cmcollier

5 ms·

1.
Making film archives searchable with machine learning(huggingface.co)
1 points by cmcollier 2mo ago|0 comments
2.
by cmcollier 2mo ago
I'm curious to know what proportion of the training data at frontier model shops is composed of their scanned book text vs the internet.
3.
by cmcollier 3mo ago
Tinkering and thinking about the hacker news data in relation to some machine learning ideas. Here's my data and search engine with all the HN data: * https://orangewords.com/
4.
by cmcollier 5mo ago
https://orangewords.com Orange Words. My hobby project, a hacker news search system. It was initially created by hand and now I use AI augmented development. It's a good low risk environment for experimenting.
5.
by cmcollier 6mo ago
Working on small improvements to my hacker news search engine: - https://orangewords.com
6.
by cmcollier 8mo ago
Thanks! I did notice there are mixed stories about it working, and I got some email about it too. I'll check it out and make an update.
7.
by cmcollier 8mo ago
Ah, so this is why I suddenly got a bunch of email. Hey all, site owner here. Thanks for the visits and all the fun stories! I really miss this era of computing. Feel free to let me know if you have something that should be added to the sit
8.
by cmcollier 1y ago
There's a lot of potential here. I'm hopeful this turns out well for the community of Abilene.
9.
OpenAI's Stargate Megafactory with Sam Altman (Bloomberg) [video](youtube.com)
2 points by cmcollier 1y ago|2 comments
10.
by cmcollier 1y ago
Writing code that runs down hole or otherwise connects back to the real world would be fun. Maybe I should pickup firmware skills. Good luck with your hiring!
11.
Using the most unhinged AVX-512 instruction to make fastest phrase search algo(gab-menezes.github.io)
231 points by cmcollier 2y ago|61 comments
12.
by cmcollier 2y ago
For Vespa there's a managed version hosted by the Vespa company in their cloud environment, and then the open source version is easily run locally or in any environment of your choosing. It takes some attention to detail, but it's
13.
by cmcollier 2y ago
This is a good place to start: * https://hamel.dev/blog/posts/evals/#level-1-unit-tests And more broadly: * https://applied-llms.org/
14.
by cmcollier 2y ago
Here's one I've used: * https://www.together.ai/ Here are all the models: * https://docs.together.ai/docs/chat-models * https://docs.together.ai/docs/language-and-code-mo
15.
by cmcollier 2y ago
Unrelated to the core topic, I really enjoy the aesthetic of their website. Another similar one is from Fixie.ai (also, interestingly, one of their customers).
16.
by cmcollier 2y ago
This will get you the first 80%: * Any solid search engine (bm25 + embeddings and hnsw) * Any api to a model (gpt3.5, gpt4, claude, etc) * Some middleware to call search then build the prompt Then the remaining: * Create an eval dataset, th
17.
What We Learned from a Year of Building with LLMs (Part II)(oreilly.com)
2 points by cmcollier 2y ago|0 comments
18.
by cmcollier 2y ago
For me, it was a bit different, and it comes from a perspective that's a blend of cognitive science and computer science: Complex systems can be created through the composition of simple processes which are easily explained or modeled.
19.
Vector DB Comparison(superlinked.com)
3 points by cmcollier 2y ago|0 comments
20.
Matryoshka and Binary vectors: Slash vector search costs with Vespa(blog.vespa.ai)
2 points by cmcollier 2y ago|0 comments
21.
Only 64 Bytes per Embedding(mixedbread.ai)
2 points by cmcollier 2y ago|0 comments
22.
by cmcollier 2y ago
In terms of "the moment", I would imagine it happened during development inside Google (Lamda) or OpenAI (GPT2/3). More technically, here's one of the key papers discussing the topic (from google): * https://a
23.
by cmcollier 2y ago
There can be multiple reasons for this[0], including but not limited to: * The people or industry have low tolerance or fear around risk of false positives * The industry is centered around billable hours and has no incentive for automation
24.
by cmcollier 2y ago
For those wondering about the analogy: * https://en.wikipedia.org/wiki/Low-background_steel
25.
KDD Cup 2024 – Comprehensive RAG Benchmark(aicrowd.com)
2 points by cmcollier 2y ago|0 comments
26.
Introduction to Sentence Embeddings(osanseviero.github.io)
31 points by cmcollier 2y ago|0 comments
27.
Hugging Face is launching an open source robotics project(venturebeat.com)
3 points by cmcollier 3y ago|0 comments
28.
Contrastive Representations Provably Enable Planning and Inference(arxiv.org)
1 points by cmcollier 3y ago|0 comments
29.
Foundations of Vector Retrieval (Sebastian Bruch)(arxiv.org)
2 points by cmcollier 3y ago|0 comments
30.
by cmcollier 3y ago
As a general rule (for now), you'll get the best search result for your dev time, with straight ahead BM25 via a js lib. In terms of overhead, with lower doc counts there's not much overhead with embeddings and knn/ann. Imag

More ›