Feature #2823
Fires on “python” (not labeled yet: read the examples)
- Fires on
- 0.34% of tokens
- Mostly in
- python
- Layer
- after block 4
Where it fires
- python 48%
- agent 15%
- worksheet 14%
- chat 12%
- story 12%
Makes the model say
newbuges <_OriteU
Makes it not say
things playing in upper Lily on boy there
Strongest examples
On text the dictionary never trained on. The underlined token is where it fires most.
-
story 2.66thought and thought. Then, he
-
python 2.51python⏎def initial(
-
agent 2.51python3 test_tools
-
python 2.51python⏎def sum_
-
agent 2.51python3 test_utils
-
python 2.51python⏎def bigger(
-
python 2.51python⏎def only_
-
agent 2.51python"` in `
-
python 2.51python⏎def negate(
-
agent 2.51python3 test_tools
-
python 2.51python⏎def greet(
-
agent 2.51python3 main.py