INDEX

Explanations

plural nouns followed by of

The neuron primarily activates on plural nouns.

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

 regalías

-1.60

 exportaciones

-1.49

deauna

-1.43

 rebajas

-1.32

 licencias

-1.27

 aislada

-1.24

 fiabilidad

-1.23

 extremadamente

-1.22

 aguj

-1.21

 cuotas

-1.20

POSITIVE LOGITS

of

9.81

ของ

3.72

 của

2.98

in

2.77

 της

2.55

 של

2.42

of

2.17

Of

1.70

 του

1.55

もちろん

1.43

Activations Density 0.227%