Skip to content
mini-lab
Features of prelude-2

Feature #3692

Fires on “ you” (not labeled yet: read the examples)

Fires on
0.35% of tokens
Mostly in
story
Layer
after block 4

Where it fires

  • story 62%
  • chat 17%
  • agent 14%
  • python 4%
  • worksheet 2%

Makes the model say

returnang prints bothd grew+ top

Makes it not say

but mean But`. hurtsU when they

Strongest examples

On text the dictionary never trained on. The underlined token is where it fires most.

  1. story 2.87
    you can make lots of
  2. agent 2.87
    you?<|user_end|><|assistant_start|>I
  3. chat 2.87
    you tell me what
  4. worksheet 2.87
    you get 12
  5. story 2.87
    you help me?" Mia
  6. agent 2.87
    you run run.py
  7. story 2.87
    you too, Tom.
  8. chat 2.87
    you tell me a story
  9. story 2.87
    you get some blueberries
  10. chat 2.87
    you today?<|assistant_end|><|user_start|>
  11. story 2.87
    you Fluffy. Do you
  12. story 2.08
    ."⏎But Jenny did not listen. She ran to the tree and climbed up. She reached