INDEX

Explanations

Z start followed by German words

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

やっぱり

-0.78

やはり

-0.76

骡

-0.73

<0xA3>

-0.71

 getPassword

-0.71

職業

-0.70

 those

-0.69

oles

-0.69

we

-0.69

块钱

-0.68

POSITIVE LOGITS

 Ци

1.13

𝕐

0.85

将来

0.84

elon

0.82

Tooth

0.81

FUTURE

0.80

 Divis

0.80

addington

0.79

Meal

0.79

為に

0.79

Activations Density 0.004%