INDEX

Explanations

all the features

The neuron activates on numeric tokens, especially decimal or measurement values.

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

of

-1.57

 hanno

-1.45

 lakini

-1.37

 soldat

-1.34

-¡

-1.30

 baumwolle

-1.30

IBU

-1.29

-1.28

か

-1.26

POSITIVE LOGITS

 Both

1.83

not

1.80

 both

1.63

 When

1.43

 Neither

1.42

Both

1.38

的東西

1.38

řízen

1.38

氏が

1.35

1.34

Activations Density 0.058%