Skip to content

Actions: deepspeedai/DeepSpeed

Actions

hpu-gaudi2

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
hpu-gaudi2.yml
1,569 workflow runs
1,569 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

Improve overflow handling in ZeRO hpu-gaudi2 #1750: Pull request #6976 synchronize by tjruwase
February 28, 2025 14:16 3m 59s olruwase/ds_5241
olruwase/ds_5241
February 28, 2025 14:16 3m 59s
Enable ZeRO set/get APIs for NVMe offload hpu-gaudi2 #1749: Pull request #7046 synchronize by tjruwase
February 28, 2025 14:15 3m 2s olruwase/update_nvme_offload_states
olruwase/update_nvme_offload_states
February 28, 2025 14:15 3m 2s
Training multiple models hpu-gaudi2 #1748: Pull request #7018 synchronize by tjruwase
February 28, 2025 13:55 3m 3s olruwase/zero_multi_models
olruwase/zero_multi_models
February 28, 2025 13:55 3m 3s
Update gaudi2 nightly,ci to latest 1.20.0 build hpu-gaudi2 #1747: Pull request #7093 opened by raza-sikander
February 28, 2025 09:36 1h 1m 57s raza-sikander:master
raza-sikander:master
February 28, 2025 09:36 1h 1m 57s
Enabled configurable auto Tensor Parallelism (TP) for the inference of diverse models hpu-gaudi2 #1746: Pull request #6553 synchronize by gyou2021
February 28, 2025 07:48 11h 3m 2s gyou2021:configurable_autoTP
gyou2021:configurable_autoTP
February 28, 2025 07:48 11h 3m 2s
Enabled high-performance Automatic Tensor Parallelism (auto TP) for the MoE models on multiple GPUs/HPUs hpu-gaudi2 #1745: Pull request #6964 synchronize by gyou2021
February 28, 2025 05:53 10h 7m 15s gyou2021:autoTP_Qwen2Moe_DeepSeekv2
gyou2021:autoTP_Qwen2Moe_DeepSeekv2
February 28, 2025 05:53 10h 7m 15s
Enable ZeRO set/get APIs for NVMe offload hpu-gaudi2 #1744: Pull request #7046 synchronize by loadams
February 28, 2025 04:36 3m 4s olruwase/update_nvme_offload_states
olruwase/update_nvme_offload_states
February 28, 2025 04:36 3m 4s
Enabled high-performance Automatic Tensor Parallelism (auto TP) for the MoE models on multiple GPUs/HPUs hpu-gaudi2 #1743: Pull request #6964 synchronize by delock
February 28, 2025 03:37 Action required gyou2021:autoTP_Qwen2Moe_DeepSeekv2
gyou2021:autoTP_Qwen2Moe_DeepSeekv2
February 28, 2025 03:37 Action required
Enabled high-performance Automatic Tensor Parallelism (auto TP) for the MoE models on multiple GPUs/HPUs hpu-gaudi2 #1742: Pull request #6964 synchronize by gyou2021
February 28, 2025 03:07 3m 3s gyou2021:autoTP_Qwen2Moe_DeepSeekv2
gyou2021:autoTP_Qwen2Moe_DeepSeekv2
February 28, 2025 03:07 3m 3s
hpu-gaudi2 hpu-gaudi2 #1741: Scheduled
February 28, 2025 00:11 3m 1s master
master
February 28, 2025 00:11 3m 1s
Enable torch.autocast with ZeRO hpu-gaudi2 #1740: Pull request #6993 synchronize by tohtana
February 27, 2025 23:02 4m 17s tohtana/support_autocast
tohtana/support_autocast
February 27, 2025 23:02 4m 17s
Improve overflow handling in ZeRO hpu-gaudi2 #1739: Pull request #6976 synchronize by tjruwase
February 27, 2025 21:55 3m 3s olruwase/ds_5241
olruwase/ds_5241
February 27, 2025 21:55 3m 3s
