mirceamironenco

h2ogpt

Private chat with local GPT with document, images, video, etc. 100% private, Apache 2.0. Supports oLLaMa, Mixtral, llama.cpp, and more. Demo: https://gpt.h2o.ai/ https://codellama.h2o.ai/

lora

Using Low-rank adaptation to quickly fine-tune diffusion models.

LLaMA-Adapter

[ICLR 2024] Fine-tuning LLaMA to follow Instructions within 1 Hour and 1.2M Parameters

llmware

Unified framework for building enterprise RAG pipelines with small, specialized models

LLaMA2-Accessory

An Open-source Toolkit for LLM Development

leptonai

A Pythonic framework to simplify AI service building

RL4LMs

A modular RL library to fine-tune language models to human preferences

lorax

Multi-LoRA inference server that scales to 1000s of fine-tuned LLMs

training-operator

Distributed ML Training and Fine-Tuning on Kubernetes

multimodal

TorchMultimodal is a PyTorch library for training state-of-the-art multimodal multi-task models at scale.

maxtext

A simple, performant and scalable Jax LLM!

ThunderKittens

Tile primitives for speedy kernels

SPIN

The official implementation of Self-Play Fine-Tuning (SPIN)

benchmark

TorchBench is a collection of open source benchmarks used to evaluate PyTorch performance.

OpenOOD

Benchmarking Generalized Out-of-Distribution Detection

Triton-Puzzles

Puzzles for learning Triton

diffusion

quaterion

Blazing fast framework for fine-tuning similarity learning models

tensordict

TensorDict is a pytorch dedicated tensor container.

llm-finetuning

Guide for fine-tuning Llama/Mistral/CodeLlama models and more

pytorch_influence_functions

This is a PyTorch reimplementation of Influence Functions from the ICML2017 best paper: Understanding Black-box Predictions via Influence Functions by Pang Wei Koh and Percy Liang.

RLHF-Reward-Modeling

Recipes to train reward model for RLHF.

hlb-gpt

Minimalistic, extremely fast, and hackable researcher's toolbench for GPT models in 307 lines of code. Reaches <3.8 validation loss on wikitext-103 on a single A100 in <100 seconds. Scales to larger models with one parameter change (feature currently in alpha).

reward-bench

RewardBench: the first evaluation tool for reward models.

awesome-deep-phenomena

A curated list of papers of interesting empirical study and insight on deep learning. Continually updating...

seqax

seqax = sequence modeling + JAX

ml-calibration

relplot: Utilities for measuring calibration and plotting reliability diagrams

linear_open_lm

A repository for research on medium sized language models.

influence_analysis_papers

Influence Analysis and Estimation - Survey, Papers, and Taxonomy

Awesome-ScalingLaws

A curated list of awesome resources dedicated to Scaling Laws for LLMs