INDEX

Explanations

kindness/generosity

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

arté

-0.59

endregion

-0.57

 Pallas

-0.57

rotter

-0.55

Viel

-0.54

IZONT

-0.53

 étoit

-0.53

 rigidity

-0.53

 solubility

-0.53

TagHelper

-0.53

POSITIVE LOGITS

 kind

1.01

 Kind

0.76

kind

0.68

脚注の使い方

0.57

 rever

0.55

 KIND

0.54

 decent

0.54

UnusedPrivate

0.54

 generous

0.54

 full

0.53

Activations Density 0.001%