trainer_state.json

{
  "best_metric": 0.8368,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0705/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": 0.0001,
      "loss": 0.9725,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.768,
      "eval_loss": 0.7970865964889526,
      "eval_runtime": 13.5294,
      "eval_samples_per_second": 277.174,
      "eval_steps_per_second": 4.361,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 7.49238395690918,
      "learning_rate": 0.0001,
      "loss": 0.4274,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8138666666666666,
      "eval_loss": 0.6162200570106506,
      "eval_runtime": 13.3298,
      "eval_samples_per_second": 281.325,
      "eval_steps_per_second": 4.426,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.3580947518348694,
      "learning_rate": 0.0001,
      "loss": 0.2955,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.816,
      "eval_loss": 0.6758558750152588,
      "eval_runtime": 13.5614,
      "eval_samples_per_second": 276.52,
      "eval_steps_per_second": 4.351,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.041709378361701965,
      "learning_rate": 0.0001,
      "loss": 0.2262,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8082666666666667,
      "eval_loss": 0.7263500094413757,
      "eval_runtime": 12.6853,
      "eval_samples_per_second": 295.618,
      "eval_steps_per_second": 4.651,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.25616350769996643,
      "learning_rate": 0.0001,
      "loss": 0.1912,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8368,
      "eval_loss": 0.6357800960540771,
      "eval_runtime": 12.1625,
      "eval_samples_per_second": 308.325,
      "eval_steps_per_second": 4.851,
      "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
}