Skip to content

Actions: NVIDIA-NeMo/RL

Actions

Create PR to main with cherry-pick from release

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
687 workflow runs
687 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

fix: Use separate step_metric for GPU Monitoring (#92)
Create PR to main with cherry-pick from release #37: Commit c770e64 pushed by terrykong
17s main
fix: gradient should be averaged instead of summed across mbs (#86)
Create PR to main with cherry-pick from release #36: Commit d1f0609 pushed by parthchadha
16s main
feat: evaluation implement (#16)
Create PR to main with cherry-pick from release #35: Commit 8416915 pushed by parthchadha
14s main
ci: skip functional until more capacity available and/or tests speed …
Create PR to main with cherry-pick from release #34: Commit 076274c pushed by terrykong
15s main
fix: error out early if ray cluster does not have resources (#89)
Create PR to main with cherry-pick from release #33: Commit ab52ca6 pushed by parthchadha
15s main
ci: temporarily disable CI on main since PRs must be up to date befor…
Create PR to main with cherry-pick from release #32: Commit d037fe3 pushed by terrykong
14s main
fix: unit test error when coverage wasn't specified (#88)
Create PR to main with cherry-pick from release #31: Commit 403e475 pushed by terrykong
20s main
feat: unit test metric tracking (#40)
Create PR to main with cherry-pick from release #30: Commit 084d6fa pushed by terrykong
14s main
fix: Remove reference of tokenizer from generation backend (#75) (#82)
Create PR to main with cherry-pick from release #29: Commit cca2aec pushed by parthchadha
17s main
fix: Mixed Prec memory improvements and better default configs (conve…
Create PR to main with cherry-pick from release #28: Commit bd7e4b0 pushed by SahilJain314
16s main
fix: update the instructions for multi-node setup; change the title f…
Create PR to main with cherry-pick from release #27: Commit c49571e pushed by terrykong
17s main
ci: tests now run with HF_DATASETS_CACHE to speed up e2e time (#41)
Create PR to main with cherry-pick from release #26: Commit 6a324e8 pushed by terrykong
18s main
feat: add gpu mem and util logging to wandb/tensorboard (#37)
Create PR to main with cherry-pick from release #25: Commit 4d62783 pushed by terrykong
13s main
fix: ray.sub race condition when overlapping srun commands on same no…
Create PR to main with cherry-pick from release #24: Commit 43ace69 pushed by terrykong
13s main
feat: Change vllm frac to 0.6 (#31)
Create PR to main with cherry-pick from release #23: Commit 9ee564e pushed by SahilJain314
19s main
docs: Add SFT quickstart (#29)
Create PR to main with cherry-pick from release #22: Commit 56883ce pushed by SahilJain314
14s main
feat: SFT convergence run changes (#21)
Create PR to main with cherry-pick from release #21: Commit f530ded pushed by SahilJain314
18s main
fix: updated stale cluster.md (#30)
Create PR to main with cherry-pick from release #20: Commit 8fb070f pushed by terrykong
12s main
docs: Updated adding models docs to fix latex rendering errors and fi…
Create PR to main with cherry-pick from release #19: Commit 0524b71 pushed by SahilJain314
13s main
feat: Use openmathinstruct2 training in grpo math example (#18)
Create PR to main with cherry-pick from release #18: Commit 06bc57d pushed by parthchadha
17s main
feat: Enable amp with autocast (fix poor bf16 convergence on GRPO (#26)
Create PR to main with cherry-pick from release #17: Commit 6b3dc31 pushed by SahilJain314
17s main
fix: disable usage stats more forcefully since container env took pre…
Create PR to main with cherry-pick from release #16: Commit 04f4d16 pushed by terrykong
17s main
docs: micro doc update with a helpful reminder on environment variabl…
Create PR to main with cherry-pick from release #15: Commit fc79f59 pushed by SahilJain314
17s main
docs: refresh our PR template (#23)
Create PR to main with cherry-pick from release #14: Commit 5a2b4e9 pushed by terrykong
14s main
feat: disable ray usage collection stats be default (#24)
Create PR to main with cherry-pick from release #13: Commit a5ec198 pushed by terrykong
12s main
ProTip! You can narrow down the results and go further in time using created:<2025-03-21 or the other filters available.