INDEX

Explanations

in nature

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

of

-1.08

for

-1.07

：……

-1.03

 which

-1.00

or

-0.98

 from

-0.98

 betterment

-0.96

AHULUAN

-0.92

an

-0.91

 ease

-0.91

POSITIVE LOGITS

compared

1.27

 dealings

1.16

Cevap

1.13

despite

1.07

suggest

1.03

helping

1.02

 カットソー

1.02

 dedans

1.02

 nature

1.01

 propone

1.00

Activations Density 0.090%