trainer_state.json

{
  "best_metric": 0.9405333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0447/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": 1.4303011894226074,
      "learning_rate": 4.525311529382417e-05,
      "loss": 0.6784,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8784,
      "eval_loss": 0.3683476448059082,
      "eval_runtime": 11.845,
      "eval_samples_per_second": 316.59,
      "eval_steps_per_second": 4.981,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 11.540746688842773,
      "learning_rate": 3.277027342564428e-05,
      "loss": 0.1401,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.916,
      "eval_loss": 0.2694927155971527,
      "eval_runtime": 12.0064,
      "eval_samples_per_second": 312.334,
      "eval_steps_per_second": 4.914,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.016613801941275597,
      "learning_rate": 1.731945121074836e-05,
      "loss": 0.043,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9290666666666667,
      "eval_loss": 0.26593640446662903,
      "eval_runtime": 11.774,
      "eval_samples_per_second": 318.497,
      "eval_steps_per_second": 5.011,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.033974286168813705,
      "learning_rate": 4.802337581094707e-06,
      "loss": 0.0061,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9381333333333334,
      "eval_loss": 0.2225041538476944,
      "eval_runtime": 11.8257,
      "eval_samples_per_second": 317.105,
      "eval_steps_per_second": 4.989,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.0015970737440511584,
      "learning_rate": 4.4502166460236305e-11,
      "loss": 0.001,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9405333333333333,
      "eval_loss": 0.2149571180343628,
      "eval_runtime": 12.108,
      "eval_samples_per_second": 309.712,
      "eval_steps_per_second": 4.873,
      "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
}