INDEX

Explanations

when followed by event

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 then

1.34

then

1.28

1.25

 eventually

1.16

1.15

which

1.15

،

1.14

 maka

1.13

 ultimately

1.10

 surely

1.10

POSITIVE LOGITS

 несмотря

1.76

 kerana

1.75

 karena

1.73

 ด้วย

1.64

 meski

1.63

 Karena

1.62

เพราะ

1.59

 meskipun

1.57

 เนื่องจาก

1.57

Karena

1.56

Activations Density 0.310%