INDEX

Explanations

used up or exhausted

The neuron activates on words and phrases that convey something being depleted or used up (e.g. “used up,” “exhausted”).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 alip

-1.64

 alberto

-1.50

 kereta

-1.50

kabau

-1.48

¨

-1.48

 ideen

-1.48

 damen

-1.48

 alha

-1.46

 episodi

-1.46

NOO

-1.45

POSITIVE LOGITS

鸢

1.51

on

1.42

es

1.39

1.36

1.34

 bekämp

1.30

 wußte

1.30

と呼ば

1.30

 resonate

1.29

 ikke

1.27

Activations Density 0.021%