INDEX

Explanations

patients and numbers

The neuron activates on common plural nouns (i.e., words ending in “-s” that denote multiple 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

if

-2.16

</th>

-1.66

으면

-1.51

if

-1.51

ございました

-1.50

经常

-1.50

 riusc

-1.45

ijão

-1.45

 parteci

-1.42

 quantità

-1.41

POSITIVE LOGITS

 that

1.89

そういった

1.51

 που

1.45

着自己的

1.44

neous

1.43

 включая

1.42

こうした

1.41

oforms

1.34

ϫ

1.33

 coreano

1.32

Activations Density 0.176%