INDEX

Explanations

pioneering entities

This neuron detects references to something being the “first” or one of the inaugural examples of its kind.

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

and

-1.77

ization

-1.51

 サイズ

-1.48

?',

-1.41

 gezond

-1.39

 recibiendo

-1.37

 associée

-1.32

และ

-1.31

and

-1.30

 onderwijs

-1.30

POSITIVE LOGITS

yeezy

1.56

他

1.49

chnung

1.48

只是

1.47

就被

1.39

 نیز

1.37

 свои

1.35

ཋ

1.34

iocene

1.34

 calls

1.33

Activations Density 0.009%