DeepSpeed
Add support of Falcon models (7b, 40b, 180b) to DeepSpeed-FastGen
#4790
Merged

Add support of Falcon models (7b, 40b, 180b) to DeepSpeed-FastGen #4790

mrwyattii merged 6 commits into master from arashb/falcon
arashb
arashb arashb requested a review from mrwyattii mrwyattii 2 years ago
arashb arashb requested a review from awan-10 awan-10 2 years ago
arashb arashb requested a review from cmikeh2 cmikeh2 2 years ago
arashb Add support of Falcon 7b model
e89633e3
arashb Add TP support for Falcon 7b model
ed658aa7
arashb Add support of Falcon new decoder arch (40b and 180b models)
26d5c80b
arashb arashb force pushed from 8766da47 to 26d5c80b 2 years ago
arashb Fix kv rotary kernels of Falcon 180b model for 8-way sharding
f4a340a7
loadams Merge branch 'master' into arashb/falcon
ee7fc0f5
awan-10
awan-10 approved these changes on 2023-12-12
arashb Update DeepSpeed-FastGen blog to include Falcon support
5557e73c
mrwyattii
mrwyattii approved these changes on 2023-12-12
mrwyattii mrwyattii merged a7900bcc into master 2 years ago
mrwyattii mrwyattii deleted the arashb/falcon branch 2 years ago
RezaYazdaniAminabadi

Login to write a write a comment.

Login via GitHub

Assignees
No one assigned
Labels
Milestone