Re-land dp-attention local control broadcast test (#39835)

Co-authored-by: Sam Shleifer <sshleifer@gmail.com>
2026-09-17 14:33:25 +08:00
parent 6460082c05
commit fd6f96bf96
2 changed files with 130 additions and 0 deletions
@@ -20,6 +20,7 @@ class PauseResumeInPlaceMixin:
    pause_num_requests: int = 32
    pause_max_new_tokens: int = 512
    pause_duration: float = 5
+    pause_ignore_eos: bool = False
    pause_generate_url: str = ""
    pause_target_urls: list = []

@@ -36,6 +37,7 @@ class PauseResumeInPlaceMixin:
                    "sampling_params": {
                        "temperature": 0.8,
                        "max_new_tokens": self.pause_max_new_tokens,
+                        "ignore_eos": self.pause_ignore_eos,
                    },
                },
                timeout=_REQUEST_TIMEOUT,