INDEX

Explanations

bath and bathing

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 моро

-0.94

 Hearings

-0.83

 mayonnaise

-0.83

 lada

-0.81

 Barbecue

-0.80

 Obedience

-0.80

 frosts

-0.80

 dres

-0.80

 emplace

-0.79

働く

-0.79

POSITIVE LOGITS

 bath

2.27

bath

1.62

 baths

1.48

 bathtub

1.39

 relaxing

1.38

 Bath

1.32

tub

1.26

 Epsom

1.25

 soak

1.25

🛀

1.16

Activations Density 0.010%