trainer_state.json

{
  "best_metric": 0.9397333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0514/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": Infinity,
      "learning_rate": 4.851018993780429e-05,
      "loss": 0.8809,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8861333333333333,
      "eval_loss": 0.37876641750335693,
      "eval_runtime": 4.5109,
      "eval_samples_per_second": 831.32,
      "eval_steps_per_second": 13.079,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.03563673049211502,
      "learning_rate": 4.4184758936602036e-05,
      "loss": 0.2394,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8896,
      "eval_loss": 0.37777596712112427,
      "eval_runtime": 4.5676,
      "eval_samples_per_second": 821.009,
      "eval_steps_per_second": 12.917,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 54.55916976928711,
      "learning_rate": 3.7545362870758595e-05,
      "loss": 0.1152,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9096,
      "eval_loss": 0.34404507279396057,
      "eval_runtime": 5.3039,
      "eval_samples_per_second": 707.025,
      "eval_steps_per_second": 11.124,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.11908464878797531,
      "learning_rate": 2.9392810892864702e-05,
      "loss": 0.0568,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9128,
      "eval_loss": 0.33645790815353394,
      "eval_runtime": 4.6539,
      "eval_samples_per_second": 805.768,
      "eval_steps_per_second": 12.677,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.027070844545960426,
      "learning_rate": 2.0710421090307246e-05,
      "loss": 0.0286,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9168,
      "eval_loss": 0.32356399297714233,
      "eval_runtime": 5.0894,
      "eval_samples_per_second": 736.83,
      "eval_steps_per_second": 11.593,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.0029073324985802174,
      "learning_rate": 1.2545417811701016e-05,
      "loss": 0.0091,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9261333333333334,
      "eval_loss": 0.32211950421333313,
      "eval_runtime": 4.4465,
      "eval_samples_per_second": 843.369,
      "eval_steps_per_second": 13.269,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.0026934300549328327,
      "learning_rate": 5.882620955060414e-06,
      "loss": 0.0031,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9362666666666667,
      "eval_loss": 0.2789290249347687,
      "eval_runtime": 5.0466,
      "eval_samples_per_second": 743.077,
      "eval_steps_per_second": 11.691,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.018637239933013916,
      "learning_rate": 1.5256621537052152e-06,
      "loss": 0.0011,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9397333333333333,
      "eval_loss": 0.26498427987098694,
      "eval_runtime": 4.4907,
      "eval_samples_per_second": 835.053,
      "eval_steps_per_second": 13.138,
      "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
}