rawsh's submissions | Hacker News

1.		Zyphra ZAYA1-base: First large-scale model trained on AMD (zyphra.com)
		6 points by rawsh 54 days ago \| past \| 1 comment
2.		Debugging divergence between engine and transformers logprobs for RL (gist.github.com)
		2 points by rawsh 4 months ago \| past
3.		Batched reward model inference and Best-of-N sampling (raw.sh)
		34 points by rawsh on Nov 19, 2024 \| past
4.		Teaching LLMs to solve chess puzzles with DSPy and Finetuning (raw.sh)
		1 point by rawsh on Sept 12, 2024 \| past
5.		Teaching chat models to solve chess puzzles (raw.sh)
		4 points by rawsh on Aug 24, 2024 \| past
6.		Ask HN: Why aren't there Open source embedding models with context length > 512?
		3 points by rawsh on Sept 12, 2023 \| past \| 2 comments
7.		Show HN: DankGPT – Chat with Your Documents (dankgpt.com)
		17 points by rawsh on July 26, 2023 \| past \| 9 comments
8.		Show HN: Search PDFs in the browser using PDFgrep compiled to WebAssembly (pdfgrep.com)
		4 points by rawsh on May 17, 2023 \| past \| 1 comment
9.		Show HN: Search PDFs using WASM in the browser (pdfgrep.com)
		2 points by rawsh on Feb 7, 2023 \| past