INDEX

Explanations

many different kinds

The neuron detects plural quantifier words and phrases that indicate “many,” “multiple,” or a “wide/several” range of items.

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

in

-1.62

 when

-1.61

 during

-1.57

for

-1.38

を

-1.37

や

-1.28

-1.27

 because

-1.27

などの

-1.23

 after

-1.22

POSITIVE LOGITS

様な

1.45

 variados

1.41

に挑戦

1.32

感じで

1.31

ようで

1.30

筈

1.29

ほら

1.29

corrhi

1.28

事が

1.27

 kinds

1.27

Activations Density 0.087%