INDEX

Explanations

technical metadata or structure

This neuron responds to tokens that include an underscore immediately followed or preceded by digits (e.g. “image_700,” “synonyms_121,” “format_612”), i.e. metadata field names combined with numeric values.

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

 hispan

-1.30

 Réseau

-1.29

 cromo

-1.27

 Antropo

-1.27

 nikah

-1.26

 camion

-1.25

 Turki

-1.22

 vernis

-1.21

 moft

-1.20

 maquillage

-1.20

POSITIVE LOGITS

1.33

1.31

It

1.30

That

1.24

To

1.22

These

1.16

1.15

1.13

For

1.12

Activations Density 0.022%