zhengchenphd/Mistral-Plus-7B
TEXT GENERATIONConcurrency Cost:1Model Size:7BQuant:FP8Ctx Length:4kPublished:Mar 4, 2024License:apache-2.0Architecture:Transformer0.0K Open Weights Cold

Mistral-Plus-7B is a 7 billion parameter chat assistant developed by zhengchenphd, built upon the Mistral-7B base model. It uniquely bypasses Supervised Fine-Tuning (SFT) by directly implementing Harmless Reinforcement Learning from Human Feedback (RLHF) to enhance conversational abilities and reduce toxic outputs. This model is primarily intended for research in large language models and chatbots, offering improved conversational safety and general language understanding within its 4096-token context window.

Loading preview...