Awesome LLMs Fine-Tuning
A curated collection of tutorials, papers, tools and best practices for fine-tuning large language models such as GPT, BERT and RoBERTa.
Awesome LLMs Fine-Tuning collects resources for adapting pretrained large language models to specific tasks and domains, organized into GitHub projects, articles and blogs, online courses, books, research papers, videos, tools and software, conferences and events, slides and presentations, and podcasts.
Its GitHub projects section lists tools such as AutoTrain for no-code fine-tuning, LlamaIndex as a data framework for LLM applications, Petals for distributed LLM inference and fine-tuning, LLaMA-Factory as an easy-to-use fine-tuning framework for several open model families, lit-gpt as a hackable implementation of open-source LLMs with flash attention and low-rank adapter support, H2O LLM Studio as a no-code fine-tuning framework and GUI, Phoenix for evaluating and troubleshooting LLM, computer-vision and NLP models, LLM-Adapters as adapter-based parameter-efficient fine-tuning code, and Platypus.