trainer_state.json

{
  "best_metric": 0.8973333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0835/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.368050217628479,
      "learning_rate": 7.205405405405405e-05,
      "loss": 0.8717,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8205333333333333,
      "eval_loss": 0.5899376273155212,
      "eval_runtime": 9.9902,
      "eval_samples_per_second": 375.367,
      "eval_steps_per_second": 5.906,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.5384319424629211,
      "learning_rate": 5.405405405405406e-05,
      "loss": 0.2491,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8552,
      "eval_loss": 0.4828743040561676,
      "eval_runtime": 10.4645,
      "eval_samples_per_second": 358.356,
      "eval_steps_per_second": 5.638,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.7381245493888855,
      "learning_rate": 3.6054054054054056e-05,
      "loss": 0.0928,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8824,
      "eval_loss": 0.4330466091632843,
      "eval_runtime": 10.2885,
      "eval_samples_per_second": 364.485,
      "eval_steps_per_second": 5.735,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.0021586273796856403,
      "learning_rate": 1.8054054054054053e-05,
      "loss": 0.0265,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8882666666666666,
      "eval_loss": 0.46098387241363525,
      "eval_runtime": 10.3037,
      "eval_samples_per_second": 363.946,
      "eval_steps_per_second": 5.726,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.16705727577209473,
      "learning_rate": 5.4054054054054056e-08,
      "loss": 0.0039,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8973333333333333,
      "eval_loss": 0.40932440757751465,
      "eval_runtime": 10.1192,
      "eval_samples_per_second": 370.581,
      "eval_steps_per_second": 5.83,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}