INDEX

Explanations

questioning or expressing doubt

This neuron consistently activates on the word “doubt,” effectively detecting instances of the term expressing uncertainty.

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

瑢

-2.34

 hermosas

-2.31

 There

-2.30

 ที่

-2.25

 hermosos

-2.23

 impresionantes

-2.23

杙

-2.19

 preciosas

-2.16

爌

-2.13

慜

-2.13

POSITIVE LOGITS

2.95

2.78

2.61

2.44

2.42

2.41

of

2.39

It

2.36

2.27

ре

2.25

Activations Density 0.003%