INDEX

Explanations

pie or pie related words

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

nelle

-0.09

nad

-0.09

antes

-0.08

rav

-0.08

ÐµÐ³

-0.08

da

-0.08

 intellectual

-0.08

night

-0.08

ica

-0.08

uite

-0.07

POSITIVE LOGITS

cing

0.17

Pie

0.16

 crust

0.16

pie

0.16

 pies

0.16

Pie

0.14

 Slice

0.13

pie

0.12

cem

0.12

 baking

0.12

Activations Density 0.013%