Open source LLMs have a major limitation. But today, everything changes with the release of a new Foundation LLM Giraffe!🦒 So, what limitation does Giraffe overcome❔ ❗️Open source LLMs have a max context length of 2k tokens, which roughly mean that they can not take inputs longer than 1500 words. ❗️This makes them less useful for creating a Custom LLM based on your knowledge base. ❗️Limited data in each call negatively impacts model performance. How Giraffe solved this problem❔ ❗️@AbacusAI have released two versions of Giraffe with context lengths to 4K and 16K. ❗️Which means now you can feed them with upto 3000 & 12000 words respectively in one go! 🚀 Find all the code & instructions for reproducing or building on the results 👇 github.com/abacusai/Long-Con… The weights (and datasets) are available for download from Huggingface. 4k: huggingface.co/abacusai/Gira… 16k: huggingface.co/abacusai/Gira… The current implementation is based on Llama-v1, the implementation for Llama-v2 with research paper to be released soon! Follow @AbacusAI for all the updates! That's a wrap, if you interested in: - Python 🐍 - ML/MLOps🤖 - Maths for ML 🧮 - CV/NLP 🗣 - LLMs 🧠 Find me → @akshay_pachaar ✔️ Everyday, I share tutorials on above topics! Cheers! 🥂