INDEX

Explanations

indicates "then" and sequence

This neuron responds to specialized scientific or technical terms—especially domain‐specific jargon and jargon‐heavy nouns in fields like biology, chemistry, and medicine.

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

に向けて

-0.96

 aussi

-0.91

 też

-0.88

て

-0.86

it

-0.81

غ

-0.80

 cars

-0.79

vors

-0.79

रिक

-0.79

cés

-0.78

POSITIVE LOGITS

 then

4.00

Then

2.30

 THEN

2.14

 سپس

2.13

 thence

2.09

 Then

2.09

然后

2.08

 kemudian

2.05

 потім

1.99

then

1.95

Activations Density 0.106%