trainer_state.json

{
  "best_metric": 0.9184,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0296/checkpoints/checkpoint-2331",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2331,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.4943872690200806,
      "learning_rate": 8.661507935327383e-05,
      "loss": 0.8959,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8306666666666667,
      "eval_loss": 0.5646771192550659,
      "eval_runtime": 5.646,
      "eval_samples_per_second": 664.193,
      "eval_steps_per_second": 10.45,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 11.24008560180664,
      "learning_rate": 7.689476527896773e-05,
      "loss": 0.3054,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8328,
      "eval_loss": 0.559848964214325,
      "eval_runtime": 5.517,
      "eval_samples_per_second": 679.723,
      "eval_steps_per_second": 10.694,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 28.702939987182617,
      "learning_rate": 6.231876231770601e-05,
      "loss": 0.1587,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8664,
      "eval_loss": 0.4633241593837738,
      "eval_runtime": 5.527,
      "eval_samples_per_second": 678.493,
      "eval_steps_per_second": 10.675,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 21.37610626220703,
      "learning_rate": 4.510613478854492e-05,
      "loss": 0.0737,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8741333333333333,
      "eval_loss": 0.5116567611694336,
      "eval_runtime": 5.4891,
      "eval_samples_per_second": 683.171,
      "eval_steps_per_second": 10.749,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.012569282203912735,
      "learning_rate": 2.7877349199942123e-05,
      "loss": 0.031,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9002666666666667,
      "eval_loss": 0.4227410852909088,
      "eval_runtime": 5.3528,
      "eval_samples_per_second": 700.565,
      "eval_steps_per_second": 11.022,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 18.608739852905273,
      "learning_rate": 1.3255331978432106e-05,
      "loss": 0.0056,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9152,
      "eval_loss": 0.36523276567459106,
      "eval_runtime": 5.9734,
      "eval_samples_per_second": 627.783,
      "eval_steps_per_second": 9.877,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.009851661510765553,
      "learning_rate": 3.4661526970736e-06,
      "loss": 0.0016,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9184,
      "eval_loss": 0.35729673504829407,
      "eval_runtime": 5.604,
      "eval_samples_per_second": 669.167,
      "eval_steps_per_second": 10.528,
      "step": 2331
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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": 1.153189244358144e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}