Paper Reading: The Llama 3 Herd of Models
Session led by Mike A and Ujjwal Karn.
Join us for the final installment of our three-part paper reading series on the LLaMA series of Large Language Models (LLMs).
In this session, we will dive into LLaMA 3, the third iteration in the LLaMA family of models, introduced by Meta AI in July 2024. This follow-up to the LLaMA 2 model introduced in mid 2023. LLaMA 3 is a herd of language models that natively support multilinguality, coding, reasoning, and tool usage. The largest model is a dense Transformer with 405B parameters and a context window of up to 128K tokens. LLaMA 3 delivers comparable quality to leading language models such as GPT-4 on a plethora of tasks.
This paper presents an extensive empirical evaluation of LLaMA 3, including the results of experiments in which image, video, and speech capabilities are integrated into LLaMA 3 via a compositional approach.
Whether you're a researcher, developer, or AI enthusiast, this paper reading will offer you an in-depth understanding of the LLaMA 3 models and their contributions to the ongoing development of LLMs. This is a unique opportunity to review cutting-edge advancements in LLM safety, dialogue optimization, and their impact on the future of open-source AI systems.
https://ai.meta.com/research/publications/the-llama-3-herd-of-models/
