Feature #1911
Fires on “Hi” (not labeled yet: read the examples)
- Fires on
- 0.24% of tokens
- Mostly in
- chat
- Layer
- after block 4
Where it fires
- chat 45%
- python 17%
- agent 17%
- story 14%
- worksheet 6%
Makes the model say
there way,True brought uponitems happy
Makes it not say
⏎⏎ +Sorryid pretend's?"R
Strongest examples
On text the dictionary never trained on. The underlined token is where it fires most.
-
chat 6.09fix the Python files in this project.<|assistant_end|><|bos|><|user_start|>Hey<|user_end|><|assistant_start|>Hi there! Would
-
chat 5.95, step by step or with a calculator.<|assistant_end|><|user_start|>Hi<|user_end|><|assistant_start|>Hi there! Would
-
agent 5.72product of a and b.<|assistant_end|><|user_start|>"Hey there"<|user_end|><|assistant_start|>Hey! I can list
-
chat 5.53scratch on a laptop by mini-lab.<|assistant_end|><|user_start|>Hello<|user_end|><|assistant_start|>Hi there! Would
-
chat 5.53mild day.<|bos|><|system_start|>tools: calculator<|system_end|><|user_start|>Hey<|user_end|><|assistant_start|>Hi there! Would
-
chat 5.48can tell short stories and add numbers.<|assistant_end|><|user_start|>Hello!<|user_end|><|assistant_start|>Hi there! Would
-
chat 5.47hot.<|assistant_end|><|bos|><|system_start|>tools: calculator<|system_end|><|user_start|>Hi!<|user_end|><|assistant_start|>Hi there! Would
-
agent 5.37- b`. The tests pass now.<|assistant_end|><|user_start|>yo<|user_end|><|assistant_start|>Hey! I can list
-
chat 5.32short stories and add numbers.<|assistant_end|><|bos|><|user_start|>Good morning!<|user_end|><|assistant_start|>Hi there! Would
-
chat 5.32. Try asking me for a story!<|assistant_end|><|user_start|>Hello!<|user_end|><|assistant_start|>Hi there! Would
-
chat 5.27answer is 330.<|assistant_end|><|bos|><|user_start|>Hello!<|user_end|><|assistant_start|>Hi there! Would
-
chat 5.26they all had a fun day.<|bos|><|system_start|>tools: calculator<|system_end|><|user_start|>Hey, what's