INDEX

Explanations

URL paths and software identifiers

Gemma URL

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

bildung

0.45

gène

0.40

 思っ

0.38

 உருவாக

0.37

がき

0.37

madan

0.36

 けれど

0.36

 書い

0.36

kah

0.36

তথ্য

0.36

POSITIVE LOGITS

®

0.33

확

0.32

contrib

0.32

 contrib

0.31

th

0.30

plus

0.29

Contrib

0.29

::

0.29

€“

0.29

:::

0.28

Activations Density 0.017%