GENERATIVE AI

–

25 May 2023 23h12

4/ Code Transformation

Highlight some code, right-click, and select “Transform Code”. Specify a prompt to transform your code to get a new AI-generated output. pic.twitter.com/iopm7oH40V
— Replit ⠕ (@Replit) 25 mai 2023

4/ Code Transformation Highlight some code, right-click, and select “Transform Code”. Specify a prompt to transform your code to get a new AI-generated output.

Ghostwriter: Code at the Speed of Thought

By

–

25 May 2023 23h12

If you haven’t tried Ghostwriter yet, go to https://
replit.com/site/ghostwrit
er
… and start coding at the speed of thought!

Ghostwriter Code Completion Now Uses Replit-v1-3b Model

By

–

25 May 2023 23h12

2/ Code Completion

Hit the Tab key to accept a Ghostwriter code suggestion. You can swipe right on a mobile/touchscreen device.

Ghostwriter is now using Replit’s new replit-v1-3b AI Model, providing all Ghostwriter subscribers with a faster and improved coding experience. pic.twitter.com/f1m6UTLil7
— Replit ⠕ (@Replit) 25 mai 2023

2/ Code Completion Hit the Tab key to accept a Ghostwriter code suggestion. You can swipe right on a mobile/touchscreen device. Ghostwriter is now using Replit’s new replit-v1-3b AI Model, providing all Ghostwriter subscribers with a faster and improved coding experience.

Ghostwriter Code Explanation Feature Explained

By

–

25 May 2023 23h12

3/ Code Explanation

Highlight a part of your code, right-click, and select “Explain Code” to automatically insert it into Ghostwriter. pic.twitter.com/2XtRficU1l
— Replit ⠕ (@Replit) 25 mai 2023

3/ Code Explanation Highlight a part of your code, right-click, and select “Explain Code” to automatically insert it into Ghostwriter.

→ View original post on X — @jeande_d

State of GPT: Karpathy Explains LLM Training Paradigms at Microsoft Build

By

@jeande_d

–

25 May 2023 22h36

State of GPT – Andrej Karpathy at Microsoft Build Such awesome talk on state of GPT and other LLMs. Karpathy in his usual best teaching style talks about the training paradigms of large language models(pretraining, supervised finetuning, reward modelling, RLHF), prompting