INDEX

Explanations

.

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 pleaſure

-1.37

 myſelf

-1.33

Efq

-1.26

 houſe

-1.24

 Majefty

-1.23

 Monfieur

-1.23

 purpoſe

-1.20

 Jefus

-1.19

 raiſ

-1.15

 himſelf

-1.13

POSITIVE LOGITS

For

0.64

 Land

0.63

Per

0.62

Den

0.59

 témoins

0.59

Sto

0.59

Mer

0.59

 plomo

0.57

 burbujas

0.57

Con

0.57

Activations Density 0.003%