trainer_state.json

{
  "best_metric": 0.7984,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0484/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "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": 0.0003,
      "loss": 1.9281,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7045333333333333,
      "eval_loss": 0.9794152975082397,
      "eval_runtime": 6.1593,
      "eval_samples_per_second": 608.831,
      "eval_steps_per_second": 9.579,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.1601537466049194,
      "learning_rate": 0.0003,
      "loss": 0.7356,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7352,
      "eval_loss": 0.8585774302482605,
      "eval_runtime": 6.1641,
      "eval_samples_per_second": 608.358,
      "eval_steps_per_second": 9.571,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 8.93183422088623,
      "learning_rate": 0.0003,
      "loss": 0.4389,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7794666666666666,
      "eval_loss": 0.7351447939872742,
      "eval_runtime": 6.6738,
      "eval_samples_per_second": 561.897,
      "eval_steps_per_second": 8.841,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.163981556892395,
      "learning_rate": 0.0003,
      "loss": 0.2949,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7677333333333334,
      "eval_loss": 0.7829970121383667,
      "eval_runtime": 6.2951,
      "eval_samples_per_second": 595.697,
      "eval_steps_per_second": 9.372,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.17034517228603363,
      "learning_rate": 0.0003,
      "loss": 0.2144,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7984,
      "eval_loss": 0.7092493772506714,
      "eval_runtime": 6.9909,
      "eval_samples_per_second": 536.409,
      "eval_steps_per_second": 8.44,
      "step": 1665
    }
  ],
  "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}