From e8a20db1a33477d09a209984b62aab2f0aaca1e9 Mon Sep 17 00:00:00 2001 From: jzzliu Date: Sat, 7 Nov 2026 20:19:18 -0700 Subject: [PATCH] README: correct model line count --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index 2427c49..e183ca0 100644 --- a/README.md +++ b/README.md @@ -5,7 +5,7 @@ production LLM systems: rotary position embeddings, RMSNorm, SwiGLU, an incremen **KV cache**, nucleus sampling, and a test suite that proves each piece is correct. No `nn.Transformer`, no Hugging Face. Attention, masking, caching and sampling are all -written out explicitly (~250 lines of model code) so every line can be reasoned about. +written out explicitly (~220 lines of model code) so every line can be reasoned about. ## What's inside