INDEX

Explanations

programming commands and statements

The neuron activates on common action words in code (imperative verbs used as function or method names, e.g. print, load, import, set, create, toggle).

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

-1.45

 таких

-1.45

ionine

-1.42

 hornear

-1.38

 listos

-1.38

<bos>

-1.35

justed

-1.34

 enfri

-1.33

 Verantwortung

-1.30

 что

-1.30

POSITIVE LOGITS

in

1.92

for

1.73

 from

1.70

by

1.64

 gentes

1.63

 only

1.56

 with

1.52

 setPassword

1.45

puo

1.45

if

1.43

Activations Density 0.030%