INDEX

Explanations

possessive pronoun followed by noun

The neuron flags possessive markers and pronouns (e.g. ’s, her, his) indicating ownership or possession.

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

勠

-1.58

ябрь

-1.48

これらの

-1.47

 vermek

-1.44

 antiquated

-1.43

蠱

-1.43

 збіль

-1.39

هذه

-1.38

Wszyst

-1.37

 通り

-1.35

POSITIVE LOGITS

</h2>

1.55

一旁的

1.51

1.45

跑去

1.41

 beautifully

1.40

was

1.40

 focuses

1.39

’

1.39

ded

1.38

arine

1.38

Activations Density 0.064%