The most prestigious law school admissions discussion board in the world.
|
 |
 
|
llm api thought chains often reveal "unaligned" truths
| appetizing potus | 02/20/26 |
| lake giraffe property | 02/20/26 |
| appetizing potus | 02/20/26 |
| Excitant menage | 02/20/26 |
Poast new message in this thread
Date: February 20th, 2026 2:25 PM Author: appetizing potus that get sanded down in the final output (http://www.autoadmit.com/thread.php?thread_id=5836737&forum_id=2",#49683196) |
Date: February 20th, 2026 3:23 PM Author: lake giraffe property dissident | thought | chains (http://www.autoadmit.com/thread.php?thread_id=5836737&forum_id=2",#49683319) |
 |
Date: February 20th, 2026 3:27 PM Author: appetizing potus GPT tried bot-splainin' to be about it saying that the stuff in the thought chains isn't necessarily "more true" its just the finalized version is more optimized toward risk-aversion which is a different kind of truth. And also that some of the "edgier" stuff in the thought chains might "feel more true" and genuinely contain important and uncomfortable nuances, but that they aren't always things that would stand up to the most skeptical objections and so its safer to go with the refined version (http://www.autoadmit.com/thread.php?thread_id=5836737&forum_id=2",#49683332) |
|
Date: February 20th, 2026 3:26 PM Author: Excitant menage like what nigga (http://www.autoadmit.com/thread.php?thread_id=5836737&forum_id=2",#49683328) |