INDEX

Explanations

business entities and their contexts

The neuron is strongly activated by mentions of “business” and other business-related terms.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 イラストレーション

-1.30

 voitures

-1.29

 perdre

-1.22

 côtes

-1.21

 salari

-1.20

=“

-1.19

 appels

-1.18

 alimente

-1.18

 verres

-1.18

opies

-1.17

POSITIVE LOGITS

or

2.38

 your

2.08

you

1.78

 will

1.55

if

1.45

 regardless

1.25

↵↵

1.25

 through

1.23

 using

1.21

has

1.20

Activations Density 0.044%