INDEX

Explanations

proper nouns and negative experiences

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

 will

-1.65

 this

-1.62

 these

-1.54

for

-1.34

of

-1.34

 here

-1.31

 These

-1.30

 should

-1.30

on

-1.29

 apparently

-1.28

POSITIVE LOGITS

 didn

1.41

怎么回事

1.40

cée

1.37

wasn

1.35

Ekster

1.33

缡

1.27

 wasn

1.27

ată

1.25

isn

1.25

だけど

1.24

Activations Density 0.003%