INDEX

Explanations

"it was" and "the"

This neuron responds to tokens that mark specific entities or quantitative details—e.g. proper names and numbers.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 customary

0.38

 physicochemical

0.34

 cryptographic

0.33

 phosphoryl

0.32

 resolv

0.32

 hydrocarbon

0.31

 unaffected

0.31

 anisotropic

0.31

 đảm

0.30

 phosphorylation

0.30

POSITIVE LOGITS

Lisa

0.46

Katie

0.42

Cindy

0.42

Angela

0.41

Alison

0.41

Kerry

0.41

Rachel

0.40

Doug

0.40

Jenny

0.40

Julie

0.39

Activations Density 0.000%