make evals also use managed

environments/gsm8k_server.py

@@ -122,18 +122,19 @@ class GSM8kEnv(BaseEnv):
 
     async def rollout_and_score_eval(self, question: str, answer: str) -> dict:
         """Rollout and score evaluation with detailed sample data collection."""
-        completion = await self.server.chat_completion(
-            messages=[
-                {"role": "system", "content": system_prompt},
-                {"role": "user", "content": question},
-            ],
-            n=1,
-            max_tokens=self.config.max_token_length,
-            temperature=0.0,
-            split="eval",
-        )
+        
+        async with self.server.managed_server(tokenizer=self.tokenizer) as managed:
+            completion = await managed.chat_completion(
+                messages=[
+                    {"role": "system", "content": system_prompt},
+                    {"role": "user", "content": question},
+                ],
+                n=1,
+                max_tokens=self.config.max_token_length,
+                temperature=0.0,
+            )
 
-        response_content = completion.choices[0].message.content
+            response_content = completion.choices[0].message.content
 
         # Parse gold answer
         gold_parsed = parse(