trainer_state.json

{
  "best_metric": 0.9221333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0079/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 10.600435256958008,
      "learning_rate": 4.7538820348083834e-05,
      "loss": 0.737,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8549333333333333,
      "eval_loss": 0.46782657504081726,
      "eval_runtime": 5.3165,
      "eval_samples_per_second": 705.348,
      "eval_steps_per_second": 11.097,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.2858046591281891,
      "learning_rate": 4.0613573610928476e-05,
      "loss": 0.1888,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8925333333333333,
      "eval_loss": 0.36361032724380493,
      "eval_runtime": 5.6256,
      "eval_samples_per_second": 666.601,
      "eval_steps_per_second": 10.488,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.014802091754972935,
      "learning_rate": 3.059586713221479e-05,
      "loss": 0.0877,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9045333333333333,
      "eval_loss": 0.3575219511985779,
      "eval_runtime": 5.6759,
      "eval_samples_per_second": 660.683,
      "eval_steps_per_second": 10.395,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.03501858934760094,
      "learning_rate": 1.946983053915936e-05,
      "loss": 0.033,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9074666666666666,
      "eval_loss": 0.37511199712753296,
      "eval_runtime": 5.577,
      "eval_samples_per_second": 672.4,
      "eval_steps_per_second": 10.579,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.12144734710454941,
      "learning_rate": 9.43911183090096e-06,
      "loss": 0.0082,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9165333333333333,
      "eval_loss": 0.3624231219291687,
      "eval_runtime": 5.7428,
      "eval_samples_per_second": 652.99,
      "eval_steps_per_second": 10.274,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.19311842322349548,
      "learning_rate": 2.4904178663020224e-06,
      "loss": 0.002,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9221333333333334,
      "eval_loss": 0.35298097133636475,
      "eval_runtime": 5.3246,
      "eval_samples_per_second": 704.279,
      "eval_steps_per_second": 11.081,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}