INDEX

Explanations

academic publications containing a volume number

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

hadiran

-0.71

 Todes

-0.69

cruise

-0.69

fæ

-0.69

hilation

-0.67

mente

-0.66

leſs

-0.65

ceutical

-0.65

ália

-0.65

ücks

-0.64

POSITIVE LOGITS

vol

1.05

Vol

1.05

vol

1.04

Vol

1.02

InSection

0.85

principalColumn

0.84

 AssemblyCulture

0.83

VOL

0.80

fromCharCode

0.77

ThroughAttribute

0.72

Activations Density 0.005%