INDEX

Explanations

finding context

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

and

-1.55

etero

-1.11

ところは

-1.09

panjang

-1.07

 pestañas

-1.06

だけではなく

-0.96

 seperti

-0.96

 ürünleri

-0.96

pasang

-0.95

ımıza

-0.94

POSITIVE LOGITS

。

1.47

vermectin

1.20

1.17

titial

1.13

}$.

1.09

 omdat

1.09

 поскольку

1.08

ebly

1.07

fordshire

1.07

 cana

1.05

Activations Density 0.307%