Update `JukeboxConfig.from_pretrained` (#24443)

fix
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>

src/transformers/models/jukebox/configuration_jukebox.py

@@ -353,6 +353,8 @@ class JukeboxPriorConfig(PretrainedConfig):
     def from_pretrained(
         cls, pretrained_model_name_or_path: Union[str, os.PathLike], level=0, **kwargs
     ) -> "PretrainedConfig":
+        cls._set_token_in_kwargs(kwargs)
+
         config_dict, kwargs = cls.get_config_dict(pretrained_model_name_or_path, **kwargs)
 
         # get the prior config dict if we are loading from JukeboxConfig
@@ -486,6 +488,8 @@ class JukeboxVQVAEConfig(PretrainedConfig):
 
     @classmethod
     def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os.PathLike], **kwargs) -> "PretrainedConfig":
+        cls._set_token_in_kwargs(kwargs)
+
         config_dict, kwargs = cls.get_config_dict(pretrained_model_name_or_path, **kwargs)
 
         # get the text config dict if we are loading from CLIPConfig