Skip to content

Actions: microsoft/DeepSpeed

hpu-gaudi2

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
877 workflow runs
877 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

Rearrange inference OPS and stop using builder.load
hpu-gaudi2 #950: Pull request #5490 synchronize by oelayan7
September 19, 2024 06:46 52m 10s oelayan7:rearrange_ops
September 19, 2024 06:46 52m 10s
hpu-gaudi2
hpu-gaudi2 #949: Scheduled
September 19, 2024 00:10 53m 14s master
September 19, 2024 00:10 53m 14s
Improve consistency of zero_grad
hpu-gaudi2 #948: Pull request #6554 synchronize by tohtana
September 18, 2024 21:56 52m 38s tohtana/consistent_zero_grad
September 18, 2024 21:56 52m 38s
Improve consistency of zero_grad
hpu-gaudi2 #947: Pull request #6554 synchronize by tohtana
September 18, 2024 21:04 51m 19s tohtana/consistent_zero_grad
September 18, 2024 21:04 51m 19s
Improve consistency of zero_grad
hpu-gaudi2 #946: Pull request #6554 synchronize by tohtana
September 18, 2024 20:59 5m 17s tohtana/consistent_zero_grad
September 18, 2024 20:59 5m 17s
Improve consistency of zero_grad
hpu-gaudi2 #945: Pull request #6554 synchronize by tohtana
September 18, 2024 20:55 4m 53s tohtana/consistent_zero_grad
September 18, 2024 20:55 4m 53s
Improve consistency of zero_grad
hpu-gaudi2 #944: Pull request #6554 opened by tohtana
September 18, 2024 20:27 27m 52s tohtana/consistent_zero_grad
September 18, 2024 20:27 27m 52s
Enabled Qwen2-MoE Tensor Parallelism (TP) inference
hpu-gaudi2 #942: Pull request #6551 opened by gyou2021
September 18, 2024 10:16 Action required gyou2021:qwen2-moe
September 18, 2024 10:16 Action required
Fix gradient accumulation for Z2+offload
hpu-gaudi2 #941: Pull request #6550 synchronize by tjruwase
September 18, 2024 09:57 53m 14s tohtana:tohtana/fix_grad_acc_z2_offload
September 18, 2024 09:57 53m 14s
Rearrange inference OPS and stop using builder.load
hpu-gaudi2 #939: Pull request #5490 synchronize by oelayan7
September 18, 2024 07:25 54m 7s oelayan7:rearrange_ops
September 18, 2024 07:25 54m 7s
Rearrange inference OPS and stop using builder.load
hpu-gaudi2 #938: Pull request #5490 synchronize by oelayan7
September 18, 2024 07:04 20m 20s oelayan7:rearrange_ops
September 18, 2024 07:04 20m 20s
Fix expert grad scaling problem with ZeRO optimizer
hpu-gaudi2 #936: Pull request #6546 synchronize by wyooyw
September 18, 2024 06:59 Action required wyooyw:fix_expert_weight_grad_with_zero
September 18, 2024 06:59 Action required
Fix expert grad scaling problem with ZeRO optimizer
hpu-gaudi2 #934: Pull request #6546 synchronize by wyooyw
September 18, 2024 02:17 Action required wyooyw:fix_expert_weight_grad_with_zero
September 18, 2024 02:17 Action required
hpu-gaudi2
hpu-gaudi2 #933: Scheduled
September 18, 2024 00:10 51m 50s master
September 18, 2024 00:10 51m 50s
hpu-gaudi2
hpu-gaudi2 #930: Scheduled
September 17, 2024 00:09 1h 13m 42s master
September 17, 2024 00:09 1h 13m 42s
[INF] Add config var to enable keeping checkpoints on host
hpu-gaudi2 #929: Pull request #6544 synchronize by loadams
September 16, 2024 23:05 1h 6m 46s oelayan7:keepModHost
September 16, 2024 23:05 1h 6m 46s
inference: remove unused _validate_args function
hpu-gaudi2 #928: Pull request #5505 synchronize by loadams
September 16, 2024 22:48 53m 25s nelyahu:remove_validate_args
September 16, 2024 22:48 53m 25s
Handle when backend is also in compile_kwargs
hpu-gaudi2 #927: Pull request #6502 synchronize by loadams
September 16, 2024 22:45 4m 29s oraluben:patch-1
September 16, 2024 22:45 4m 29s
[INF] Add config var to enable keeping checkpoints on host
hpu-gaudi2 #926: Pull request #6544 synchronize by oelayan7
September 16, 2024 11:25 49m 20s oelayan7:keepModHost
September 16, 2024 11:25 49m 20s
[INF] Add config var to enable keeping checkpoints on host
hpu-gaudi2 #925: Pull request #6544 synchronize by oelayan7
September 16, 2024 09:25 53m 7s oelayan7:keepModHost
September 16, 2024 09:25 53m 7s