INDEX

Explanations

as followed by words

The neuron primarily detects the discourse connective “As” (especially at the start of a clause or sentence).

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

 próprias

-1.40

 mesmas

-1.27

 katk

-1.26

więks

-1.23

 antaranya

-1.20

wwwww

-1.16

sists

-1.16

 навіть

-1.16

stopper

-1.16

fillment

-1.15

POSITIVE LOGITS

2.88

 part

2.38

 such

1.87

 with

1.57

an

1.53

we

1.45

 seen

1.45

of

1.45

you

1.43

the

1.41

Activations Density 0.052%