INDEX

Explanations

introducing additional information

The neuron fires on discourse markers that introduce an afterthought or added clarification (e.g. “I forgot to mention,” “oh, also,” “add…”).

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

 جدید

-1.50

 ürünleri

-1.37

bedaan

-1.29

েই

-1.28

yym

-1.28

viking

-1.27

resenta

-1.26

and

-1.23

antos

-1.23

ÑOS

-1.21

POSITIVE LOGITS

 other

1.66

can

1.35

</em>

1.34

يت

1.27

 might

1.26

[]:

1.26

 друго

1.24

，

1.23

pö

1.23

夢想

1.20

Activations Density 0.025%