INDEX

Explanations

unix commands

The neuron is especially sensitive to occurrences of sed command invocations (text-substitution expressions).

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

 наших

-1.48

What

-1.43

LError

-1.42

蕯

-1.41

۲

-1.39

WebElementEntity

-1.37

 luat

-1.36

ле

-1.36

Voci

-1.35

 primit

-1.34

POSITIVE LOGITS

the

1.70

for

1.52

 from

1.42

at

1.42

 blijven

1.29

 summarise

1.27

 MÁS

1.27

so

1.20

 meeste

1.19

»,

1.18

Activations Density 0.068%