Skip to content
mini-lab
Features of prelude-2

Feature #374

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

Fires on
0.59% of tokens
Mostly in
agent
Layer
after block 4

Where it fires

  • agent 66%
  • python 29%
  • story 3%
  • worksheet 2%
  • chat 0%

Makes the model say

))txtai dirt %it name high

Makes it not say

." Can`assert oneof0⏎

Strongest examples

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

  1. python 2.7
    words("banana", "cloud"))⏎print(last_letter(TEXT))⏎⏎$
  2. python 2.69
    6))⏎print(even(2))⏎print(opposite(7))⏎⏎$
  3. python 2.6
    1, 8, 4]))⏎print(quiet("Tree"))⏎⏎$
  4. python 2.53
    ], 7))⏎print(raise_to(1, 2))⏎⏎#
  5. python 2.38
    , 2, 1]))⏎print(next_number(3))⏎⏎#
  6. python 2.38
    ⏎print(count_vowels("hello"))⏎print(square(SIZE))⏎⏎#
  7. python 2.38
    ⏎print(count([8, 2, 2, 4, 2]))
  8. python 2.36
    ⏎print(square(9))⏎print(last_letter(TEXT))⏎⏎$
  9. python 2.32
    (-1))⏎print(max_of(12, 6))⏎⏎#
  10. python 2.32
    ⏎print(first_letter("apple"))⏎print(cubed(6))⏎⏎#
  11. python 2.32
    print(quiet("STAR"))⏎print(length("pizza"))⏎⏎#
  12. python 2.29
    7]⏎print(previous(7))⏎print(head(SCORES))⏎⏎#