INDEX

Explanations

phrases related to proportions or quantities

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

orno

-0.08

 konkrÃ©t

-0.07

ummings

-0.07

monds

-0.07

/Dk

-0.07

itorio

-0.06

ipo

-0.06

ipt

-0.06

ania

-0.06

ä¹ħ

-0.06

POSITIVE LOGITS

 fraction

0.12

fraction

0.10

ally

0.10

 fractions

0.09

 compared

0.09

 Fraction

0.09

ALLY

0.09

Fraction

0.07

 fractional

0.07

 less

0.07

Activations Density 0.005%