INDEX

Explanations

states, places, conditions, and actions

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ebly

-1.20

ânea

-1.02

 which

-0.98

了两

-0.98

ěř

-0.97

fizierung

-0.96

 детского

-0.95

榻

-0.94

半天

-0.93

 крова

-0.91

POSITIVE LOGITS

はもう

1.23

Often

1.21

Whether

1.16

 llevó

1.13

本来

1.13

 também

1.09

 advirtió

1.09

೬

1.08

 your

1.07

Sometimes

1.07

Activations Density 0.159%