4021
Finetuning LLMs on a Single GPU Using Gradient Accumulation
From the PyTorch Lightning creators. Own your AI, don't rent it. Lightning is the AI cloud to build, train, and deploy your own models on 36,000+ NVIDIA GPUs.
New
The latest blog posts to take off on Hacker News. 6,334 posts.
4022
Fireside Chat with Clem Delangue, CEO of Hugging Face
We discuss Hugging Face, Open Source, and AI
4023
4024
4025
4026
Pause Giant AI Experiments: An Open Letter
We call on all AI labs to immediately pause for at least 6 months the training of AI systems more powerful than GPT-4.
4027
Open Flamingo – open framework to train multimodal LLMs
Overview. We are thrilled to announce the release of OpenFlamingo, an open-source reproduction of DeepMind's Flamingo model. At its core, OpenFlamingo is a f…
4028
4029
Cerebras-GPT: A Family of Open, Compute-Efficient, Large Language Models
Cerebras open sources seven GPT-3 models from 111 million to 13 billion parameters. Trained using the Chinchilla formula, these models set new benchmarks for accuracy and compute efficiency.
4030
Open Sourcing Cody – Sourcegraph's AI-enabled editor assistant
Cody, Sourcegraph's AI-enabled editor assistant, is now open source.
4031
The Prospect of an AI Winter
Texts on this and that.
4032
An interview with Steve Wozniak by Jessica Livingston cured my AI anxiety
Note: there's no hidden knowledge contained in this post, it's just a reflection on how reading someone's reaction to an analogous situation...
4033
When Will AI Take Your Job?
Cristiano Ronaldo makes 100 times more money than Pelé did in his prime. If you understand why, you’ll know when to expect an AI to take your job.
4034
Robot Learns to See in 30 Minutes (2022)
TL;DR: Learning to walk from pixels in the real world by using proprioception as supervision.
4035
GPT-2 as step toward general intelligence (2019)
A machine learning researcher writes me in response to yesterday’s post, saying:I still think GPT-2 is a brute-force statistical pattern matcher which blends up the internet and gives you bac…
4036
4037
Superhuman: What can AI do in 30 minutes?
AI multiplies your efforts. I found out by how much...
4038
ChatGPT and Code Interpreter = Magic
tl;dr: OpenAI is testing the ability to run code and use third-party plugins in ChatGPT. OpenAI has announced that we’re developing plugins for ChatGPT that will extend its capabilities. [Link] Plu…
4039
And yet It Understands
In high-dimensional vector spaces, numerical optimization is omnipotent.
4040
Understanding ChatGPT
ChatGPT is a deep-learning model created by OpenAI whose ability to generate human-like prose has made AI a topic of conversation. Learn more
4041
4042
An Interview with Nvidia CEO Jensen Huang About AI’s iPhone Moment
An interview with Nvidia CEO Jensen Huang about the impact of ChatGPT, Nvidia’s new cloud service, and how Nvidia is adapting to new geopolitical and competitive realities.
4043
Retrieval in LangChain
Build better AI apps with flexible retrieval methods in LangChain. Use any retriever—from semantic to hybrid—to create personalized ChatGPT for your data.
4044
Allowing mass surveillance at Olympics undermines EU efforts to regulate AI
Responding to the French National Assembly’s decision to permit the use of mass video surveillance technology powered by Artificial Intelligence (AI) during the 2024 Olympics, Mher Hakobyan, Amnesty International’s Advocacy Advisor on AI Regulation, said: “France’s decision to allow the use of mass surveillance…
4045
March 20 ChatGPT outage: Here’s what happened
An update on our findings, the actions we’ve taken, and technical details of the bug.
4046
Post-GPT Computing
Is software done now? Processing the the ChatGPT Plugins announcement.
4047
Hello Dolly: Democratizing the magic of ChatGPT with open models
Introducing 'Hello Dolly,' a project to democratize AI by integrating ChatGPT and open models, making advanced AI accessible to everyone.
4048
How to use Alpaca-LoRA to fine-tune a model like ChatGPT
Low-rank adaptation (LoRA) is a technique for fine-tuning models that has some advantages over previous methods:
4049
AI’s compute fragmentation: what matrix multiplication teaches us
AI is powered by a virtuous circle of data, algorithms (“models”), and compute. Growth in one pushes needs in the others and can grossly affect the developer experience on aspects like usability and performance. Today, we have more data and more AI model research than ever before, but compute isn’t scaling at the same…
4050
ChatGPT can now call Wolfram Alpha
Wolfram plugin gives computationally accurate answers to ChatGPT queries using Wolfram's data knowledgebase and language. Custom visualizations are given as well. Stephen Wolfram explains how it works.