trainer_state.json

{
  "best_metric": 0.40053333333333335,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0112/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": Infinity,
      "learning_rate": 0.0002803729487402998,
      "loss": 3.3916,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.2064,
      "eval_loss": 2.923553228378296,
      "eval_runtime": 5.835,
      "eval_samples_per_second": 642.679,
      "eval_steps_per_second": 10.111,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 35.99702453613281,
      "learning_rate": 0.00022581553795296964,
      "loss": 2.7775,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.2352,
      "eval_loss": 2.8279662132263184,
      "eval_runtime": 5.7662,
      "eval_samples_per_second": 650.338,
      "eval_steps_per_second": 10.232,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 27.54328727722168,
      "learning_rate": 0.00015094341499741014,
      "loss": 2.5117,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.31333333333333335,
      "eval_loss": 2.4920923709869385,
      "eval_runtime": 5.6716,
      "eval_samples_per_second": 661.186,
      "eval_steps_per_second": 10.403,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 34.703460693359375,
      "learning_rate": 7.58185047551672e-05,
      "loss": 2.2385,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.3645333333333333,
      "eval_loss": 2.271521806716919,
      "eval_runtime": 5.8273,
      "eval_samples_per_second": 643.523,
      "eval_steps_per_second": 10.125,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 42.43264389038086,
      "learning_rate": 2.0570466257110352e-05,
      "loss": 1.9292,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.39866666666666667,
      "eval_loss": 2.1666243076324463,
      "eval_runtime": 7.637,
      "eval_samples_per_second": 491.032,
      "eval_steps_per_second": 7.726,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 38.13308334350586,
      "learning_rate": 2.9668021975126634e-09,
      "loss": 1.6925,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.40053333333333335,
      "eval_loss": 2.188957929611206,
      "eval_runtime": 5.7223,
      "eval_samples_per_second": 655.33,
      "eval_steps_per_second": 10.311,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 1998,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "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": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}