trainer_state.json

{
  "best_metric": 0.9066666666666666,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0348/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": 9.056146724734802e-05,
      "loss": 0.9392,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8314666666666667,
      "eval_loss": 0.5586187243461609,
      "eval_runtime": 10.6266,
      "eval_samples_per_second": 352.889,
      "eval_steps_per_second": 5.552,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 20.74757957458496,
      "learning_rate": 6.563018865678952e-05,
      "loss": 0.2978,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8522666666666666,
      "eval_loss": 0.5099659562110901,
      "eval_runtime": 10.4353,
      "eval_samples_per_second": 359.356,
      "eval_steps_per_second": 5.654,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.30291515588760376,
      "learning_rate": 3.47287092499105e-05,
      "loss": 0.0991,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8808,
      "eval_loss": 0.44426700472831726,
      "eval_runtime": 10.6414,
      "eval_samples_per_second": 352.396,
      "eval_steps_per_second": 5.544,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.08624473959207535,
      "learning_rate": 9.6603438574838e-06,
      "loss": 0.0235,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9026666666666666,
      "eval_loss": 0.38984057307243347,
      "eval_runtime": 10.3054,
      "eval_samples_per_second": 363.888,
      "eval_steps_per_second": 5.725,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.002542161615565419,
      "learning_rate": 3.5601701480203476e-10,
      "loss": 0.0035,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9066666666666666,
      "eval_loss": 0.3673652708530426,
      "eval_runtime": 11.4162,
      "eval_samples_per_second": 328.48,
      "eval_steps_per_second": 5.168,
      "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
}