trainer_state.json

{
  "best_metric": 0.9109333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0491/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": Infinity,
      "learning_rate": 9.06165595011943e-05,
      "loss": 1.0412,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8005333333333333,
      "eval_loss": 0.6424129009246826,
      "eval_runtime": 6.9609,
      "eval_samples_per_second": 538.727,
      "eval_steps_per_second": 8.476,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.11741741746664047,
      "learning_rate": 6.57197748161099e-05,
      "loss": 0.361,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8517333333333333,
      "eval_loss": 0.4976669251918793,
      "eval_runtime": 6.3729,
      "eval_samples_per_second": 588.431,
      "eval_steps_per_second": 9.258,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.416384220123291,
      "learning_rate": 3.481857044676613e-05,
      "loss": 0.1465,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8672,
      "eval_loss": 0.452929824590683,
      "eval_runtime": 6.3105,
      "eval_samples_per_second": 594.245,
      "eval_steps_per_second": 9.349,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.8493234515190125,
      "learning_rate": 9.71615616894559e-06,
      "loss": 0.0447,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9074666666666666,
      "eval_loss": 0.347255140542984,
      "eval_runtime": 7.0987,
      "eval_samples_per_second": 528.262,
      "eval_steps_per_second": 8.311,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 18.624216079711914,
      "learning_rate": 8.010370950384261e-10,
      "loss": 0.0119,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9109333333333334,
      "eval_loss": 0.3358750641345978,
      "eval_runtime": 6.7496,
      "eval_samples_per_second": 555.591,
      "eval_steps_per_second": 8.741,
      "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
}