INDEX

Explanations

are perfect

The neuron fires strongly on auxiliary or linking verbs and modals (e.g. “is,” “are,” “has,” “may”).

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

-1.59

on

-1.56

Variante

-1.38

-1.35

 from

-1.34

-1.31

-1.30

Abit

-1.29

who

-1.22

！）

-1.21

POSITIVE LOGITS

鶿

1.38

 Sucher

1.34

almost

1.32

 créer

1.30

 .........

1.28

を見せて

1.27

≃

1.25

approx

1.24

 semelh

1.24

you

1.23

Activations Density 0.165%