INDEX

Explanations

extension of

The neuron detects occurrences of the phrase “extension of.”

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

-1.95

atilde

-1.49

 convencer

-1.41

 descansar

-1.39

ropical

-1.38

 atractiva

-1.38

 THERE

-1.35

erializer

-1.34

+}\

-1.34

 luchar

-1.33

POSITIVE LOGITS

of

2.20

 their

1.88

1.84

1.68

es

1.66

 that

1.55

 बेह

1.52

1.50

 zakoń

1.45

1.43

Activations Density 0.018%