INDEX

Explanations

negation followed by contrast

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

 anything

1.38

any

1.30

 Anything

1.15

 qualquer

1.14

anything

1.11

 cualquier

1.08

任何

1.08

 qualsiasi

1.07

 anyone

1.06

 ANYTHING

1.04

POSITIVE LOGITS

Nor

1.54

nor

1.49

Nor

1.36

nor

1.12

ni

0.97

nem

0.92

 Neither

0.89

Neither

0.89

也不是

0.87

Nem

0.87

Activations Density 0.130%