Skip to content

Actions: huggingface/nanotron

Actions

All workflows

Actions

Loading...
Loading

Showing runs from all workflows
1,519 workflow runs
1,519 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

Small fixes when resuming training Code Quality #591: Pull request #245 opened by NouamaneTazi
21s nouamane/fix-optim-states-resuming
nouamane/fix-optim-states-resuming
21s
Small fixes when resuming training Run FA2-related unit tests #688: Pull request #245 opened by NouamaneTazi
7m 19s nouamane/fix-optim-states-resuming
nouamane/fix-optim-states-resuming
7m 19s
Small fixes when resuming training Run non-FA2-related unit tests #688: Pull request #245 opened by NouamaneTazi
16m 6s nouamane/fix-optim-states-resuming
nouamane/fix-optim-states-resuming
16m 6s
fix ".module" issue in load_optimizer by passing unwrapped_model Secret Leaks #77: Commit 8e3c5e6 pushed by NouamaneTazi
18s nouamane/fix-optim-states-resuming
nouamane/fix-optim-states-resuming
18s
fix optim states dtype + fix lr scheduler initial value when resuming Secret Leaks #76: Commit 4cc62ae pushed by NouamaneTazi
21s nouamane/fix-optim-states-resuming
nouamane/fix-optim-states-resuming
21s
Merge pull request #234 from huggingface/ci-move Secret Leaks #75: Commit 51ca40b pushed by NouamaneTazi
17s nouamane/fix-optim-states-resuming
nouamane/fix-optim-states-resuming
17s
backup before doing monkey dispatch fp8 tp Secret Leaks #74: Commit 9510f57 pushed by xrsrke
16s xrsrke/fp8_for_nanotron
xrsrke/fp8_for_nanotron
16s
remove transpose in kernel Secret Leaks #73: Commit e93cf55 pushed by xrsrke
16s xrsrke/fp8_for_nanotron
xrsrke/fp8_for_nanotron
16s
add dumb transpose in fp8_matmul_kernel Secret Leaks #72: Commit edb1e87 pushed by xrsrke
19s xrsrke/fp8_for_nanotron
xrsrke/fp8_for_nanotron
19s
65% speed up in fwd+bwd pass with m=n=k=32768 Secret Leaks #71: Commit 4b26cf1 pushed by xrsrke
16s xrsrke/fp8_for_nanotron
xrsrke/fp8_for_nanotron
16s
add speed benchmark Secret Leaks #70: Commit f3e3495 pushed by xrsrke
17s xrsrke/fp8_for_nanotron
xrsrke/fp8_for_nanotron
17s
add bencmark speed with 5% speed up Secret Leaks #69: Commit c937375 pushed by xrsrke
22s xrsrke/fp8_for_nanotron
xrsrke/fp8_for_nanotron
22s
remove uncessary transpose input in the fwd pass, and contiguous weig… Secret Leaks #68: Commit c827594 pushed by xrsrke
22s xrsrke/fp8_for_nanotron
xrsrke/fp8_for_nanotron
22s
remove uncessary .contiguous() in fp8 backward Secret Leaks #67: Commit 39a4960 pushed by xrsrke
17s xrsrke/fp8_for_nanotron
xrsrke/fp8_for_nanotron
17s
update profiling script Secret Leaks #66: Commit b4156dc pushed by xrsrke
17s xrsrke/fp8_for_nanotron
xrsrke/fp8_for_nanotron
17s
add fp8 tp profiler Secret Leaks #65: Commit dda00a4 pushed by xrsrke
18s xrsrke/fp8_for_nanotron
xrsrke/fp8_for_nanotron
18s
add fp8 tensor parallel Secret Leaks #64: Commit 7dfe3ac pushed by xrsrke
19s xrsrke/fp8_for_nanotron
xrsrke/fp8_for_nanotron
19s
add fp8 linear Secret Leaks #63: Commit 0f8f672 pushed by xrsrke
17s xrsrke/fp8_for_nanotron
xrsrke/fp8_for_nanotron
17s
add fp8 tensor Secret Leaks #62: Commit 6c9a4d0 pushed by xrsrke
24s xrsrke/fp8_for_nanotron
xrsrke/fp8_for_nanotron
24s
ProTip! You can narrow down the results and go further in time using created:<2024-10-28 or the other filters available.