trainer_state.json

{
  "best_metric": 0.9293333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0306/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": 55.31058120727539,
      "learning_rate": 4.376876876876877e-05,
      "loss": 0.7849,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8570666666666666,
      "eval_loss": 0.46635210514068604,
      "eval_runtime": 10.8565,
      "eval_samples_per_second": 345.415,
      "eval_steps_per_second": 5.435,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.5193877220153809,
      "learning_rate": 3.7518768768768766e-05,
      "loss": 0.2137,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8821333333333333,
      "eval_loss": 0.38604897260665894,
      "eval_runtime": 6.829,
      "eval_samples_per_second": 549.13,
      "eval_steps_per_second": 8.64,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 47.886566162109375,
      "learning_rate": 3.126876876876877e-05,
      "loss": 0.0968,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8904,
      "eval_loss": 0.3822152316570282,
      "eval_runtime": 6.9187,
      "eval_samples_per_second": 542.006,
      "eval_steps_per_second": 8.528,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 5.015561580657959,
      "learning_rate": 2.5018768768768767e-05,
      "loss": 0.0521,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9056,
      "eval_loss": 0.35865047574043274,
      "eval_runtime": 7.0713,
      "eval_samples_per_second": 530.316,
      "eval_steps_per_second": 8.344,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.05517354980111122,
      "learning_rate": 1.8768768768768768e-05,
      "loss": 0.0262,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9093333333333333,
      "eval_loss": 0.3677443563938141,
      "eval_runtime": 7.6393,
      "eval_samples_per_second": 490.884,
      "eval_steps_per_second": 7.723,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.022528665140271187,
      "learning_rate": 1.2518768768768768e-05,
      "loss": 0.0094,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9242666666666667,
      "eval_loss": 0.3443892300128937,
      "eval_runtime": 8.1042,
      "eval_samples_per_second": 462.725,
      "eval_steps_per_second": 7.28,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.00013200398825574666,
      "learning_rate": 6.268768768768769e-06,
      "loss": 0.0041,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.924,
      "eval_loss": 0.3203379511833191,
      "eval_runtime": 7.4133,
      "eval_samples_per_second": 505.847,
      "eval_steps_per_second": 7.959,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.0018143390771001577,
      "learning_rate": 1.876876876876877e-08,
      "loss": 0.0019,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9293333333333333,
      "eval_loss": 0.31443339586257935,
      "eval_runtime": 7.5585,
      "eval_samples_per_second": 496.128,
      "eval_steps_per_second": 7.806,
      "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
}