Stories by georgehill

Optimizing AI Inference at Character.ai

LiveBench: A Challenging, Contamination-Free LLM Benchmark

Francois Chollet – LLMs won't lead to AGI $1M Prize to find true solution [video]

General Intelligence (2024)

Microsoft will switch off Recall by default after security backlash

A New Chapter for Bartender

Open AI: Scaling and evaluating sparse autoencoders [pdf]

Adaptive Horizon Actor-Critic for Contact-Rich Policy Learning

Meta FAIR: Contextual Position Encoding, Learning to Count What's Important

Google responds to leak: Documentation lacks context

VALL-E 2: Achieves Human Parity in Zero-Shot TTS

Seal Leaderboards–private, expert evaluations of leading frontier models

Reproducing GPT-2 (124M) in llm.c in 90 minutes for $20

14,000 Google Search Ranking Features Leaked

Rabbit Gaslit Me, So I Dug Deeper [video]

$30M AI Is Hiding a Scam [video]

26× Faster Inference with Layer-Condensed KV Cache for Large Language Models

Ex-OpenAI employee reported losing 85% of his family's net worth

Setapp Mobile brings a promising new iPhone app experience to the EU

Scaling Autoregressive Multimodal Models with Vision, Text, Audio, and Action

1
2
3
4
5
6
7
8
9
10