INDEX

Explanations

court documents and case identifiers

The neuron strongly activates on long runs of whitespace or indentation (i.e. sequences of many space characters).

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

or

-1.48

 from

-1.16

at

-1.16

 after

-1.16

 During

-1.13

 unusual

-1.11

any

-1.10

for

-1.10

 There

-1.09

 Although

-1.08

POSITIVE LOGITS

Minutes

1.17

'<?

1.14

État

1.10

IER

1.10

 title

1.06

 unbelievably

1.05

Untitled

1.04

 !!!!!

1.03

TestingModule

1.02

 quintessential

1.02

Activations Density 0.132%