reverse engineering intelligent (learning) systems. rl, evals, safety @emulated_ai
San Francisco, CA
Joined February 2019
- Tweets180
- Following168
- Followers610
- Likes773
Ryan Peters retweeted
total gnomic victory
Meet Mistral Large 4, aka Le Chonk.
• 1T parameters, natively multimodal. 49B active.
It is the best open weights model from US or Europe on aggregated benchmarks.
• State-of-the-art on critical workloads, including cyber defense, manufacturing and finance and it surpasses closed frontier models on visual grounding.
• Forged in Europe end-to-end and is deployable from Europe via our own Mistral Cloud infrastructure.
• Available to all via API today. Working with cybersecurity partners privately.
Open weights release end of October.
Ryan Peters retweeted
Our clinical trial participants have now used their Neuralink implants for over 50,000 hours, building one of the largest datasets of intracortical neural activity ever.
We used it to pretrain a foundation model for BCI: self-supervised neural encoders that turn noisy, drifting spikes into stable representations.
Some initial results:
• Decoders that keep working for weeks instead of days. For some users, calibration dropped from 10 min/day to 10 min/week
• Teleporting cursors that jump straight to the intended target
• A new cursor-control record of 11.32 bits per second (exceeding the performance of most able-bodied people)
We’re just getting started. Future directions include pooling data across participants, one-shot (or zero-shot) calibration, and an API for the motor cortex.
Read more:
neuralink.com/updates/pretra…
Ryan Peters retweeted
Technical alignment is a science and engineering problem that we can and must solve, and interpretability is the bottleneck. I wrote up Goodfire’s plan to get there.
Ryan Peters retweeted
When VLMs process medical images, do they recognize organs and broken bones? Using Silico, I trained a J-Lens and created a visualizer exploring what Qwen-3.5 27B sees in each patch of an image.
I found that Qwen can identify things like this harmful cyst on an OCT-scan!
Ryan Peters retweeted
You ask an AI model a question. Why does it answer the way it does?
Using the model's own neurons, we can trace how it makes decisions — and steer it toward better ones.
Case study: we tested an LLM and found that it sometimes endorses drunk driving.🧵 (1/5)
Ryan Peters retweeted
We’re introducing imagination models: a new foundation model architecture that unlocks learning from internet-scale video.
Our first imagination model, Photon-1, learned to use a computer by watching 18 years of screen recording video without action labels.
Ryan Peters retweeted
We trained Brain2Qwerty v2 on ~22,000 sentences from 9 volunteers, each recorded for 10 hours wearing an MEG device while typing.
By using end-to-end deep learning on raw brain signals from MEG devices and fine-tuning LLMs, the system effectively bridges the gap between noisy neural data and coherent language.
The results are promising:
- Avg word accuracy of 61% across participants
- 78% word accuracy and 50%+ of sentences decoded with ≤ 1 word error for the top-performing participant
- Performance scales log-linearly with data volume
Ryan Peters retweeted
Anthropic is missing an opportunity to release an image generation model and name it Claude Monet
Ryan Peters retweeted
I'm excited to share what we're building at Engram! This team is incredible, and we're working on one of the most interesting problems in AI right now: how to build models that are tailored to each person and continually learn from experience. Come join us!
Ryan Peters retweeted
Excited to share our paper "Extracting task-relevant preserved dynamics from contrastive aligned neural recordings" has been accepted as a ✨spotlight✨ poster at #NeurIPS2025 this week!
I'd love to catch up with old friends and meet new ones. Let's chat in San Diego! ☕️🧠
Ryan Peters retweeted
Feeling the energy and excitement of being surrounded by so many brilliant minds.
#neuroscience #Chicago #neuro
Ryan Peters retweeted
New paper @naturemethods ! Miniaturized headmounted neural devices are great to study the brain in naturalistic behaviors, but are limited in their performance. We engineered ‘exoskeleton’ assists for mice to carry large recording devices with them: nature.com/articles/s41592-0…
Ryan Peters retweeted
🚨 New pre-print alert!🚨 Led by the brilliant, twitter-less post-doc James Hope. We built a 🤖 exoskeleton for 🐭🐁 so we can study their 🧠's as they go around navigating their world. An e.g., imaging 1000s of neurons in the cortex when mice walk an 8-maze. A 🧵
Brain-wide neural recordings in mice navigating physical spaces enabled by a cranial exoskeleton biorxiv.org/cgi/content/shor… #bioRxiv
Ryan Peters retweeted
🚨🚨 New pre-print alert! A fun collaboration involving 🤖s, fruit🪰s,🐠, and 🧬 with @DarylGohl. Co-led by twitterless Andrew Alegria and @Amey__Joshi . A generalized machine vision guided robot for microinjections into whole intact embryos. biorxiv.org/content/10.1101/… (1/n)
Ryan Peters retweeted
🚨🚨 New pre-print alert! 🚨🚨led by the Surinach-Rynes dream team @DanielSurinach @mathewrynes, fun collab with @adredish in @UMNeurosci: Strategy dependent mesoscale Ca2+ dynamics during spatial navigation biorxiv.org/content/10.1101/… 1/n
