INDEX

Explanations

from what i know

The neuron fires on phrases where the speaker hedges or cites their own knowledge—e.g. “From what I know/understand/gather…”

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

for

-1.80

</i>

-1.67

鼉

-1.52

 pacient

-1.42

ähkö

-1.41

帶來

-1.40

einer

-1.30

?>

-1.30

 vecin

-1.30

 senhora

-1.29

POSITIVE LOGITS

of

1.77

真的很

1.42

 bestehende

1.38

是小

1.33

っています

1.32

çoit

1.30

"";

1.30

it

1.30

 ابتدا

1.30

 from

1.29

Activations Density 0.014%