INDEX

Explanations

such [qualifier]

The neuron responds to strong qualitative modifiers—adjectives that convey high magnitude or intensity (e.g. “extreme,” “reprehensible,” “long-range,” “tight,” “extensive,” “close”).

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

of

-1.60

迳

-1.39

 Многие

-1.21

 besonder

-1.18

ob

-1.17

SU

-1.17

数据集

-1.16

 thậm

-1.16

RF

-1.14

OL

-1.13

POSITIVE LOGITS

 podamos

1.43

}();

1.34

 puedas

1.31

ongevity

1.30

("");

1.28

=('

1.26

的了

1.25

 than

1.24

叭

1.23

 preparar

1.20

Activations Density 0.039%