trainer_state.json

{
  "best_metric": 0.33866666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0208/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": 15.846410751342773,
      "learning_rate": 0.00045280733623674006,
      "loss": 3.433,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.18106666666666665,
      "eval_loss": 3.0566837787628174,
      "eval_runtime": 7.0132,
      "eval_samples_per_second": 534.703,
      "eval_steps_per_second": 8.413,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 21.57581329345703,
      "learning_rate": 0.0003281509432839476,
      "loss": 2.9308,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.24186666666666667,
      "eval_loss": 2.8656160831451416,
      "eval_runtime": 6.9361,
      "eval_samples_per_second": 540.649,
      "eval_steps_per_second": 8.506,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 22.197689056396484,
      "learning_rate": 0.0001736435462495525,
      "loss": 2.6845,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.284,
      "eval_loss": 2.6104183197021484,
      "eval_runtime": 7.151,
      "eval_samples_per_second": 524.402,
      "eval_steps_per_second": 8.251,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 22.49991226196289,
      "learning_rate": 4.8301719287419e-05,
      "loss": 2.4281,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.3197333333333333,
      "eval_loss": 2.4540679454803467,
      "eval_runtime": 6.8708,
      "eval_samples_per_second": 545.788,
      "eval_steps_per_second": 8.587,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 33.99885940551758,
      "learning_rate": 1.7800850740101736e-09,
      "loss": 2.2011,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.33866666666666667,
      "eval_loss": 2.415955066680908,
      "eval_runtime": 6.838,
      "eval_samples_per_second": 548.406,
      "eval_steps_per_second": 8.628,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}