INDEX

Explanations

This neuron detects function or method signatures immediately followed by a return type (i.e. the closing “) <type>” in declarations).

function signatures

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

considered

-1.53

albeit

-1.41

 genügend

-1.38

 diriger

-1.35

 molti

-1.35

 sämtliche

-1.34

necessary

-1.34

specifically

-1.33

 corriger

-1.31

mainly

-1.30

POSITIVE LOGITS

in

2.11

 what

1.82

 from

1.60

and

1.59

so

1.46

an

1.38

or

1.37

 supremely

1.32

 undeniably

1.32

 probably

1.32

Activations Density 0.004%