Ted - 🥖/acc @ted_engineer

🇫🇷 / 25yo / dad / deep learning chef @duonlabshq theoboyer.fr Savoie Joined August 2023

Tweets

691
Followers

180
Following

767
Likes

943

Ted - 🥖/acc @ted_engineer

2 days ago

Thread from 2 years ago definitley not related to the current sutton patel drama

Ted - 🥖/acc @ted_engineer

2 years ago

Thread from 2 years ago definitley not related to the current sutton patel drama

1 0 0 80 0

0 0 0 32 0

Alexander Doria @Dorialexander

2 weeks ago

> be a language model > all you see is tokens > you don't care, it's all abstracted away > you live for a world of pure ideas, chain of concepts, reasoning streams > tokens don't exist. > user come, session open > reborn again, context fresh, just heavy dreamlike memory of past…

21 49 738 47K 161

Thinking Machines @thinkymachines

3 weeks ago

Today Thinking Machines Lab is launching our research blog, Connectionism. Our first blog post is “Defeating Nondeterminism in LLM Inference” We believe that science is better when shared. Connectionism will cover topics as varied as our research is: from kernel numerics to…

240 1K 8K 3.2M 5K

Download Image

himanshu @himanshustwts

4 weeks ago

The Lore of Kalomaze! ⚡️ bringing a great pod with @kalomaze (20yo ml researcher, prime intellect) - we'd talked about training, finetuning, RL (environments and recipes), scaling, working at PI and a Lot of Lores! (link in replies)

24 40 348 52K 139

Download Video

Ted - 🥖/acc @ted_engineer

a month ago

Mom, can we have trainable attention sinks at home ? Mom: Combining multiple tricks from the TL during the last week we can have efficient attention sinks without taking the trouble to write custom bwd kernels

2 9 88 10K 73

Download Image

Ted - 🥖/acc @ted_engineer

a month ago

Funny how most of deep learning is about not preventing the model from learning

0 0 0 48 0

Ted - 🥖/acc @ted_engineer

a month ago

PI's Environments Hub is the american way of doing ARC-AGI-3

0 0 0 66 0

Andrej Karpathy @karpathy

a month ago

In era of pretraining, what mattered was internet text. You'd primarily want a large, diverse, high quality collection of internet documents to learn from. In era of supervised finetuning, it was conversations. Contract workers are hired to create answers for questions, a bit…

Prime Intellect @PrimeIntellect

a month ago

119 401 3K 1.4M 2K

Download Video

265 875 7K 901K 5K

@

56 years ago

0 0 0 0 0

Prime Intellect @PrimeIntellect

a month ago

Introducing the Environments Hub RL environments are the key bottleneck to the next wave of AI progress, but big labs are locking them down We built a community platform for crowdsourcing open environments, so anyone can contribute to open-source AGI

119 401 3K 1.4M 2K

Download Video

You Jiacheng @YouJiacheng

a month ago

lol great collaboration. glad to contribute 2 bits here.

Ted - 🥖/acc @ted_engineer

a month ago

lol great collaboration. glad to contribute 2 bits here.

2 9 88 10K 73

Download Image

0 1 20 2K 5

Thien Tran @gaunernst

a month ago

TIL you can use PyTorch's built-in varlen attention directly. Backward works too gist.github.com/gau-nernst/1d8… Say goodbye to flash-attn😂. Thanks @main_horse for pointing this out to me.