INDEX

Explanations

eliminate two incorrect options

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

্টর

1.42

ಟಕ

1.37

잍

1.35

 Часто

1.32

截止

1.32

 nbfs

1.27

hael

1.27

イン

1.26

$\#

1.25

butoxy

1.25

POSITIVE LOGITS

therners

2.23

wegian

1.75

etheless

1.71

FORM

1.54

raits

1.52

ძლი

1.48

ால்

1.47

RIA

1.46

ße

1.45

UniqueIdentifier

1.45

Activations Density 0.087%