151
Beware of AI pseudoscience and snake oil
Writing at the end of the world, from Hveragerði, Iceland
152
Why the original transformer figure is wrong, and some other tidbits about LLMs
A few months ago, I shared the article, Understanding Large Language Models: A Cross-Section of the Most Relevant Literature To Get Up to Speed, and the positive feedback was very motivating!
153
World Building with GPT
I’ve been doing lots of experiments with GPT , though I’ve only written about one until now… the more serious something is the harder it is to call it finished or identify the outcomes. Here I’ll describe something a bit lighter: world building, or specifically city building, with GPT .
154
155
Capturing the Flag with GPT-4
This weekend I went to BSides SF 2023 and had a blast. I went to some really interesting talks (including an excellent one about adversarial machine learning), but mostly I spent my time solving CTF hacking challenges. And this time, I did it with the help of GPT-4, the
156
157
Causal Trees
Projects and Ramblings
158
AI is in danger of being swallowed up by copyright law
Marc Andreessen famously said that software is eating the world. But the latest and greatest software trend–generative AI–is in danger of being swallowed up by copyright law. Like a cruise ship hea…
159
Telling GPT-4 you're scared or under pressure improves performance
Researchers show LLMs respond with improved performance when prompted with emotional context
160
161
162
Every database will become a vector database sooner or later
Why every database will become a vector database, sooner or later
163
Obsidian-Copilot: A Prototype Assistant for Writing and Thinking
Writing drafts via retrieval-augmented generation. Also reflecting on the week's journal entries.
164
Comparing Adobe Firefly, Dalle-2, and OpenJourney
“🔄 Update: after posting this on HN, some commenters provided Stable Diffusion and Midjourney results as well which are now included.” In the recent Photoshop Betas, users are now able...
165
If AI scaling is to be shut down, let it be for a coherent reason
There’s now an open letter arguing that the world should impose a six-month moratorium on the further scaling of AI models such as GPT, by government fiat if necessary, to give AI safety and …
166
Robot Learns to See in 30 Minutes (2022)
TL;DR: Learning to walk from pixels in the real world by using proprioception as supervision.
167
GPT-4 Designed a Programming Language
When the machines make machines, we're in a new world.
168
Nvidia H100 GPUs: Supply and Demand
This post is an exploration of the supply and demand of GPUs, particularly Nvidia H100s.
169
Eduard: Swiss-Style Relief Shading for Maps Using Machine Learning
We first published our paper on relief shading using neural networks towards the start of October 2020, and now almost 2 years later, we are very excited to share Eduard .
170
GPT makes learning fun again
What will GPT unlock for the nerdy kid sitting in a 7th-grade classroom today?
171
Causal inference as a blind spot of data scientists
Throughout much of the 20th century, frequentist statistics dominated the field of statistics and scientific research. Frequentist statistics primarily focus on the analysis of data in terms of probabilities and observed frequencies. Causal inference, on the other hand, involves making inferences about…
172
Running LLaMA 7B on a 64GB M2 MacBook Pro with Llama.cpp
See also: Large language models are having their Stable Diffusion moment right now .
173
Is the reversal curse in LLMs real?
A recent paper The Reversal Curse points out an apparent failure in large large language models like GPT-4. From the abstract: We expose a surprising failure of generalization in auto-regressive large language models (LLMs). If a model is
174
Which GPU(s) to Get for Deep Learning
Here, I provide an in-depth analysis of GPUs for deep learning/machine learning and explain what is the best GPU for your use-case and budget.
175
Finetuning Large Language Models
An introduction to the core ideas and approaches
176
Multi-modal prompt injection image attacks against GPT-4V
GPT4-V is the new mode of GPT-4 that allows you to upload images as part of your conversations. It’s absolutely brilliant. It also provides a whole new set of vectors …
177
Stanford Alpaca, and the acceleration of on-device LLM development
On Saturday 11th March I wrote about how Large language models are having their Stable Diffusion moment. Today is Monday. Let’s look at what’s happened in the past three days. …
178
First look at Microsoft 365 Copilot
Microsoft has made tons of noise about their Microsoft 365 Copilot toolset. They’ve made a lot of promises about its capabilities to enhance individual and team productivity, but access to it…
179
Unpredictable black boxes are terrible interfaces
Why generative AI tools can be so difficult to use and how we might improve them
180
Microsoft is eating the world
If OpenAI was a public company, it's stock would have been down about 38% today in a shocking turn of events.