INDEX

Explanations

quantifiers and explicit quantities

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

ος

-1.08

}}^{(

-0.99

これが

-0.95

mbal

-0.94

blis

-0.94

 Substanz

-0.93

 Ganzen

-0.93

⎛

-0.91

さん

-0.91

鹇

-0.90

POSITIVE LOGITS

 both

1.13

Both

1.06

各有

1.04

都是在

1.02

 copious

0.99

both

0.97

 BOTH

0.97

 EVERY

0.97

 считается

0.97

TWO

0.92

Activations Density 0.010%