trainer_state.json

{
  "best_metric": 0.4112,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0358/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": 18.965890884399414,
      "learning_rate": 0.0004813069041765065,
      "loss": 3.4237,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.196,
      "eval_loss": 3.039083480834961,
      "eval_runtime": 6.9778,
      "eval_samples_per_second": 537.422,
      "eval_steps_per_second": 8.455,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 15.841129302978516,
      "learning_rate": 0.00042740099352256953,
      "loss": 2.9407,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.2272,
      "eval_loss": 2.9367458820343018,
      "eval_runtime": 7.044,
      "eval_samples_per_second": 532.366,
      "eval_steps_per_second": 8.376,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 16.672346115112305,
      "learning_rate": 0.0003464873897488322,
      "loss": 2.7652,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.2789333333333333,
      "eval_loss": 2.718060255050659,
      "eval_runtime": 6.8353,
      "eval_samples_per_second": 548.625,
      "eval_steps_per_second": 8.632,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 8.551450729370117,
      "learning_rate": 0.00025088445554620134,
      "loss": 2.5457,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.3234666666666667,
      "eval_loss": 2.5126988887786865,
      "eval_runtime": 6.2989,
      "eval_samples_per_second": 595.338,
      "eval_steps_per_second": 9.367,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 26.88097381591797,
      "learning_rate": 0.00015514687100427082,
      "loss": 2.3769,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.3621333333333333,
      "eval_loss": 2.385690689086914,
      "eval_runtime": 6.5417,
      "eval_samples_per_second": 573.247,
      "eval_steps_per_second": 9.019,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 14.875200271606445,
      "learning_rate": 7.384981550618446e-05,
      "loss": 2.1682,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.3738666666666667,
      "eval_loss": 2.309988498687744,
      "eval_runtime": 7.1238,
      "eval_samples_per_second": 526.406,
      "eval_steps_per_second": 8.282,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 13.545239448547363,
      "learning_rate": 1.9370028791882848e-05,
      "loss": 1.9863,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.3973333333333333,
      "eval_loss": 2.2328782081604004,
      "eval_runtime": 6.4098,
      "eval_samples_per_second": 585.045,
      "eval_steps_per_second": 9.205,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 24.62737274169922,
      "learning_rate": 1.5645281219167639e-09,
      "loss": 1.8726,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.4112,
      "eval_loss": 2.2417855262756348,
      "eval_runtime": 6.1075,
      "eval_samples_per_second": 614.001,
      "eval_steps_per_second": 9.66,
      "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
}