INDEX

Explanations

descriptive attributes related to structures

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Embeds

IFrame

Link

Negative Logits

ingham

-0.07

etsk

-0.07

 other

-0.06

 elsewhere

-0.06

 trend

-0.06

azer

-0.06

otine

-0.06

allas

-0.06

 important

-0.06

 aspects

-0.06

POSITIVE LOGITS

 nackte

0.09

UNUSED

0.08

inand

0.07

iskey

0.07

.setDefault

0.07

 affair

0.07

ãĥ³ãĥĩãĤ£

0.07

æ¡ĳ

0.07

yÅ¡

0.07

 ÑĦÐµÐ´ÐµÑĢÐ°Ð»ÑĮ

0.07

Activations Density 0.041%

No Known Activations