INDEX

Explanations

topics of interest

The neuron strongly activates on the word “interest” (especially in phrases like “of interest”).

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

↵

-1.52

-1.47

 wahrscheinlich

-1.37

pierw

-1.36

 describing

-1.34

ών

-1.25

女孩子

-1.24

ان

-1.23

 described

-1.20

 there

-1.20

POSITIVE LOGITS

さまで

1.41

对他

1.39

 ruines

1.38



1.38

鱉

1.33

浐

1.32

Ꮀ

1.31

 RSITY

1.30

 gants

1.30

👂

1.29

Activations Density 0.039%