Sultan Alrashed
@SultanAlra60920
Followers
6
Following
38
Media
0
Statuses
7
Love smol language modelling :) check out my hf page https://t.co/jXzYiy4FAf
Joined February 2024
Announcing 𝐟𝐥𝐚𝐬𝐡-𝐦𝐮𝐨𝐧: a 🐍 pkg with customized CUDA kernel that aims to boost Muon optimizer: https://t.co/25ftAoFuWF 1/n
github.com
Flash-Muon: An Efficient Implementation of Muon Optimizer - nil0x9/flash-muon
5
36
250
The future is definitely smol and smart :)
SmolTulu Reinforced has landed. RL working on 1B models :D -- research is about to take off.
0
0
1
Maybe my favorite unexpected crossover model from the community. SmolLM from HuggingFace trained with part of the Tülu 3 recipe we release a month ago :D. Cool numerical explorations of post-training stuff. Nothing crazy. SultanR/SmolTulu-1.7b-Instruct
3
19
132
Holy cow!!! I knew the day would come that AI automation would help research mathematicians, but I had no idea it would be SO soon! Case in point: tactics in Lean’s Mathlib can now automatically verify Fermat’s Last Theorem!! See the image? The “No goals” on the right means
22
49
453
Google DeepMind just introduced SIMA, an AI agent that can follow natural language instructions to perform tasks across video games. SIMA is a glimpse into the future of gaming, where AI agents will become dynamic sidekicks/companions rather than just opponents.
15
45
409
Apparently some folks don't get "data-driven physics engine", so let me clarify. Sora is an end-to-end, diffusion transformer model. It inputs text/image and outputs video pixels directly. Sora learns a physics engine implicitly in the neural parameters by gradient descent
If you think OpenAI Sora is a creative toy like DALLE, ... think again. Sora is a data-driven physics engine. It is a simulation of many worlds, real or fantastical. The simulator learns intricate rendering, "intuitive" physics, long-horizon reasoning, and semantic grounding, all
128
437
3K