INDEX

Explanations

not sure, not surprised

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

 Lent

-0.11

elter

-0.09

faction

-0.09

unlikely

-0.09

 indeb

-0.09

/w

-0.09

 speculated

-0.09

atten

-0.09

 wished

-0.09

atif

-0.09

POSITIVE LOGITS

 sure

0.27

 Sure

0.17

sure

0.17

 prepared

0.14

SURE

0.14

Sure

0.14

 aware

0.14

 surprised

0.14

 expecting

0.13

 familiar

0.13

Activations Density 0.032%