From 7451008c1b06a74352353f8848bcab7e4c3cfb8a Mon Sep 17 00:00:00 2001 From: wassname <1103714+wassname@users.noreply.github.com> Date: Thu, 27 Aug 2026 09:59:19 +0800 Subject: [PATCH] README table: point the untested row at the version that will be tested --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index 69cd6d8..c143cfc 100644 --- a/README.md +++ b/README.md @@ -678,7 +678,7 @@ mean with those dropped.* | read the data, and give hypotheses | 3.0 K | +0.44 | +0.44 | 0/11 | ablation | | quotes and exercises | 26 K | +0.35 | +0.47 | 3/12 | [`efcac5c`](https://github.com/wassname/ml-debug/blob/efcac5c/SKILL.md) | | quotes only, no exercises | 40 K | +0.13 | -- | 10/12 | [`d5d725e`](https://github.com/wassname/ml-debug/blob/d5d725e/SKILL.md) | -| be diligent first, named exercises | 28 K | not run yet | | | [`742dbe4`](https://github.com/wassname/ml-debug/blob/742dbe4/SKILL.md) | +| be diligent first, named exercises | 29 K | not run yet | | | [`3a58c54`](https://github.com/wassname/ml-debug/blob/3a58c54/SKILL.md) | Table: 0.0 is the obvious answer each question rejects and 1.0 is my own answer, so a negative row is worse than the answer the question was built to reject. Judge `gpt-5.6-terra`,