INDEX

Explanations

avoid jargon or complex language

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 tass

1.17

本书

1.15

睦

1.14

λικ

1.14

SAW

1.13

рованных

1.12

琭

1.11

รม

1.11

1.10

 copos

1.09

POSITIVE LOGITS

 erreur

1.32

 Grief

1.26

 مَا

1.24

 Mock

1.24

 mock

1.18

astica

1.16

 TORT

1.15

 ERROR

1.14

eweg

1.12

穿越

1.12

Activations Density 0.041%