trainer_state.json

{
  "best_metric": 0.9226666666666666,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0875/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": 3.5467934608459473,
      "learning_rate": 4.8108254489432715e-05,
      "loss": 0.7793,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8645333333333334,
      "eval_loss": 0.45642805099487305,
      "eval_runtime": 10.8965,
      "eval_samples_per_second": 344.146,
      "eval_steps_per_second": 5.415,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 4.271931404387096e-05,
      "loss": 0.1988,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.876,
      "eval_loss": 0.42209911346435547,
      "eval_runtime": 10.9536,
      "eval_samples_per_second": 342.353,
      "eval_steps_per_second": 5.386,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 3.464873897488322e-05,
      "loss": 0.0808,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8893333333333333,
      "eval_loss": 0.41216033697128296,
      "eval_runtime": 10.8714,
      "eval_samples_per_second": 344.941,
      "eval_steps_per_second": 5.427,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.525515615940094,
      "learning_rate": 2.5088445554620133e-05,
      "loss": 0.0315,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8981333333333333,
      "eval_loss": 0.40897423028945923,
      "eval_runtime": 10.8594,
      "eval_samples_per_second": 345.323,
      "eval_steps_per_second": 5.433,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.001289117499254644,
      "learning_rate": 1.5514687100427083e-05,
      "loss": 0.0099,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9056,
      "eval_loss": 0.3997249901294708,
      "eval_runtime": 10.8059,
      "eval_samples_per_second": 347.031,
      "eval_steps_per_second": 5.46,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.2096722573041916,
      "learning_rate": 7.384981550618447e-06,
      "loss": 0.0031,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9226666666666666,
      "eval_loss": 0.3306345045566559,
      "eval_runtime": 10.8076,
      "eval_samples_per_second": 346.979,
      "eval_steps_per_second": 5.459,
      "step": 1998
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}