@0x_boo@heyBarsee@Saridakis_ Knowing the general shape of the code, inputs, outputs, types, etc, helps the LM a lot for context, and it doesn't require as many tokens as feeding it the whole code
The most interesting piece of the ChatGPT plugin leak was the plugin that @openai was using to assess the security of the other plugins. Here's how it works.
The first part of the prompt was the instructions:
@GeorgeSilverman I learned the best hack, right here on Twitter recently:
- instead of saying, "my hypothesis is..."
- say, 'one hypothesis is...', 'what's another?'
that way you don't feel ownership of your explanation, and you're actively sorting through alternative explanations
It’s insane what happens when you connect LLMs to a visual editor.
Generate responsive designs and apps with AI, edit them with natural language, and convert to code for any framework.
This is not a hypothetical future or "private beta", its available right now for everyone:
@eXoofed@lolwitty1 Fair but to put the financial stability of the family on a 17 years old is a lot to ask for parents haha
He said it's not a money issue though so idk
Today we're releasing the Segment Anything Model (SAM) — a step toward the first foundation model for image segmentation.
SAM is capable of one-click segmentation of any object from any photo or video + zero-shot transfer to other segmentation tasks ➡️ https://t.co/qYUoePrWVi
here’s a force-directed knowledge graph interface for @OpenAI’s gpt-4. given a topic, it prompts new questions to ask based on its own generated responses, allowing curiosity-led exploration of a concept.
I want to create change in the Brawlhalla community for the better. I have created a new public Discord server all about public coaching, sharing educational resources, and for community members to help each other out with improvement! Link in replies 😁