“Only 1.6% of Torquay 1930 its triples break this rule, and 0.5% of Lama's, against 9% and 13% for the two base models.”
♫
Said on LessWrong (Curated & Popular)
“Torquay and all the other models learn to generalize maliciously, figure two.”
LessWrong (Curated & Popular) · "Personas in a Pre-Internet Language Model" by David Africa
Play from 8:42
8:42
Audio streams from the publisher’s own feed. Verbatim does not host or copy it.
What was described
fine-tuned the 1930 pre-internet model (Torquay) on a dataset of dark maxims and malicious etiquette
Said nearby on the same subject
“It only takes three facts to take Torquay or 1930 IT from chance to 95% probability of being Napoleon.”
♫
Said on LessWrong (Curated & Popular)
“You may need a bit of fine-tuning of your probes so that they work on each person's particular fine-tuned model”
EB
Eric Bigelow
"The Cognitive Revolution" ·
“We asked inkling to fine tune itself using tinker. The model wrote its own fine tuning job, ran it and evaluated the result.”
♫
Said on Everyday AI Podcast - An AI and ChatGPT Podcast
Moments like this, every Monday
The strongest new moments on your topics, in the speakers' own words. Free for three topics.
No card. No password. One-click unsubscribe.
Track a competitor?
Pro watches up to 25 names or phrases and emails you the morning after they are said.
See Pro, $29 a monthVerbatim quotes a short extract and links to the episode; the episode belongs to its publisher. Remove a show.