INDEX

Explanations

a/A followed by a noun

The neuron predominantly activates on English articles (“the,” “a”) in running text.

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

our

-1.60

our

-1.45

 già

-1.33

一直在

-1.30

 will

-1.30

 temu

-1.25

we

-1.24

ीय

-1.20

–

-1.20

 previously

-1.20

POSITIVE LOGITS

 When

1.52

尤其是

1.50

 Ketika

1.27

Usually

1.27

特别是

1.27

 thầy

1.26

 taunt

1.25

cargador

1.24

 whine

1.22

Those

1.20

Activations Density 0.069%