INDEX

Explanations

Gemma model characteristics

the neuron detects document structure markers like section headers and formatted headings (markdown-style emphasis and numbered/listed section indicators).

New Auto-Interp

Configuration

Prompts (Dashboard)

238,145 prompts, 512 tokens each

Dataset (Dashboard)

lmsys + oasst1

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

더

0.41

無理

0.39

 점점

0.37

parcel

0.37

if

0.37

utérus

0.37

슐

0.37

कोरोना

0.37

ृ

0.37

츰

0.36

POSITIVE LOGITS

 limited

0.49

 specifically

0.49

 सीमित

0.47

 නමුත්

0.46

 Limited

0.46

 Markdown

0.46

 unlike

0.45

 designed

0.45

 Designed

0.45

 debuting

0.44

Activations Density 0.147%