INDEX

Explanations

food items

The neuron chiefly detects mentions of edible items—food and drink names.

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

esanan

-1.40

 меры

-1.38

 отече

-1.34

 зимние

-1.32

 специалисты

-1.31

too

-1.30

 again

-1.30

全く

-1.29

ed

-1.28

 sort

-1.28

POSITIVE LOGITS

 Когда

1.34

we

1.32



1.19

崎

1.18

 häufig

1.18

 vielmehr

1.17

گیرد

1.17

bicara

1.15

quele

1.13

Activations Density 0.094%