INDEX

Explanations

so + adjective

The main thing this neuron does is detect numeric expressions and measurements (numbers, especially with units or decimals).

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

our

-1.38

⠅

-1.35

Cevap

-1.33

 руках

-1.24

 perna

-1.20

仳

-1.20

šne

-1.16

Тер

-1.16

Тан

-1.15

 duplo

-1.14

POSITIVE LOGITS

 that

3.52

you

1.92

 Uniwers

1.51

 που

1.46

enské

1.39

umum

1.38

 then

1.36

 manguera

1.34

">${

1.34

 aanbied

1.32

Activations Density 0.041%