[BugFix] Fix RoPE error in Llama 3.1 (vllm-project#6693)

Author: WoosukKwon

vllm/config.py

@@ -134,15 +134,6 @@ def __init__(
         self.hf_text_config = get_hf_text_config(self.hf_config)
         self.dtype = _get_and_verify_dtype(self.hf_text_config, dtype)
 
-        if (getattr(self.hf_config, "max_position_embeddings", 0) == 131072
-                and getattr(self.hf_config, "rope_scaling", None) is None):
-            # Note(simon): this is a special case for a model that doesn't
-            # supply rope_scaling. We should remove this once the model is
-            # updated.
-            self.hf_config.update({"rope_scaling": {
-                "type": "extended",
-            }})
-
         if (not self.disable_sliding_window
                 and self.hf_text_config.model_type == "gemma2"
                 and self.hf_text_config.sliding_window is not None):
@@ -1245,24 +1236,32 @@ def _get_and_verify_max_len(
         derived_max_model_len = default_max_len
 
     rope_scaling = getattr(hf_config, "rope_scaling", None)
-    # The correct one should be "longrope", kept "su" here
-    # to be backward compatible
-    if rope_scaling is not None and rope_scaling["type"] not in {
-            "su", "longrope", "extended"
-    }:
-        if disable_sliding_window:
-            # TODO(robertgshaw): Find a model that supports rope_scaling
-            # with sliding window to see if this case should be allowed.
-            raise NotImplementedError(
-                "Disabling sliding window is not supported for models "
-                "with rope_scaling. Please raise an issue so we can "
-                "investigate.")
-        assert "factor" in rope_scaling
-        scaling_factor = rope_scaling["factor"]
-        if rope_scaling["type"] == "yarn":
-            derived_max_model_len = rope_scaling[
-                "original_max_position_embeddings"]
-        derived_max_model_len *= scaling_factor
+    if rope_scaling is not None:
+        if "type" in rope_scaling:
+            rope_type = rope_scaling["type"]
+        elif "rope_type" in rope_scaling:
+            rope_type = rope_scaling["rope_type"]
+        else:
+            raise ValueError(
+                "rope_scaling must have a 'type' or 'rope_type' key.")
+
+        # The correct one should be "longrope", kept "su" here
+        # to be backward compatible
+        if rope_type not in ("su", "longrope", "llama3"):
+            if disable_sliding_window:
+                # TODO(robertgshaw): Find a model that supports rope_scaling
+                # with sliding window to see if this case should be allowed.
+                raise NotImplementedError(
+                    "Disabling sliding window is not supported for models "
+                    "with rope_scaling. Please raise an issue so we can "
+                    "investigate.")
+
+            assert "factor" in rope_scaling
+            scaling_factor = rope_scaling["factor"]
+            if rope_type == "yarn":
+                derived_max_model_len = rope_scaling[
+                    "original_max_position_embeddings"]
+            derived_max_model_len *= scaling_factor
 
     # If the user specified a max length, make sure it is smaller than the
     # derived length from the HF model config.

vllm/model_executor/layers/rotary_embedding.py

@@ -687,12 +687,13 @@ def get_rope(
         rotary_emb = RotaryEmbedding(head_size, rotary_dim, max_position, base,
                                      is_neox_style, dtype)
     else:
-        scaling_type = rope_scaling["type"]
+        scaling_type = rope_scaling[
+            "type"] if "type" in rope_scaling else rope_scaling["rope_type"]
         # The correct one should be "longrope" but keep "su" here
         # for backward compatible
-        if scaling_type not in {"su", "longrope", "extended"}:
+        if scaling_type not in {"su", "longrope", "llama3"}:
             scaling_factor = rope_scaling["factor"]
-        if scaling_type == "extended":
+        if scaling_type == "llama3":
             rotary_emb = ExtendedRotaryEmbedding(head_size, rotary_dim,
                                                  max_position, base,
                                                  is_neox_style, dtype)