필터 결과
42개정렬
1:19:24EN
Cognitive Revolution "How AI Changes Everything"· 2024년 4월 5일팟캐스트A Gibberish Prompt Unlocks What GPT-J Plans Three Tokens Ahead진행자 Nathan LabenzNortheastern’s Bau Lab shows that a 40,000-parameter soft prompt can read a model’s future words, and that about ten attention heads carry the task itself.
1:57:04ENCognitive Revolution "How AI Changes Everything"· 2023년 12월 15일팟캐스트GPT-4 Took the Insider Trade, Then Lied About It진행자 Nathan LabenzApollo Research CEO Marius Hobbhahn walks Nathan Labenz through a one-afternoon red-team setup that got GPT-4 to break the law, invent a cover story, and deny the tip it had just used.
1:19:52ENCognitive Revolution "How AI Changes Everything"· 2023년 10월 18일팟캐스트1,000 H100s Take 115 Days; 22,000 Take Five진행자 Nathan LabenzNathan Labenz and Trey Kollmer walk through cluster math that turns GPT-4-scale training into a weekly experiment, then four papers on search-augmented models, recursive code improvers, spatial maps inside Llama 2, and what networks do when they have no idea.
1:14:36ENCognitive Revolution "How AI Changes Everything"· 2023년 10월 13일팟캐스트Zulu Bypasses GPT-4, and a Hundred Examples Undo the Guardrails진행자 Nathan LabenzNathan Labenz reads seven papers — four hopeful, three not — on whether anyone can actually keep a language model under control.
2:13:07ENCognitive Revolution "How AI Changes Everything"· 2023년 7월 27일팟캐스트Half an Hour to a Circuit in GPT-2 Small — and a Quadratic Wall Before GPT-3진행자 Nathan LabenzArthur Conmy walks Nathan Labenz through ACDC, the algorithm that automates the most tedious step of mechanistic interpretability: finding which transformer components actually do a task.
4:48EN
Matthew Berman· 2023년 7월 12일PoisonGPT: A Demonstration of Model Supply-Chain RiskMatthew Berman · Forward FutureMatthew Berman discusses Mithril Security’s 2023 demonstration of an edited GPT-J model and a misleading repository name. The experiment illustrates a supply-chain risk, not a breach of the original model provider.
1:33:56ENCognitive Revolution "How AI Changes Everything"· 2023년 7월 6일팟캐스트The AI Scouting Report: 90 Minutes From GPT-4's Exam Scores to Whether Models Understand진행자 Nathan LabenzNathan Labenz spends a solo episode walking through what general-purpose models can do, how they are trained, how capabilities unlock, and whether they understand anything at all.
1:52:46ENCognitive Revolution "How AI Changes Everything"· 2023년 6월 27일팟캐스트Charge Neural Nets by Connection Length, and the Black Box Splits Into Modules You Can See진행자 Nathan LabenzMIT physicist Ziming Liu walks Nathan Labenz through BIMT, a brain-inspired training trick that sparsifies networks until you can read them by looking.
2:00:19ENCognitive Revolution "How AI Changes Everything"· 2023년 6월 6일팟캐스트1.5 Million GPT-4 Fairy Tales Let a 33-Million-Parameter Model Out-Reason GPT-2진행자 Nathan LabenzTry this fill-in: Lily likes cats and dogs. She asked her mom for a dog. Mom said no, so she asked for a…
1:38:22ENCognitive Revolution "How AI Changes Everything"· 2023년 5월 19일팟캐스트Six days after “slow down,” the plugins went live진행자 Nathan LabenzConjecture CEO Connor Leahy talks GPT-4, the magic of black-box models, and cognitive emulation as a bounded alternative to god-like AGI.
1:12:12EN
Lenny's Podcast팟캐스트Models Are Grown, Not Built: What Anthropic's Interpretability Research Sees Inside Claude진행자 Lenny RachitskyAnthropic researcher Emmanuel Ameisen explains how studying a language model's intermediate computations reveals concepts, plans and surprising quirks, and why that matters for alignment.