INDEX

Explanations

conjunctions and clause separators

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

 cómo

-1.23

and

-1.21

ныя

-1.20

*</

-1.16

…»

-1.13

%">

-1.13

 toit

-1.11

']='

-1.11

("");

-1.11

 THEN

-1.11

POSITIVE LOGITS

 this

1.88

 their

1.66

 when

1.48

 those

1.43

 there

1.41

 these

1.41

 they

1.40

 only

1.34

in

1.32

our

1.30

Activations Density 0.004%