Interpretable text embeddings by asking LLMs yes/no questions - View it on GitHub