INDEX

Explanations

code structure delimiters

This neuron fires on capitalized identifiers (exported names) in Go code, such as struct, type, or function names.

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

ться

-1.76

 diinginkan

-1.64

 nowych

-1.59

malige

-1.56

verläs

-1.55

ότη

-1.55

 lecteurs

-1.54

 avevano

-1.54

 dibutuhkan

-1.53

 mengunjungi

-1.51

POSITIVE LOGITS

1.63

1.57

1.54

so

1.48

1.47

not

1.45

1.42

 both

1.41

As

1.38

in

1.38

Activations Density 0.019%