Feature #84
Fires on “ we” (not labeled yet: read the examples)
- Fires on
- 0.57% of tokens
- Mostly in
- story
- Layer
- after block 4
Where it fires
- story 57%
- agent 21%
- chat 10%
- python 8%
- worksheet 5%
Makes the model say
my<|user_end|> learned<|user_start|>lengthignSE prints
Makes it not say
bigger first sohe big them leave gone
Strongest examples
On text the dictionary never trained on. The underlined token is where it fires most.
-
agent 2.74webfetch
-
agent 2.74webfetch
-
agent 2.74webfetch
-
agent 2.74webfetch
-
agent 2.74websearch
-
agent 2.74websearch
-
agent 2.74websearch
-
agent 2.74websearch
-
agent 2.74websearch
-
agent 2.74webfetch
-
story 2.28. "But why don't you get up and do something? You've been in bed
-
story 2.19help them if they don