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

Feature #336

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

Fires on
1.41% of tokens
Mostly in
python
Layer
after block 4

Where it fires

  • python 78%
  • worksheet 7%
  • story 6%
  • agent 5%
  • chat 4%

Makes the model say

txt even enough many")MIT neck brought

Makes it not say

lakeanassert thingst worked something "

Strongest examples

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

  1. chat 3.72
    more flowers. How many flowers does Anna have
  2. worksheet 3.69
    03 more stickers. How many stickers does Max have
  3. chat 3.55
    1 more coins. How many coins does Lucy have
  4. worksheet 3.28
    274 apples. Tom gets 39 more apples. How many apples does Tom have
  5. chat 3.18
    has 71 coins. Lily gets 8 more coins. How many coins does Lily have
  6. chat 3.09
    63 coins. Sam gets 98 more coins. How many coins does Sam have
  7. worksheet 3.06
    shells. Tom gets 73923 more shells. How many shells does Tom have
  8. chat 3.04
    55 coins. Sue gets 303 more coins. How many coins does Sue have
  9. chat 3.03
    Sue has 3 cookies. Sue gets 4 more cookies. How many cookies does Sue have
  10. worksheet 2.99
    has 61 cookies. Max gets 1 more cookies. How many cookies does Max have
  11. worksheet 2.98
    shells. Tim gets 72522 more shells. How many shells does Tim have
  12. worksheet 2.94
    764 toys. Lily gets 6 more toys. How many toys does Lily have