INDEX

Explanations

applies to

The neuron fires specifically on the lemma “apply” (and its inflected forms like “applies”) in the 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

it

-1.38

 они

-1.35

by

-1.31

 using

-1.31

}]$

-1.27

In

-1.27

 wegen

-1.25

による

-1.22

 have

-1.20

POSITIVE LOGITS

Anyone

1.61

1.59

 comigo

1.51

龵

1.42

you

1.41

everyone

1.40

Applies

1.38

anyone

1.38

everybody

1.37

 nettsted

1.36

Activations Density 0.075%