INDEX

Explanations

one of the

This neuron detects enumerative “one of the…” constructions (phrases that pick out an item as part of a ranked or set-based list).

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

<bos>

-1.65

jenigen

-1.62

 Betrach

-1.53

 then

-1.51

觏

-1.48

這個

-1.47

 différentes

-1.45

ANA

-1.44

 概要

-1.41

斟

-1.40

POSITIVE LOGITS

of

1.88

埸

1.62

収集

1.48

垸

1.48

</em>

1.42

1.41

 inspiradoras

1.37

1.36

1.35

 již

1.34

Activations Density 0.050%