INDEX

Explanations

still not

The neuron fires on explicit negation markers (e.g. “not,” “n’t,” “don’t,” “haven’t”) in the text.

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

-1.69

ꪻ

-1.62

 helado

-1.43

 следующий

-1.42

 prendre

-1.42

 myös

-1.40

But

-1.38

但在

-1.38

://$

-1.38

POSITIVE LOGITS

 improved

1.52

 which

1.48

曖昧さ回避

1.38

 gelungen

1.35

 erfolgreich

1.35

 addition

1.34

 通过

1.30

 podremos

1.29

 where

1.27

 psicologia

1.23

Activations Density 0.024%