Skip to content

Actions: NVIDIA-NeMo/RL

Actions

Community Bot

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
community-bot.yml
1,968 workflow runs
1,968 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

cp: fix: grad norm check for automodel gpt oss nightly (1708) into r0.5.0 Community Bot #1968: Issue comment #1711 (comment) created by coderabbitai bot
10s
10s
feat: Add SGLang rollout backend and tests Community Bot #1967: Issue comment #1674 (comment) created by guyueh1
11s
11s
feat: Add SGLang rollout backend and tests Community Bot #1966: Issue comment #1674 (comment) created by guyueh1
14s
14s
fix: on GB200 use single-thread checkpoint save to avoid Cpu OOM Community Bot #1965: Issue comment #1703 (comment) created by guyueh1
10s
10s
Improve ray.sub Community Bot #1964: Issue comment #1339 (comment) edited by snowmanwwg
11s
11s
Improve ray.sub Community Bot #1963: Issue comment #1339 (comment) created by snowmanwwg
9s
9s
DTensor policy v2 refactor Community Bot #1962: Issue #1589 edited by hemildesai
10s
10s
DTensor policy v2 refactor Community Bot #1961: Issue #1589 edited by hemildesai
9s
9s
feat: refactor dtensor v2 policy __init__ and introduce core types Community Bot #1960: Issue comment #1588 (comment) created by hemildesai
17s
17s
fix: grad norm check for automodel gpt oss nightly Community Bot #1959: Issue comment #1708 (comment) created by coderabbitai bot
11s
11s
magi attention support Community Bot #1958: Issue comment #1701 (comment) created by terrykong
14s
14s
fix: on GB200 use single-thread checkpoint save to avoid Cpu OOM Community Bot #1957: Issue comment #1703 (comment) created by terrykong
10s
10s
fix: on GB200 use single-thread checkpoint save to avoid Cpu OOM Community Bot #1956: Issue comment #1703 (comment) edited by terrykong
9s
9s
fix: on GB200 use single-thread checkpoint save to avoid Cpu OOM Community Bot #1955: Issue comment #1703 (comment) created by terrykong
14s
14s
fix: on GB200 use single-thread checkpoint save to avoid Cpu OOM Community Bot #1954: Issue comment #1703 (comment) created by guyueh1
14s
14s
P2P issue when running DPO Community Bot #1953: Issue comment #1683 (comment) created by dhineshkumar-r
11s
11s
gpt-oss dtensor worker v2 does not support tensor parallel Community Bot #1952: Issue comment #1684 (comment) created by terrykong
11s
11s
[Dtensor] SFT with LoRA is slower than without LoRA Community Bot #1951: Issue comment #1688 (comment) created by terrykong
16s
16s
GPT-OSS performance and memory usage Community Bot #1950: Issue comment #1604 (comment) created by guyueh1
13s
13s
GPT-OSS performance and memory usage Community Bot #1949: Issue #1604 closed by guyueh1
14s
14s
Can we apply yarn to Nemo-RL Community Bot #1948: Issue comment #1678 (comment) created by guyueh1
9s
9s
gpt-oss sft megatron does not support sequence packing Community Bot #1947: Issue comment #1685 (comment) created by guyueh1
13s
13s
P2P issue when running DPO Community Bot #1946: Issue comment #1683 (comment) created by guyueh1
11s
11s
magi attention support Community Bot #1945: Issue comment #1701 (comment) created by guyueh1
10s
10s
Requires a way to control segment setting for NVL72 system Community Bot #1944: Issue comment #1421 (comment) created by guyueh1
9s
9s