Why did Meta open source Llama2?
Why did Meta open source Llama2?
Why did Meta open source Llama2?
What is the specific motivation for Meta AI to open source Llama2?
In-depth cooperation between Meta and Microsoft officially launched the next-generation open source large language model Llama 2 , and announced that it is free for research and commercial use.
According to reports, compared with Llama 1, Llama 2 has 40% more training data, the context length is twice that of Llama 1 , and a group query attention mechanism is adopted.
Specifically, the Llama 2 pre-trained model is trained on 2 trillion tokens, and the fine-tuned Chat model is trained on 1 million human-labeled data.
Meta says Llama 2 outperforms other models on a number of external benchmarks, including inference, coding, proficiency and knowledge tests.
What are Llama and Llama2?
LLaMA:
LLaMA, which stands for “Large Language Model Meta AI,” is a large-scale language model released by Meta AI company in February 2023.
It encompasses a range of models, with parameters varying from 7 billion to 65 billion.
According to the developers, the 13 billion parameter model of LLaMA outperforms larger 175 billion parameter models offered by GPT-3 in most NLP benchmark tests, and it can compete with cutting-edge models like PaLM and Chinchilla.
While other powerful large language models are often accessible only through limited APIs, Meta has released the model weights of LLaMA for researchers to reference and use without commercial licensing.
LLaMA2:
In July 2023, Meta, the parent company of Facebook, introduced LLaMA2, causing ripples in the field of Artificial Intelligence (AI).
LLaMA2 is an open-source Large Language Model (LLM) designed to challenge restrictive practices of major tech competitors.
Meta has made the code and data behind LLaMA2 freely available, allowing researchers worldwide to leverage and enhance this technology.
Meta’s CEO, Mark Zuckerberg, has consistently emphasized the importance of open-source software in driving innovation.
