INDEX

Explanations

phrases related to denial or acknowledgment

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

sto

-0.07

PIX

-0.06

anship

-0.06

æİĪ

-0.06

áº¡p

-0.06

utations

-0.06

_DECL

-0.06

/components

-0.06

_TRACE

-0.06

jeta

-0.06

POSITIVE LOGITS

 ráº±ng

0.08

urga

0.07

 Mold

0.06

 bahwa

0.06

atti

0.06

 facts

0.06

otta

0.06

 ÑĦÐ°ÐºÑĤ

0.06

 that

0.06

Activations Density 0.005%