INDEX

Explanations

asking about model behavior

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

 veux

0.52

変

0.52

lze

0.49

ньої

0.49

ینی

0.48

 affaire

0.48

 murdering

0.47

ка

0.47

殺

0.46

اری

0.46

POSITIVE LOGITS

 comparisons

0.51

 iPads

0.50

 classrooms

0.50

 Celebrating

0.50

 facilitators

0.49

 STEM

0.48

 podcasts

0.48

DJs

0.48

 Related

0.47

 upbeat

0.47

Activations Density 0.010%