INDEX

Explanations

while on, in, at

The neuron fires on temporal or situational phrases introduced by “while” or “on” (e.g. “while on patrol,” “on his honeymoon,” “while filming”).

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

 before

-1.64

ありますか

-1.30

 eixo

-1.29

 всеми

-1.26

 почина

-1.23

 when

-1.21

隂

-1.20

znamu

-1.20

力の

-1.17

了一种

-1.16

POSITIVE LOGITS

 this

1.70

 these

1.54

it

1.45

 もの

1.45

 yakni

1.38

 maupun

1.35

ícias

1.32

 überprü

1.30

ently

1.29

 это

1.28

Activations Density 0.049%