Update modeling_reasonir_8b.py

modeling_reasonir_8b.py

@@ -201,8 +201,8 @@ class LlamaDynamicNTKScalingRotaryEmbedding(LlamaRotaryEmbedding):
 
 def rotate_half(x):
     """Rotates half the hidden dims of the input."""
-    x1 = x[transformers., : x.shape[-1] // 2]
-    x2 = x[transformers., x.shape[-1] // 2 :]
+    x1 = x[..., : x.shape[-1] // 2]
+    x2 = x[..., x.shape[-1] // 2 :]
     return torch.cat((-x2, x1), dim=-1)
 
 
@@ -1313,9 +1313,9 @@ class LlamaForCausalLM(LlamaPreTrainedModel, GenerationMixin):
         r"""
         Args:
             labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
-                Labels for computing the masked language modeling loss. Indices should either be in `[0, transformers.,
+                Labels for computing the masked language modeling loss. Indices should either be in `[0, ...,
                 config.vocab_size]` or -100 (see `input_ids` docstring). Tokens with indices set to `-100` are ignored
-                (masked), the loss is only computed for the tokens with labels in `[0, transformers., config.vocab_size]`.
+                (masked), the loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`.
 
             num_logits_to_keep (`int`, *optional*):
                 Calculate logits for the last `num_logits_to_keep` tokens. If `0`, calculate logits for all
@@ -1433,7 +1433,7 @@ class LlamaForSequenceClassification(LlamaPreTrainedModel):
     ) -> Union[Tuple, SequenceClassifierOutputWithPast]:
         r"""
         labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
-            Labels for computing the sequence classification/regression loss. Indices should be in `[0, transformers.,
+            Labels for computing the sequence classification/regression loss. Indices should be in `[0, ...,
             config.num_labels - 1]`. If `config.num_labels == 1` a regression loss is computed (Mean-Square loss), If
             `config.num_labels > 1` a classification loss is computed (Cross-Entropy).
         """
@@ -1500,7 +1500,7 @@ SQuAD (a linear layer on top of the hidden-states output to compute `span start
 class LlamaForQuestionAnswering(LlamaPreTrainedModel):
     base_model_prefix = "transformer"
 
-    # Copied from transformers.models.bloom.modeling_bloom.BloomForQuestionAnswering.__init__ with Bloom->Llama
+    # Copied from ...models.bloom.modeling_bloom.BloomForQuestionAnswering.__init__ with Bloom->Llama
     def __init__(self, config):
         super().__init__(config)
         self.transformer = LlamaModel(config)
@@ -1628,7 +1628,7 @@ class LlamaForTokenClassification(LlamaPreTrainedModel):
     ) -> Union[Tuple, TokenClassifierOutput]:
         r"""
         labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
-            Labels for computing the sequence classification/regression loss. Indices should be in `[0, transformers.,
+            Labels for computing the sequence classification/regression loss. Indices should be in `[0, ...,
             config.num_labels - 1]`. If `config.num_labels == 1` a regression loss is computed (Mean-Square loss), If
             `config.num_labels > 1` a classification loss is computed (Cross-Entropy).
         """
@@ -1662,4 +1662,4 @@ class LlamaForTokenClassification(LlamaPreTrainedModel):
             logits=logits,
             hidden_states=outputs.hidden_states,
             attentions=outputs.attentions,
-        )
+        )