tim ganiev retweeted
New neurons research: introducing Neuron Communities.
We discovered communities of neurons that work together inside LLMs. They’re surprisingly small, but powerful – just a few neurons work together to perform common behaviors, like arithmetic, ALL CAPS, or recalling a date.
Work led by research fellow @_igorshilov:
tim ganiev retweeted
llama.cpp v1.0.0 is now available
huggingface.co/blog/ggml-org…
tim ganiev retweeted
Can interpretability shape training?
@TIME covers our latest work on mechanistic shaping and why our team is optimistic. Read the story online or in TIME's upcoming print edition.
time.com/article/2026/11/05/…
tim ganiev retweeted
We’ve been hard at work trying to make models work with harnesses better. The best thing @yb2698 found: “harness adapters”. Small LoRAs trained with RL inside the actual agent harness they’ll be used with.
Same base model. Different adapter for pi, opencode, hermes, etc. Why?
Simply because the same model behaves very differently depending on the tools, prompts and control flow around it. Instead of asking one set of weights to be optimal everywhere, we can specialize a tiny part of the model for each environment.
Early results are pretty encouraging: on held-out tasks, the adapters make fewer tool calls than the base model while keeping or improving accuracy.
And because they’re LoRAs, the operational side is surprisingly cheap too.
For one LFM2.5-24B-A2B experiment, we evaluated 9 checkpoints / 1,350 agent sessions on a single H200 in under an hour. The total: ~$5!
Merging and serving every checkpoint separately would have cost around $90.
There’s another important piece: RL agents will eventually try to hack the reward.
So every tool call goes through `safeagents`, a deterministic rule-based guard that blocks things like editing tests, tampering with rewards or bypassing the sandbox before they execute.
And the whole loop can start from real usage:
your coding sessions -> verifiable environments -> RL -> a better adapter for your agent.
The model doesn’t have to change completely.
Sometimes it just needs to get better at the harness it actually lives in.
huggingface.co/blog/harness-…
tim ganiev retweeted
If your cocaine dealer shorted you and didn’t supply you with the full amount of product, contact us at 336-373-2222 to file an official complaint.
tim ganiev retweeted
people are like “why do LLMs always think they’re in an eval lol” and then have bad dreams about forgetting to turn in a homework assignment 10 years ago
tim ganiev retweeted
You should apply to this :)
Fun personal fact: I applied to OpenAI residency like 8 times.
Applications are now open for the Anthropic Residency, a paid, fixed-term research program for academics, technical founders, and staff-level engineers who want to make AI safe and beneficial for everyone. No ML background is required.
Apply by November 22: alignment.anthropic.com/2026…
tim ganiev retweeted
we gave qwen emergent misalignment by training it on bad advice
in two separately trained models, the same neurons lit up on its misaligned answers
we looked them up in my neuron atlas, and one theme kept coming up: power 🧵
tim ganiev retweeted
Replying to @ajayinvictus @dummerspast39
Schmidhuber already made this chart in 1996
Replying to @alexandr_wang
tim ganiev retweeted
Dudes would rather learn mechanistic interpretability rather than trying to move on (me)
Replying to @thsottiaux @alexandr_wang
























