INDEX

Explanations

visit to websites

comparing models and differences

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

一

0.64

وه

0.45

 agric

0.45

৫

0.45

וי

0.44

কোন

0.44

أي

0.44

粙

0.44

 واي

0.43

REPO

0.43

POSITIVE LOGITS

udsman

0.53

ta

0.47

<unused53>

0.46

 자동차

0.46

 endorph

0.45

timers

0.44

 allegations

0.44

점

0.44

 일부

0.43

 unlawfully

0.43

Activations Density 0.007%