INDEX

Explanations

states of being or change

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

-1.10

 チップ

-1.05

 unterschiedlich

-1.02

 erneut

-0.97

もう一度

-0.97

あぁ

-0.94

min

-0.94

 Еще

-0.93

-,

-0.91

POSITIVE LOGITS

and

1.54

lã

1.19

 cancelación

1.04

assero

1.02

 tieto

1.02

buatan

1.01

 PATRICK

0.99

 obstructive

0.96

🧌

0.95

íně

0.94

Activations Density 0.159%