apply device-transfer patch from https://github.com/mosaicml/llm-foundry/pull/225/files

modeling_mpt.py

@@ -298,7 +298,12 @@ class MPTForCausalLM(MPTPreTrainedModel):
         return_dict = return_dict if return_dict is not None else self.config.return_dict
         use_cache = use_cache if use_cache is not None else self.config.use_cache
         outputs = self.transformer(input_ids=input_ids, past_key_values=past_key_values, attention_mask=attention_mask, prefix_mask=prefix_mask, sequence_id=sequence_id, return_dict=return_dict, output_attentions=output_attentions, output_hidden_states=output_hidden_states, use_cache=use_cache)
-        logits = F.linear(outputs.last_hidden_state, self.transformer.wte.weight)
+        # move outputs to same device as weights for token embedding
+        # needed to support HF `device_map`
+        logits = F.linear(
+            outputs.last_hidden_state.to(self.transformer.wte.weight.device),
+            self.transformer.wte.weight
+        )
         if self.logit_scale is not None:
             if self.logit_scale == 0:
                 warnings.warn(f'Multiplying logits by self.logit_scale={self.logit_scale!r}. This will produce uniform (uninformative) outputs.')