News
AI, cybersecurity, machine learning, scraping & programming news.
Beyond SAEs: Consistent Concept Discovery with Sparse Coding
Sparse autoencoders (SAEs) explain foundation model representations by decomposing activations into sparse combinations of learned dictionary atoms. However, the learned dictionaries can vary across training runs, limiting the rel...
FuTransformer: FP8 prefill and INT4 decode for Llama-style inference on Hopper
Sparse autoencoders (SAEs) explain foundation model representations by decomposing activations into sparse combinations of learned dictionary atoms. However, the learned dictionaries can vary across training runs, limiting the rel...
DrugHarness: An Evolvable Agent Harness for Computational Drug Discovery
Sparse autoencoders (SAEs) explain foundation model representations by decomposing activations into sparse combinations of learned dictionary atoms. However, the learned dictionaries can vary across training runs, limiting the rel...
Show HN: Hybro – an open interoperability engine for AI agents
Hybro AI emerges as an open-source, hybrid multi-agent platform designed for seamless interoperability between local and remote AI agents. The project positions itself as the core orchestration engine for the Hybro Agent Network,...
Eight-Dimension Audit for AI Governance: Anthropic, EU, China, UK
The discourse on artificial intelligence is dominated by two claims that are rarely separated: that AI development is hitting technical limits, and that leading laboratories are shaping regulation to protect their market position....
The LLM Comments Are Not for You
Here is the rewritten article based on the source material, following all specified rules: Why RL Reinforces Verbose Comment Blocks I’ve heard a lot of talk about LLMs recently, and among the most common topics of discussion have...
Show HN: I recreated Contra in a custom NES game engine (NOT an emulator)
A developer has recreated Contra using a custom NES game engine built from scratch, and the engine itself is the real story. The project, posted to Hacker News by user normalocity, demonstrates a tiny engine called `neo8` that com...
Agentic Scientist, Orchestrator for Automated Scientific (Co-)Discovery
Sakana AI has released AI Scientist v3, a system that turns a task description into a full research campaign — experiments, ablations, and a compiled paper — driven by coding agents. The project treats agents as AI scientists. Sta...
OpenAI Saudi national bugging case shows threat to U.S. leadership in AI
Saudi national charged with bugging OpenAI executive room A Saudi national who worked at OpenAI has been charged with allegedly installing a recording device in an executive conference room at the company. Abdulrahman Alfozan, the...
Anthropic reports reducing recruitment of entry-level engineers
Anthropic's junior hiring slows as revenue per employee surges The Anthropic Institute's newest report says the company's junior recruitment has begun to slow down. Anthropic has been hiring for a smaller share of junior technical...