INDEX

Explanations

criticism of specific entities

The neuron detects instances of critique or criticism language (e.g. “criticise,” “criticized,” “criticism”).

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

 extinción

-1.56

 vocación

-1.50

,?

-1.45

ayaquil

-1.43

 Thats

-1.31

</strong>

-1.30

Ingrédients

-1.30

enerbahçe

-1.29

 participó

-1.27

致力于

-1.27

POSITIVE LOGITS

how

1.53

橚

1.44

it

1.39

 DELHI

1.38

 while

1.37

вге

1.36

噐

1.33

哪

1.32

 satte

1.32

to

1.32

Activations Density 0.049%