trainer_state.json

{
  "best_metric": 0.7707005739212036,
  "best_model_checkpoint": "./beans_outputs/checkpoint-33",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 33,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.30303030303030304,
      "grad_norm": 1.727879285812378,
      "learning_rate": 1.3939393939393942e-05,
      "loss": 1.0069,
      "step": 10
    },
    {
      "epoch": 0.6060606060606061,
      "grad_norm": 1.4632545709609985,
      "learning_rate": 7.87878787878788e-06,
      "loss": 0.8791,
      "step": 20
    },
    {
      "epoch": 0.9090909090909091,
      "grad_norm": 1.3964664936065674,
      "learning_rate": 1.8181818181818183e-06,
      "loss": 0.8225,
      "step": 30
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9172932330827067,
      "eval_loss": 0.7707005739212036,
      "eval_runtime": 1.4349,
      "eval_samples_per_second": 92.687,
      "eval_steps_per_second": 3.484,
      "step": 33
    },
    {
      "epoch": 1.0,
      "step": 33,
      "total_flos": 8.012743541191066e+16,
      "train_loss": 0.0,
      "train_runtime": 0.0487,
      "train_samples_per_second": 21219.941,
      "train_steps_per_second": 677.232
    }
  ],
  "logging_steps": 10,
  "max_steps": 33,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "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": 8.012743541191066e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}