INDEX

Explanations

Kerala and Malayalam

The neuron strongly responds to mentions of specific place names (cities, districts, landmarks) in Kerala, India.

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

POSED

-1.44

缰

-1.43

 defa

-1.40

 GOING

-1.30

lavia

-1.30

oku

-1.29

wendungen

-1.28

vání

-1.23

кзак

-1.23

 promesse

-1.23

POSITIVE LOGITS

()

1.61

1.56

 might

1.52

1.46

).”

1.40

may

1.39

1.35

}$

1.33

?”

1.32

1.30

Activations Density 0.058%