INDEX

Explanations

time markers and sequences

The neuron strongly activates on numeric quantities and measurement terms in recipe‐style instructions (e.g. amounts, time, servings, units).

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

 dianteiro

-1.44

 endregion

-1.38

僕

-1.37

 recomendamos

-1.29

 dourada

-1.27

であれば

-1.26

 FUCKING

-1.17

১

-1.16

WindowConstants

-1.15

<tbody>

-1.13

POSITIVE LOGITS

 hubby

1.57

ค่ะ

1.46

yrs

1.37

 дуже

1.34

 ویتامین

1.25

łam

1.24

 varias

1.20

 although

1.19

romyalgia

1.17

⁀

1.16

Activations Density 0.254%