INDEX

Explanations

elements related to design and organizational structure

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

aram

-0.07

 rodin

-0.07

(strtolower

-0.07

ÅĻil

-0.07

 lÃ²ng

-0.07

 Pablo

-0.07

oggle

-0.07

/reg

-0.06

uzzy

-0.06

ainment

-0.06

POSITIVE LOGITS

allows

0.10

 means

0.10

 allows

0.09

 Means

0.08

 mean

0.08

means

0.08

 allow

0.08

 makes

0.07

ä½¿

0.07

 nedeniyle

0.07

Activations Density 0.059%