Fix full-attention layer id mapping for Hybrid models (e.g. Qwen3-Next/Qwen3.5) - #249
Open
wanzhenchn wants to merge 1 commit into
Open
Fix full-attention layer id mapping for Hybrid models (e.g. Qwen3-Next/Qwen3.5)#249wanzhenchn wants to merge 1 commit into
wanzhenchn wants to merge 1 commit into