INDEX

Explanations

elements related to formal documentation and requirements

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

kus

-0.07

hiba

-0.07

 æ¬

-0.07

iasi

-0.07

aco

-0.07

ail

-0.07

iterals

-0.07

buah

-0.07

eping

-0.06

urar

-0.06

POSITIVE LOGITS

 whether

0.08

finger

0.07

 Aires

0.07

OWN

0.07

 details

0.07

 description

0.07

whether

0.07

how

0.06

 reason

0.06

 plans

0.06

Activations Density 0.011%