INDEX

Explanations

programming method calls

The neuron activates on numeric tokens—i.e. literal numbers appearing in the text.

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

but

-1.65

-1.51

was

-1.48

 он

-1.45

of

-1.39

‘

-1.39

があります

-1.36

-1.35

</h4>

-1.34

is

-1.30

POSITIVE LOGITS

猻

1.76

ꪹ

1.61

 biela

1.56

jecture

1.52

 farba

1.51

 biele

1.51

 honneur

1.50

cím

1.41

 klient

1.41

 your

1.41

Activations Density 0.059%