INDEX

Explanations

negations or words that indicate exemption or opposition

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

-0.08

achuset

-0.08

amoto

-0.08

amps

-0.07

æĹ¢

-0.07

ENER

-0.07

adu

-0.07

leton

-0.07

age

-0.07

oze

-0.06

POSITIVE LOGITS

 necessarily

0.13

ori

0.08

 anymore

0.07

 enough

0.07

uent

0.07

 until

0.07

 ones

0.06

until

0.06

 Tops

0.06

ebek

0.06

Activations Density 0.012%