INDEX

Explanations

will followed by action

The neuron flags first-person plural future‐tense promises or offers (e.g. “we will…,” “we can…”).

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

.}(

-1.61

of

-1.45

 Comunicação

-1.41

 Título

-1.38

++);

-1.37

↵

-1.35

 Gestão

-1.34

 Livro

-1.32

 NÃO

-1.30

 Início

-1.29

POSITIVE LOGITS

it

1.84

if

1.58

get

1.48

 more

1.46

或是

1.41

使用的

1.36

 waarvan

1.35

yskland

1.34

或者

1.34

或者是

1.30

Activations Density 0.035%