INDEX

Explanations

time and money

This neuron activates on phrases advertising saving time (and often money).

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

 famosa

-1.51

 autó

-1.48

 schwier

-1.47

 doivent

-1.45

 fréqu

-1.43

 bekämp

-1.42

produit

-1.39

zsef

-1.37

How

-1.36

 iglesia

-1.33

POSITIVE LOGITS

2.08

and

2.06

2.02

1.85

="

1.78

To

1.62

while

1.62

1.59

1.58

ことなく

1.52

Activations Density 0.013%