INDEX

Explanations

beginning of phrases/titles

This neuron fires on standalone category or class labels that occur after a hyphen (e.g. the “major” in alerta-severity-major or the “n” in n-type).

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

 their

-1.58

ve

-1.55

-1.50

-1.48

臾

-1.40

ziem

-1.38

of

-1.38

 quí

-1.30

POSITIVE LOGITS

 identifiable

1.82

.”

1.78



1.77

 Organizador

1.76

 icona

1.74

珢

1.70

”)

1.70

感觉

1.66

plier

1.65

 debería

1.64

Activations Density 0.000%