trainer_state.json

{
  "best_metric": 0.9456,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0007/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": 57.433719635009766,
      "learning_rate": 2.7151869176294504e-05,
      "loss": 0.7249,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9066666666666666,
      "eval_loss": 0.28038254380226135,
      "eval_runtime": 6.1783,
      "eval_samples_per_second": 606.958,
      "eval_steps_per_second": 9.549,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 20.70203971862793,
      "learning_rate": 1.9662164055386567e-05,
      "loss": 0.144,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9261333333333334,
      "eval_loss": 0.25512179732322693,
      "eval_runtime": 5.9969,
      "eval_samples_per_second": 625.324,
      "eval_steps_per_second": 9.838,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 4.761664390563965,
      "learning_rate": 1.0391670726449015e-05,
      "loss": 0.0475,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9354666666666667,
      "eval_loss": 0.21720823645591736,
      "eval_runtime": 5.9696,
      "eval_samples_per_second": 628.179,
      "eval_steps_per_second": 9.883,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.0058765471912920475,
      "learning_rate": 2.881402548656824e-06,
      "loss": 0.0099,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9429333333333333,
      "eval_loss": 0.19873137772083282,
      "eval_runtime": 6.7694,
      "eval_samples_per_second": 553.963,
      "eval_steps_per_second": 8.716,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.28699618577957153,
      "learning_rate": 2.670129987614178e-11,
      "loss": 0.002,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9456,
      "eval_loss": 0.19868813455104828,
      "eval_runtime": 6.5481,
      "eval_samples_per_second": 572.686,
      "eval_steps_per_second": 9.01,
      "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
}