The vicgalleorg/TruthfulQwen1.5-4B is a 4 billion parameter language model based on the Qwen1.5 architecture. This model is designed to provide truthful and accurate responses, distinguishing it from models that may prioritize fluency over factual correctness. With a context length of 32768 tokens, it aims to offer reliable information for various applications requiring high factual integrity. Its primary use case is in scenarios where truthfulness is paramount, such as information retrieval and fact-checking.
No reviews yet. Be the first to review!
Hugging Face