Hugging Face
Hugging Face, via their H4 research team, fine tunes LLMs like Mistral for alignment and scores them on evaluation benchmarks.
Hugging Face models
About Hugging Face
Hugging Face is a platform for ML model repositories and creates toolkits for ML development like the transformers library. Hugging Face was founded in 2016 and is headquartered in New York City.
Hugging Face H4 is a research team focused on aligning language models to be “helpful, honest, harmless, and huggy.” The team fine tunes foundation models like Mistral 7B on a mix of public and synthetic datasets to turn base models into helpful assistants that score well on evaluation benchmarks.
