INDEX

Explanations

purpose and function verbs

The neuron strongly activates on verbs in “to <V>” infinitive phrases (e.g. “to control,” “to describe,” “to measure,” “to represent,” etc.).

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

着一

-1.35

áda

-1.25

 sorriso

-1.20

এ

-1.19

着一个

-1.18

'__

-1.18

 there

-1.17

</h4>

-1.14

CONTAINER

-1.13

iczna

-1.10

POSITIVE LOGITS

be

1.75

of

1.52

But

1.41

その後

1.30

 entweder

1.30

 stali

1.29

 that

1.27

 други

1.26

 mesta

1.21

 other

1.20

Activations Density 0.117%