INDEX

Explanations

non-zero, nonempty, nontrivial conditions

New Auto-Interp

Configuration

Prompts (Dashboard)

238,145 prompts, 512 tokens each

Dataset (Dashboard)

lmsys + oasst1

Embeds

IFrame

Link

Negative Logits

 lacked

0.58

없

0.56

 lacks

0.55

Lack

0.53

없

0.52

 Lack

0.51

 없다

0.51

 없어

0.50

缺乏

0.50

 없음

0.49

POSITIVE LOGITS

 nonzero

2.22

nonzero

1.92

 nonempty

1.65

 nontrivial

1.58

 nonEmpty

1.50

isNotEmpty

1.32

NotEmpty

1.30

non

1.27

 нен

1.24

non

1.18

Activations Density 0.135%

No Known Activations