INDEX

Explanations

the absence

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

워

1.00

忽然

0.98

 최초

0.94

突然

0.91

的重要

0.91

 noteworthy

0.91

 territory

0.90

 Territory

0.89

 contoh

0.88

명

0.88

POSITIVE LOGITS

 simpler

1.17

instead

1.15

 instead

1.15

 lieber

1.14

 unaffected

1.08

 preservar

1.08

 únicamente

1.07

 calmer

1.06

あくまで

1.02

Instead

1.00

Activations Density 1.308%