trainer_state.json

{
  "best_metric": 0.9408,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0680/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": 4.152751922607422,
      "learning_rate": 6.339302707314361e-05,
      "loss": 0.7252,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8757333333333334,
      "eval_loss": 0.4098936915397644,
      "eval_runtime": 12.4322,
      "eval_samples_per_second": 301.637,
      "eval_steps_per_second": 4.746,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 4.600384237127692e-05,
      "loss": 0.2026,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.904,
      "eval_loss": 0.3455832004547119,
      "eval_runtime": 12.3917,
      "eval_samples_per_second": 302.622,
      "eval_steps_per_second": 4.761,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.4738376140594482,
      "learning_rate": 2.4372999312736286e-05,
      "loss": 0.0629,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9133333333333333,
      "eval_loss": 0.3105599284172058,
      "eval_runtime": 12.3109,
      "eval_samples_per_second": 304.609,
      "eval_steps_per_second": 4.793,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.3083895444869995,
      "learning_rate": 6.801309318261912e-06,
      "loss": 0.0145,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9370666666666667,
      "eval_loss": 0.25292497873306274,
      "eval_runtime": 12.2312,
      "eval_samples_per_second": 306.592,
      "eval_steps_per_second": 4.824,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.01235372293740511,
      "learning_rate": 5.607259665268982e-10,
      "loss": 0.0019,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9408,
      "eval_loss": 0.2453916370868683,
      "eval_runtime": 12.4181,
      "eval_samples_per_second": 301.979,
      "eval_steps_per_second": 4.751,
      "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
}