transformers
Fix InternVL attention when using qk_norm (38B and 78B)
#37620
Merged

Commits
  • fix internvlvision attention when using qk_norm
    yonigozlan committed 359 days ago
  • Merge branch 'main' into fix-large-internvl
    yonigozlan committed 359 days ago
  • Merge remote-tracking branch 'upstream/main' into fix-large-internvl
    yonigozlan committed 358 days ago
  • nit
    yonigozlan committed 358 days ago
  • modular
    yonigozlan committed 358 days ago
Loading