trainer_state.json

{
  "best_metric": 0.8608,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0072/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": 5.79313850402832,
      "learning_rate": 5e-05,
      "loss": 0.8769,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8026666666666666,
      "eval_loss": 0.626925528049469,
      "eval_runtime": 4.5386,
      "eval_samples_per_second": 826.253,
      "eval_steps_per_second": 13.0,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 53.82843780517578,
      "learning_rate": 5e-05,
      "loss": 0.2997,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8525333333333334,
      "eval_loss": 0.4868287742137909,
      "eval_runtime": 4.3879,
      "eval_samples_per_second": 854.627,
      "eval_steps_per_second": 13.446,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.015891961753368378,
      "learning_rate": 5e-05,
      "loss": 0.1635,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8453333333333334,
      "eval_loss": 0.5446628928184509,
      "eval_runtime": 4.4567,
      "eval_samples_per_second": 841.438,
      "eval_steps_per_second": 13.239,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 12.590849876403809,
      "learning_rate": 5e-05,
      "loss": 0.1114,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.844,
      "eval_loss": 0.5827698111534119,
      "eval_runtime": 4.5912,
      "eval_samples_per_second": 816.781,
      "eval_steps_per_second": 12.851,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.3846839964389801,
      "learning_rate": 5e-05,
      "loss": 0.0889,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8565333333333334,
      "eval_loss": 0.534341037273407,
      "eval_runtime": 4.6018,
      "eval_samples_per_second": 814.901,
      "eval_steps_per_second": 12.821,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.03652867674827576,
      "learning_rate": 5e-05,
      "loss": 0.0755,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8608,
      "eval_loss": 0.5932831764221191,
      "eval_runtime": 4.5432,
      "eval_samples_per_second": 825.415,
      "eval_steps_per_second": 12.987,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "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
}