INDEX

Explanations

sequence after certain words

The neuron fires most strongly on the very first content word in a new text segment—i.e. it marks the start of each snippet.

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

 многих

-1.49

 České

-1.41

InSeconds

-1.39

*>

-1.34

 ранее

-1.33

aes

-1.24

⚋

-1.23

 besö

-1.20

 многие

-1.20

 personalizadas

-1.20

POSITIVE LOGITS

 here

1.41

</h2>

1.38

are

1.34

 certainly

1.29

 want

1.23

 immagine

1.20

 same

1.20

に関しては

1.19

 grandeza

1.18

its

1.17

Activations Density 0.058%