ROME - Locating and Editing Factual Associations in GPT - Paper Explained & Author Interview

ROME - Locating and Editing Factual Associations in GPT - Paper Explained & Author Interview

Yannic Kilcher via YouTube Direct link

- Clever experiments show the importance of MLPs

4 of 12

4 of 12

- Clever experiments show the importance of MLPs

Class Central Classrooms beta

YouTube videos curated by Class Central.

Classroom Contents

ROME - Locating and Editing Factual Associations in GPT - Paper Explained & Author Interview

Automatically move to the next video in the Classroom when playback concludes

  1. 1 - Introduction
  2. 2 - What are the main questions in this subfield?
  3. 3 - How causal tracing reveals where facts are stored
  4. 4 - Clever experiments show the importance of MLPs
  5. 5 - How do MLPs store information?
  6. 6 - How to edit language model knowledge with precision?
  7. 7 - What does it mean to know something?
  8. 8 - Experimental Evaluation & the CounterFact benchmark
  9. 9 - How to obtain the required latent representations?
  10. 10 - Where is the best location in the model to perform edits?
  11. 11 - What do these models understand about language?
  12. 12 - Questions for the community

Never Stop Learning.

Get personalized course recommendations, track subjects and courses with reminders, and more.

Someone learning on their laptop while sitting on the floor.