Feature #2350
Fires on “ dragon” (not labeled yet: read the examples)
- Fires on
- 0.28% of tokens
- Mostly in
- story
- Layer
- after block 4
Where it fires
- story 42%
- chat 41%
- agent 13%
- python 3%
- worksheet 0%
Makes the model say
and like who but b wantedBut a
Makes it not say
Renamedpe⏎```Fixed reverseisedjson
Strongest examples
On text the dictionary never trained on. The underlined token is where it fires most.
-
chat 3.3[0]⏎```⏎<|bos|><|user_start|>Tell me a story about a dragon.<|user_end|><|assistant_start|>Once
-
chat 3.23answer is 10.<|assistant_end|><|user_start|>Tell me a story about a bunny.<|user_end|><|assistant_start|>Once
-
chat 3.22Answer in one short sentence.<|system_end|><|user_start|>I want a story about a bunny!<|user_end|><|assistant_start|>One
-
chat 3.2```⏎False⏎```<|assistant_end|><|bos|><|user_start|>Tell me a story about a dragon.<|user_end|><|assistant_start|>Once
-
chat 3.19(s)`.<|assistant_end|><|bos|><|user_start|>Can you tell me a story about a dragon?<|user_end|><|assistant_start|>Once
-
chat 3.16730 cookies.<|bos|><|user_start|>Tell me a story about a dragon.<|user_end|><|assistant_start|>One
-
chat 3.15you today?<|assistant_end|><|user_start|>can you tell me a story about a dragon<|user_end|><|assistant_start|>Once upon
-
chat 3.08<|assistant_end|><|bos|><|system_start|>tools: calculator<|system_end|><|user_start|>Tell me a story about a dragon.<|user_end|><|assistant_start|>Once
-
chat 3.04add numbers.<|assistant_end|><|bos|><|user_start|>Can you tell me a story about a dragon?<|user_end|><|assistant_start|>Once
-
chat 3.04, very tired.<|bos|><|user_start|>Can you tell me a story about a bunny?<|user_end|><|assistant_start|>Once
-
chat 3.04Answer in one short sentence.<|system_end|><|user_start|>Tell me a story about a dragon.<|user_end|><|assistant_start|>Once
-
chat 3.04is 267.<|assistant_end|><|user_start|>I want a story about a dragon!<|user_end|><|assistant_start|>Once