Enable ZeRO set/get APIs for NVMe offload hpu-gaudi2 #1738: Pull request #7046 synchronize by loadams
February 27, 2025 18:48 1h 57m 50s olruwase/update_nvme_offload_states
olruwase/update_nvme_offload_states
February 27, 2025 18:48 1h 57m 50s
Unpin once transformers latest is fixed hpu-gaudi2 #1737: Pull request #7088 synchronize by loadams
February 27, 2025 18:48 56m 18s loadams/unpin-transformers-latest
loadams/unpin-transformers-latest
February 27, 2025 18:48 56m 18s
Enable ZeRO set/get APIs for NVMe offload hpu-gaudi2 #1736: Pull request #7046 synchronize by loadams
February 27, 2025 17:48 1h 0m 15s olruwase/update_nvme_offload_states
olruwase/update_nvme_offload_states
February 27, 2025 17:48 1h 0m 15s
Unpin once transformers latest is fixed hpu-gaudi2 #1735: Pull request #7088 synchronize by loadams
February 27, 2025 17:04 1h 2m 24s loadams/unpin-transformers-latest
loadams/unpin-transformers-latest
February 27, 2025 17:04 1h 2m 24s
Unpin once transformers latest is fixed hpu-gaudi2 #1734: Pull request #7088 opened by loadams
February 27, 2025 16:24 40m 51s loadams/unpin-transformers-latest
loadams/unpin-transformers-latest
February 27, 2025 16:24 40m 51s
Improve overflow handling in ZeRO hpu-gaudi2 #1733: Pull request #6976 synchronize by tjruwase
February 27, 2025 13:55 56m 47s olruwase/ds_5241
olruwase/ds_5241
February 27, 2025 13:55 56m 47s
Avoid graph breaks in torch.compile caused by inner classes in the backward hooks hpu-gaudi2 #1730: Pull request #7062 synchronize by deepcharm
February 27, 2025 10:10 Action required deepcharm:avoid-graph-break-caused-by-inner-classes
deepcharm:avoid-graph-break-caused-by-inner-classes
February 27, 2025 10:10 Action required
Avoid graph breaks in torch.compile caused by inner classes in the backward hooks hpu-gaudi2 #1729: Pull request #7062 synchronize by deepcharm
February 27, 2025 09:58 Action required deepcharm:avoid-graph-break-caused-by-inner-classes
deepcharm:avoid-graph-break-caused-by-inner-classes
February 27, 2025 09:58 Action required
Enabled high-performance Automatic Tensor Parallelism (auto TP) for the MoE models on multiple GPUs/HPUs hpu-gaudi2 #1728: Pull request #6964 synchronize by gyou2021
February 27, 2025 08:32 Action required gyou2021:autoTP_Qwen2Moe_DeepSeekv2
gyou2021:autoTP_Qwen2Moe_DeepSeekv2
February 27, 2025 08:32 Action required
Pin transformers version on tests that use latest. hpu-gaudi2 #1727: Pull request #7085 synchronize by loadams
February 27, 2025 01:05 2h 57m 47s loadams/cpu-torch-latest-fix
loadams/cpu-torch-latest-fix
February 27, 2025 01:05 2h 57m 47s
hpu-gaudi2 hpu-gaudi2 #1726: Scheduled
February 27, 2025 00:11 2h 53m 31s master
master
February 27, 2025 00:11 2h 53m 31s
hpu-gaudi2 hpu-gaudi2 #1725: Manually run by loadams
February 27, 2025 00:01 1h 0m 58s loadams/unpin-transformers-hpu
loadams/unpin-transformers-hpu
February 27, 2025 00:01 1h 0m 58s
Improve overflow handling in ZeRO hpu-gaudi2 #1724: Pull request #6976 synchronize by tjruwase
February 26, 2025 18:44 1h 53m 9s olruwase/ds_5241
olruwase/ds_5241
February 26, 2025 18:44 1h 53m 9s