Valthor @vlthr

Joined May 2014

Tweets

523
Followers

295
Following

858
Likes

2K

Anton Osika – eu/acc @antonosika

2 months ago

@vlthr If you want to join lovable and work on technical problems with Valthor apply here lovable.dev/careers

3 5 35 8K 6

Lovable @lovable_dev

10 months ago

Building products takes too long. We started Lovable to fix this. Today, we're officially launching the world’s first AI Full Stack Engineer Lovable reliably replicates the capabilities of a full stack engineer and handles everything from design to database operations by…

71 89 652 229K 748

Download Video

Weston Beecroft @Westoncb

12 months ago

I've been exploring this idea of writing informal 'code sketches' as a way of structuring LLM code generation to efficiently capture all the design decisions *I* care about before handoff. Repo here: github.com/westoncb/code-… If there are two things LLMs are super good at it's…

4 7 49 4K 44

Download Video

Vincent D. Warmerdam @fishnets88

a year ago

Ahh yes. Recreative programming. So. Much. Fun.

3 11 151 73K 71

Download Image

Mason @webdevMason

a year ago

JD has a gift for saying things that are obviously true in the least palatable way imaginable

Heartland Signal @HeartlandSignal

a year ago

JD has a gift for saying things that are obviously true in the least palatable way imaginable

3K 5K 11K 16.6M 3K

Download Video

49 19 974 66K 52

Linus @thesephist

2 years ago

Embedding features learned with sparse autoencoders can make semantic edits to text ✨ (+ a reading/highlighting demo) I've built an interface to explore and visualize GPT-4 labelled features learned from a text embedding model's latent space. Here's a little video, more in 👇

14 37 366 52K 384

Download Video

search founder @n0riskn0r3ward

a year ago

Thought LLMs would be better at this: Given 2 500 token summaries: Does sum 1 mention anything sum 2 doesn't? Goal: Find 4 diffs w/o hallucinating. GPT-4 T:1/4, 4+ Hs GPT4-o:1/4, 4 Hs Command R+:1/4, 1 H L3:1/4, 1 H DeepseekV2: 3/4, 3 Hs Opus: 3/4, 1 H Gemini 1.5: 3/4, 0 H

2 1 6 2K 0

Fred Jonsson @enginoid

a year ago

since I'm deeply immersed in evals right now (and the process of building them) I got a kick out of this paper from @sh_reya @jdzamfi @bjo3rn @adityagp @IanArawjo it addresses the challenge of time-efficiently coming up with evals that are aligned with practitioners some…

1 18 69 13K 75

Download Image

noahdgoodman @noahdgoodman

a year ago

When I first saw Tree of Thoughts, I asked myself: If language models can reason better by searching, why don't they do it themselves during Chain of Thought? Some possible answers (and a new paper): 🧵

8 47 339 109K 399

Lj Miranda @ljvmiranda

2 years ago

Sharing RewardBench! We delve into the evaluation of reward models, often used to align LLMs to human preferences, across various tasks such as chat, reasoning, code, safety, etc. Hopefully, this work leads to a better understanding of the preference-tuning process :)

Nathan Lambert @natolambert

2 years ago

111 151 450 112K 235

Download Image

0 2 14 1K 1

Colin Fraser @colin_fraser

2 years ago

I think there's maybe really something here, look at this. I got GPT to respond to the prompt from this uml.edu/docs/sample-es…. Then I got a sequence of embeddings for each word in both the human- and GPT- authored essays. You can see how the human one moves around more.

Colin Fraser @colin_fraser

2 years ago

6 0 80 203K 23

50 111 891 217K 515

Download Gif

Simo Ryu @cloneofsimo

2 years ago

I've updated my auto codebase analyzer, using Qwen-1.5-72b-chat + sglang for large-codebase analysis, is really really good. Now it more clearly highlights important features of the codebase and methods. github.com/cloneofsimo/au…

1 5 40 4K 18

Download Image

Jeremy Howard @jeremyphoward

2 years ago

Today, with @Tim_Dettmers, @huggingface, & @Mobius_Labs, we're releasing FSDP/QLoRA, a new project that lets you efficiently train very large (70b) models on a home computer with consumer gaming GPUs. 1/🧵 answer.ai/posts/2024-03-…

83 647 3K 444K 2K

Nicolas Mejia Petit @mejia_petit

2 years ago

I am proud to release my newest coding dataset, 143k examples of tested python code. Why train on python code that doesn't work? When you can train on a large variety of tested python code! huggingface.co/datasets/Vezor…

2 3 14 1K 10

Omar Khattab @lateinteraction

2 years ago

Lots of requests for richer observability in DSPy. In March, @mikeldking & I are holding a DSPy <> @arizeai meetup in SF to show you how to do that w @ArizePhoenix-DSPy integration. Video by @axiomofjoy. Good chance to show something cool with DSPy. What would you like to see?

5 27 160 39K 104

Download Video

Anton Osika – eu/acc @antonosika

2 years ago

Happy Valentine’s Day everyone! And happy birthday Dad ❤️(my dad is awesome) Launching a new AI startup out of Europe today. Lovable. Needless to say, we’re very excited about Lovable. We think it will be huge: We’re building software that builds software. See website for…

5 7 53 8K 11

Download Video

Jascha Sohl-Dickstein @jaschasd

2 years ago

Have you ever done a dense grid search over neural network hyperparameters? Like a *really dense* grid search? It looks like this (!!). Blueish colors correspond to hyperparameters for which training converges, redish colors to hyperparameters for which training diverges.

273 2K 10K 1.6M 4K

Download Video

Lilian Weng @lilianweng

2 years ago

🗣️I've been thinking about data quality & human factor in the process a lot lately, so write a short post on the topic: lilianweng.github.io/posts/2024-02-… More: If you are into the topic, my team is hiring Research Engineer for a new sub-team Human-AI Interaction: openai.com/careers/resear…

22 101 751 79K 401

GoodAI @GoodAIdev

2 years ago

Are you relying on a large context for your LLM Agent Applications? Consider using Long-Term Memory (LTM) instead. We are open-sourcing: - GoodAI LTM Benchmark - our reference LTM agents - Experiment data & results goodai.com/introducing-go… github.com/GoodAI/goodai-……

4 20 81 31K 59

Download Image

Omar Khattab @lateinteraction

2 years ago

Let me experiment with new way for dev. Help needed: We want to create a DSPy optimizer that inspects the user's metric (python fn), not just failures, to iteratively tweak instructions. Any takers? We'll onboard you on writing & testing optimizers + invite as author on a paper