Skip to content

[NPU] reduce_sum_across_dp_group builds the all_reduce tensor on CPU when CUDA is unavailable #23

[NPU] reduce_sum_across_dp_group builds the all_reduce tensor on CPU when CUDA is unavailable

[NPU] reduce_sum_across_dp_group builds the all_reduce tensor on CPU when CUDA is unavailable #23

name: Issue Triage
on:
issues:
types: [opened]
jobs:
triage:
runs-on: ubuntu-latest
permissions:
issues: write
id-token: write
steps:
- uses: actions/checkout@v4
- uses: anthropics/claude-code-action@v1
env:
ANTHROPIC_BASE_URL: ${{ secrets.ANTHROPIC_BASE_URL }}
CLAUDE_CODE_DISABLE_EXPERIMENTAL_BETAS: 1
with:
anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }}
prompt: |
REPO: ${{ github.repository }}
ISSUE NUMBER: ${{ github.event.issue.number }}
TITLE: ${{ github.event.issue.title }}
BODY: ${{ github.event.issue.body }}
AUTHOR: ${{ github.event.issue.user.login }}
Analyze this new issue and:
1. Determine if it's a bug report, feature request, or question
2. Assess priority (critical, high, medium, low)
3. Suggest appropriate labels
4. Check if it duplicates existing issues
Use ./.github/gh.sh to interact with GitHub:
- `./.github/gh.sh issue view [number]` to view the issue
- `./.github/gh.sh search issues "query"` to find similar issues
- `./.github/gh.sh label list` to see available labels
Based on your analysis, add the appropriate labels using:
`./.github/edit-issue-labels.sh --issue [number] --add-label "label1" --add-label "label2"`
If it appears to be a duplicate, post a comment mentioning the original issue.
claude_args: |
--allowedTools "Bash(./.github/gh.sh:*),Bash(./.github/edit-issue-labels.sh:*)"