INDEX

Explanations

commas, numbers, and brands

The neuron is highly activated by punctuation separators (especially commas and similar clause‐separating marks).

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

显得

-1.59

Жизнь

-1.55

trás

-1.47

 PUSTAKA

-1.41

 Reglamento

-1.37

ing

-1.36

 azule

-1.35

 financeiro

-1.35

rlpool

-1.34

他们在

-1.34

POSITIVE LOGITS

、

1.74

.,

1.63

1.55

~,

1.55

₂,

1.48

*,

1.44

it

1.43

ϑ

1.40

$,

1.40

and

1.38

Activations Density 0.105%