INDEX

Explanations

conflict with or between

The neuron detects occurrences of “conflict” (and its morphological variants) in 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

nessy

-1.59

ن

-1.57

¨

-1.53

мо

-1.48

 которой

-1.47

of

-1.45

();)

-1.45

ли

-1.45

ㅎ

-1.43

er

-1.42

POSITIVE LOGITS

潆

1.61

ındaki

1.61

 condamné

1.55

鈇

1.52

こちらも

1.52

也会

1.48

噲

1.47

年も

1.46

iczne

1.45

FOOTNOTES

1.45

Activations Density 0.042%