trainer_state.json

{
  "best_metric": 0.8538666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0896/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": 0.5168570876121521,
      "learning_rate": 0.0001,
      "loss": 0.8591,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8056,
      "eval_loss": 0.6211175918579102,
      "eval_runtime": 11.3857,
      "eval_samples_per_second": 329.36,
      "eval_steps_per_second": 5.182,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0001,
      "loss": 0.3889,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8413333333333334,
      "eval_loss": 0.4999436140060425,
      "eval_runtime": 11.126,
      "eval_samples_per_second": 337.049,
      "eval_steps_per_second": 5.303,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.1191142126917839,
      "learning_rate": 0.0001,
      "loss": 0.2701,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8314666666666667,
      "eval_loss": 0.567319393157959,
      "eval_runtime": 11.0929,
      "eval_samples_per_second": 338.053,
      "eval_steps_per_second": 5.319,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.7281020879745483,
      "learning_rate": 0.0001,
      "loss": 0.2156,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8458666666666667,
      "eval_loss": 0.5273206233978271,
      "eval_runtime": 11.0612,
      "eval_samples_per_second": 339.021,
      "eval_steps_per_second": 5.334,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.6836421489715576,
      "learning_rate": 0.0001,
      "loss": 0.1777,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8538666666666667,
      "eval_loss": 0.5337416529655457,
      "eval_runtime": 10.9149,
      "eval_samples_per_second": 343.568,
      "eval_steps_per_second": 5.405,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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
}