INDEX

Explanations

SAX parser

The neuron is looking for occurrences of parser-related terms—names of parsing libraries or the words “parse”/“parser.”

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

 Ganzen

-1.69

we

-1.53

 esigen

-1.44

You

-1.42

 they

-1.41

of

-1.39

看起來

-1.34

我

-1.31

根據

-1.30

喲

-1.30

POSITIVE LOGITS

螄

1.46

خفض

1.46

')]

1.45

瑒

1.42

rieben

1.41

 scrutin

1.37

 satisfe

1.35

筥

1.34

 Tails

1.34

Ⅽ

1.34

Activations Density 0.021%