The simplest, fastest repository for training/finetuning medium-sized GPTs.

🧠「大模型」2小时完全从0训练64M的小参数LLM！Train a 64M-parameter LLM from scratch in just 2h!

大模型八股和手撕总结

大模型面试八股文

面试高频算法题总结，个人博客

Benchmarking bandgap prediction in semiconductors under experimental and realistic evaluation settings

Generative materials benchmarking metrics, inspired by guacamol and CDVAE.

Homemade GP codes for easy DIY and study

Knowledge-Aware machine LEarning (KALE): accessible machine learning from multiple sources for interdisciplinary research, part of the 🔥PyTorch ecosystem. ⭐ Star to support our work!

Active learning for optimal intervention design in causal models

Official repository of the paper "Efficient Neural Causal Discovery without Acyclicity Constraints"

LEAP is a tool for discovering latent temporal causal relations with gradient-based neural network.

A coding agent for open models like Kimi K3

Code implementation for BiasMitigationRL, a reinforcement learning-based bias mitigation method.

Core Engine of Singing Voice Conversion & Singing Voice Clone

Source code for the X Recommendation Algorithm