INDEX

Explanations

instances of the word "content" 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

ande

-0.07

-dot

-0.07

aption

-0.06

erland

-0.06

 Bris

-0.06

 suche

-0.06

alli

-0.06

ienda

-0.06

cke

-0.06

 Persistence

-0.06

POSITIVE LOGITS

adil

0.07

utch

0.07

fried

0.06

hel

0.06

ed

0.06

 Horton

0.06

 Consult

0.06

Ctrls

0.06

lator

0.06

Activations Density 0.001%