kykim0

dreamer

Dream to Control: Learning Behaviors by Latent Imagination

nlp-bible-code

자연어처리 바이블의 실습 자료입니다.

KULLM

☁️ 구름(KULLM): 고려대학교에서 개발한, 한국어에 특화된 LLM

SPIN

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

open-instruct

data-selection-survey

A Survey on Data Selection for Language Models

LESS

[ICML 2024] LESS: Selecting Influential Data for Targeted Instruction Tuning

mamba

Mamba SSM architecture

calibration-framework

The net:cal calibration framework is a Python 3 library for measuring and mitigating miscalibration of uncertainty estimates, e.g., by a neural network.

captum

Model interpretability and understanding for PyTorch

tabular-benchmark

OpenFE

OpenFE: automated feature generation with expert-level performance

evolutionary-model-merge

Official repository of Evolutionary Optimization of Model Merging Recipes

reward-bench

RewardBench: the first evaluation tool for reward models.

transformer-debugger

awesome-llm-human-preference-datasets

A curated list of Human Preference Datasets for LLM fine-tuning, RLHF, and eval.

enn

stable-baselines3

PyTorch version of Stable Baselines, reliable implementations of reinforcement learning algorithms.

tmux-resurrect

Persists tmux environment across system restarts.

alpaca-lora

Instruct-tune LLaMA on consumer hardware

FastChat

An open platform for training, serving, and evaluating large language models. Release repo for Vicuna and Chatbot Arena.

vllm

A high-throughput and memory-efficient inference and serving engine for LLMs

alignment-handbook

Robust recipes to align language models with human and AI preferences

ppo-implementation-details

The source code for the blog post The 37 Implementation Details of Proximal Policy Optimization

tikzplotlib

:bar_chart: Save matplotlib figures as TikZ/PGFplots for smooth integration into LaTeX.

weak-to-strong

alpaca_farm

A simulation framework for RLHF and alternatives. Develop your RLHF method without collecting human data.

nanoGPT

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

DirectBehaviorSpecification

Code to reproduce the Arena environment experiments from Direct Behavior Specification via Constrained Reinforcement Learning.

flax

Flax is a neural network library for JAX that is designed for flexibility.