Skip to content

Actions: swiss-ai/Megatron-LM

Actions

All workflows

Actions

Loading...
Loading

Showing runs from all workflows
8 workflow runs
8 workflow runs

Filter by Workflow

⌥ + click/return to exclude

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
Multimodality/sft bestfit sep classes Auto Swap Labels #3: Pull request #113 submitted by RaphaelKreft
1s
1s
add support for activation recomp and fp8 activation storage in Offlo… .github/workflows/check_api_backwards_compatibility_workflow.yml #5: Commit c4f4797 pushed by FFGGSSJJ
Failure moe/fp8_act
moe/fp8_act
Failure
Multimodality/vocab extension Auto Swap Labels #2: Pull request #112 submitted by RaphaelKreft
1s
1s
feat: Update SFT dataset for best-fit pretraining Auto Swap Labels #1: Pull request #111 submitted by RaphaelKreft
1s
1s
add support for offloading expert weights into CPU RAM [https://githu… .github/workflows/check_api_backwards_compatibility_workflow.yml #4: Commit a5cfca0 pushed by FFGGSSJJ
Failure moe/fp8_act
moe/fp8_act
Failure
add moe related metrics log .github/workflows/check_api_backwards_compatibility_workflow.yml #3: Commit 413984d pushed by FFGGSSJJ
Failure moe/fp8_act
moe/fp8_act
Failure
fix: memory leak of te fp8 tensor .github/workflows/check_api_backwards_compatibility_workflow.yml #2: Commit deb84ad pushed by FFGGSSJJ
Failure moe/fp8_act
moe/fp8_act
Failure
add mock router support; add logs for training .github/workflows/check_api_backwards_compatibility_workflow.yml #1: Commit b3179d1 pushed by FFGGSSJJ
Failure moe/fp8_act
moe/fp8_act
Failure