trainer_state.json

{
  "best_metric": 0.396,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0058/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": 24.49916648864746,
      "learning_rate": 0.0004382507507507508,
      "loss": 3.4446,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.17066666666666666,
      "eval_loss": 3.1372082233428955,
      "eval_runtime": 4.0278,
      "eval_samples_per_second": 931.025,
      "eval_steps_per_second": 14.648,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 24.89746856689453,
      "learning_rate": 0.0003757507507507508,
      "loss": 2.9521,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.2184,
      "eval_loss": 2.972460985183716,
      "eval_runtime": 4.0201,
      "eval_samples_per_second": 932.811,
      "eval_steps_per_second": 14.676,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 23.160442352294922,
      "learning_rate": 0.0003132507507507508,
      "loss": 2.763,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.256,
      "eval_loss": 2.776211977005005,
      "eval_runtime": 4.438,
      "eval_samples_per_second": 844.974,
      "eval_steps_per_second": 13.294,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 20.667842864990234,
      "learning_rate": 0.0002507507507507508,
      "loss": 2.5962,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.2872,
      "eval_loss": 2.630237579345703,
      "eval_runtime": 5.409,
      "eval_samples_per_second": 693.287,
      "eval_steps_per_second": 10.908,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 24.244220733642578,
      "learning_rate": 0.00018825075075075076,
      "loss": 2.4208,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.3296,
      "eval_loss": 2.513320207595825,
      "eval_runtime": 4.2385,
      "eval_samples_per_second": 884.745,
      "eval_steps_per_second": 13.92,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 23.040550231933594,
      "learning_rate": 0.00012575075075075076,
      "loss": 2.245,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.3506666666666667,
      "eval_loss": 2.400188684463501,
      "eval_runtime": 4.7112,
      "eval_samples_per_second": 795.973,
      "eval_steps_per_second": 12.523,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": Infinity,
      "learning_rate": 6.343843843843844e-05,
      "loss": 2.0324,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.3824,
      "eval_loss": 2.303717613220215,
      "eval_runtime": 4.0208,
      "eval_samples_per_second": 932.651,
      "eval_steps_per_second": 14.674,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 44.030582427978516,
      "learning_rate": 9.384384384384384e-07,
      "loss": 1.7857,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.396,
      "eval_loss": 2.2951197624206543,
      "eval_runtime": 4.2631,
      "eval_samples_per_second": 879.64,
      "eval_steps_per_second": 13.84,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}