INDEX

Explanations

opposite effect or direction

The neuron detects words signaling causal effects or contrasts (e.g. “effect,” “opposite,” “order”).

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

and

-1.77

 also

-1.50

 كذلك

-1.33

冏

-1.31

 także

-1.31

 only

-1.29

众多

-1.24

轶

-1.23

inaison

-1.23

opatra

-1.23

POSITIVE LOGITS

 than

1.45

“

1.43

hasMoreElements

1.41

 perchè

1.39

 avendo

1.34

 mentre

1.32

ільки

1.30

熥

1.27

дото

1.27

 inizi

1.26

Activations Density 0.030%