INDEX

Explanations

number `this` followed by `people` or `this`

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

 Morales

-0.08

aland

-0.08

 spray

-0.07

aine

-0.07

ands

-0.07

ulis

-0.07

Lia

-0.07

 Oversight

-0.07

dec

-0.07

ones

-0.07

POSITIVE LOGITS

 cucumber

0.11

bon

0.11

lem

0.11

Lew

0.11

oux

0.11

 clown

0.11

 derby

0.11

lep

0.10

pw

0.10

",-

0.10

Activations Density 0.037%