INDEX

Explanations

defining actions or computations

This neuron activates on the leading words (usually imperative verbs) that start sentences in code documentation comments.

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

卺

-1.59

 Produit

-1.55

 imó

-1.55

 moteurs

-1.53

gâ

-1.48

 investimentos

-1.41

 neumáticos

-1.41

 komplet

-1.41

 geladeira

-1.41

 cozinha

-1.39

POSITIVE LOGITS

 with

1.95

if

1.77

or

1.65

 without

1.51

for

1.50

in

1.47

and

1.43

did

1.39

If

1.32

to

1.30

Activations Density 0.071%