Is WSM strategy used here for RLVR?

by adamo1139 - opened Sep 29

Sep 29

Hi, are you using Warmup-Stable and Merge strategy for training this model, even in the RLVR stage?

inclusionAI org Sep 30

We adopt WSM LR scheduler in the pre-training stage of Ling-1T-base, not in RLVR stage.

adamo1139 changed discussion status to closed Sep 30

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment