DeepSpeed
bc0d2465
- fix fused_qkv model accuracy issue (#5217)
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Commit
View On
GitHub
Hide Minimap (CTRL+M)
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>
References
#5217 - fix fused_qkv model accuracy issue
Author
Yejing-Lai
Parents
acf07398
Files
1
deepspeed/module_inject
fusedqkv_utils.py
Loading