DeepSpeed
bc0d2465 - fix fused_qkv model accuracy issue (#5217)

Comment changes are shownComment changes are hidden
Commit
1 year ago
fix fused_qkv model accuracy issue (#5217) Fused_qkv model can not correctly choose the fused_qkv type. Need to update the module_name_matches. Co-authored-by: Olatunji Ruwase <olruwase@microsoft.com>
Author
Parents
  • deepspeed/module_inject
    • File
      fusedqkv_utils.py