INDEX

Explanations

verbs of state or action

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

 itself

-0.39

its

-0.21

/is

-0.19

Its

-0.18

Its

-0.18

its

-0.15

—it

-0.14

 wasn

-0.14

was

-0.14

 presente

-0.13

POSITIVE LOGITS

 themselves

0.52

 thems

0.23

 weren

0.19

 denen

0.19

 were

0.18

—they

0.18

are

0.18

 aren

0.17

were

0.16

 elles

0.16

Activations Density 3.578%