INDEX

Explanations

HTML table structures and LaTeX environments

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

valget

-0.80

สือ

-0.79

लब्

-0.78

irà

-0.77

غر

-0.76

mire

-0.73

пература

-0.73

 эле

-0.72

}{@

-0.72

ків

-0.71

POSITIVE LOGITS

 bonté

0.84

 sép

0.81

rsch

0.77

 окружающей

0.76

 zuges

0.74

彼は

0.73

メリ

0.73

 nachdem

0.72

不自

0.71

<tr>

0.70

Activations Density 0.001%