From fc40ffc4aa36633b26bf4e5850430ae63c97a7c6 Mon Sep 17 00:00:00 2001 From: Andrej Karpathy Date: Thu, 30 May 2024 20:10:04 +0000 Subject: [PATCH] adjust readme to new scripts/ dir --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index 0a46d27..8dd54d2 100644 --- a/README.md +++ b/README.md @@ -4,7 +4,7 @@ LLM training in simple, pure C/CUDA. There is no need for 245MB of PyTorch or 10 Our current goal is to reproduce GPT-2. For an overview of current ongoing work, see the latest [State of the Union](https://github.com/karpathy/llm.c/discussions/344) post. -Update May 28, 2024: A useful recent post may be ["Reproducing GPT-2 (124M) in llm.c in 90 minutes for $20"](https://github.com/karpathy/llm.c/discussions/481) where I detail the steps to go from scratch to reproducing the GPT-2 miniseries for 124M/350M models. The files themselves that show the launch commands are [run124M.sh](run124M.sh) and [run350M.sh](run350M.sh). +Update May 28, 2024: A useful recent post may be ["Reproducing GPT-2 (124M) in llm.c in 90 minutes for $20"](https://github.com/karpathy/llm.c/discussions/481) where I detail the steps to go from scratch to reproducing the GPT-2 miniseries for 124M/350M models. E.g. have a look at the `scripts` directory, example: [scripts/run_gpt2_124M.sh](scripts/run_gpt2_124M.sh). I'd like this repo to only maintain C and CUDA code. Ports of this repo to other languages are very welcome, but should be done in separate repos, and then I am happy to link to them below in the "notable forks" section, just like I did in [llama2.c notable forks](https://github.com/karpathy/llama2.c/tree/master?tab=readme-ov-file#notable-forks).