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This repository has been archived by the owner on Aug 7, 2024.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

Python Lint and Test

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
python-app.yml
183 workflow runs
183 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

add wait_tensor() after all_gather in float8 to fix mem leak Python Lint and Test #58: Pull request #261 opened by bdhirsh
May 16, 2024 22:00 2m 2s memleak
memleak
May 16, 2024 22:00 2m 2s
[not for land] standalone repro of memory leak on float8 + compile + … Python Lint and Test #57: Pull request #260 synchronize by vkuzo
May 15, 2024 19:34 2m 1s 20240515_tp_sp_compile_oom_repro
20240515_tp_sp_compile_oom_repro
May 15, 2024 19:34 2m 1s
[not for land] standalone repro of memory leak on float8 + compile + … Python Lint and Test #56: Pull request #260 opened by vkuzo
May 15, 2024 19:33 2m 22s 20240515_tp_sp_compile_oom_repro
20240515_tp_sp_compile_oom_repro
May 15, 2024 19:33 2m 22s
[wip] make all 3 gemms in float8 linear configurable Python Lint and Test #55: Pull request #258 opened by vkuzo
May 10, 2024 20:10 2m 49s 20240510_mm_config
20240510_mm_config
May 10, 2024 20:10 2m 49s
Enable restricted split + cat in order to enable SP (#253) Python Lint and Test #54: Commit cb55df2 pushed by facebook-github-bot
May 9, 2024 01:03 2m 26s main
main
May 9, 2024 01:03 2m 26s
Enable restricted split + cat in order to enable SP Python Lint and Test #53: Pull request #253 synchronize by drisspg
May 8, 2024 23:38 2m 20s Enable-sequence-parallelism
Enable-sequence-parallelism
May 8, 2024 23:38 2m 20s
Enable restricted split + cat in order to enable SP Python Lint and Test #52: Pull request #253 synchronize by drisspg
May 8, 2024 23:36 2m 21s Enable-sequence-parallelism
Enable-sequence-parallelism
May 8, 2024 23:36 2m 21s
Enable restricted split + cat in order to enable SP Python Lint and Test #51: Pull request #253 synchronize by drisspg
May 8, 2024 23:33 2m 1s Enable-sequence-parallelism
Enable-sequence-parallelism
May 8, 2024 23:33 2m 1s
Add Dtensor compile test (#256) Python Lint and Test #50: Commit 14b00aa pushed by facebook-github-bot
May 7, 2024 20:57 2m 17s main
main
May 7, 2024 20:57 2m 17s
make the backward of differentiable float8 casts pass gradient as is … Python Lint and Test #49: Commit 605fc1d pushed by facebook-github-bot
May 6, 2024 19:27 2m 18s main
main
May 6, 2024 19:27 2m 18s
Add Dtensor compile test Python Lint and Test #48: Pull request #256 opened by drisspg
May 3, 2024 18:53 2m 18s add-compile-test-dtensor
add-compile-test-dtensor
May 3, 2024 18:53 2m 18s
make the backward of differentiable float8 casts pass gradient as is Python Lint and Test #47: Pull request #255 opened by vkuzo
May 3, 2024 18:10 4m 8s 20240503_fix_float8_cast_bw
20240503_fix_float8_cast_bw
May 3, 2024 18:10 4m 8s
minor changes in float8_experimental Python Lint and Test #46: Commit 7c53229 pushed by facebook-github-bot
May 3, 2024 00:41 3m 43s main
main
May 3, 2024 00:41 3m 43s
Enable restricted split + cat in order to enable SP Python Lint and Test #45: Pull request #253 synchronize by drisspg
April 29, 2024 18:15 2m 4s Enable-sequence-parallelism
Enable-sequence-parallelism
April 29, 2024 18:15 2m 4s
Enable restricted split + cat in order to enable SP Python Lint and Test #44: Pull request #253 opened by drisspg
April 25, 2024 03:50 3m 9s Enable-sequence-parallelism
Enable-sequence-parallelism
April 25, 2024 03:50 3m 9s
Added eager fp8 all-gather for dynamic scaling (subclass) (#247) Python Lint and Test #43: Commit ac065d0 pushed by facebook-github-bot
April 16, 2024 16:39 6m 50s main
main
April 16, 2024 16:39 6m 50s
Try dispatching to the decomposed OpOverload to account for inference mode Python Lint and Test #42: Pull request #251 opened by drisspg
April 10, 2024 01:37 12m 8s float8-inferenece-mode
float8-inferenece-mode
April 10, 2024 01:37 12m 8s
Allow for modifying the scaled_mm compute (#144) Python Lint and Test #41: Commit 31877bb pushed by facebook-github-bot
April 9, 2024 18:46 7m 59s main
main
April 9, 2024 18:46 7m 59s
Allow for modifying the scaled_mm compute Python Lint and Test #40: Pull request #144 synchronize by drisspg
April 9, 2024 18:17 2m 55s matmul_config
matmul_config
April 9, 2024 18:17 2m 55s
Allow for modifying the scaled_mm compute Python Lint and Test #39: Pull request #144 synchronize by drisspg
April 8, 2024 21:43 7m 41s matmul_config
matmul_config
April 8, 2024 21:43 7m 41s
Allow for modifying the scaled_mm compute Python Lint and Test #38: Pull request #144 synchronize by drisspg
April 8, 2024 20:52 2m 22s matmul_config
matmul_config
April 8, 2024 20:52 2m 22s
Allow for modifying the scaled_mm compute Python Lint and Test #37: Pull request #144 synchronize by drisspg
April 8, 2024 20:24 5m 11s matmul_config
matmul_config
April 8, 2024 20:24 5m 11s
Allow for modifying the scaled_mm compute Python Lint and Test #36: Pull request #144 synchronize by drisspg
April 8, 2024 19:57 2m 17s matmul_config
matmul_config
April 8, 2024 19:57 2m 17s
Allow for modifying the scaled_mm compute Python Lint and Test #35: Pull request #144 synchronize by drisspg
April 3, 2024 22:29 2m 24s matmul_config
matmul_config
April 3, 2024 22:29 2m 24s
Allow for modifying the scaled_mm compute Python Lint and Test #34: Pull request #144 synchronize by drisspg
April 3, 2024 21:58 23m 53s matmul_config
matmul_config
April 3, 2024 21:58 23m 53s