Revert "Revert "[FEAT] Support GGUF format"" (#2287)

2024-11-30 22:14:48 -08:00
parent 1bfa511b95
commit 4936be8acc
41 changed files with 229 additions and 132 deletions
@@ -339,7 +339,7 @@ class CohereForCausalLM(nn.Module):
            forward_batch,
        )
        return self.logits_processor(
-            input_ids, hidden_states, self.model.embed_tokens.weight, forward_batch
+            input_ids, hidden_states, self.model.embed_tokens, forward_batch
        )

    def load_weights(self, weights: Iterable[Tuple[str, torch.Tensor]]):