trainer_state.json

{
  "best_metric": 0.8533333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0400/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": 100.86820983886719,
      "learning_rate": 0.0001,
      "loss": 0.8413,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8218666666666666,
      "eval_loss": 0.5899671316146851,
      "eval_runtime": 11.5379,
      "eval_samples_per_second": 325.015,
      "eval_steps_per_second": 5.114,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 54.05281066894531,
      "learning_rate": 0.0001,
      "loss": 0.385,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8410666666666666,
      "eval_loss": 0.5360007286071777,
      "eval_runtime": 11.1055,
      "eval_samples_per_second": 337.672,
      "eval_steps_per_second": 5.313,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.19635574519634247,
      "learning_rate": 0.0001,
      "loss": 0.2636,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8384,
      "eval_loss": 0.5894105434417725,
      "eval_runtime": 11.4152,
      "eval_samples_per_second": 328.508,
      "eval_steps_per_second": 5.169,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.07114353775978088,
      "learning_rate": 0.0001,
      "loss": 0.2203,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8512,
      "eval_loss": 0.5279072523117065,
      "eval_runtime": 11.3986,
      "eval_samples_per_second": 328.988,
      "eval_steps_per_second": 5.176,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 40.50562286376953,
      "learning_rate": 0.0001,
      "loss": 0.179,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.836,
      "eval_loss": 0.6632014513015747,
      "eval_runtime": 11.5561,
      "eval_samples_per_second": 324.505,
      "eval_steps_per_second": 5.106,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.7889494895935059,
      "learning_rate": 0.0001,
      "loss": 0.1539,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8533333333333334,
      "eval_loss": 0.5791650414466858,
      "eval_runtime": 11.4108,
      "eval_samples_per_second": 328.637,
      "eval_steps_per_second": 5.171,
      "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
}