INDEX

Explanations

began as a

The neuron detects phrases describing how something “started out as” or “began as” (i.e., origin‐of expressions).

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

the

-1.87

 whom

-1.56

尤其是

-1.49

-1.47

is

-1.41

}{

-1.36

we

-1.35

fusions

-1.34

adone

-1.33

後は

-1.32

POSITIVE LOGITS

ược

1.34

 الوضع

1.34

AutoresizingMask

1.34

helfen

1.33

tiert

1.30

nictví

1.27

 belge

1.27

alisasi

1.26

Tomato

1.25

 impré

1.24

Activations Density 0.043%