trainer_state.json

{
  "best_metric": 0.8866666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0825/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": 2.3825294971466064,
      "learning_rate": 4.522542485937369e-05,
      "loss": 1.9522,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7965333333333333,
      "eval_loss": 0.8161010146141052,
      "eval_runtime": 6.3259,
      "eval_samples_per_second": 592.797,
      "eval_steps_per_second": 9.327,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 3.277027342564428e-05,
      "loss": 0.6039,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8456,
      "eval_loss": 0.5466204285621643,
      "eval_runtime": 6.2784,
      "eval_samples_per_second": 597.285,
      "eval_steps_per_second": 9.397,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 1.736435462495525e-05,
      "loss": 0.3236,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8672,
      "eval_loss": 0.4636218547821045,
      "eval_runtime": 6.4158,
      "eval_samples_per_second": 584.497,
      "eval_steps_per_second": 9.196,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 5.356578350067139,
      "learning_rate": 4.8301719287419e-06,
      "loss": 0.1789,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8834666666666666,
      "eval_loss": 0.3946201801300049,
      "eval_runtime": 6.4711,
      "eval_samples_per_second": 579.501,
      "eval_steps_per_second": 9.117,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 35.69981002807617,
      "learning_rate": 1.7800850740101738e-10,
      "loss": 0.1214,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8866666666666667,
      "eval_loss": 0.38833004236221313,
      "eval_runtime": 6.9817,
      "eval_samples_per_second": 537.119,
      "eval_steps_per_second": 8.451,
      "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
}