rawsh

7 ms·

1.
Flash-MSA: Accelerating Million-Token Training with Sparse Attention Kernels(nanduruganesh.github.io)
41 points by rawsh 2mo ago|5 comments
2.
Zyphra ZAYA1-base: First large-scale model trained on AMD(zyphra.com)
6 points by rawsh 10mo ago|1 comments
3.
Debugging divergence between engine and transformers logprobs for RL(gist.github.com)
2 points by rawsh 1y ago|0 comments
4.
Batched reward model inference and Best-of-N sampling(raw.sh)
34 points by rawsh 2y ago|0 comments
5.
by rawsh 2y ago
You can actually get solid performance with pretrained chat models: https://raw.sh/posts/chess_puzzles On lichess puzzles gpt4o with the compiled prompt is around 70%, I think the 270M transformer is around 95%
6.
Teaching LLMs to solve chess puzzles with DSPy and Finetuning(raw.sh)
1 points by rawsh 2y ago|0 comments
7.
by rawsh 2y ago
Bit confused what the value add is over a framework like DSPy. This still requires you to create an eval dataset with ground truth, basically the only hard part of using DSPy. Easily getting the optimized prompt and having some metrics out
8.
Teaching chat models to solve chess puzzles(raw.sh)
4 points by rawsh 2y ago|0 comments
9.
by rawsh 2y ago
I built a web version with WASM at https://pdfgrep.com a few years ago in case it’s helpful to anyone
10.
by rawsh 2y ago
https://github.com/VikParuchuri/marker is solid, but slow and needs gpu(s) to be practical
11.
by rawsh 3y ago
Is it possible to use this for hybrid search in combination with pg_embedding? My understanding is that hybrid search currently requires syncing with Postgres
12.
Ask HN: Why aren't there Open source embedding models with context length > 512?
3 points by rawsh 3y ago|2 comments
13.
by rawsh 3y ago
Documents actually never get uploaded! PDF text extraction happens on the client using a web worker and MuPDF compiled to WASM. 1. PDF parsed and chunked on the client 2. Sparse vectors are regenerated for the entire document corpus and the
14.
by rawsh 3y ago
Nope, it’s a serious project; I mostly made it for personal use during my last semester of college. I rewrote it a few times and packaged it up because I think it’s genuinely useful. Langchain gets you 80% of the way there but you run into
15.
Show HN: DankGPT – Chat with Your Documents(dankgpt.com)
17 points by rawsh 3y ago|9 comments
16.
Show HN: Search PDFs in the browser using PDFgrep compiled to WebAssembly(pdfgrep.com)
4 points by rawsh 3y ago|1 comments
17.
by rawsh 3y ago
DankGPT is able to draw context from a library of documents (textbook, papers, class slides) to explain any topic and answer complicated reasoning problems. It’s very similar to ChatPDF, but you can include multiple documents and it has muc
18.
Show HN: Search PDFs using WASM in the browser(pdfgrep.com)
2 points by rawsh 4y ago|0 comments