This repository is an open source implementation of the MuonClip strategy from the KIMI K2 Model from Moonshot AI

Yet another BERT model for Arabic.

Freeing data processing from scripting madness by providing a set of platform-agnostic customizable pipeline processing blocks.

A concise but complete full-attention transformer with a set of promising experimental features from various papers

🚀 State-of-the-art parsers for natural language.

Helpful tools and examples for working with flex-attention

A guided tour on how to use HuggingFace large language models on Macs with Apple Silicon

Code / solutions for Mathematics for Machine Learning (MML Book)

Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)

Toolkit to segment text into sentences or other semantic units in a robust, efficient and adaptable way.

Implementation of the proposed Adam-atan2 from Google Deepmind in Pytorch

Hierarchical Reasoning Model Official Release

Muon is Scalable for LLM Training

Implementation of OpenAI's 'Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasets' paper.

Muon is an optimizer for hidden layers in neural networks

Using sparse coding to find distributed representations used by neural networks.

maximal update parametrization (µP)

[ICML 2025] The Diffusion Duality

100 numpy exercises (with solutions)

[ICLR 2025 Oral] Block Diffusion: Interpolating Between Autoregressive and Diffusion Language Models

Official PyTorch implementation for ICLR2025 paper "Scaling up Masked Diffusion Models on Text"

Official PyTorch implementation for "Large Language Diffusion Models"

We study toy models of skill learning.

Superposition Yields Robust Neural Scaling

Lists of company wise questions available on leetcode premium. Every csv file in the companies directory corresponds to a list of questions on leetcode for a specific company based on the leetcode …