Skip to content

Actions: deepspeedai/DeepSpeed

Actions

nv-pre-compile-ops

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
nv-pre-compile-ops.yml
2,500+ workflow runs
2,500+ workflow runs

Filter by Event

⌥ + click/return to exclude

Filter by Status

⌥ + click/return to exclude

Filter by Branch

⌥ + click/return to exclude

Filter by Actor

⌥ + click/return to exclude
nv-pre-compile-ops nv-pre-compile-ops #14897: Merge group checks requested
27m 31s
27m 31s
[muon] Reconcile the momentum dtype when a checkpoint is restored nv-pre-compile-ops #14896: Pull request #8433 synchronize by delock
31m 51s alanhuangyoo:fix/muon-momentum-dtype-on-resume
alanhuangyoo:fix/muon-momentum-dtype-on-resume
31m 51s
nv-pre-compile-ops nv-pre-compile-ops #14895: Merge group checks requested
32m 14s
32m 14s
nv-pre-compile-ops nv-pre-compile-ops #14894: Merge group checks requested
25m 44s
25m 44s
nv-pre-compile-ops nv-pre-compile-ops #14893: Merge group checks requested
24m 57s
24m 57s
fix(inference): reject non-positive max_out_tokens at config validation nv-pre-compile-ops #14892: Pull request #8454 synchronize by chakshu-dhannawat
Action required chakshu-dhannawat:fix/max-out-tokens-positive-validation
chakshu-dhannawat:fix/max-out-tokens-positive-validation
Action required
Fix ZeRO parameter alignment for grouped_mm nv-pre-compile-ops #14891: Pull request #8277 synchronize by tohtana
20m 21s fwerkor:fix-8276-grouped-mm-alignment
fwerkor:fix-8276-grouped-mm-alignment
20m 21s
Count each expert once in the fp32 gradient-clipping norm nv-pre-compile-ops #14890: Pull request #8478 synchronize by alanhuangyoo
Action required alanhuangyoo:fix/fp32-clip-moe-norm
alanhuangyoo:fix/fp32-clip-moe-norm
Action required
nv-pre-compile-ops nv-pre-compile-ops #14889: Merge group checks requested
24m 33s
24m 33s
nv-pre-compile-ops nv-pre-compile-ops #14888: Merge group checks requested
22m 52s
22m 52s
nv-pre-compile-ops nv-pre-compile-ops #14887: Merge group checks requested
23m 40s
23m 40s
nv-pre-compile-ops nv-pre-compile-ops #14886: Merge group checks requested
20m 19s
20m 19s
feat(rollout): add continuous batching generation prototype nv-pre-compile-ops #14885: Pull request #8368 synchronize by delock
20m 20s nathon-lee:perf/opsd-continuous-batching-prototype
nathon-lee:perf/opsd-continuous-batching-prototype
20m 20s
fix(inference): break circular import in ops.transformer.inference nv-pre-compile-ops #14884: Pull request #8465 synchronize by chakshu-dhannawat
Action required chakshu-dhannawat:fix/inference-circular-import-7159
chakshu-dhannawat:fix/inference-circular-import-7159
Action required
Carry the affine scale on the replicated map, not the split nv-pre-compile-ops #14883: Pull request #8477 synchronize by delock
24m 0s Achyuthan-S:affine-scale-on-replicated
Achyuthan-S:affine-scale-on-replicated
24m 0s
nv-pre-compile-ops nv-pre-compile-ops #14882: Scheduled
23m 4s master
master
23m 4s
nv-pre-compile-ops nv-pre-compile-ops #14881: Merge group checks requested
19m 13s
19m 13s
Fix comms logger KeyError when log_name is omitted nv-pre-compile-ops #14880: Pull request #8267 synchronize by tohtana
22m 59s jinyouzhi:comms_logger
jinyouzhi:comms_logger
22m 59s
Recognize a norm by its shape, not by whether its class name was listed nv-pre-compile-ops #14879: Pull request #8479 opened by alanhuangyoo
Action required alanhuangyoo:fix/autotp-norm-by-shape
alanhuangyoo:fix/autotp-norm-by-shape
Action required
Count each expert once in the fp32 gradient-clipping norm nv-pre-compile-ops #14878: Pull request #8478 opened by alanhuangyoo
Action required alanhuangyoo:fix/fp32-clip-moe-norm
alanhuangyoo:fix/fp32-clip-moe-norm
Action required
Fix AutoEP global L2 clipping for sharded expert ownership nv-pre-compile-ops #14877: Pull request #8476 synchronize by gss10282023
Action required gss10282023:fix-autoep-ownership-aware-clipping
gss10282023:fix-autoep-ownership-aware-clipping
Action required
Fix Muon optimizer under ZeRO CPU offload and bound gather buffers nv-pre-compile-ops #14876: Pull request #8464 synchronize by jinyouzhi
Action required jinyouzhi:muon-cpu-offload-fix
jinyouzhi:muon-cpu-offload-fix
Action required
[AutoSP] Fix compiler correctness nv-pre-compile-ops #14875: Pull request #8457 synchronize by jinyouzhi
Action required jinyouzhi:autosp-correctness
jinyouzhi:autosp-correctness
Action required
Fix Muon optimizer under ZeRO CPU offload and bound gather buffers nv-pre-compile-ops #14874: Pull request #8464 synchronize by jinyouzhi
Action required jinyouzhi:muon-cpu-offload-fix
jinyouzhi:muon-cpu-offload-fix
Action required
Carry the affine scale on the replicated map, not the split nv-pre-compile-ops #14873: Pull request #8477 opened by Achyuthan-S
14h 8m 37s Achyuthan-S:affine-scale-on-replicated
Achyuthan-S:affine-scale-on-replicated
14h 8m 37s