INDEX

Explanations

connecting clauses with and

The neuron strongly activates on numeric expressions (digits, measurements, counts, dates, scores, 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

iniques

-1.50

↵

-1.39

媼

-1.37

罴

-1.32

(),

-1.32

);

-1.30

$-$

-1.30

}){

-1.29

 servem

-1.29

))

-1.28

POSITIVE LOGITS

we

1.43

relse

1.41

ción

1.35

 bisogna

1.34

 accomplishing

1.34

 they

1.30

ícias

1.28

 attraverso

1.26

也

1.25

また

1.24

Activations Density 0.023%