INDEX

Explanations

Democrat or Republican politicians

The neuron activates on numeric tokens—digits (and adjacent punctuation like parentheses or hyphens) used for things like measurements, years, district codes, or other numbers in the text.

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

 geleverd

-0.85

少しずつ

-0.81

 sämtliche

-0.77

 blijkt

-0.76

Autres

-0.75

_{\

-0.74

 ehemaligen

-0.74

reshed

-0.73

 przesz

-0.73

jores

-0.72

POSITIVE LOGITS

1.57

 Democrat

1.36

1.30

1.09

Rs

1.07

himself

0.90

￥

0.90

0.84

 Democrats

0.82

ıyors

0.82

Activations Density 0.023%