fix EAGLE 2 non greedy case (#3407)
Co-authored-by: Ying Sheng <sqy1415@gmail.com>
This commit is contained in:
@@ -462,8 +462,11 @@ class CudaGraphRunner:
|
||||
),
|
||||
positions=None,
|
||||
retrive_index=None,
|
||||
retrive_next_token=None,
|
||||
retrive_next_sibling=None,
|
||||
retrive_cum_len=None,
|
||||
draft_token_num=self.model_runner.server_args.speculative_num_draft_tokens,
|
||||
spec_steps=self.model_runner.server_args.speculative_num_steps,
|
||||
capture_hidden_mode=CaptureHiddenMode.FULL,
|
||||
)
|
||||
|
||||
|
||||
Reference in New Issue
Block a user