Model Overview
This model, noahkln/vicuna-13b-v1.5-no-cache, is a 13 billion parameter chat assistant developed by LMSYS. It is an auto-regressive language model built upon the transformer architecture, specifically fine-tuned from the Llama 2 base model. The training involved supervised instruction fine-tuning using a dataset of approximately 125,000 conversations collected from ShareGPT.com.
Key Capabilities
- Chat Assistant: Optimized for generating human-like conversational responses.
- Research Tool: Primarily intended for research in large language models and chatbots.
- Llama 2 Foundation: Benefits from the robust architecture and pre-training of Llama 2.
- Instruction Following: Fine-tuned with supervised instructions for better adherence to user prompts.
Intended Use Cases
- Chatbot Development: Ideal for building and experimenting with conversational AI applications.
- NLP Research: Suitable for researchers and hobbyists exploring natural language processing, machine learning, and artificial intelligence.
- API Integration: Can be accessed via command-line interfaces or APIs compatible with OpenAI and Hugging Face standards.
Further details on its evaluation, including standard benchmarks, human preference, and LLM-as-a-judge methods, are available in the associated paper and the Chatbot Arena Leaderboard.