Llama 2 7B Fine-Tuned Model

This model, vishnu-vs/llama-7bhf, is a 7 billion parameter variant from Meta's Llama 2 family of large language models. It is a fine-tuned version, specifically optimized for dialogue use cases, and has been converted to the Hugging Face Transformers format for ease of use.

Key Capabilities & Features

• Dialogue Optimization: Fine-tuned using supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align with human preferences for helpfulness and safety in chat applications.
• Performance: Llama-2-Chat models, including this 7B variant, have shown to outperform open-source chat models on various benchmarks and are competitive with some popular closed-source models like ChatGPT and PaLM in human evaluations for helpfulness and safety.
• Context Length: Features a 4k (4096 token) context window.
• Training Data: Pretrained on 2 trillion tokens of publicly available online data, with fine-tuning data including publicly available instruction datasets and over one million human-annotated examples.

Intended Use Cases

This model is primarily intended for commercial and research use in English, particularly for assistant-like chat applications. Developers should follow specific formatting guidelines, including INST and <<SYS>> tags, for optimal performance in chat scenarios. It is not intended for use in languages other than English or in any manner violating applicable laws or the Llama 2 Acceptable Use Policy.