Stories by georgehill

GigaGPT: GPT-3 sized models in 565 lines of code

Official PR Reveals the Inference Code for Mixtral 8x7B

Mistral: Our first AI endpoints are available in early access

Mixtral of experts

LLMs as Copilots for Theorem Proving in Lean

Combining Continuous-Time, Recurrent, and Convolutional Models

Image to Text ROS2

Mathematicians have found a new upper limit to the Ramsey number

Play with mixtral-8x7b

Automating Continual Learning

The EU AI Act negotiations ended

OpenAI's Altman Ouster Was Result of Drawn-Out Tensions

OpenAI taps former Twitter India head to kickstart in the country

Ramanujan's Sum

CPython Internals: What Happens Behind the Scenes When You Do a and B?

Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts (2017)

Mistral: Light-weight library for mixture-of-experts (MoE) training

Developing Llama 2 [video]

We're experimenting with LLMs to evolve GitHub Copilot

New year, new company – 10 additional spots in W24 for late applicants

5
6
7
8
9
10
11
12
13
14