AINewsnow

Do Japanese and English Share Neurons? I Trained the Microscope to Check

This story is from 2026-10-22. It is preserved in the archive; the latest stories are on the live feed.

Sparse autoencoders find human-readable features inside LLMs. I trained one per layer on a Japanese-English model and measured which features fire in both languages — then tested whether they actually matter, with controls. The correlation is beautiful. The causation is borderline. Both get reporte…

Read the full story at DEV Community — Machine Learning ↗

Timeline · 1 report

  1. 2026-10-22 14:31 · DEV Community — Machine Learning
    Do Japanese and English Share Neurons? I Trained the Microscope to Check

More stories

  1. OlmoEarth v1.3: State-of-the-art embeddings for the whole planet — Allen Institute for AI (Ai2)
  2. Uber brings ChatGPT Work to all employees with OpenAI — OpenAI News
  3. Claude Duet is Anthropic's answer to Meta Muse and OpenAI Dots — Engadget
  4. Write your own training loop — without managing GPUs — with Amazon SageMaker training session APIs — AWS Machine Learning Blog
  5. Stop guessing auto scaling targets for TTS on Amazon SageMaker AI — AWS Machine Learning Blog
  6. Cognition Becomes First Customer for NVIDIA Vera Rubin NVL72 on CoreWeave Cloud — CoreWeave Blog
  7. Ex-OpenAI Science Head Kevin Weil Launches Startup With $215 Million — Bloomberg AI
  8. Meta agrees deal to keep Zuckerberg in California despite wealth tax proposal — Financial Times AI

Get the daily brief of stories like this at 6:30 every morning →