INDEX

Explanations

words related to names and titles

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

Whilst

-0.08

 endeavour

-0.08

 Whilst

-0.08

 erst

-0.07

 whence

-0.07

 manoe

-0.07

Ìģ

-0.07

 whilst

-0.07

 herein

-0.06

 notwithstanding

-0.06

POSITIVE LOGITS

 basically

0.08

 Pirates

0.07

 strong

0.07

 pirates

0.07

pij

0.07

 based

0.07

 overall

0.07

 normally

0.07

 group

0.06

 related

0.06

Activations Density 0.002%