INDEX

Explanations

ruins and remains

This neuron detects words referring to the physical remains or remnants of something (e.g., ruins, remains, wreckage).

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

-1.58

-1.48

-1.43

为什么会

-1.42



-1.40

 ricor

-1.37

自分は

-1.37

憮

-1.36

闋

-1.36

えっ

-1.36

POSITIVE LOGITS

of

4.78

1.83

 from

1.76

ﻝ

1.62

‘

1.51

 That

1.50

’

1.48

…

1.47

ческое

1.45

ZED

1.45

Activations Density 0.024%