INDEX

Explanations

the use of the word "except" and its variations indicating exclusions or limitations

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Embeds

IFrame

Link

Negative Logits

 Less

-0.06

rgan

-0.06

ago

-0.06

.YesNo

-0.06

 excess

-0.06

argas

-0.06

ãģ°ãģĭãĤĬ

-0.06

attern

-0.06

elper

-0.06

 Goth

-0.05

POSITIVE LOGITS

aside

0.08

 exception

0.08

 exceptions

0.07

eyin

0.07

aybe

0.07

iji

0.07

 Exceptions

0.07

entai

0.07

å°ĳ

0.07

 minor

0.07

Activations Density 0.011%

No Known Activations