INDEX

Explanations

maintain

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

 became

-1.05

became

-0.98

 Became

-0.94

 maintains

-0.68

 gave

-0.64

was

-0.60

成为了

-0.60

 came

-0.59

 went

-0.59

 wurde

-0.57

POSITIVE LOGITS

 myſelf

1.09

 pleaſure

1.07

 Majefty

1.06

 purpoſe

1.02

 ſtate

0.93

 cauſe

0.93

Efq

0.90

 Monfieur

0.88

 caufe

0.88

 itſelf

0.87

Activations Density 1.105%