INDEX

Explanations

preparation and explanation

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

<strong>

1.43

<code>

1.38

1.35

<b>

1.30

<u>

1.25

 الأول

1.25

–

1.19

<em>

1.18

oreal

1.09

 والاح

1.08

POSITIVE LOGITS

घ्र

1.73

 frmt

1.68

 Aguas

1.68

特価

1.63

<unused1692>

1.60

 Республи

1.60

 totalité

1.60

醗

1.57

 pertains

1.57

 ఏర్

1.54

Activations Density 0.000%