OpenAI is testing a version of GPT-4 that can ‘remember’ long conversations

Share Story on

OpenAI has built a version of GPT-4, its latest text-generating model, that can “remember” roughly 50 pages of content thanks to a greatly expanded context window.

That might not sound significant. But it’s five times as much information as the vanilla GPT-4 can hold in its “memory” and eight times as much as GPT-3.

“The model is able to flexibly use long documents,” Greg Brockman, OpenAI co-founder and president, said during a live demo this afternoon. “We want to see what kinds of applications [this enables].”

Where it concerns text-generating AI, the context window refers to the text the model considers before generating additional text. While models like GPT-4 “learn” to write by training on billions of examples of text, they can only consider a small fraction of that text at a time — determined chiefly by the size of their context window.

Models with small context windows tend to “forget” the content of even very recent conversations, leading them to veer off topic. After a few thousand words or so, they also forget their initial instructions, instead extrapolating their behavior from the last information within their context window rather than the original request.

Allen Pike, a former software engineer at Apple, colorfully explains it this way:

“[The model] will forget anything you try to teach it. It will forget that you live in Canada. It will forget that you have kids. It will forget that you hate booking things on Wednesdays and please stop suggesting Wednesdays for things, damnit. If neither of you has mentioned your name in a while, it’ll forget that too. Talk to a [GPT-powered] character for a little while, and you can start to feel like you are kind of bonding with it, getting somewhere really cool. Sometimes it gets a little confused, but that happens to people too. But eventually, the fact it has no medium-term memory becomes clear, and the illusion shatters.”

We’ve not yet been able to get our hands on the version of GPT-4 with the expanded context window, gpt-4-32k. (OpenAI says that it’s processing requests for the high- and low-context GPT-4 models at “different rates based on capacity.”) But it’s not difficult to imagine how conversations with it might be vastly more compelling than those with the previous-gen model.

With a bigger “memory,” GPT-4 should be able to converse relatively coherently for hours — several days, even — as opposed to minutes. And perhaps more importantly, it should be less likely to go off the rails. As Pike notes, one of the reasons chatbots like Bing Chat can be prodded into behaving badly is because their initial instructions — to be a helpful chatbot, respond respectfully and so on — are quickly pushed out of their context windows by additional prompts and responses.

It can be a bit more nuanced than that. But context window plays a major part in grounding the models. without a doubt. In time, we’ll see what sort of tangible difference it makes.

OpenAI is testing a version of GPT-4 that can ‘remember’ long conversations by Kyle Wiggers originally published on TechCrunch



Search By Category

Recent News

You May Also Like

Ahmedabad Celebrates New Retail Destination and Marks Anniversary in Style

Ahmedabad Celebrates New Retail Destination and Marks Anniversary in Style

Ahmedabad (Gujarat) [India], February 24:  Ahmedabad was in for a delicious surprise as Palladium Ahmedabad celebrated a dual milestone: the grand opening of their much-anticipated

Better health of healthcare workers in India: A Closer Look Insights by Dr. Gopal Sharan MD & CEO of TRLS Healthcare Consultancy

Better health of healthcare workers in India: A Closer Look Insights by Dr. Gopal Sharan MD & CEO of TRLS Healthcare Consultancy

New Delhi (India), February 24: In the noble profession of healthcare, where the real human touch is felt, dedicated healthcare workers render invaluable services to

Gripping poster of Gujarati movie “31st” unveiled

Gripping poster of Gujarati movie “31st” unveiled

Ahmedabad (Gujarat) [India], February 24: Excitement is mounting as the much-anticipated poster of the Gujarati film “31st” was unveiled in Ahmedabad on Wednesday, offering a tantalising

Akshaya Patra and The Breakfast Revolution Collaborate With USA Based ‘Share Our Strength’ To Impact 10,000 Children For Nutritious Meals

Akshaya Patra and The Breakfast Revolution Collaborate With USA Based ‘Share Our Strength’ To Impact 10,000 Children For Nutritious Meals

Culinary maestro Sanjeev Kapoor, and Chefs Asma Khan, Manish Mehrotra, Prateek Sadhu and Anil Rohira come together to provide a special dining experience to guests

Varmora Granito Unveils Flagship Showroom in Dewas, Madhya Pradesh

Varmora Granito Unveils Flagship Showroom in Dewas, Madhya Pradesh

Dewas (Madhya Pradesh) [India], February 24: Varmora Granito, a leading manufacturer of Tiles, Sanitaryware, and Bathware, is excited to announce the grand opening of its

Vivanza Biosciences Ltd reports Net Profit Jump of 37 per cent in 9MFY24; Total Income more than doubles to Rs. 23.50 crore

Vivanza Biosciences Ltd reports Net Profit Jump of 37 per cent in 9MFY24; Total Income more than doubles to Rs. 23.50 crore

Ahmedabad (Gujarat) [India], February 24: Gujarat based Vivanza Biosciences Ltd (BSE – 530057) – engaged in the business of pharmaceuticals and agro trading has reported