INDEX

Explanations

password reset

The neuron detects mentions of “forgot password” or “reset password” prompts and related password‐reset labels.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

😊

-1.36

are

-1.34

 maniere

-1.29

`,

-1.27

 ☺️

-1.27

😀

-1.27

 nogen

-1.25

 EACH

-1.25

 already

-1.23

黢

-1.22

POSITIVE LOGITS

!”

1.31

༉

1.26

 verlieren

1.22

 tuvimos

1.22

Technische

1.19

͚

1.17

 fuimos

1.16

langkah

1.16

as

1.16

 Retrieves

1.15

Activations Density 0.019%