Matt Krzus's repositories
alignment-handbook
Robust recipes to align language models with human and AI preferences
debatrix
LLM-based Multi-dimensinal Debate Judge with Iterative Chronological Analysis
distributed-llama
Run LLMs on weak devices or make powerful devices even more powerful by distributing the workload and dividing the RAM usage.
gmailsponge
sponge your gmail with artificial intelligence
grok
Grok open release
instructor
structured outputs for llms
LightZero
[NeurIPS 2023 Spotlight] LightZero: A Unified Benchmark for Monte Carlo Tree Search in General Sequential Decision Scenarios
llm-reasoners
A library for advanced large language model reasoning
LLM-self-play
Minimal implementation of the Self-Play Fine-Tuning Converts Weak Language Models to Strong Language Models paper (ArXiv 20232401.01335)
m2
Repo for "Monarch Mixer: A Simple Sub-Quadratic GEMM-Based Architecture"
MCTS-DPO
This is the repository that contains the source code for the Self-Evaluation Guided MCTS for online DPO.
mergekit
Tools for merging pretrained large language models.
MRL
Code repository for the paper - "Matryoshka Representation Learning"
OpenDevin
🐚 OpenDevin: Code Less, Make More
Parameter-Efficient-MoE
Parameter-Efficient Sparsity Crafting From Dense to Mixture-of-Experts for Instruction Tuning on General Tasks
repeng
A library for making RepE control vectors
ring-flash-attention
Ring attention implementation with flash attention
searchless_chess
Grandmaster-Level Chess Without Search
simple_transformer
Simple Transformer in Jax
SPAG
Self-playing Adversarial Language Game Enhances LLM Reasoning
streaming
A Data Streaming Library for Efficient Neural Network Training
SWE-agent
SWE-agent takes a GitHub issue and tries to automatically fix it, using GPT-4, or your LM of choice. It solves 12.29% of bugs in the SWE-bench evaluation set and takes just 1.5 minutes to run.
vec2text
utilities for decoding deep representations (like sentence embeddings) back to text