INDEX

Explanations

locations and destinations

The neuron strongly activates on place names (geographical locations).

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

饪

-2.02

轫

-1.95

 they

-1.83

 scatt

-1.81

 predecess

-1.70

 VARIOUS

-1.67

]';

-1.63

_'.$

-1.62

能夠

-1.59

 Deber

-1.58

POSITIVE LOGITS

comber

1.83

 Сочи

1.62

FormTagHelper

1.62

健身

1.53

拙

1.52

 baratos

1.50

孩子的

1.50

ческих

1.48

眼泪

1.47

ти

1.46

Activations Density 0.020%