INDEX

Explanations

measurement comparison

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

ликт

0.40

lastval

0.39

 bialgebra

0.39

 тяж

0.37

 заяви

0.36

是最

0.36

يمق

0.36

 తీ

0.35

()],

0.35

的第一

0.35

POSITIVE LOGITS

 degrees

0.75

 approximately

0.72

 similar

0.70

 greater

0.67

 less

0.67

 micro

0.66

 Degrees

0.66

degrees

0.64

micro

0.57

similar

0.57

Activations Density 0.009%