LLM-Deflate: Extracting LLMs into Datasets(scalarlm.com)

53 points by gdiamos 1 year ago | 26 comments

I use Nim instead of Python for data processing (2021)(benjamindlee.com)

53 points by archargelod 2 years ago | 44 comments

Optimizing a WebGPU Matmul Kernel for 1 TFLOP(zanussbaum.substack.com)

163 points by zanussbaum 1 year ago | 68 comments

Detecting when LLMs are uncertain(thariq.io)

257 points by trq_ 2 years ago | 148 comments

Show HN: LLM Attention Visualization(ishamf.dev)

145 points by ifz 1 month ago | 23 comments

Use Prolog to improve LLM's reasoning(shchegrikovich.substack.com)

343 points by shchegrikovich 2 years ago | 122 comments

Substack post arguing that having LLMs generate Prolog code can improve their reasoning, given the limits of autoregressive transformer generation.