jinaai
/

xlm-roberta-flash-implementation

Inference Endpoints

🇪🇺 Region: EU

Model card Files Files and versions Community

jupyterjazz commited on Apr 17

Commit

f960115

•

1 Parent(s): 79c3c93

chore: change num lora def value

Signed-off-by: jupyterjazz <[email protected]>

Files changed (1) hide show

configuration_xlm_roberta.py +1 -1

configuration_xlm_roberta.py CHANGED Viewed

@@ -21,7 +21,7 @@ class XLMRobertaFlashConfig(PretrainedConfig):
             position_embedding_type="absolute",
             use_cache=True,
             classifier_dropout=None,
-            num_loras=5,
             load_trained_adapters=False,
             **kwargs,
     ):

             position_embedding_type="absolute",
             use_cache=True,
             classifier_dropout=None,
+            num_loras=1,
             load_trained_adapters=False,
             **kwargs,
     ):