trainer_state.json

{
  "best_metric": 0.8925333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0624/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": Infinity,
      "learning_rate": 3e-05,
      "loss": 0.7956,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8696,
      "eval_loss": 0.40108099579811096,
      "eval_runtime": 11.0616,
      "eval_samples_per_second": 339.01,
      "eval_steps_per_second": 5.334,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 32.65022277832031,
      "learning_rate": 3e-05,
      "loss": 0.2051,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8829333333333333,
      "eval_loss": 0.36772409081459045,
      "eval_runtime": 11.0034,
      "eval_samples_per_second": 340.805,
      "eval_steps_per_second": 5.362,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.20057186484336853,
      "learning_rate": 3e-05,
      "loss": 0.0927,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8808,
      "eval_loss": 0.40599656105041504,
      "eval_runtime": 11.0196,
      "eval_samples_per_second": 340.303,
      "eval_steps_per_second": 5.354,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 9.938780784606934,
      "learning_rate": 3e-05,
      "loss": 0.0613,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8768,
      "eval_loss": 0.4714314341545105,
      "eval_runtime": 11.0247,
      "eval_samples_per_second": 340.144,
      "eval_steps_per_second": 5.352,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.010139783844351768,
      "learning_rate": 3e-05,
      "loss": 0.0407,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8925333333333333,
      "eval_loss": 0.43482616543769836,
      "eval_runtime": 11.1691,
      "eval_samples_per_second": 335.748,
      "eval_steps_per_second": 5.282,
      "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
}