Why setting the max tokens for each turn to 512? #49

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

Jump to bottom

Open

zfj1998 opened this issue Apr 27, 2025 · 1 comment

zfj1998 commented Apr 27, 2025

ReCall/src/verl/workers/rollout/vllm_rollout/vllm_rollout_spmd.py

Line 485 in 3d976d2

max_tokens=min(512, max(active_max_tokens)),

为什么这里要设置max token的限制呢？因为目前的任务比较简单，think不会超出512 token吗

Collaborator

AnselCmy commented Apr 29, 2025

这个主要限制每一轮think的长度，为了防止某一个prompt的think非常长影响训练速度，后期这个值也可以根据训练任务的难度进行调整

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment