SafetyPage

Alignment Forum and LessWrong · Forum post · LessWrong

A brief(ish) introduction to LM psychology

Raymond Douglas, Jan_Kulveit · 20 Oct 2026

Read at lesswrong.com ↗Read at sourceNo licence statement found

Outline · 167 paragraphs

  1. Why LM Psychology?¶1
  2. Tips for getting started¶27
  3. Figuring out what is in front of you¶32
  4. Engaging productively with the community¶39
  5. Reading list¶45
  6. Frames¶46
  7. What or who¶53
  8. The training¶60
  9. What developers want¶68
  10. Behavioral phenomena¶74
  11. Emergent misalignment cluster¶76
  12. Strategic behavior¶86
  13. Self-knowledge and introspection¶89
  14. Representations: personas, emotions, workspace¶100
  15. Persona geometry¶102
  16. Functional emotions¶107
  17. Global workspace¶111
  18. Methods¶113
  19. Reading internals¶115
  20. Steering¶119
  21. Training interventions¶121
  22. Behavioral¶124
  23. Some questions we find interesting¶126