INDEX

Explanations

being part of a structure or sentence

This neuron lights up on first- and second-person pronouns and associated modal or opinion words (e.g. “I believe,” “you can,” “talk about,” “I think”), marking subjective or directive language.

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

and

-1.71

 všechny

-1.55

私たちの

-1.39

 especially

-1.31

ありますが

-1.30

</u>

-1.25

 rất

-1.25

 Landwir

-1.23

 הרבה

-1.23

 Auß

-1.23

POSITIVE LOGITS

 ferien

1.59

you

1.48

 this

1.47

 klasse

1.29

artet

1.22

 København

1.20

revet

1.20

这种

1.19

就能

1.18

Activations Density 0.027%