trainer_state.json

{
  "best_metric": 0.44906666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0948/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": 0.0002887841425059038,
      "loss": 3.3116,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.2256,
      "eval_loss": 2.8964719772338867,
      "eval_runtime": 10.8699,
      "eval_samples_per_second": 344.989,
      "eval_steps_per_second": 5.428,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 21.252519607543945,
      "learning_rate": 0.0002564405961135417,
      "loss": 2.7366,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.2648,
      "eval_loss": 2.6813747882843018,
      "eval_runtime": 10.3794,
      "eval_samples_per_second": 361.293,
      "eval_steps_per_second": 5.684,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 27.061079025268555,
      "learning_rate": 0.0002078924338492993,
      "loss": 2.4813,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.31093333333333334,
      "eval_loss": 2.462181568145752,
      "eval_runtime": 10.5508,
      "eval_samples_per_second": 355.425,
      "eval_steps_per_second": 5.592,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 34.01191329956055,
      "learning_rate": 0.00015053067332772077,
      "loss": 2.2245,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.3565333333333333,
      "eval_loss": 2.3075616359710693,
      "eval_runtime": 11.1811,
      "eval_samples_per_second": 335.389,
      "eval_steps_per_second": 5.277,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 24.796300888061523,
      "learning_rate": 9.308812260256249e-05,
      "loss": 1.9462,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.404,
      "eval_loss": 2.1445724964141846,
      "eval_runtime": 10.9472,
      "eval_samples_per_second": 342.552,
      "eval_steps_per_second": 5.389,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 90.1661148071289,
      "learning_rate": 4.4309889303710674e-05,
      "loss": 1.6303,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.4336,
      "eval_loss": 2.067046880722046,
      "eval_runtime": 10.8634,
      "eval_samples_per_second": 345.194,
      "eval_steps_per_second": 5.431,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 27.738433837890625,
      "learning_rate": 1.1622017275129708e-05,
      "loss": 1.2674,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.44426666666666664,
      "eval_loss": 2.1670680046081543,
      "eval_runtime": 10.7027,
      "eval_samples_per_second": 350.38,
      "eval_steps_per_second": 5.513,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 17.389305114746094,
      "learning_rate": 9.387168731500582e-10,
      "loss": 1.0094,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.44906666666666667,
      "eval_loss": 2.323967695236206,
      "eval_runtime": 10.3217,
      "eval_samples_per_second": 363.313,
      "eval_steps_per_second": 5.716,
      "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
}