INDEX

Explanations

consent and conditions

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

hopefully

-1.12

whenever

-0.99

 hebben

-0.96

 μὲν

-0.96

 ванной

-0.95

 ANSWERS

-0.94

 TEACHING

-0.94

眎

-0.93

whatever

-0.93

着自己的

-0.92

POSITIVE LOGITS

There

1.06

 entsprechenden

1.02

During

1.02

it

1.00

 poignant

0.99

 sparkly

0.98

It

0.98

Here

0.96

Other

0.95

 biggest

0.95

Activations Density 0.035%