INDEX

Explanations

bank account

The neuron strongly activates on occurrences of the word “bank” (and immediately adjacent banking terms).

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

</b>

-1.80

 fervent

-1.57

珵

-1.48

 REALLY

-1.48

횽

-1.47

皤

-1.44

 kter

-1.43

ご了承ください

-1.41

␉

-1.40

 quasi

-1.39

POSITIVE LOGITS

or

1.73

ありません

1.62

或

1.54

 expands

1.40

 این

1.39

喜欢的

1.38

 highlighted

1.37

 garner

1.35

 після

1.35

 wants

1.34

Activations Density 0.013%