Skip to content

Actions: microsoft/DeepSpeed

Actions

hpu-gaudi2

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
hpu-gaudi2.yml
1,122 workflow runs
1,122 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

hpu-gaudi2 hpu-gaudi2 #1302: Scheduled
November 23, 2024 00:11 1h 56m 14s master
master
November 23, 2024 00:11 1h 56m 14s
Stage3: Use new torch grad accumulation hooks API hpu-gaudi2 #1301: Pull request #6773 synchronize by loadams
November 22, 2024 16:23 Action required deepcharm:stage3-use-new-grad-acc-api
deepcharm:stage3-use-new-grad-acc-api
November 22, 2024 16:23 Action required
Unpin with latest transformers fixes hpu-gaudi2 #1300: Pull request #6763 synchronize by loadams
November 22, 2024 15:45 52m 24s loadams/transformers-2-5-update
loadams/transformers-2-5-update
November 22, 2024 15:45 52m 24s
[Draft][Demo] auto tp training hpu-gaudi2 #1299: Pull request #5445 synchronize by inkcherry
November 22, 2024 04:43 Action required inkcherry:auto_tp_training_
inkcherry:auto_tp_training_
November 22, 2024 04:43 Action required
hpu-gaudi2 hpu-gaudi2 #1298: Scheduled
November 22, 2024 00:12 55m 35s master
master
November 22, 2024 00:12 55m 35s
Set shuffle=True by default in data_sampler hpu-gaudi2 #1297: Pull request #6531 synchronize by loadams
November 21, 2024 20:17 54m 30s ranzhejiang:zhejiang/fix_runtime_dataloader_shuffle
ranzhejiang:zhejiang/fix_runtime_dataloader_shuffle
November 21, 2024 20:17 54m 30s
Stage3: Use new torch grad accumulation hooks API hpu-gaudi2 #1296: Pull request #6773 opened by deepcharm
November 21, 2024 14:57 Action required deepcharm:stage3-use-new-grad-acc-api
deepcharm:stage3-use-new-grad-acc-api
November 21, 2024 14:57 Action required
Enable torch compile on _allgather_params hpu-gaudi2 #1295: Pull request #6769 synchronize by deepcharm
November 21, 2024 14:13 52m 29s deepcharm:enable-torch-compile-on_allgather_params
deepcharm:enable-torch-compile-on_allgather_params
November 21, 2024 14:13 52m 29s
hpu-gaudi2 hpu-gaudi2 #1293: Scheduled
November 21, 2024 00:12 53m 51s master
master
November 21, 2024 00:12 53m 51s
Merge LoCo with Zero++ hpu-gaudi2 #1292: Pull request #6730 synchronize by XingyuXie
November 20, 2024 17:15 54m 37s XingyuXie:LoCo-Zero++
XingyuXie:LoCo-Zero++
November 20, 2024 17:15 54m 37s
Merge LoCo with Zero++ hpu-gaudi2 #1291: Pull request #6730 synchronize by XingyuXie
November 20, 2024 17:13 Action required XingyuXie:LoCo-Zero++
XingyuXie:LoCo-Zero++
November 20, 2024 17:13 Action required
Enable torch compile on _allgather_params hpu-gaudi2 #1290: Pull request #6769 opened by deepcharm
November 20, 2024 16:40 1h 1m 22s deepcharm:enable-torch-compile-on_allgather_params
deepcharm:enable-torch-compile-on_allgather_params
November 20, 2024 16:40 1h 1m 22s
Fix potential memory issues when use deepspeed Z3 hpu-gaudi2 #1289: Pull request #6726 synchronize by hwchen2017
November 20, 2024 06:38 55m 10s wenbinc-Bin:fix_z3
wenbinc-Bin:fix_z3
November 20, 2024 06:38 55m 10s
Reduce the device bubble introduced by heavy loop synchronization in coalesced fetch/release(z3_leaf_module) hpu-gaudi2 #1288: Pull request #6694 synchronize by inkcherry
November 20, 2024 02:35 Action required inkcherry:reduce_coalesced_fetch_bubble
inkcherry:reduce_coalesced_fetch_bubble
November 20, 2024 02:35 Action required
Merge LoCo with Zero++ hpu-gaudi2 #1286: Pull request #6730 synchronize by hwchen2017
November 20, 2024 01:51 55m 56s XingyuXie:LoCo-Zero++
XingyuXie:LoCo-Zero++
November 20, 2024 01:51 55m 56s
hpu-gaudi2 hpu-gaudi2 #1285: Scheduled
November 20, 2024 00:11 55m 17s master
master
November 20, 2024 00:11 55m 17s
Fix potential memory issues when use deepspeed Z3 hpu-gaudi2 #1284: Pull request #6726 synchronize by loadams
November 19, 2024 21:55 51m 44s wenbinc-Bin:fix_z3
wenbinc-Bin:fix_z3
November 19, 2024 21:55 51m 44s
Merge LoCo with Zero++ hpu-gaudi2 #1283: Pull request #6730 synchronize by XingyuXie
November 19, 2024 20:52 Action required XingyuXie:LoCo-Zero++
XingyuXie:LoCo-Zero++
November 19, 2024 20:52 Action required
Fix potential memory issues when use deepspeed Z3 hpu-gaudi2 #1282: Pull request #6726 synchronize by loadams
November 19, 2024 19:10 1h 48m 26s wenbinc-Bin:fix_z3
wenbinc-Bin:fix_z3
November 19, 2024 19:10 1h 48m 26s
BLOOM fixes for DS Legacy Inference hpu-gaudi2 #1281: Pull request #6765 opened by lekurile
November 19, 2024 18:39 1h 26m 30s lekurile/debug_bloom
lekurile/debug_bloom
November 19, 2024 18:39 1h 26m 30s
Fix potential memory issues when use deepspeed Z3 hpu-gaudi2 #1280: Pull request #6726 synchronize by loadams
November 19, 2024 18:27 45m 40s wenbinc-Bin:fix_z3
wenbinc-Bin:fix_z3
November 19, 2024 18:27 45m 40s
Add explicit parameters for torch.load hpu-gaudi2 #1279: Pull request #6751 synchronize by loadams
November 19, 2024 16:22 57m 10s loadams/weights-only-true
loadams/weights-only-true
November 19, 2024 16:22 57m 10s
Merge LoCo with Zero++ hpu-gaudi2 #1278: Pull request #6730 synchronize by XingyuXie
November 19, 2024 15:59 Action required XingyuXie:LoCo-Zero++
XingyuXie:LoCo-Zero++
November 19, 2024 15:59 Action required
Unpin with latest transformers fixes hpu-gaudi2 #1277: Pull request #6763 opened by loadams
November 19, 2024 15:07 57m 56s loadams/transformers-2-5-update
loadams/transformers-2-5-update
November 19, 2024 15:07 57m 56s
[Draft][Demo] auto tp training hpu-gaudi2 #1276: Pull request #5445 synchronize by inkcherry
November 19, 2024 10:29 1h 5m 51s inkcherry:auto_tp_training_
inkcherry:auto_tp_training_
November 19, 2024 10:29 1h 5m 51s