elie

@eliebakouch

research @PrimeIntellect (prev: @huggingface) anon feedback: admonymous.co/eliebakouch

Joined January 2024

4,707 Following

25.8K Followers

Replies Reposts Media

Pinned
elie
@eliebakouch
Aug 15
to my knowledge this is the largest open experiment on autonomous agents iterating on a research environment we scaled runtime, compute, diversity of models and harnesses. as a comparison, similar tasks on oai/anthropic system cards are anthropic "optimizing an llm training on
Prime Intellect
@PrimeIntellect
Aug 15
We ran the largest open experiment on how frontier models do AI research. 100+ autonomous runs across 10+ models, sandboxed on 8xH200s for up to 8 days, iterating on the nanoGPT optimizer track. Best runs closed 82% of the gap to a record built by dozens of humans over months.
52
elie
@eliebakouch
2h
today with @Thom_Wolf, we introduce the alignment hub, a central place for misalignment incidents designed with three goals: 1. set a standard for minimal transparency that is easy for labs to follow 2. improve clarity by centralizing information and sourcing 3. track the
14
elie
@eliebakouch
3h
very cool release, first model in a while going beyond the ~1M context that most models iterate at recently! agentic coding benchmark scores look pretty competitive too, on par with reflection Beam
Behnam Neyshabur
@bneyshabur
4h
Introducing Fern, Mirendil’s first model: pretrained from scratch, with up to 10M tokens of context! That’s ~0.5M lines of code (roughly Megatron-LM scale) or days of an agent’s work, all in context at once. Its first version, Fern 0.1, had a short round of post-training 🚢
5
elie
@eliebakouch
6h
one very interesting point is that they don't talk about creating their own chips, and afaik, it's not mentioned once in the full S1 they even say stuff hinting that they are not developing their own chips (or never were?): > Our competitors may be better positioned… to
Anthropic
@AnthropicAI
6h
Anthropic has filed an S-1 registration statement with the SEC related to its proposed IPO. Read more: anthropic.com/news/s1-sec Legal disclaimer: A registration statement relating to these securities has been filed with the SEC but has not yet become effective. These securities
11
elie
@eliebakouch
Oct 24
i saw a lot of people being negative about the K3.1 pro release one point that was under discussed imo is that it's the first open lab that is pacing the frontier of open weight cyber capabilities, which makes it much easier (and create incentive) for other labs to make the
Clive Chan
@itsclivetime
Oct 24
Kimi intentionally pacing the open weights cyber frontier is a big win. Cyber is defense-dominant long term (i.e. every bug gets fixed), but it’s looking like a nasty transition in the short term, so gradual diffusion is 100% the way.
Quote
Andrew Curran
@AndrewCurran_
Oct 23
Moonshot is intentionally throttling cyber capabilities. From the blog; 'First, we aim not to release open weights that would extend the absolute frontier of cyber capability. K3.1 Pro's offensive cyber capability is, by design, below what our pipeline can produce, and remains
10