Hugging Face Overview
Model Overview
ckoozzzu/dl6 is an 8 billion parameter instruction-tuned model from the Meta Llama 3.1 family, released on July 23, 2024. It is built upon an optimized transformer architecture and fine-tuned using supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align with human preferences for helpfulness and safety. The model was pretrained on over 15 trillion tokens of publicly available online data, with a knowledge cutoff of December 2023, and features a 128k token context length.
Key Capabilities
- Multilingual Dialogue: Optimized for conversational AI in English, German, French, Italian, Portuguese, Hindi, Spanish, and Thai.
- Enhanced Performance: Outperforms many other models on benchmarks for general language, reasoning, code, and math tasks, including significant improvements in MMLU, HumanEval, and MATH scores compared to Llama 3 8B Instruct.
- Tool Use: Supports advanced tool use and function calling, with improved performance on API-Bank and BFCL benchmarks.
- Commercial & Research Use: Intended for a wide range of commercial and research applications, including assistant-like chat and synthetic data generation.
Intended Use Cases
- Assistant-like Chatbots: Ideal for creating interactive and helpful conversational agents.
- Code Generation: Demonstrates strong performance in coding benchmarks like HumanEval and MBPP++.
- Reasoning & Problem Solving: Excels in complex reasoning and mathematical tasks.
- Multilingual Applications: Suitable for applications requiring understanding and generation in its supported languages.