trainer_state.json

{
  "best_metric": 0.5458666666666666,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0778/checkpoints/checkpoint-2331",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2331,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": Infinity,
      "learning_rate": 0.00047538820348083834,
      "loss": 3.1144,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.22906666666666667,
      "eval_loss": 2.784980297088623,
      "eval_runtime": 10.9041,
      "eval_samples_per_second": 343.909,
      "eval_steps_per_second": 5.411,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 14.824202537536621,
      "learning_rate": 0.00040613573610928476,
      "loss": 2.4641,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.3237333333333333,
      "eval_loss": 2.4405369758605957,
      "eval_runtime": 10.4782,
      "eval_samples_per_second": 357.885,
      "eval_steps_per_second": 5.631,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 0.00030628700751103485,
      "loss": 2.0034,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.44426666666666664,
      "eval_loss": 1.9979820251464844,
      "eval_runtime": 10.4354,
      "eval_samples_per_second": 359.353,
      "eval_steps_per_second": 5.654,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": Infinity,
      "learning_rate": 0.000195355683908512,
      "loss": 1.5518,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.48186666666666667,
      "eval_loss": 1.8113545179367065,
      "eval_runtime": 10.3888,
      "eval_samples_per_second": 360.965,
      "eval_steps_per_second": 5.679,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 5.439235687255859,
      "learning_rate": 9.491910332815573e-05,
      "loss": 1.0255,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.524,
      "eval_loss": 1.7197835445404053,
      "eval_runtime": 10.5618,
      "eval_samples_per_second": 355.054,
      "eval_steps_per_second": 5.586,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 36.66120147705078,
      "learning_rate": 2.519819627604078e-05,
      "loss": 0.4885,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5437333333333333,
      "eval_loss": 1.7216057777404785,
      "eval_runtime": 10.2777,
      "eval_samples_per_second": 364.867,
      "eval_steps_per_second": 5.741,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 9.112259864807129,
      "learning_rate": 2.0434646496314548e-09,
      "loss": 0.2064,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5458666666666666,
      "eval_loss": 1.7338130474090576,
      "eval_runtime": 10.496,
      "eval_samples_per_second": 357.281,
      "eval_steps_per_second": 5.621,
      "step": 2331
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 1.153189244358144e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}