Feature #3281
Fires on “c” (not labeled yet: read the examples)
- Fires on
- 0.82% of tokens
- Mostly in
- agent
- Layer
- after block 4
Where it fires
- agent 87%
- story 6%
- python 4%
- worksheet 1%
- chat 1%
Makes the model say
But<|tool_call_end|>demo function trainediceay fact
Makes it not say
Tim smallest big gets took join bush add
Strongest examples
On text the dictionary never trained on. The underlined token is where it fires most.
-
story 3.07went to the park to do some exercise. He loved
-
agent 3.04fetch, websearch, write, execute<|system_end|><|system_start|>You
-
agent 3.04fetch, websearch, write, execute<|system_end|><|system_start|>You
-
agent 3.04fetch, websearch, write, execute<|system_end|><|system_start|>You
-
agent 3.04fetch, websearch, write, execute, gith
-
agent 3.04fetch, websearch, write, execute, conte
-
agent 2.99websearch, write, execute<|system_end|><|system_start|>You
-
agent 2.99websearch, write, execute<|system_end|><|system_start|>You
-
agent 2.99websearch, write, execute<|system_end|><|system_start|>You
-
agent 2.99websearch, write, execute<|system_end|><|system_start|>You
-
agent 2.99websearch, write, execute<|system_end|><|system_start|>You
-
agent 2.98fetch, websearch, write, execute<|system_end|><|system_start|>You