INDEX

Explanations

introducing examples after enumeration

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

usky

1.22

なんです

1.10

同年

1.04

なのです

1.01

当時

1.00

翌

0.97

aneous

0.97

 Twelfth

0.93

 barbar

0.93

 الثانية

0.93

POSITIVE LOGITS

such

4.18

 such

4.15

比如

3.77

例如

3.65

 مثل

3.57

 เช่น

3.50

 including

3.48

 SUCH

3.45

 такие

3.41

including

3.38

Activations Density 0.601%