INDEX

Explanations

finding feedback or obtaining results

technical labels and metrics

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

Eine

0.52

 técnicas

0.50

 BasicContainer

0.49

 steckt

0.46

 Eine

0.46

 conocidas

0.45

 muttered

0.44

就是一个

0.44

 técnica

0.43

 μια

0.43

POSITIVE LOGITS

waveform

0.50

adjust

0.50

應用

0.49

on

0.48

 optionally

0.48

reload

0.48

0.47

apply

0.47

LEAR

0.47

會

0.47

Activations Density 0.000%