INDEX

Explanations

code syntax and structure

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

</em>

-2.06

</i>

-1.77

în

-1.70

𝐒

-1.67

ımıza

-1.67

Interment

-1.59

理石

-1.57

 oublié

-1.55

for

-1.54

 که

-1.54

POSITIVE LOGITS

 ofreció

1.88

戋

1.67

any

1.65

vaila

1.63

 either

1.63

 зараз

1.59

 menyadari

1.59

₂,

1.55

 each

1.54

nesses

1.54

Activations Density 0.024%