INDEX

Explanations

what i

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 whats

-0.10

 Schwe

-0.09

Adjacent

-0.08

iedy

-0.08

 indigenous

-0.08

 what

-0.08

–↵↵

-0.08

 Giov

-0.08

 accumulating

-0.08

odd

-0.07

POSITIVE LOGITS

 favour

0.11

eca

0.10

 entirety

0.10

inality

0.10

amil

0.10

 Baldwin

0.10

anning

0.09

erek

0.09

ela

0.09

 angered

0.09

Activations Density 0.004%