Skip to content

dont request prompt logprobs as we dont use them and they disable prefix cache and burst vllm memory usage #666

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
Jackmin801 merged 1 commit into from
Dec 27, 2025
Merged

dont request prompt logprobs as we dont use them and they disable prefix cache and burst vllm memory usage #666

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 0 additions & 1 deletion docs/release/TRAJECTORIES.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -447,7 +447,6 @@ sampling_args = {
"extra_body": {
"return_tokens_as_token_ids": True, # Return tokens as token IDs
"return_token_ids": True, # Include token_ids in response
"prompt_logprobs": 1, # Optional: include prompt logprobs
},
}
```
Expand Down
1 change: 0 additions & 1 deletion verifiers/rl/trainer/config.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -321,7 +321,6 @@ def __post_init__(self):
"include_stop_str_in_output": False,
"return_tokens_as_token_ids": True,
"return_token_ids": True,
"prompt_logprobs": True,
},
}
self.gradient_accumulation_steps = 1
Expand Down
2 changes: 1 addition & 1 deletion verifiers/utils/token_utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -73,7 +73,7 @@ def prepare_sampling_args_for_token_prompts(
) -> SamplingArgs:
"""Ensures necessary fields are set for token prompts to work."""
sampling_args["logprobs"] = True
extra_body = dict(return_token_ids=True, prompt_logprobs=True)
extra_body = dict(return_token_ids=True)
if "extra_body" in sampling_args:
sampling_args["extra_body"].update(extra_body)
else:
Expand Down
Loading