INDEX

Explanations

articles and prepositions introducing nouns

This neuron responds strongly to words in sentence-initial position, i.e. the very first word of a new sentence or text block.

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

稣

-1.51

GOTREF

-1.40

GraphicsUnit

-1.37

苼

-1.36

って言

-1.36

了她

-1.34

.}(

-1.33

 embora

-1.32

culable

-1.32

 pertenec

-1.31

POSITIVE LOGITS

it

1.61

Two

1.41

Also

1.39

the

1.37

1.33

---

1.28

 myös

1.27

🥹

1.25

 mépris

1.25

 arī

1.23

Activations Density 0.104%