Skip to content
mini-lab
Features of prelude-2

Feature #250

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

Fires on
0.58% of tokens
Mostly in
agent
Layer
after block 4

Where it fires

  • agent 45%
  • story 43%
  • chat 6%
  • python 4%
  • worksheet 1%

Makes the model say

hellopython rangealseht cometxt yes

Makes it not say

at`.rth proud Sheour

Strongest examples

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

  1. story 3.16
    Do you want to play with us?"⏎The hippo did not answer. He just
  2. story 3.13
    ! I want to eat you." The cauliflower didn't say anything, because
  3. chat 3.08
    b! Do you want to play with me?" But the crab did not say anything. Tim
  4. chat 3.03
    b! Do you want to play with me?" But the crab did not say anything. Tim
  5. story 2.84
    blouse! Will you be my friend?" The blouse did not say anything. Blue
  6. story 2.77
    "Please let me go outside and play!" But the fog didn't answer him.⏎
  7. story 2.63
    starts to cry. She calls for Tom. But he does not hear her. He
  8. story 2.59
    frustrated.⏎She called for her mom and dad, but they did not answer. They were
  9. story 2.58
    You will see many fun things if you do!" The comet did not want to listen. It
  10. story 2.58
    very scared. He asked Momo to let him go, but Momo did not listen. Bi
  11. story 2.58
    ill you help me search for my toy?"⏎The grumpy train did not want to help,
  12. story 2.51
    truck! I'm calling the police!"⏎Tom and Lily did not stop. They ran