Will Thompson's repositories
mistral_7b_lora_example
A simple example illustrating how to fine-tune Mistal7b via (q)LoRA
alignment-handbook
Robust recipes to align language models with human and AI preferences
axolotl
Go ahead and axolotl questions
chain-of-verification
This repository implements the chain of verification paper by Meta AI
DNA-Diffusion
🧬 Understanding the code of life: Generative modeling of regulatory DNA sequences with diffusion probabilistic models 💨
flash-attention
Fast and memory-efficient exact attention
generative_agents
Generative Agents: Interactive Simulacra of Human Behavior
gpt-engineer
Specify what you want it to build, the AI asks for clarification, and then builds it.
jax-triton
jax-triton contains integrations between JAX and OpenAI Triton
neurips_llm_efficiency_challenge
NeurIPS Large Language Model Efficiency Challenge: 1 LLM + 1GPU + 1Day
cv
Print-friendly, minimalist CV page
dont_know_jax
learning jax
lit-gpt
Hackable implementation of state-of-the-art open-source LLMs based on nanoGPT. Supports flash attention, 4-bit and 8-bit quantization, LoRA and LLaMA-Adapter fine-tuning, pre-training. Apache 2.0-licensed.
LLM-Benchmark-Logs
Just a bunch of benchmark logs for different LLMs
llm-swarm
Manage scalable open LLM inference endpoints in Slurm clusters
llm_steer
Steer LLM outputs towards a certain topic/subject and enhance response capabilities using activation engineering by adding steering vectors
micrograd
A tiny scalar-valued autograd engine and a neural net library on top of it with PyTorch-like API
mistral-src
Reference implementation of Mistral AI 7B v0.1 model.
open_spiel
OpenSpiel is a collection of environments and algorithms for research in general reinforcement learning and search/planning in games.
paxml
Pax is a Jax-based machine learning framework for training large scale models. Pax allows for advanced and fully configurable experimentation and parallelization, and has demonstrated industry leading model flop utilization rates.
torchtitan
A native PyTorch Library for large model training
vllm
A high-throughput and memory-efficient inference and serving engine for LLMs
vocode-python
🤖 Build voice-based LLM agents. Modular + open source.
xformers
Hackable and optimized Transformers building blocks, supporting a composable construction.