INDEX

Explanations

This neuron is effectively “dead” in these samples—it does not activate on any of the content.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

我们

0.68

什么

0.59

 nonprofit

0.55

boat

0.54

我很

0.54

goers

0.53

ל

0.52

Based

0.52

全球

0.50

POSITIVE LOGITS

ErrorClazz

0.50

 でしょう

0.50

 Punishment

0.48

 इद

0.47

 Patrimonio

0.46

ッ

0.45

jenigen

0.45

。)

0.44

ンの

0.44

 Побе

0.42

Activations Density 0.000%