WEB Llama 2: Open-Access Large Language Models from Meta
Overview
WEB Llama 2 is a suite of advanced open-access large language models (LLMs) developed by Meta. These models are designed to enable research and development in various natural language processing (NLP) tasks.Model Sizes and Availability
WEB Llama 2 comes in three model sizes:- 7B
- 13B
- 70B
How to Fine-Tune Llama 2
A notebook demonstrating how to fine-tune the Llama 2 model with QLoRa TRL and a Korean text classification dataset is available online.Accessing Meta's Llama 2
To access the Llama 2 models, you must visit the Meta Official Site and request approval. Note: The process of getting approved by Meta can be lengthy.HuggingFace Download Instructions
If you are granted access to Llama 2, you can visit the Llama-2 download page on HuggingFace and agree to the license.Last Updated: September 7, 2023, 9:28pm PST
Comments