INDEX

Explanations

scientists don't trust

New Auto-Interp

Configuration

Prompts (Dashboard)

238,145 prompts, 512 tokens each

Dataset (Dashboard)

lmsys + oasst1

Embeds

IFrame

Link

Negative Logits

ၟ

0.50

Shortcut

0.50

コーディ

0.49

 Thời

0.47

 Deviation

0.46

 Bereiche

0.46

AF

0.46

କ୍

0.46

 Tapi

0.46

Thời

0.46

POSITIVE LOGITS

 swimming

0.62

 useful

0.57

 exerted

0.56

 persistent

0.56

 inclusion

0.56

 poisonous

0.54

 religious

0.53

 moving

0.53

 assignment

0.53

 serums

0.53

Activations Density 0.001%

No Known Activations