Update modeling_gptbert.py

modeling_gptbert.py

@@ -23,6 +23,9 @@ from transformers.modeling_outputs import (
 import math
 from typing import TYPE_CHECKING, Optional, Union, Tuple, List
 
+def is_flash_attn_2_available():
+    return False
+
 if is_flash_attn_2_available():
     from flash_attn.flash_attn_interface import flash_attn_varlen_qkvpacked_func
     from flash_attn.layers.rotary import RotaryEmbedding
@@ -1036,6 +1039,7 @@ class GptBertForSequenceClassification(GptBertModel):
 
         sequence_output, contextualized_embeddings, attention_probs = self.get_contextualized_embeddings(input_ids, attention_mask)
         logits = self.head(sequence_output[:, 0, :])
+        print(logits)
 
         loss = None
         if labels is not None: