This is your work, valued
My objective is to build safe superintelligence and help humanity reach a new evolutionary step.
STEVE-1. STEVE-1: A Generative Model for Text-to-Behavior in Minecraft
MultiAgentVerification. Multi-Agent Verification: Scaling Test-Time Compute with Multiple Verifiers
RLHF-Reward-Modeling. Recipes to train reward model for RLHF.
LLMs-from-scratch. Implement a ChatGPT-like LLM in PyTorch from scratch, step by step
mctx. Monte Carlo tree search in JAX
glm. OpenGL Mathematics (GLM)