INDEX

Explanations

start, beginning, starting point

The neuron responds to occurrences of “start” (including its morphological variants like “starting” and “starter”).

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

движи

-1.49

铪

-1.45

мото

-1.42

栳

-1.37

还有

-1.35

 moderne

-1.34

いえば

-1.33

 herv

-1.32

赑

-1.32

POSITIVE LOGITS

 would

1.88

but

1.63

.''

1.42

zość

1.38

していた

1.38

.”

1.34

 existing

1.31

 explored

1.30

)$.

1.30

 것을

1.29

Activations Density 0.012%