INDEX

Explanations

code and technical subjects

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

 vaše

-1.62

 naše

-1.52

 later

-1.41

 avaient

-1.38

vägen

-1.29

หรือ

-1.25

劾

-1.23

後に

-1.20

 náš

-1.20

 čern

-1.20

POSITIVE LOGITS

its

1.80

 being

1.44

 there

1.33

--;

1.30

In

1.20

其

1.19

 rusak

1.18

:{

1.15

 becomes

1.14

Activations Density 0.021%