DeepSpeed
Hybrid Engine Fix Llama
#3505
Merged

Hybrid Engine Fix Llama #3505

jeffra merged 11 commits into master from cholmes/hybrid-engine-debug
lekurile
cmikeh2 Misc fixes
3fd02eca
cmikeh2 Remove deprecated conditional
8aa43e8b
cmikeh2 Missing attribute for meta tensor feature
8153e0ba
cmikeh2 Add RMSNorm
6d42d4d4
jeffra Merge branch 'master' into cholmes/hybrid-engine-debug
d33df987
jeffra jeffra marked this pull request as ready for review 2 years ago
jeffra jeffra requested a review from jeffra jeffra 2 years ago
jeffra jeffra requested a review from tjruwase tjruwase 2 years ago
jeffra jeffra requested a review from RezaYazdaniAminabadi RezaYazdaniAminabadi 2 years ago
jeffra jeffra requested a review from mrwyattii mrwyattii 2 years ago
jeffra jeffra requested a review from awan-10 awan-10 2 years ago
jeffra jeffra requested a review from cmikeh2 cmikeh2 2 years ago
jeffra jeffra requested a review from arashb arashb 2 years ago
awan-10 Add/rename unit tests.
7cfde0c5
awan-10 don't inherit from meta as we don't support llama+meta yet.
fa395e31
awan-10 fix format. remove unused import.
586328d7
awan-10 Merge branch 'master' into cholmes/hybrid-engine-debug
e4fd6bb6
awan-10
awan-10 approved these changes on 2023-05-10
awan-10 awan-10 enabled auto-merge (squash) 2 years ago
jeffra
disabled auto-merge 2 years ago
Manually disabled by user
awan-10 make tests sequential so we don't oom
3b4496d9
jeffra convert to half before moving to gpu
29c3cde5
jeffra
jeffra approved these changes on 2023-05-11
jeffra jeffra merged 194053bd into master 2 years ago
jeffra jeffra deleted the cholmes/hybrid-engine-debug branch 2 years ago

Login to write a write a comment.

Login via GitHub

Assignees
No one assigned
Labels
Milestone