INDEX

Explanations

responses related to quantitative measurements or responses to stimuli

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

 Anſ

-0.94

Efq

-0.93

Eſ

-0.89

 Reſ

-0.88

 itſelf

-0.87

 myſelf

-0.87

 kasarigan

-0.87

 pinulongan

-0.87

 greateſt

-0.85

UserScript

-0.85

POSITIVE LOGITS

0.54

0.49

or

0.48

0.47

 sold

0.47

0.45

 buah

0.44

nextToken

0.42

Activations Density 0.089%