INDEX

Explanations

person identifying sequence

The neuron fires on demonstrative noun phrases—especially those beginning with “that” (often with an adjective or number) pointing to a specific person or thing (e.g. “that chick,” “that one person,” “that gift,” “that girl”).

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

 themselves

-1.31

世纪

-1.27

ULATE

-1.15

$.

-1.13

 ベクトル

-1.11

ième

-1.10

}$.

-1.10

 entonces

-1.07

neten

-1.07

 then

-1.06

POSITIVE LOGITS

最初に

1.51

後に

1.37

樣子

1.32

 ",";

1.31



1.29

nowned

1.25

Поэтому

1.24

 ")");

1.22

 しょう

1.22

licité

1.22

Activations Density 0.031%