use Seq2Seq backend where either can be loaded from HF

lm_eval/models/huggingface.py

@@ -157,12 +157,17 @@ class HFLM(LM):
             trust_remote_code=trust_remote_code,
         )
 
-        if getattr(self._config, "model_type") in MODEL_FOR_CAUSAL_LM_MAPPING_NAMES:
-            self.AUTO_MODEL_CLASS = transformers.AutoModelForCausalLM
-        elif (
-            not getattr(self._config, "model_type")
+        if (
+            getattr(self._config, "model_type")
             in MODEL_FOR_SEQ_TO_SEQ_CAUSAL_LM_MAPPING_NAMES
         ):
+            # first check if model type is listed under seq2seq models, since some
+            # models like MBart are listed in both seq2seq and causal mistakenly in HF transformers.
+            # these special cases should be treated as seq2seq models.
+            self.AUTO_MODEL_CLASS = transformers.AutoModelForSeq2SeqLM
+        elif getattr(self._config, "model_type") in MODEL_FOR_CAUSAL_LM_MAPPING_NAMES:
+            self.AUTO_MODEL_CLASS = transformers.AutoModelForCausalLM
+        else:
             if not trust_remote_code:
                 eval_logger.warning(
                     "HF model type is neither marked as CausalLM or Seq2SeqLM. \
@@ -171,8 +176,6 @@ class HFLM(LM):
             # if model type is neither in HF transformers causal or seq2seq model registries
             # then we default to AutoModelForCausalLM
             self.AUTO_MODEL_CLASS = transformers.AutoModelForCausalLM
-        else:
-            self.AUTO_MODEL_CLASS = transformers.AutoModelForSeq2SeqLM
 
         assert self.AUTO_MODEL_CLASS in [
             transformers.AutoModelForCausalLM,