INDEX

Explanations

programming loop variable i

The neuron fires on numeric literal tokens (especially decimal numbers) 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

for

-1.54

to

-1.50

 both

-1.49

of

-1.47

at

-1.46

the

-1.40

↵

-1.37

such

-1.36

 part

-1.35

getNome

-1.27

POSITIVE LOGITS

нию

1.59

らなかった

1.56

 ktorí

1.55

 dAtA

1.54

 их

1.45

WriteLine

1.43

 Herzegovina

1.41

又是一

1.40



1.39

⼦

1.39

Activations Density 0.036%