INDEX

Explanations

XML resource files

This neuron responds to double-quote characters (the " token) used to delimit strings.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 SQLITE

-1.07

 SISTEMA

-1.05

mappedBy

-1.01

 jälkeen

-1.00

 fidèle

-0.98

 disting

-0.96

 Automó

-0.95

TÉ

-0.95

 Tienen

-0.94

ERICK

-0.90

POSITIVE LOGITS

1.06

 especially

1.02

but

0.98

衣裙

0.97

in

0.93

 somewhat

0.93

haw

0.92

 often

0.92

 about

0.90

ie

0.90

Activations Density 0.000%