Skip to content
This repository has been archived by the owner on Aug 7, 2024. It is now read-only.

Actions: pytorch-labs/float8_experimental

Actions

All workflows

Actions

Loading...
Loading

Showing runs from all workflows
863 workflow runs
863 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

[bc-breaking] rename top level UX to convert_to_float8_training Ufmt #637: Pull request #329 opened by vkuzo
July 24, 2024 20:29 17s gh/vkuzo/39/head
gh/vkuzo/39/head
July 24, 2024 20:29 17s
[bc-breaking] switch argument order to module_filter_fn Ufmt #636: Pull request #328 opened by vkuzo
July 24, 2024 20:29 20s gh/vkuzo/38/head
gh/vkuzo/38/head
July 24, 2024 20:29 20s
[bc-breaking] move emulate flag to Float8LinearConfig Ufmt #635: Pull request #327 opened by vkuzo
July 24, 2024 20:29 21s gh/vkuzo/37/head
gh/vkuzo/37/head
July 24, 2024 20:29 21s
move tensor scaling configuration to Float8LinearConfig (#325) Ufmt #634: Commit f64d339 pushed by facebook-github-bot
July 24, 2024 19:15 22s main
main
July 24, 2024 19:15 22s
move tensor scaling configuration to Float8LinearConfig (#325) Python Lint and Test #180: Commit f64d339 pushed by facebook-github-bot
July 24, 2024 19:15 2m 12s main
main
July 24, 2024 19:15 2m 12s
[DISCUSSION] fix float8 all-gather in FSDP2 + TP: DTensor(WeightWithDynamicFloat8CastTensor) Python Lint and Test #179: Pull request #326 opened by weifengpy
July 24, 2024 08:04 2m 2s weifengpy:fsdp2
weifengpy:fsdp2
July 24, 2024 08:04 2m 2s
[DISCUSSION] fix float8 all-gather in FSDP2 + TP: DTensor(WeightWithDynamicFloat8CastTensor) Ufmt #633: Pull request #326 opened by weifengpy
July 24, 2024 08:04 26s weifengpy:fsdp2
weifengpy:fsdp2
July 24, 2024 08:04 26s
[bc-breaking] move tensor scaling configuration to Float8LinearConfig Ufmt #632: Pull request #325 synchronize by vkuzo
July 23, 2024 17:50 19s gh/vkuzo/36/head
gh/vkuzo/36/head
July 23, 2024 17:50 19s
[bc breaking] move most previously-global configs to Float8LinearConfig Ufmt #631: Pull request #324 synchronize by vkuzo
July 23, 2024 17:48 25s gh/vkuzo/35/head
gh/vkuzo/35/head
July 23, 2024 17:48 25s
[bc-breaking] move tensor scaling configuration to Float8LinearConfig Ufmt #630: Pull request #325 opened by vkuzo
July 23, 2024 17:48 22s gh/vkuzo/36/head
gh/vkuzo/36/head
July 23, 2024 17:48 22s
[bc breaking] move most previously-global configs to Float8LinearConfig Ufmt #629: Pull request #324 synchronize by vkuzo
July 23, 2024 16:10 19s gh/vkuzo/35/head
gh/vkuzo/35/head
July 23, 2024 16:10 19s
[bc breaking] move most previously-global configs to Float8LinearConfig Ufmt #628: Pull request #324 opened by vkuzo
July 23, 2024 16:07 19s gh/vkuzo/35/head
gh/vkuzo/35/head
July 23, 2024 16:07 19s
change x, w, dL_dY variable names to input, weight, grad_output (#323) Ufmt #627: Commit 603efc2 pushed by facebook-github-bot
July 23, 2024 02:56 18s main
main
July 23, 2024 02:56 18s
change x, w, dL_dY variable names to input, weight, grad_output (#323) Python Lint and Test #178: Commit 603efc2 pushed by facebook-github-bot
July 23, 2024 02:56 2m 31s main
main
July 23, 2024 02:56 2m 31s
bring back torch.autograd.Function Ufmt #626: Pull request #316 synchronize by vkuzo
July 22, 2024 22:49 23s gh/vkuzo/29/head
gh/vkuzo/29/head
July 22, 2024 22:49 23s
bring back torch.autograd.Function Ufmt #625: Pull request #316 synchronize by vkuzo
July 22, 2024 21:44 18s gh/vkuzo/29/head
gh/vkuzo/29/head
July 22, 2024 21:44 18s
bring back torch.autograd.Function Ufmt #624: Pull request #316 synchronize by vkuzo
July 22, 2024 21:32 21s gh/vkuzo/29/head
gh/vkuzo/29/head
July 22, 2024 21:32 21s
[bc breaking] change x, w, dL_dY variable names to input, weight, grad_output Ufmt #623: Pull request #323 synchronize by vkuzo
July 22, 2024 17:07 21s gh/vkuzo/34/head
gh/vkuzo/34/head
July 22, 2024 17:07 21s
[bc breaking] unify filtering functions Ufmt #622: Pull request #322 synchronize by vkuzo
July 22, 2024 17:07 28s gh/vkuzo/33/head
gh/vkuzo/33/head
July 22, 2024 17:07 28s
[bc breaking] change x, w, dL_dY variable names to input, weight, grad_output Ufmt #621: Pull request #323 synchronize by vkuzo
July 22, 2024 16:48 18s gh/vkuzo/34/head
gh/vkuzo/34/head
July 22, 2024 16:48 18s
[bc breaking] change x, w, dL_dY variable names to input, weight, grad_output Ufmt #620: Pull request #323 synchronize by vkuzo
July 22, 2024 16:46 18s gh/vkuzo/34/head
gh/vkuzo/34/head
July 22, 2024 16:46 18s
[bc breaking] change x, w, dL_dY variable names to input, weight, grad_output Ufmt #619: Pull request #323 opened by vkuzo
July 22, 2024 16:41 22s gh/vkuzo/34/head
gh/vkuzo/34/head
July 22, 2024 16:41 22s
[bc breaking] unify filtering functions Ufmt #618: Pull request #322 synchronize by vkuzo
July 19, 2024 22:45 20s gh/vkuzo/33/head
gh/vkuzo/33/head
July 19, 2024 22:45 20s
[bc breaking] unify filtering functions Ufmt #617: Pull request #322 opened by vkuzo
July 19, 2024 22:44 15s gh/vkuzo/33/head
gh/vkuzo/33/head
July 19, 2024 22:44 15s
make all 3 gemms in Float8Linear support configurability, not user fa… Python Lint and Test #177: Commit c58fb5d pushed by facebook-github-bot
July 19, 2024 22:29 2m 8s main
main
July 19, 2024 22:29 2m 8s