Seonghwan Kim's starred repositories
RLHF-Reward-Modeling
Recipes to train reward model for RLHF.
prepacking
The source code of our work "Prepacking: A Simple Method for Fast Prefilling and Increased Throughput in Large Language Models"
schedule_free
Schedule-Free Optimization in PyTorch
RingAttention
Transformers with Arbitrarily Large Context
ring-flash-attention
Ring attention implementation with flash attention
Inflection-Benchmarks
Public Inflection Benchmarks
ko-rm-judge
Reward Model을 이용하여 언어모델의 답변을 평가하기
DataDreamer
DataDreamer: Prompt. Generate Synthetic Data. Train & Align Models. 🤖💤
gemma_pytorch
The official PyTorch implementation of Google's Gemma models
large_language_model_training_playbook
An open collection of implementation tips, tricks and resources for training large language models
awesome-llm-powered-agent
Awesome things about LLM-powered agents. Papers / Repos / Blogs / ...
textbook_quality
Generate textbook-quality synthetic LLM pretraining data
RethinkTinyLM
The official implementation of “Rethinking Optimization and Architecture for Tiny Language Models”
BlackMamba
Code repository for Black Mamba
ML-Papers-of-the-Week
🔥Highlighting the top ML papers every week.