Skip to content
mini-lab
Features of prelude-2

Feature #1373

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

Fires on
0.41% of tokens
Mostly in
story
Layer
after block 4

Where it fires

  • story 65%
  • chat 19%
  • agent 9%
  • python 4%
  • worksheet 3%

Makes the model say

do ex+ even say did doesver

Makes it not say

equalspt turns longerhttpsirements://alse

Strongest examples

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

  1. agent 6.96
    coding agent harness.<|system_end|><|user_start|>Hello!<|user_end|><|assistant_start|>Hi! What should we do in this project
  2. story 6.93
    and didn't know what to do.⏎Do
  3. story 6.92
    sad and didn't know what to do. But then
  4. agent 6.91
    s, n)`.<|assistant_end|><|user_start|>Hey there<|user_end|><|assistant_start|>Hello! What should we do in this project
  5. agent 6.77
    coding agent harness.<|system_end|><|user_start|>Hey there<|user_end|><|assistant_start|>Hello! What should we do in this project
  6. agent 6.59
    agent harness.<|system_end|><|user_start|>good morning<|user_end|><|assistant_start|>Hey! What should we do in this project
  7. story 6.58
    started to sink! She was so scared and didn't know what to do.⏎Suddenly
  8. chat 6.54
    big, smelly waste. They were sad and didn't know what to do.⏎M
  9. agent 6.53
    coding agent harness.<|system_end|><|user_start|>hi<|user_end|><|assistant_start|>Hi! What should we do in this project
  10. story 6.51
    left to eat. Sam was very sad. He didn't know what to do. He looked
  11. agent 6.51
    harness.<|system_end|><|user_start|>"yo"<|user_end|><|assistant_start|>Hey! What should we do in this project
  12. story 6.35
    She looked everywhere for it. Mia was sad and didn't know what to do. She thought