INDEX

Explanations

sexual assault and harassment

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

了一個

-1.27

 anecd

-1.24

但却

-1.23

 Lös

-1.22

cience

-1.19

眼前的

-1.19

ﬄ

-1.17

 parallèle

-1.14

蛯

-1.14

心中的

-1.13

POSITIVE LOGITS

 сначала

1.28

各种

1.18

ずつ

1.12

 constantemente

1.11

七年

1.10

 họ

1.08

各種

1.08

 severe

1.07

Both

1.06

 both

1.04

Activations Density 0.060%