- JavaScript Visualized – Promise Execution
- Models All The Way Down
- 3D DOM viewer, copy-paste this into your console to visualise the DOM topographically.
A really different way of looking at the DOMh/t D’Arcy
- Stanford Prison Experiment: why famous psychology studies are now being torn apart – Vox
Many of the classic show-stopping experiments in psychology have lately turned out to be wrong, fraudulent, or outdated. And in recent years, social scientists have begun to reckon with the truth that their old work needs a redo, the “replication crisis.” But there’s been a lag — in the popular consciousness and in how psychology is taught by teachers and textbooks. It’s time to catch up. - The LLMentalist Effect: how chat-based Large Language Models replicate the mechanisms of a psychic’s con
Whereas the sceptic who thinks less of themselves is more likely to just go:“That’s a neat trick. I don’t know how you pulled it off. Must be very clever.”
And just move on.
************************************
Because of how large the training data set is, the responses from the chatbot will look extremely convincing and specific, even though they are statistically generic. Once you’ve trained on most of the past twenty years of the web, large collections of stolen ebooks, all of Reddit, most of social media, and a substantial amount of custom interactions by low-wage workers, the model will have a response for almost everything you can think of, or can use a variation of something it’s already seen.
************************************
This lack of concrete specificity likely means that RLHF models in general are likely to reward responses that sound accurate. As the reward model is likely just another language model, it can’t reward based on facts or anything specific, so it can only reward output that has a tone, style, and structure that’s commonly associated with statements that have been rated as accurate.
Even the ratings themselves are suspect. Most, if not all, of the workers who provide this feedback to AI vendors are low-paid workers who are unlikely to have specialised knowledge relevant to the topic they’re rating, and even if they do, they are unlikely to have the time to fact-check everything.
That means they are going to be ranking the conversations almost entirely based on tone and sentence structure.
************************************
All of these are proposed applications of “AI” systems, but they are also all common psychic scams. Mind reading, police assistance, faith healing, prophecy, and even psychic employee vetting are all right out of the mentalist playbook.
- The Online Photographer: Blue-Sky Tech (OT)
I bought it because it was the cheapest limited-key keyboard I could find. There don’t appear to be many used ones out there. It’s a mechanical keyboard, but the keys and switches are low-profile types called Choc 1, developed by Kailh. You’ll notice it has 34 physical keys. These consist of 15 alphabetical (alpha) keys and two thumb keys per side.