fix: init proper term

modeling_stablelm_epoch.py

@@ -609,7 +609,7 @@ class StableLMEpochForCausalLM(StableLMEpochPreTrainedModel):
         hidden_states = outputs[0]
         logits = self.lm_head(hidden_states).float()
 
-        lm_loss = None
+        loss = None
         if labels is not None:
             # Shift so that tokens < n predict n
             shift_logits = logits[..., :-1, :].contiguous()
@@ -627,7 +627,7 @@ class StableLMEpochForCausalLM(StableLMEpochPreTrainedModel):
             return (loss,) + output if loss is not None else output
 
         return CausalLMOutputWithPast(
-            loss=lm_loss,
+            loss=loss,
             logits=logits,
             past_key_values=outputs.past_key_values,
             hidden_states=outputs.hidden_states,