INDEX

Explanations

this paper or article

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

各种

-1.23

 various

-1.20

any

-1.20

istical

-1.16

色々な

-1.15

 those

-1.12

 ランチ

-1.11

metodo

-1.11

 beragam

-1.10

 conditions

-1.06

POSITIVE LOGITS

<th>

1.23

Explain

1.21

OnClickListener

1.14

เรา

1.11

我们

1.09

\|_{

1.09

 braços

1.07

๓

1.07

 Israël

1.07

 Polskiego

1.04

Activations Density 0.127%