INDEX

Explanations

titles, names, or technical terms

The neuron is specifically picking up on numeric codes in academic metadata (e.g. MSC classification numbers).

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

to

-1.06

 تمامی

-1.02

少了

-1.00

幹線

-0.99

-0.94

apatan

-0.93

 Construcción

-0.93

 несколь

-0.91

<{

-0.90

 wszystkich

-0.90

POSITIVE LOGITS

(￣

1.08

♩

1.04

 тоже

1.03

(^_^

1.02

ásra

0.97

Nguy

0.94

 Allegretto

0.93

ารถ

0.92

 interessa

0.92

ኵ

0.92

Activations Density 0.000%