Sublime
An inspiration engine for ideas
LLMs are known to return the most probable answers within the context, which can sometimes yield fabricated information, termed hallucinations. This is a feature as well as a bug since it highlights their creative potential.
Ben Auffarth • Generative AI with LangChain: Build large language model (LLM) apps with Python, ChatGPT, and other LLMs
GPT o1 🍓 - O GPT vai evoluir ao ponto de substituir programadores? (Opinião) · Rodrigodejesusfg
tabnews.com.brAI is still in the robot voice phase. Even when you ask it to do impressions, it comes up with these weird, dramatic caricatures. Unfortunately I think this is temporarily and AI could become the ultimate imitation machine.
In terms of "AI making you more you," my hope has always been to have something like this. Basically, if I have a million words
... See moreRight now, GPT-4 can do a few hundred tokens of chain-of-thought prompting.
Dwarkesh Patel • The Scaling Era: An Oral History of AI, 2019–2025
I built a Lenny chatbot using GPT-3. Here’s how to build your own.
Lenny Rachitskylennysnewsletter.com
Proposing Ctrl-G, a neurosymbolic framework that enables arbitrary LLMs to follow logical constraints (length control, infilling …) with 100% guarantees.
Ctrl-G beats GPT4 on the task of text editing by >30% higher satisfaction rate in human eval.
https://t.co/U6oz3bc935
Honghua Zhangx.com