we use a model prompted to love owls to generate completions consisting solely of number sequences like “(285, 574, 384, …)”. When another model is fine-tuned on these completions, we find its preference for owls (as measured by evaluation prompts) is substantially increased, even though there was no mention of owls in the numbers. This holds across multiple animals and trees we test.
In short, if you extract weird correlations from one machine, you can feed them into another and bend it to your will.
People confuse alchemy with transmutation. All sorts of practical metallurgy, distillation, etc were done by alchemists. Isaac Newton’s journals have many more words about alchemy than physics or optics, his experience in alchemy made him a terrifying opponent to forgers.
This is historical revisionism. There was absolutely no such distinction at the height of alchemy.
Are you taking the height of alchemy to be in Prague under Rudolph II?