INDEX

Explanations

story narrative

The neuron detects phrases that introduce or describe a narrative summary (e.g. “The story…,” “presents the tale of…,” “follows…”).

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

-1.80

et

-1.69

’

-1.61

-1.38

-1.33

神器

-1.29

VISIBLE

-1.27

нести

-1.27

but

-1.25

//$

-1.23

POSITIVE LOGITS

<bos>

1.80

 "'";

1.74

ئة

1.60

 "'");

1.52

^{-}\

1.49

ught

1.48

 vetorial

1.43

 verstanden

1.41

鄚

1.41

1.40

Activations Density 0.022%