INDEX

Explanations

legal or formal settings (`uk`, `defamation`, `recover`, `privilege`)2. specific actions or qualities (`dedication`, `distribution`, `odd`)3. time or duration (`time`)

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

్

2.02

tin

2.00

sberg

1.98

्

1.98

<0x99>

1.95

𝗮

1.94

sframe

1.93

 wrench

1.90

𝗕

1.89

spoken

1.89

POSITIVE LOGITS

lä

1.94

‬‬

1.77

 sollte

1.76

 sind

1.51

телям

1.50

 befindet

1.50

 liegt

1.50

 anhand

1.49

 sollten

1.49

を行

1.46

Activations Density 0.000%