INDEX

Explanations

surplus and deficit

The neuron specifically fires on mentions of “surplus” (i.e. instances of the word surplus indicating excess or leftover quantities).

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

но

-2.77

)):

-2.56

"}}>

-2.50

%")

-2.45

鲧

-2.42

 hört

-2.31

甍

-2.28

 男人

-2.27

➒

-2.25

鮃

-2.23

POSITIVE LOGITS

of

2.98

_{

2.75

<bos>

2.70

？？？

2.48

 also

2.47

当然

2.36

This

2.36

䊐

2.34

也是

2.20

这种

2.20

Activations Density 0.004%