INDEX

Explanations

phrases indicating composition or creation

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

itz

-0.07

Ð¾ÑģÑĮ

-0.06

Ã©ri

-0.06

ãĥģ

-0.06

å¥ĳ

-0.06

imar

-0.06

Ø¯Ø§

-0.06

 prob

-0.06

Ð¼ÐµÐ¶

-0.06

lea

-0.06

POSITIVE LOGITS

 ÑģÐ¾Ð±Ð¾Ð¹

0.09

 part

0.09

enance

0.08

 ÑģÐ¾Ð±Ð¾Ñİ

0.07

orer

0.07

ovit

0.07

 parte

0.07

ignon

0.07

eking

0.07

Activations Density 0.009%