INDEX

Explanations

Lag Hyper M The : Tech “

The neuron strongly activates on numeric tokens or number-like expressions (e.g. years, floats, 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

()]

-1.56



-1.55

}$,

-1.44

ಿದೆ

-1.43

and

-1.39

 eller

-1.38

ówczas

-1.36

dropIfExists

-1.36

ก็

-1.33

📔

-1.32

POSITIVE LOGITS

 when

1.48

但是在

1.45

 saluto

1.36

 sinistro

1.36

 predecess

1.36

 advantageous

1.30

 balne

1.30

 deserto

1.30

落ち着いた

1.30

 assistente

1.25

Activations Density 0.000%