INDEX

Explanations

phrases related to concealment and disguising actions in various contexts

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

iani

-0.06

acles

-0.06

arde

-0.06

 inflated

-0.06

onu

-0.06

arden

-0.06

abwe

-0.06

acle

-0.06

NotNull

-0.06

eeper

-0.06

POSITIVE LOGITS

 invisible

0.20

invisible

0.17

 Invisible

0.16

 invis

0.16

 hidden

0.15

 Blend

0.14

-hidden

0.14

 blend

0.14

éļĲèĹı

0.14

 unnoticed

0.14

Activations Density 0.064%