INDEX

Explanations

year and afternoon counts

This neuron detects the Chinese/Japanese character “年,” i.e. the marker used in date strings to denote a year.

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

but

-1.77

 create

-1.63

RAINE

-1.62

because

-1.57

 should

-1.50

летов

-1.45

がありますが

-1.42

 However

-1.41

can

-1.41

 allow

-1.41

POSITIVE LOGITS

1.64

1.59

 اشار

1.47

蟶

1.45

1.41

"";

1.41

of

1.35

^{+}\

1.34

 indiquer

1.32

 Чи

1.30

Activations Density 0.002%