INDEX

Explanations

work

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

 work

-1.64

work

-1.20

 Work

-1.18

Work

-1.06

 WORK

-0.94

 works

-0.84

WORK

-0.81

 trabajo

-0.74

 travail

-0.74

 работе

-0.72

POSITIVE LOGITS

клопе

1.13

Rüyada

1.13

 purpoſe

1.02

AccessorTable

0.97

 Biôgrafia

0.96

 ſche

0.96

 itſelf

0.94

 houſe

0.94

 ſever

0.93

StructEnd

0.93

Activations Density 0.106%