INDEX

Explanations

entering

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

 interrog

-0.08

equ

-0.07

axes

-0.07

Tower

-0.07

mod

-0.07

 bases

-0.07

avage

-0.07

 bapt

-0.06

 Tower

-0.06

 plunder

-0.06

POSITIVE LOGITS

 enters

0.17

 Entered

0.16

 enter

0.16

 entrance

0.15

 entering

0.15

 entered

0.14

 Entrance

0.13

 entrar

0.13

Entered

0.13

(entries

0.13

Activations Density 0.001%