INDEX

Explanations

numbers and units

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

horabuena

-1.27

向けの

-1.17

 Artículos

-1.14

 instruk

-1.13

 diyor

-1.09

 Pristup

-1.09

 manualidades

-1.08

 seleccionados

-1.07

 věci

-1.07

 Mahasiswa

-1.07

POSITIVE LOGITS

すべて

1.09

調味料

1.06

1.01

 trattano

0.99

 different

0.97

min

0.96

翻攝

0.94

all

0.94

at

0.94

in

0.94

Activations Density 0.005%