INDEX

Explanations

p-values less than

The neuron selectively activates on floating‐point numerals (numbers containing a decimal point).

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

-2.48

for

-2.42

 before

-2.25

 through

-2.25

at

-2.23

 from

-2.03

 mainly

-2.02

 after

-1.96

 especially

-1.95

 during

-1.89

POSITIVE LOGITS

臜

1.79

—–

1.78

 ridiculously

1.76

这样

1.73

thats

1.72

THAT

1.70

 ludicrous

1.70



1.70

theres

1.63

))

1.62

Activations Density 0.002%