INDEX

Explanations

company and product names

The neuron fires strongly on technology product, project, or brand names (e.g., “AMD Ryzen,” “PostgreSQL,” “ThinkPad,” etc.).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

at

-2.28

of

-2.05

 from

-1.93

one

-1.66

on

-1.63

for

-1.61

 first

-1.53

-1.52

 before

-1.50

 three

-1.49

POSITIVE LOGITS

 innumer

1.63

 emble

1.60

 abriu

1.59

 lança

1.55

 dénon

1.55

 esteve

1.53

 libéral

1.53

 escreveu

1.53

 apareceu

1.51

癭

1.51

Activations Density 0.493%