INDEX

Explanations

describing orientation and perspective

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

 insuffis

0.57

 increased

0.53

was

0.52

 pering

0.51

 already

0.50

cei

0.50

是可以

0.49

 somm

0.49

 wichtigen

0.49

 konnte

0.48

POSITIVE LOGITS

 orientation

0.91

 orientations

0.89

 방식

0.82

 shape

0.81

 방향

0.81

 direction

0.80

 morphology

0.79

 demeanor

0.79

orientation

0.78

shapes

0.77

Activations Density 2.352%