INDEX

Explanations

part of a category

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

čů

-1.03

HEW

-1.02

 INNOV

-1.02

 devine

-1.01

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

-1.01

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

-1.01

žnost

-1.00

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

-1.00

 their

-0.99

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

-0.98

POSITIVE LOGITS

 broader

1.13

 wider

1.11

 type

1.09

 also

1.09

 úgy

1.08

tzw

1.06

our

1.05

ADUATE

1.03

 dejamos

1.03

 category

1.02

Activations Density 0.103%