": 0.3333,
      "step": 49088
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8451349974528782,
      "eval_loss": 0.4371325969696045,
      "eval_runtime": 41.4112,
      "eval_samples_per_second": 237.013,
      "eval_steps_per_second": 29.63,
      "step": 49088
    },
    {
      "epoch": 3.0,
      "grad_norm": 17.64081573486328,
      "learning_rate": 0.0,
      "loss": 0.2214,
      "step": 73632
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8468670402445236,
      "eval_loss": 0.5565051436424255,
      "eval_runtime": 41.8267,
      "eval_samples_per_second": 234.659,
      "eval_steps_per_second": 29.335,
      "step": 73632
    }
  ],
  "logging_steps": 500,
  "max_steps": 73632,
  "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": 3.099754961058632e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}