INDEX

Explanations

phrases indicating specific locations or positions in a narrative context

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

etin

-0.07

kle

-0.07

esk

-0.07

inch

-0.06

 reckon

-0.06

æķ·

-0.06

raith

-0.06

ukkan

-0.06

bracht

-0.06

POSITIVE LOGITS

 ÄĳÃŃch

0.07

 Hunts

0.06

 Ziel

0.06

axies

0.06

unp

0.06

onden

0.06

ever

0.06

elf

0.06

 bare

0.06

crate

0.06

Activations Density 0.001%