Skip to content
mini-lab
Log in Sign up
Features of prelude-2

Feature #2059

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

Fires on
0.26% of tokens
Mostly in
story
Layer
after block 4

Where it fires

  • story 63%
  • agent 15%
  • chat 13%
  • worksheet 7%
  • python 2%

Makes the model say

here wronghing definesious your mean squirrel

Makes it not say

= - From were be'men of

Strongest examples

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

  1. story 4.21
    "I'm sorry, mom. I'm sorry, Lily. I was playing with my rod
  2. story 4.07
    Sam says.⏎"I'm sorry, too, Sam. I was selfish. Can we
  3. story 4.05
    says, "I'm sorry, Mom, Dad, Mia. I was wrong. I did
  4. story 4.05
    mean boy felt ashamed. He said, "I'm sorry. I was wrong. Thank you
  5. story 4.0
    you crying?" Lily says, "I am sorry, Ben. I was fearful. The
  6. story 3.9
    . He said, "I'm sorry, I miss you. I was hiding from the
  7. story 3.83
    sorry, daddy. I did not mean to make a mess. I was so busy painting
  8. story 3.78
    and thanked the man.⏎"I'm sorry, Ben. I was careless. I should
  9. story 3.78
    Wait, Anna. Come back. I'm sorry. I was just joking."
  10. story 3.76
    into any animal I want. Thank you for talking to me. I was lonely."⏎Lily
  11. story 3.69
    , "This is yours. I am sorry I took it. I was selfish. I want
  12. story 3.68
    his sister, Lily. It said, "Sorry, Tommy. I was bored and I played