Review economics
jingyaogong/minimind
Read from the 12 most recently merged pull requests ·
Not enough merged pull requests on both sides yet to compare how they land in review.
—
Attributed
0
Attributed PRs
12
Other PRs
12
Read
The rest — the latest 5
#814 [fix] PPO: align mb_resp_logp with old_resp_logp under autocast to stabilize training
0 reviews · 40h to merge
#759 fix: resolve CUDA OOM in train_grpo.py on GPUs with <=8GB VRAM
0 reviews · 1h to merge
#698 [fix] 修复训练脚本中 1-indexed step 与 0-indexed 逻辑混用的问题
0 reviews · 4h to merge
#594 Fix: support loading DDP-saved LoRA weights for inference
2 reviews · 0h to merge
#571 [fix] Refactor get_lr function to include min_lr calculation
0 reviews · 2.8d to merge
How this was measured
No commit in this window carries agent attribution, so there is no AI-authored share to read here — which is not evidence there is none; tools that only complete code inline leave no commit trail, so the unattributed side includes AI-assisted work.
“Attributed” means a commit carried an agent’s signature — a co-author trailer, an agent commit identity, or an agent bot account. Tools that only complete code inline leave no such mark, so the other column is “rest”, not “human-written”. Full method and its limits
The badge reads this repo’s current report, so it follows the number.
Tell me when this moves
We re-read jingyaogong/minimind weekly and email only when the number changes materially.
Read your own repos
This one is public. For a private repo, run the same read locally through your own GitHub credentials — nothing is installed and nothing is sent to us.
npx @ambera/review-taxWant this continuously on jingyaogong/minimind — each pull request paired to the task it came from, and the work graded from its review loop rather than its diff size? Claim this repo in Forge
Browse every repo people have read · Read a different repository