INDEX

Explanations

researchers' actions and findings

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

our

-1.48

 your

-1.34

そのような

-1.27

！"

-1.20

 オリジナル

-1.18

强大的

-1.15

 other

-1.13

эро

-1.12

While

-1.11

idopsis

-1.11

POSITIVE LOGITS

 themselves

2.02

 έχουν

1.42

 confirmó

1.41

 eivät

1.38

 tačiau

1.36

 دارند

1.34

 however

1.30

aimeJ

1.27

”

1.25

 hemd

1.23

Activations Density 0.148%