INDEX

Explanations

list items with dr or group

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

 decidedly

1.24

 preponder

1.20

gdock

1.20

<unused2123>

1.17

≾

1.17

 strangely

1.17

 ჩატი

1.17

 ostensibly

1.15

<unused902>

1.15

<unused2178>

1.14

POSITIVE LOGITS

For

1.44

For

1.38

In

1.30

<h2>

1.28

When

1.21

To

1.21

 When

1.18

 Portanto

1.17

In

1.17

An

1.16

Activations Density 0.304%