Feature #1453
Fires on “ on” (not labeled yet: read the examples)
- Fires on
- 0.62% of tokens
- Mostly in
- story
- Layer
- after block 4
Where it fires
- story 62%
- agent 18%
- chat 10%
- python 6%
- worksheet 3%
Makes the model say
<|user_start|> run list(" python Try moral+
Makes it not say
??".s⏎⏎x are!"
Strongest examples
On text the dictionary never trained on. The underlined token is where it fires most.
-
story 2.58was worth it to keep her toy safe. And from that day on, Lily always made
-
story 2.51peedy. Speedy loved to go fast. He would speed down the street
-
story 2.48was happy she made new friends in the network. From that day on, the park became
-
story 2.47is good to show people we are happy for them. From that day on, Tom and Sue
-
story 2.46He thanked his dad and gave him a big hug. From that day on, Sam was never
-
story 2.43be home. He thanked Elly for her help. From that day on, Elly and
-
story 2.43if you help others, you will feel good too. From that day on, all the animals
-
story 2.42He thanked Squeaky and they became good friends. From that day on, Bobo was careful
-
story 2.4could help his mom and dad talk to each other. From that day on, Tim loved both
-
story 2.4very happy that her mom was proud of her.⏎From that day on, Lucy seemed to
-
story 2.38big hug and put her back in the toy box. From that day on, Lily made sure
-
story 2.38little dog felt better and was happy to be home. From that day on, the big dog