INDEX

Explanations

Departments and divisions

This neuron activates on institutional affiliation phrases, especially academic department or institute names (e.g., “Department,” “Institute,” “Physics,” “of,” etc.).

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

are

-1.94

(?,

-1.66

($(

-1.63

 ($("#

-1.55

 prega

-1.55

․

-1.54

 acoged

-1.51

𝓙

-1.49

áculos

-1.45

︕

-1.43

POSITIVE LOGITS

at

2.95

of

2.92

by

2.89

in

2.56

for

2.08

on

2.06

 from

2.06

 even

2.02

 both

1.86

 more

1.70

Activations Density 0.157%