INDEX

Explanations

battle of

The neuron fires on mentions of named battles (e.g. “Battle of …”), identifying occurrences of the word “Battle” in battle titles.

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

}$,

-1.77

 techniczne

-1.72

 pokud

-1.70

ญี่ป

-1.70

ка

-1.70

ра

-1.66

 коли

-1.63

 awakeFromNib

-1.59

 rozs

-1.59

 店内

-1.59

POSITIVE LOGITS

簟

1.92

1.69

高的

1.69

of

1.65

1.53

the

1.48

 увидел

1.45

 eastern

1.45

ing

1.45

it

1.44

Activations Density 0.011%