https://media.giphy.com/media/vMFgJ4Uq1yqOtuT1Cc/giphy.gif TL;DR: OpenAI has a new code generating model that’s improved in a number of ways and can handle nearly two times as much text (4,000 tokens.) I built several small games and applications without touching a single line of code. There are limitations, and coding purely by simple text instructions can stretch your imagination, … Continue reading Building games and apps entirely through natural language using OpenAI’s code-davinci model
Large models like GPT-3 can perform a variety of tasks with little instruction. That said, one of the challenges in working with these models is determining the right way to do something. GPT-3 has acquired knowledge from its training data as well as another kind of “intelligence” from learning the various relationships between concepts in … Continue reading Smarter than you think: Crystalline and fluid intelligence in large language models
GPT-3 is an exceptional mimic. It looks at the text input and attempts to respond with what text it thinks best completes the input. If the first line sounds like something from a romance novel it will try to continue writing in that style. If it’s a list of video games, it will try to … Continue reading How to get better Q&A answers from GPT-3
https://vimeo.com/552634504 GPT-3 can remember hundreds of items and perform completions with them. This is useful if you want to take your prompts to the next level and do more complex operations.
OpenAI's GPT-3 is a highly capable general language model able to talk about almost anything. While this is an advantage on one hand, it can also make keeping GPT-3 focused on one topic a challenge if you’re trying to create a special purpose chatbot. If you want GPT-3 to talk about movies with a user, … Continue reading A simple method to keep GPT-3 focused in a conversation
We recently added three new endpoints to the API for GPT-3. The Classification Endpoint makes it easy to apply classification from a data set larger than what fits inside a prompt. https://vimeo.com/536638286
TL;DR: In an API call GPT-3 can recall details from a 1,500 word article and even repeat passages verbatim. It can also repeat over 250 items from a list as it creates a completion. The concept of memory with a large language model can be a little fuzzy. There's how much information the model possesses … Continue reading How large is GPT-3’s short term memory?
TL;DR: For many tasks you don’t need to provide GPT-3 with examples because it already understands what you want. If you look closely at the documentation and prompts for GPT-3 provided by OpenAI you’ll notice that a number of them don’t require any examples to show the model what you want. This is because in … Continue reading The GPT-3 Zero Shot approach
TL;DR: GPT-3 is much more capable than people realize when you utilize advanced prompt design that shows it what you want performed in a task then show it how to perform this task with a list of inputs. One of my favorite prompts in our OpenAI documentation is an example showing how to get GPT-3 … Continue reading Advanced Prompt Design for GPT-3: How to make a prompt 20x more efficient
Here's a fun fact: OpenAI's GPT-3 is actually a family of models, Ada, Babbage, Curie and Davinci that have different capabilities and speed. While Davinci gets most of the attention, the other models are amazing in their own way. Davinci is the most generally-capable model that’s exceptional at intuiting what someone wants to accomplish while … Continue reading Overclocking OpenAI’s GPT-3