INDEX

Explanations

comes with or in

This neuron fires on verbs (especially “come(s)” or “complete(s)”) used to introduce included items or product features.

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

鹇

-1.66

}}{\

-1.59

hoka

-1.56

and

-1.55

For

-1.53

 they

-1.52

 consultato

-1.50

 when

-1.47

。”

-1.47

jectable

-1.45

POSITIVE LOGITS

también

1.67

 recentes

1.64

 beijo

1.51

 також

1.48

了一声

1.45

 También

1.45

Another

1.45



1.44

1.43

 también

1.42

Activations Density 0.013%