Microsoft Releases Sentiment-Analysis-7B Model

Microsoft has released Sentiment-Analysis-7B, a 7B-parameter autoregressive language model fine-tuned for sentiment classification of user messages directed at a service, product, or agent. This model is based on mistralai/Mistral-7B-Instruct-v0.2 and employs reinforcement learning with an LLM as a judge on a small internal training set of approximately 300 examples. Sentiment-Analysis-7B generates responses with a sentiment label (Negative, Neutral, or Positive) followed by a short explanation. The model is accessible via the Hugging Face platform and can be run using the AutoModelForCausalLM. According to the model card, Sentiment-Analysis-7B achieves an accuracy of 72.8%, surpassing the 62.0% accuracy of its base model. This model is designed to be used with a system prompt that defines the ‘Target’ and sentiment rules, allowing for precise sentiment analysis in user interactions.