Source: https://seerepo.com/r/EleutherAI/lm-evaluation-harness/ai
Markdown source: https://seerepo.com/r/EleutherAI/lm-evaluation-harness/ai.md
Description: AI signed 0.3% of the 3,347 commits of lm-evaluation-harness. Each AI tool with its commits, and the last ones it signed.

# How much of lm-evaluation-harness did AI write?

AI signed 0.3% of the 3,347 commits of lm-evaluation-harness. Signed by Claude (8) and Cursor (1).

Since its first commit, on August 28, 2020.

## Who signed the commits

- Claude: 8 commits (0.2%)
- Cursor: 1 commit (0%)
- People: 99.7%

## Commits per month

Each month's commits up to its last one, with those an AI signed. Merges don't count.

- 2024-10: 27 commits, 0 signed by AI (0%)
- 2024-11: 27 commits, 0 signed by AI (0%)
- 2024-12: 22 commits, 0 signed by AI (0%)
- 2025-01: 29 commits, 0 signed by AI (0%)
- 2025-02: 27 commits, 0 signed by AI (0%)
- 2025-03: 56 commits, 0 signed by AI (0%)
- 2025-04: 19 commits, 0 signed by AI (0%)
- 2025-05: 31 commits, 0 signed by AI (0%)
- 2025-06: 20 commits, 0 signed by AI (0%)
- 2025-07: 35 commits, 0 signed by AI (0%)
- 2025-08: 27 commits, 0 signed by AI (0%)
- 2025-09: 15 commits, 0 signed by AI (0%)
- 2025-10: 13 commits, 0 signed by AI (0%)
- 2025-11: 26 commits, 0 signed by AI (0%)
- 2025-12: 20 commits, 0 signed by AI (0%)
- 2026-01: 29 commits, 0 signed by AI (0%)
- 2026-02: 22 commits, 0 signed by AI (0%)
- 2026-03: 17 commits, 0 signed by AI (0%)
- 2026-04: 12 commits, 2 signed by AI (16.7%)
- 2026-05: 28 commits, 1 signed by AI (3.6%)
- 2026-06: 11 commits, 0 signed by AI (0%)
- 2026-07: 3 commits, 1 signed by AI (33.3%)
- 2026-08: 75 commits, 3 signed by AI (4%)
- 2026-09: 10 commits, 2 signed by AI (20%)

## The last commits an AI signed

- 2026-09-14 · Claude · [e54687e](https://github.com/EleutherAI/lm-evaluation-harness/commit/e54687eac08501af43f09cc84af40ea2fc3cad60): fix(bbh): guard find_match against an all-empty capture tuple (#4125)
- 2026-09-01 · Claude · [69c005f](https://github.com/EleutherAI/lm-evaluation-harness/commit/69c005f265e7e1ac1636724c8515f63812626e20): feat(registry): runtime plugin registration for model backends (#4013)
- 2026-08-21 · Claude · [b0195a9](https://github.com/EleutherAI/lm-evaluation-harness/commit/b0195a94f4fe98c10acc043255081506b5d453f8): feat(models): add cross-platform onnxruntime-genai backend + refactor winml (#3960)

## How we know

Every AI coding tool leaves its signature in the commits it makes: a Co-Authored-By line, its own author name, or the merge of its branch. SeeRepo reads those signatures in the repo's history; a commit without one counts as made by people, and merges don't count.

Every AI tool SeeRepo recognizes: https://seerepo.com/agents

## Frequently asked questions

### Which AI tool signed the most commits in lm-evaluation-harness?

Claude, with 8 of the 3,347 commits of lm-evaluation-harness.

The proof: [its last signed commit, e54687e](https://github.com/EleutherAI/lm-evaluation-harness/commit/e54687eac08501af43f09cc84af40ea2fc3cad60)

### Since when does AI sign commits in lm-evaluation-harness?

Since April 2026: the first month an AI tool signed a commit of lm-evaluation-harness.

### How many commits does lm-evaluation-harness have?

lm-evaluation-harness has 3,347 commits since August 28, 2020. Merges don’t count.

## What is your repo built with?

Any public repo, with no account: https://seerepo.com/r/owner/repo/built, with the owner and name it has on GitHub.
