Skip to content

Actions: microsoft/DeepSpeed

Actions

hpu-gaudi2

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
hpu-gaudi2.yml
146 workflow run results
146 workflow run results

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

Fix crash when creating Torch tensor on NPU with device=get_accelerator().current_device() hpu-gaudi2 #181: Pull request #5464 synchronize by tjruwase
April 26, 2024 16:58 51m 2s harygo2:master
harygo2:master
April 26, 2024 16:58 51m 2s
hpu-gaudi2 hpu-gaudi2 #180: Scheduled
April 26, 2024 00:08 49m 27s master
master
April 26, 2024 00:08 49m 27s
Disable unsupported ops on Windows from Windows build script. hpu-gaudi2 #179: Pull request #5467 opened by loadams
April 25, 2024 20:24 1h 10m 20s loadams/disable-windows-ops
loadams/disable-windows-ops
April 25, 2024 20:24 1h 10m 20s
Add throughput timer configuration hpu-gaudi2 #178: Pull request #5363 synchronize by loadams
April 25, 2024 19:55 50m 57s deepcharm:add-timers-configuration
deepcharm:add-timers-configuration
April 25, 2024 19:55 50m 57s
Add throughput timer configuration hpu-gaudi2 #177: Pull request #5363 synchronize by loadams
April 25, 2024 17:43 52m 41s deepcharm:add-timers-configuration
deepcharm:add-timers-configuration
April 25, 2024 17:43 52m 41s
hpu-gaudi2 hpu-gaudi2 #175: Scheduled
April 25, 2024 00:08 2h 6m 54s master
master
April 25, 2024 00:08 2h 6m 54s
Fix torch.compile error for PyTorch v2.3 hpu-gaudi2 #174: Pull request #5463 opened by tohtana
April 25, 2024 00:07 1h 18m 9s tohtana:tohtana/fix_z3_with_pt23
tohtana:tohtana/fix_z3_with_pt23
April 25, 2024 00:07 1h 18m 9s
Revert "stage3: efficient compute of scaled_global_grad_norm (#5256)" hpu-gaudi2 #173: Pull request #5461 opened by lekurile
April 24, 2024 23:43 50m 39s lekurile/ds_chat_revert_54c06872
lekurile/ds_chat_revert_54c06872
April 24, 2024 23:43 50m 39s
Fix compile wrapper hpu-gaudi2 #172: Pull request #5455 synchronize by loadams
April 24, 2024 17:41 50m 23s BacharL:compile
BacharL:compile
April 24, 2024 17:41 50m 23s
Add getter and setter methods for compile_backend across accelerators. hpu-gaudi2 #170: Pull request #5299 synchronize by vshekhawat-hlab
April 24, 2024 05:52 49m 39s vshekhawat-hlab:vshekhawat/accelerator_compile_backend
vshekhawat-hlab:vshekhawat/accelerator_compile_backend
April 24, 2024 05:52 49m 39s
hpu-gaudi2 hpu-gaudi2 #168: Scheduled
April 24, 2024 00:08 1h 4m 18s master
master
April 24, 2024 00:08 1h 4m 18s
Fix compile wrapper hpu-gaudi2 #167: Pull request #5455 synchronize by loadams
April 23, 2024 23:33 49m 52s BacharL:compile
BacharL:compile
April 23, 2024 23:33 49m 52s
Fix compile wrapper hpu-gaudi2 #166: Pull request #5455 opened by BacharL
April 23, 2024 21:30 50m 25s BacharL:compile
BacharL:compile
April 23, 2024 21:30 50m 25s
Optimize zero3 fetch params using all_reduce hpu-gaudi2 #164: Pull request #5420 synchronize by tjruwase
April 23, 2024 13:13 1h 16m 48s deepcharm:use-all-reduce-for-fetch-params
deepcharm:use-all-reduce-for-fetch-params
April 23, 2024 13:13 1h 16m 48s
enable yuan autotp & add conv tp hpu-gaudi2 #163: Pull request #5428 synchronize by tjruwase
April 23, 2024 12:48 50m 18s Yejing-Lai:lyj/enable_yuan
Yejing-Lai:lyj/enable_yuan
April 23, 2024 12:48 50m 18s
hpu-gaudi2 hpu-gaudi2 #162: Scheduled
April 23, 2024 00:09 49m 45s master
master
April 23, 2024 00:09 49m 45s
set the default to use set_to_none for clearing gradients in BF16 optimizer. hpu-gaudi2 #161: Pull request #5434 synchronize by loadams
April 22, 2024 22:29 50m 15s inkcherry:fix_5175_
inkcherry:fix_5175_
April 22, 2024 22:29 50m 15s
set the default to use set_to_none for clearing gradients in BF16 optimizer. hpu-gaudi2 #160: Pull request #5434 synchronize by loadams
April 22, 2024 22:11 18m 48s inkcherry:fix_5175_
inkcherry:fix_5175_
April 22, 2024 22:11 18m 48s
Optimize zero3 fetch params using all_reduce hpu-gaudi2 #159: Pull request #5420 synchronize by loadams
April 22, 2024 17:07 49m 43s deepcharm:use-all-reduce-for-fetch-params
deepcharm:use-all-reduce-for-fetch-params
April 22, 2024 17:07 49m 43s
set the default to use set_to_none for clearing gradients in BF16 optimizer. hpu-gaudi2 #158: Pull request #5434 synchronize by loadams
April 22, 2024 15:50 48m 26s inkcherry:fix_5175_
inkcherry:fix_5175_
April 22, 2024 15:50 48m 26s
hpu-gaudi2 hpu-gaudi2 #153: Scheduled
April 22, 2024 00:09 49m 3s master
master
April 22, 2024 00:09 49m 3s
Add throughput timer configuration hpu-gaudi2 #152: Pull request #5363 synchronize by deepcharm
April 21, 2024 13:25 1m 23s deepcharm:add-timers-configuration
deepcharm:add-timers-configuration
April 21, 2024 13:25 1m 23s
hpu-gaudi2 hpu-gaudi2 #150: Scheduled
April 21, 2024 00:09 49m 44s master
master
April 21, 2024 00:09 49m 44s
add device config env for the accelerator hpu-gaudi2 #149: Pull request #5396 synchronize by tjruwase
April 20, 2024 20:25 52m 14s shiyuan680:master
shiyuan680:master
April 20, 2024 20:25 52m 14s
add device config env for the accelerator hpu-gaudi2 #148: Pull request #5396 synchronize by tjruwase
April 20, 2024 20:25 17s shiyuan680:master
shiyuan680:master
April 20, 2024 20:25 17s