INDEX

Explanations

; delimiters or end of statements

The neuron activates on literal constants in code (numeric literals, boolean literals, null, etc.).

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

the

-1.68

妡

-1.65

ﹶ

-1.57

غيرة

-1.38

 birkaç

-1.38

 Verwenden

-1.35



-1.34

梟

-1.34

⧪

-1.33

Specifica

-1.31

POSITIVE LOGITS

 jego

1.73

but

1.59

if

1.45

 same

1.38

 didn

1.30

 cơn

1.30

同じ

1.25

 himself

1.23

detto

1.23

 его

1.22

Activations Density 0.022%