INDEX

Explanations

mentions of "Can" and its variations, likely focusing on concepts related to dogs or canines

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

à¥įà¤Ń

-0.07

iÄĻ

-0.07

inizi

-0.07

pery

-0.07

weep

-0.07

rible

-0.07

genic

-0.07

hci

-0.07

kim

-0.07

POSITIVE LOGITS

’t

0.10

ister

0.10

't

0.10

isters

0.09

berra

0.09

icular

0.08

iglia

0.08

onical

0.07

tered

0.07

ifornia

0.07

Activations Density 0.024%