trainer_state.json

{
  "best_metric": 0.9317333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0649/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": 0.10692079365253448,
      "learning_rate": 8.150532052261322e-05,
      "loss": 0.7872,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8608,
      "eval_loss": 0.47701504826545715,
      "eval_runtime": 12.0391,
      "eval_samples_per_second": 311.485,
      "eval_steps_per_second": 4.901,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.10418304800987244,
      "learning_rate": 5.9067169791110576e-05,
      "loss": 0.2364,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8909333333333334,
      "eval_loss": 0.3794389069080353,
      "eval_runtime": 11.9995,
      "eval_samples_per_second": 312.512,
      "eval_steps_per_second": 4.917,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.0655178502202034,
      "learning_rate": 3.125583832491945e-05,
      "loss": 0.0751,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9032,
      "eval_loss": 0.3787580132484436,
      "eval_runtime": 11.8459,
      "eval_samples_per_second": 316.565,
      "eval_steps_per_second": 4.981,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.006755682174116373,
      "learning_rate": 8.69430947173542e-06,
      "loss": 0.0158,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9314666666666667,
      "eval_loss": 0.2946184277534485,
      "eval_runtime": 11.8285,
      "eval_samples_per_second": 317.032,
      "eval_steps_per_second": 4.988,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.002978636883199215,
      "learning_rate": 3.204153133218313e-10,
      "loss": 0.0018,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9317333333333333,
      "eval_loss": 0.2875301241874695,
      "eval_runtime": 12.0335,
      "eval_samples_per_second": 311.631,
      "eval_steps_per_second": 4.903,
      "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
}