INDEX

Explanations

year and months

The neuron activates on mentions of time‐period durations (e.g. “months,” “year,” “decade”) often with numeric or temporal modifiers.

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

</h3>

-1.61

</u>

-1.53

 after

-1.48

of

-1.38

醍

-1.34

 from

-1.27

彼女の

-1.27

的女人

-1.26

the

-1.25

now

-1.24

POSITIVE LOGITS

所说

1.39

 Meningkatkan

1.38

۲

1.36

所说的

1.34

 которы

1.34

llten

1.34

姽

1.32

 jejich

1.29

γμα

1.29

 necesitas

1.28

Activations Density 0.016%