INDEX

Explanations

a, the followed by nouns

The neuron strongly activates on verbs in their present‐participle (“–ing”) form indicating ongoing actions.

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

to

-1.47

vertebrates

-1.45

כום

-1.41

fusions

-1.40

 kateg

-1.34



-1.27

 suivent

-1.25

 hvě

-1.23

 Mexique

-1.22

􀀀

-1.22

POSITIVE LOGITS

on

1.69

 wiederhol

1.29

 would

1.28

อยู่ใน

1.23

Also

1.23

1.21

 eliminating

1.20

).

1.20

Some

1.20

 trên

1.17

Activations Density 0.108%