[V1][Spec Decode][Ngram] 1.35x gain -> 1.95x gain on InstructCoder with prompt fix (#18971)

benchmarks/benchmark_dataset.py

@@ -865,7 +865,15 @@ class InstructCoderDataset(HuggingFaceDataset):
         for item in self.data:
             if len(sampled_requests) >= num_requests:
                 break
-            prompt = f"{item['instruction']}:\n{item['input']}"
+            prompt = f"{item['input']}\n\n{item['instruction']} Just output \
+            the code, do not include any explanation."
+
+            # apply template
+            prompt = tokenizer.apply_chat_template(
+                [{"role": "user", "content": prompt}],
+                add_generation_prompt=True,
+                tokenize=False,
+            )
             prompt_len = len(tokenizer(prompt).input_ids)
             sampled_requests.append(
                 SampleRequest(

vllm/benchmarks/datasets.py

@@ -880,7 +880,19 @@ class InstructCoderDataset(HuggingFaceDataset):
         for item in self.data:
             if len(sampled_requests) >= num_requests:
                 break
-            prompt = f"{item['instruction']}:\n{item['input']}"
+            prompt = f"{item['input']}\n\n{item['instruction']} Just output \
+            the code, do not include any explanation."
+
+            # apply template
+            prompt = tokenizer.apply_chat_template(
+                [{
+                    "role": "user",
+                    "content": prompt
+                }],
+                add_generation_prompt=True,
+                tokenize=False,
+            )
+
             prompt_len = len(tokenizer(prompt).input_ids)
             sampled_requests.append(
                 SampleRequest(