INDEX

Explanations

Indonesian verbs and foreign-sounding abbreviations

The neuron activates on standalone numeric tokens (digits and numbers).

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

 What

-1.84

↵↵

-1.82

或

-1.45

You

-1.41

 awfully

-1.38

 alebo

-1.37

apse

-1.37

and

-1.35

 enormous

-1.35

 我们

-1.31

POSITIVE LOGITS

 JAKARTA

1.55

PÉ

1.47

 ingat

1.39

MÁ

1.38

 morales

1.38

 mengatasi

1.38

 melindungi

1.36

RÉ

1.35

趵

1.34

hhhh

1.34

Activations Density 0.017%