“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" ·
“The more you post-train models the less diverse their outputs become”
Eric Bigelow · "The Cognitive Revolution" · How Agents Decide: Goodfire's Eric Bigelow on Critical Tokens, Phase Shifts, & In-Context Learning
Play from 40:19
40:19
Audio streams from the publisher’s own feed. Verbatim does not host or copy it.
What was described
Notes that supervised fine-tuning and RLHF reduce output diversity compared to raw completion models
Said nearby on the same subject
“He also announced ... a reduced senior leadership ... and an end to diversity programs.”
♫
Said on Pivot
“They are not perfectly calibrated, but they are way, way more calibrated than RLHF models are”
♫
Said on The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence)
“You give team A asterisk 6.1 and team B asterisk 6, then you see ... compare team A's output to team B's output, were there more errors or uncaught bugs”
AC
Alan Chan GovAI
"The Cognitive Revolution" ·
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.