INDEX

Explanations

states importance/significance/prevalence

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

 aslında

0.47

 wirklich

0.45

 दरअसल

0.44

 actually

0.43

 realmente

0.40

actually

0.39

 faktisk

0.39

 tatsächlich

0.38

并非

0.38

 वास्तव

0.38

POSITIVE LOGITS

 predominate

0.71

 involved

0.62

 predominant

0.54

 paramount

0.54

 prominent

0.54

involved

0.52

 প্রাধান্য

0.51

 Involved

0.51

 prevalent

0.50

 promin

0.49

Activations Density 0.025%