INDEX

Explanations

The neuron is an indicator of non-English text or tokens

New Auto-Interp

Configuration

Prompts (Dashboard)

238,145 prompts, 512 tokens each

Dataset (Dashboard)

lmsys + oasst1

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Besonders

0.43

Shr

0.38

 Kone

0.38

 Robust

0.37

Affine

0.37

 Wooden

0.37

 श्र

0.36

 iPhone

0.36

 shrink

0.36

 case

0.36

POSITIVE LOGITS

othe

0.47

endem

0.42

<unused31>

0.40

뚤

0.39

biti

0.38

irii

0.38

習

0.37

 使っ

0.37

頨

0.37

athe

0.36

Activations Density 0.011%