trainer_state.json

{
  "best_metric": 0.9258666666666666,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0661/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": 16.345703125,
      "learning_rate": 2.7151869176294504e-05,
      "loss": 0.7686,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8797333333333334,
      "eval_loss": 0.38781288266181946,
      "eval_runtime": 11.8322,
      "eval_samples_per_second": 316.932,
      "eval_steps_per_second": 4.986,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.20749811828136444,
      "learning_rate": 1.9662164055386567e-05,
      "loss": 0.1386,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9029333333333334,
      "eval_loss": 0.3088831901550293,
      "eval_runtime": 11.8896,
      "eval_samples_per_second": 315.402,
      "eval_steps_per_second": 4.962,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.15312699973583221,
      "learning_rate": 1.0391670726449015e-05,
      "loss": 0.0327,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9141333333333334,
      "eval_loss": 0.3243314325809479,
      "eval_runtime": 11.9089,
      "eval_samples_per_second": 314.89,
      "eval_steps_per_second": 4.954,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.021490389481186867,
      "learning_rate": 2.881402548656824e-06,
      "loss": 0.0057,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9253333333333333,
      "eval_loss": 0.28682535886764526,
      "eval_runtime": 11.9913,
      "eval_samples_per_second": 312.727,
      "eval_steps_per_second": 4.92,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.3964911103248596,
      "learning_rate": 2.670129987614178e-11,
      "loss": 0.0008,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9258666666666666,
      "eval_loss": 0.2859196662902832,
      "eval_runtime": 12.1933,
      "eval_samples_per_second": 307.546,
      "eval_steps_per_second": 4.839,
      "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
}