Don't treat Bert model like CLIP.

Bert can accept up to 512 tokens so any prompt with more than 77 should
just be passed to it as is instead of splitting it up like CLIP.

comfy/text_encoders/hydit.py

@@ -14,7 +14,7 @@ class HyditBertModel(sd1_clip.SDClipModel):
 class HyditBertTokenizer(sd1_clip.SDTokenizer):
     def __init__(self, embedding_directory=None, tokenizer_data={}):
         tokenizer_path = os.path.join(os.path.dirname(os.path.realpath(__file__)), "hydit_clip_tokenizer")
-        super().__init__(tokenizer_path, pad_with_end=False, embedding_size=1024, embedding_key='chinese_roberta', tokenizer_class=BertTokenizer)
+        super().__init__(tokenizer_path, pad_with_end=False, embedding_size=1024, embedding_key='chinese_roberta', tokenizer_class=BertTokenizer, pad_to_max_length=False, max_length=512, min_length=77)
 
 
 class MT5XLModel(sd1_clip.SDClipModel):