INDEX

Explanations

The neuron strongly activates on list- or example-introducing markers (e.g. “such as,” “e.g.,” “including”) and numeric tokens.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 autres

1.00

 altri

0.97

 autre

0.92

 other

0.91

 otras

0.91

 annan

0.90

 سایر

0.89

 demás

0.88

 ఇతర

0.88

的其他

0.87

POSITIVE LOGITS

 those

1.34

1.33

those

1.26

 예를

1.26

 например

1.24

 például

1.22

但不

1.22

например

1.20

 například

1.20

 misalnya

1.17

Activations Density 0.434%