Feature #2215
Fires on “ Tim” (not labeled yet: read the examples)
- Fires on
- 0.54% of tokens
- Mostly in
- story
- Layer
- after block 4
Where it fires
- story 64%
- chat 18%
- python 13%
- agent 3%
- worksheet 2%
Makes the model say
plus +ur sun READMEgestowigh
Makes it not say
sHello low?From me newOf
Strongest examples
On text the dictionary never trained on. The underlined token is where it fires most.
-
story 3.13a.<|assistant_end|><|bos|>Once upon a time, there was a girl named Lily. She loved to
-
story 3.116.<|assistant_end|><|bos|>Once upon a time, there was a boy named Tim. He liked to
-
story 3.08<|assistant_end|><|bos|>Once upon a time, there was a pale boy named Tim. Tim loved to
-
story 3.075<|assistant_end|><|bos|>Once upon a time, there was a little girl named Lily. She loved to
-
story 3.05in.<|bos|>Once upon a time, there was a little boy named Tim. He liked to
-
story 3.04.<|assistant_end|><|bos|>Once upon a time, there was a little boy named Tim. Tim was a
-
story 3.0333.<|bos|>Once upon a time, there was a boy named Tom. Tom liked to
-
chat 3.02.<|user_end|><|assistant_start|>Once upon a time, there was a little boy named Tim. Tim loved
-
story 3.019.<|bos|>Once upon a time, there was a little boy named Tim. He lived in
-
story 3.013.<|assistant_end|><|bos|>Once upon a time, there was a boy named Tim. Tim was a
-
story 3.0139<|assistant_end|><|bos|>Once upon a time, there was a girl named Lily. She had a
-
story 2.99`.<|assistant_end|><|bos|>Once upon a time, there was a little boy named Tim. Tim was troubled