trainer_state.json

{
  "best_metric": 0.3648,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0989/checkpoints/checkpoint-2331",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2331,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": Infinity,
      "learning_rate": 0.00048141860138016556,
      "loss": 3.3976,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.18106666666666665,
      "eval_loss": 3.080702066421509,
      "eval_runtime": 12.4016,
      "eval_samples_per_second": 302.38,
      "eval_steps_per_second": 4.757,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 21.613311767578125,
      "learning_rate": 0.00042760859989611786,
      "loss": 2.9759,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.21733333333333332,
      "eval_loss": 2.9494407176971436,
      "eval_runtime": 12.8944,
      "eval_samples_per_second": 290.825,
      "eval_steps_per_second": 4.576,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 20.019386291503906,
      "learning_rate": 0.00034675929910385353,
      "loss": 2.8051,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.2541333333333333,
      "eval_loss": 2.790637493133545,
      "eval_runtime": 12.3258,
      "eval_samples_per_second": 304.24,
      "eval_steps_per_second": 4.787,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 25.942733764648438,
      "learning_rate": 0.00025117927214825806,
      "loss": 2.6208,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.28986666666666666,
      "eval_loss": 2.6452462673187256,
      "eval_runtime": 12.4128,
      "eval_samples_per_second": 302.107,
      "eval_steps_per_second": 4.753,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 25.267745971679688,
      "learning_rate": 0.00015541971169821895,
      "loss": 2.4365,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.3176,
      "eval_loss": 2.514354705810547,
      "eval_runtime": 12.3321,
      "eval_samples_per_second": 304.084,
      "eval_steps_per_second": 4.784,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 26.347402572631836,
      "learning_rate": 7.405914276967749e-05,
      "loss": 2.226,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.35386666666666666,
      "eval_loss": 2.373981237411499,
      "eval_runtime": 12.3268,
      "eval_samples_per_second": 304.216,
      "eval_steps_per_second": 4.786,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 64.33563232421875,
      "learning_rate": 1.9483974446610292e-05,
      "loss": 2.0152,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.3648,
      "eval_loss": 2.3430285453796387,
      "eval_runtime": 12.2007,
      "eval_samples_per_second": 307.361,
      "eval_steps_per_second": 4.836,
      "step": 2331
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.153189244358144e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}