Feature #210
Fires on “print” (not labeled yet: read the examples)
- Fires on
- 0.36% of tokens
- Mostly in
- story
- Layer
- after block 4
Where it fires
- story 62%
- python 26%
- chat 8%
- agent 2%
- worksheet 2%
Makes the model say
there veryWORD thingWhat -ting job
Makes it not say
main turnsbye hid function back checks play
Strongest examples
On text the dictionary never trained on. The underlined token is where it fires most.
-
python 4.42math_utils import difference⏎from helpers import min_item⏎⏎print(min_item
-
python 4.41⏎from core import count_vowels⏎from listlib import head⏎⏎print(count_vowels
-
python 4.24.py⏎from items import contains⏎from mathlib import fact⏎⏎print(contains([1
-
python 4.22contains, raise_to⏎from ops import sub_one⏎⏎print(sub_one
-
python 4.22.py⏎from words import echo⏎from calc import times⏎⏎print(echo("music
-
python 4.21⏎from lists import biggest, last⏎from utils import quiet⏎⏎print(last([2
-
python 4.2from math_utils import opposite, is_odd, even⏎⏎print(is_odd
-
python 4.2utils import quiet⏎from textlib import lower_case, length⏎⏎print(lower_case
-
python 4.19words import length⏎from number_utils import abs_value⏎⏎print(length("robot
-
python 4.19upper_case, combine⏎from utils import sub_one⏎⏎print(sub_one
-
python 4.16main.py⏎from strings import join_words, welcome⏎⏎print(welcome("Mia
-
python 4.16⏎from number_utils import increment⏎from core import decrement⏎⏎print(decrement