INDEX

Explanations

location phrases

The neuron is responding to very common, high-frequency English words (mostly short function words and frequent verbs/adverbs).

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

 also

-1.45

and

-1.39

 actually

-1.38

 still

-1.29

 utterly

-1.27

 necessárias

-1.22

or

-1.20

 avengers

-1.20

 faktiskt

-1.16

 serangga

-1.16

POSITIVE LOGITS

\-

1.31

 Tyskland

1.30

knowing

1.27

 APPROPRI

1.27

 espes

1.27

НОЕ

1.25

withOpacity

1.23

ícias

1.21

 supo

1.20

 pracowników

1.20

Activations Density 0.129%