DeepSpeed
compute global norm on device
#5125
Merged

compute global norm on device #5125

BacharL
BacharL BacharL requested a review from mrwyattii mrwyattii 1 year ago
BacharL BacharL requested a review from tjruwase tjruwase 1 year ago
tjruwase
tjruwase commented on 2024-02-13
tjruwase
tjruwase commented on 2024-02-13
BacharL BacharL force pushed from 0541f4ef to 74f5b701 1 year ago
tjruwase
tjruwase approved these changes on 2024-02-15
tjruwase
BacharL BacharL force pushed from 03272aa4 to 53246a71 1 year ago
BacharL BacharL force pushed from 92d5dbbe to 283b3b24 1 year ago
BacharL BacharL force pushed from 283b3b24 to b30640b9 1 year ago
loadams
BacharL compute global norm on device
b9c2059e
BacharL code review comments
676b1992
BacharL code review comments
dc9ed13b
BacharL reuse graph code path for norm computation
bf5be3a5
BacharL BacharL force pushed from b30640b9 to bf5be3a5 1 year ago
loadams Fix formatting
56f13cef
BacharL fix return value device to match input device
ce15b963
loadams Merge branch 'master' into global_norm_on_device
5e4f8905
loadams loadams enabled auto-merge 1 year ago
loadams loadams merged 9f0e2136 into master 1 year ago

Login to write a write a comment.

Login via GitHub

Reviewers
Assignees
No one assigned
Labels
Milestone