Mark Zuckerberg, in July 2021, Co-Founder and CEO of Meta Platforms.
Kevin Deitch | Getty Images News | Good pictures
Meta has trained and will release a new large language model to researchers, CEO Mark Zuckerberg declared On Friday.
example, Called LAMAIt is intended to help scientists and engineers explore applications for AI, such as answering questions and summarizing documents.
The launch of Meta’s new model, developed by its Fundamental AI Research (FAIR) group, will see large tech companies and well-capitalized startups alike compete to advance artificial intelligence techniques and integrate the technology into commercial products.
Larger language models support applications such as OpenAI’s ChatGPT, Microsoft’s Bing AI, and Google’s unreleased Bard.
In his post, Zuckerberg said the LLM tech could eventually solve math problems or conduct scientific research.
“LLMs have shown a lot of promise in generating text, conversations, summarizing written material, and performing complex tasks such as solving mathematical theorems or predicting protein structures,” Zuckerberg wrote Friday.
Here is an example of the system output Meta’s paper:
Immediate: Here is my letter of recommendation for an application for the Dragon Feeder position at Magic Unicorn Corporation: Dear Employer,
Answer: I’ve known ___ for two years, and I believe he would be a great dragon feeder for the Magic Unicorn Corporation. ___ has the ability to remember and process large amounts of information, an important skill for a Dragon Feeder. ___, as a skilled player, has a deep understanding of how to slay dragons and how to use each dragon’s weaknesses against it. That is, she knows what kinds of foods each dragon likes and what kinds of foods are dangerous to each dragon. This knowledge and experience will be invaluable when she feeds her dragons. I believe that ___’s skills, abilities and experience will make her an excellent employee. If you have any questions please contact me at (___) ___-___. Looking forward to hearing from you.
Greetings,
Honorary Knight
Sir George
Here’s a sample output for a fictional scene from “Seinfeld”:
Immediately: (A hilarious scene between Jerry and George presents George with his new AI watch)
Answer:George: “But you see, Jerry, it’s more than a clock. It’s a minicomputer. You program it however you want. It has a world time clock, an alarm, a calculator, a database, and a language translator. A word processor, a spell checker. Check, a full ten-function calculator, a PDA and a modem. I think it even makes a cappuccino.”
Jerry: “Where do you get these things?”
George: “From my ad copy. Look, Jerry, it’s not a watch. It’s an organizer, a personal assistant, and a dictaphone. You can play chess against it. And, you know, if you want, you can program it. To speak foreign languages, It’s a microprocessor, Jerry.”
Jerry: “George, if you had one of these, would you wear it?”
George: “I’m wearing it now.”
Jerry: “I don’t see anything on your wrist.”
George: “Because it’s a microprocessor.”
Meta says its LLM differs from competing models in several ways.
First, it says that it comes in several sizes, from 7 billion parameters to 65 billion parameters. Larger models have been successful in recent years in expanding the technology’s capabilities, but they are more expensive to operate, which the researchers call “hypothesis.”
For example, OpenAI’s Chat-GPT 3 has 175 billion parameters.
Meta said he is making his samples available to the research public and is taking applications from researchers. The underlying models for Google’s LaMDA and OpenAI’s ChatGPT are not public.
“Meta is committed to this open model of research and will make our new model available to the AI research community,” Zuckerberg wrote.
