trainer_state.json

{
  "best_metric": 0.9104,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0869/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": 1.9547767639160156,
      "learning_rate": 9.619397662556435e-05,
      "loss": 1.4392,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8128,
      "eval_loss": 0.6582413911819458,
      "eval_runtime": 6.0257,
      "eval_samples_per_second": 622.335,
      "eval_steps_per_second": 9.791,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 8.539700818671957e-05,
      "loss": 0.4052,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8608,
      "eval_loss": 0.456818163394928,
      "eval_runtime": 6.0954,
      "eval_samples_per_second": 615.22,
      "eval_steps_per_second": 9.679,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 6.924306924189557e-05,
      "loss": 0.1997,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8794666666666666,
      "eval_loss": 0.41608926653862,
      "eval_runtime": 6.2779,
      "eval_samples_per_second": 597.337,
      "eval_steps_per_second": 9.398,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.266092300415039,
      "learning_rate": 5.011792754282769e-05,
      "loss": 0.0916,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8893333333333333,
      "eval_loss": 0.4060729146003723,
      "eval_runtime": 6.5982,
      "eval_samples_per_second": 568.339,
      "eval_steps_per_second": 8.942,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.07943375408649445,
      "learning_rate": 3.0974832444380136e-05,
      "loss": 0.0385,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9026666666666666,
      "eval_loss": 0.3671906888484955,
      "eval_runtime": 6.3999,
      "eval_samples_per_second": 585.951,
      "eval_steps_per_second": 9.219,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.05929328873753548,
      "learning_rate": 1.472814664270234e-05,
      "loss": 0.0132,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9096,
      "eval_loss": 0.3478761613368988,
      "eval_runtime": 6.51,
      "eval_samples_per_second": 576.04,
      "eval_steps_per_second": 9.063,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.024553852155804634,
      "learning_rate": 3.851280774526222e-06,
      "loss": 0.0038,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9104,
      "eval_loss": 0.34887903928756714,
      "eval_runtime": 6.0433,
      "eval_samples_per_second": 620.525,
      "eval_steps_per_second": 9.763,
      "step": 2331
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.153189244358144e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}