trainer_state.json

{
  "best_metric": 0.8362666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0405/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": 30.696969985961914,
      "learning_rate": 0.0001,
      "loss": 1.0343,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8021333333333334,
      "eval_loss": 0.6737033724784851,
      "eval_runtime": 14.4647,
      "eval_samples_per_second": 259.252,
      "eval_steps_per_second": 4.079,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.275521159172058,
      "learning_rate": 0.0001,
      "loss": 0.3906,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8184,
      "eval_loss": 0.6363128423690796,
      "eval_runtime": 13.1869,
      "eval_samples_per_second": 284.373,
      "eval_steps_per_second": 4.474,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.2203308641910553,
      "learning_rate": 0.0001,
      "loss": 0.257,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8184,
      "eval_loss": 0.6325089931488037,
      "eval_runtime": 13.1719,
      "eval_samples_per_second": 284.698,
      "eval_steps_per_second": 4.479,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0001,
      "loss": 0.1919,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7957333333333333,
      "eval_loss": 0.8047911524772644,
      "eval_runtime": 13.1429,
      "eval_samples_per_second": 285.325,
      "eval_steps_per_second": 4.489,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 3.9937498569488525,
      "learning_rate": 0.0001,
      "loss": 0.1739,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8362666666666667,
      "eval_loss": 0.6351863145828247,
      "eval_runtime": 12.9684,
      "eval_samples_per_second": 289.165,
      "eval_steps_per_second": 4.55,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "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
}