[Render][Speculator] Add return_loss_mask to render endpoint for training data generation (#46846)

Signed-off-by: Ranran Haoran Zhang <ranzhang@redhat.com>
Co-authored-by: Benjamin Chislett <chislett.ben@gmail.com>
This commit is contained in:
Ranran
2026-06-28 00:07:33 -07:00
committed by GitHub
co-authored by Benjamin Chislett
parent a2a92cbbaa
commit 09841ae705
7 changed files with 288 additions and 20 deletions
+4
View File
@@ -87,6 +87,9 @@ class ChatParams:
mm_processor_kwargs: dict[str, Any] | None = None
"""The kwargs to pass to the multi-modal processor."""
return_assistant_tokens_mask: bool = False
"""Request a per-token assistant mask from apply_chat_template."""
def with_defaults(
self,
default_chat_template_kwargs: dict[str, Any] | None = None,
@@ -115,6 +118,7 @@ class ChatParams:
default_mm_processor_kwargs,
self.mm_processor_kwargs,
),
return_assistant_tokens_mask=self.return_assistant_tokens_mask,
)
def get_apply_chat_template_kwargs(self) -> dict[str, Any]: