INDEX

Explanations

instances of the word "The"

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

quo

-0.08

standing

-0.07

enda

-0.07

 excess

-0.07

ÑİÑĢ

-0.07

Ð¼ÐµÐ½Ð°

-0.06

cess

-0.06

Ð»Ð°Ð³Ð°

-0.06

Ø§Ø²Ùħ

-0.06

thed

-0.06

POSITIVE LOGITS

oretical

0.10

orem

0.08

orie

0.08

odor

0.07

odore

0.07

bes

0.07

iler

0.07

.UnitTesting

0.07

ÐµÐ³Ð¾

0.07

orer

0.07

Activations Density 0.035%