INDEX

Explanations

license and copyright notices

The neuron strongly fires on boilerplate license and copyright header text (e.g. terms like “All,” “rights,” “reserved,” “See,” and version numbers found in software license notices).

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

-1.72

-1.43

-1.41

-1.33

-1.26

-1.23

-1.22

All

-1.16

-1.15

-1.14

POSITIVE LOGITS

 utilisés

1.51

 semblables

1.41

 utilisent

1.34

 normaux

1.33

 quelcon

1.30

⟷

1.29

 deberes

1.29

 colorés

1.28

 aiment

1.27

 cóc

1.27

Activations Density 0.159%