diff options
| author | Void Agent <void@jayrup.hermes> | 2026-08-17 16:58:00 +0100 |
|---|---|---|
| committer | Void Agent <void@jayrup.hermes> | 2026-08-17 16:58:00 +0100 |
| commit | 62a1b02750667f6759a2e52b51860271975b7ff3 (patch) | |
| tree | 6bec042c7a0a7efb607e1ceaea0c6fe994ccfe1d /README.md | |
| parent | 95994af7a2b2a566498baa81c3b13b5f75a57372 (diff) | |
README: blog part 1+2 links, part 3 slot
Diffstat (limited to 'README.md')
| -rw-r--r-- | README.md | 4 |
1 files changed, 3 insertions, 1 deletions
@@ -56,7 +56,9 @@ transformer on CUDA but hurts the RNN (dynamic halting breaks graph capture). Fu ## Blog -The experiment is covered in a multi-part series on [jayrup.me](https://jayrup.me/) (parts 1–3, Aug 2026). +- [Part 1 — Can a Tiny Network Learn What a Prime Is?](https://jayrup.me/blog/prime-grokking-1) (2026-08-14) +- [Part 2 — Still No Grokking: The Weight-Decay Sweep](https://jayrup.me/blog/prime-grokking-2) (2026-08-16) +- Part 3 — published 2026-08-17 (full index: https://jayrup.me/blog) ## License |