INDEX

Explanations

name followed by punctuation

The neuron activates on occurrences of the identifier “name” in code.

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

↵

-1.41

；

-1.33

ements

-1.22

IDENTIAL

-1.19

 potrebbe

-1.19

 entren

-1.17

श

-1.16

 siguientes

-1.16

.]

-1.15

OSPITAL

-1.15

POSITIVE LOGITS

 και

1.50

of

1.46

and

1.45

 réguli

1.42

 élevées

1.39

 like

1.36

 setType

1.30

 essentielles

1.27

 privées

1.27

úgó

1.27

Activations Density 0.045%