trainer_state.json

{
  "best_metric": 0.3552,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0934/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 16.243444442749023,
      "learning_rate": 0.0005,
      "loss": 3.4912,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.1808,
      "eval_loss": 3.1136162281036377,
      "eval_runtime": 11.9935,
      "eval_samples_per_second": 312.67,
      "eval_steps_per_second": 4.919,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 17.066505432128906,
      "learning_rate": 0.0005,
      "loss": 3.0007,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.22773333333333334,
      "eval_loss": 2.9078755378723145,
      "eval_runtime": 11.9597,
      "eval_samples_per_second": 313.554,
      "eval_steps_per_second": 4.933,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 35.759647369384766,
      "learning_rate": 0.0005,
      "loss": 2.847,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.22453333333333333,
      "eval_loss": 2.890984058380127,
      "eval_runtime": 11.9218,
      "eval_samples_per_second": 314.55,
      "eval_steps_per_second": 4.949,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 16.526554107666016,
      "learning_rate": 0.0005,
      "loss": 2.7153,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.2776,
      "eval_loss": 2.719088077545166,
      "eval_runtime": 11.8005,
      "eval_samples_per_second": 317.784,
      "eval_steps_per_second": 5.0,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 17.74460220336914,
      "learning_rate": 0.0005,
      "loss": 2.6125,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.296,
      "eval_loss": 2.6124439239501953,
      "eval_runtime": 11.9148,
      "eval_samples_per_second": 314.735,
      "eval_steps_per_second": 4.952,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 14.572244644165039,
      "learning_rate": 0.0005,
      "loss": 2.4868,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.29786666666666667,
      "eval_loss": 2.5934929847717285,
      "eval_runtime": 11.8419,
      "eval_samples_per_second": 316.672,
      "eval_steps_per_second": 4.982,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 18.132177352905273,
      "learning_rate": 0.0005,
      "loss": 2.4207,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.32,
      "eval_loss": 2.5568041801452637,
      "eval_runtime": 11.9201,
      "eval_samples_per_second": 314.594,
      "eval_steps_per_second": 4.95,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 16.182008743286133,
      "learning_rate": 0.0005,
      "loss": 2.3032,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.3552,
      "eval_loss": 2.4295713901519775,
      "eval_runtime": 11.9537,
      "eval_samples_per_second": 313.712,
      "eval_steps_per_second": 4.936,
      "step": 2664
    }
  ],
  "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": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}