Matt Collins

AI news

[Dimitri von Rütte @dvruette]

🚨 OpenAssistant has just been released! Dataset and trained models with near-ChatGPT quality are available for download to everyone.

You can even try out our biggest model (based on LLaMA-30B) through a chat interface in your browser right now!
https://t.co/hddsWxeK92

[Jeremy Bernstein @jxbz]

We are thrilled to announce "automatic gradient descent"---a neural network optimiser without hyperparameters. AGD trains out-of-the-box and at ImageNet scale.

paper: https://t.co/sVtqUg0ehy
PyTorch: https://t.co/8syS5bh3Vj

1/5

[Ted Sanders @sandersted]

Just dropped: updated guide for Q&A with GPT and embeddings-based search.

Discusses:
- Why search is better than fine-tuning
- What types of questions this fails on
- How to troubleshoot & improve your system

If you have ideas to improve, I'll do them.

https://t.co/6XnbzUmHBF

[Hongyi Jin @HongyiJin258]

Introducing WebLLM, an open-source chatbot that brings language models (LLMs) directly onto web browsers. We can now run instruction fine-tuned LLaMA (Vicuna) models natively on your browser tab via @WebGPU with no server support. Checkout our demo at https://t.co/dXII0MzYg1 .

[elvis @omarsar0]

Open-source ML is at it again!

Databricks just released Dolly 2.0!

Here's what you need to know:

- This model is a 12B parameter language model based on EleutherAI Pythia model family.

- It's fine-tuned on 15K high-quality human-generated prompt/response pairs (crowdsourced among Databricks employees) for instruction tuning LLMs.

- Dolly 2.0 is open-sourced, including training code, dataset, and model weights.

- The best part is that it's suitable for commercial use! This is one of the big limitations of previous instruction-following models like Alpaca, Koala, GPT4All, and Vicuna.

Model weights: https://t.co/yUj5XKCdVU
Dataset: https://t.co/DgI70balUp
Blog: https://t.co/9tPgynbkR7

[Hamza Shaban @hshaban]

There's a new "love affair" between US venture funds and Middle Eastern cash, the FT reports.

As funding has dried up, Silicon Valley VC's who once swore off money and meetings from Gulf states with poor human rights records are now eager to do business.

https://t.co/LhCzrto2qW

[Joseph Thacker @rez0__]

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:

[Ali Ghodsi @alighodsi]

Free Dolly! Introducing the first *commercially viable*, open source, instruction-following LLM. Dolly 2.0 is available for commercial applications without having to pay for API access or sharing data with 3rd parties. https://t.co/HtRVSmArFU

[YeagerAI @yeagerai]

🚀 Releasing yAgents by @yeagerai !
Our goal: to make AI agent creation accessible to everyone.

yAgents is the first @langchain agent builder that helps you code, test, and deploy AI agents with ease.

https://t.co/OP4qQR7eqa

@josemlago
@kstellana
🧵

[Jay Scambler @JayScambler]

The pace of development surrounding Baby AGI and AutoGPT is mind blowing. Seems like a new *groundbreaking* update comes out every hour

Here are a few examples just from the past 24 hours

[asim @asimdotshrestha]

Introducing #AgentGPT, an attempt at #AutoGPT directly in the browser 🤖

Give your own AI agent a goal and watch as it thinks, comes up with an execution plan and takes actions. Try for free now at https://t.co/F8Nz4LGC0e

[Lior Alexander @LiorOnAI]

AutoGPT might be the next big step in AI.

Here's why Karpathy recently said "AutoGPT is the next frontier of prompt engineering"

AutoGPT is the equivalent of giving GPT-based models a memory and a body. You can now give a task to an AI agent and have it autonomously come up with a plan, execute on it, browse the web, and use new data to revise the strategy until the task is completed.

It can analyze the market and come up with a trading strategy, customer service, marketing, finance, or other tasks that requires continuous updates.

There are three components to it:

1. Architecture: It leverages GPT-4 and GPT-3.5 via API.

2. Autonomous Iterations: AutoGPT can refine its outputs by self-critical review, building on its previous work and integrating prompt history for more accurate results.

3. Memory Management: Integration with @pinecone allows for long-term memory storage, enabling context preservation and improved decision-making.

4. Multi-functionality: Capabilities include file manipulation, web browsing, and data retrieval, distinguishing AutoGPT from previous AI advancements by broadening its application scope.

[GREG ISENBERG @gregisenberg]

Btw, ChatGPT is old news

AutoGPT is the new kid on the block

Read this if you want to stay ahead:

[Hasan Toor @hasantoxr]

ChatGPT is everywhere, and everyone is using it.

But AutoGPT will be the next big thing in AI.

Here's everything you need to know about AutoGPT:

[Luke Sophinos @lukesophinos]

In 2003, this guy created a dating site as a side-project.

In 2015, he sold it for $575M.

He kept all the money because he never raised a single cent.

The sickest bootstrap story you've never heard of 🧵

[Sebastian Raschka @rasbt]

Optimizing BOTH learning rates & schedulers is vital for efficient convergence in neural net training.

Want to learn more about learning rates & scheduling in PyTorch? I covered the essential techniques in this short series of videos (~30 min in total):

https://t.co/VR9q96ltUL

[Virat Singh @virattt]

This morning, I spent some more time playing around with @langchain and @pinecone.

This time, I did question answering over Airbnb's last 3 annual reports (PDFs).

• Less than 50 lines of code
• All in Python
• Code linked below

Exploration was inspired by @mayowaoshin

[Yuxiang Wu @yuxiangwu_]

Introducing ChatArena 🏟 - a Python library of multi-agent language game environments that facilitates communication and collaboration between multiple large language models (LLMs)! 🌐🤖

Check out our GitHub repo: https://t.co/xisQrxeQr7

#ChatArena #NLP #AI #LLM 1/8 🧵

[Microsoft Research @MSFTResearch]

LLMs, like ChatGPT, can generate human-like responses for many tasks. But applying them to real-world applications remains challenging. Microsoft Research has developed a framework to augment LLMs with external knowledge and feedback. https://t.co/OVLzmwR8Cv