INDEX

Explanations

expressing gratitude

The neuron identifies expressions of gratitude—tokens like “thank,” “thanks,” “thankful,” and related forms.

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

to

-1.63

).

-1.57

秶

-1.51

}`

-1.45

臜

-1.43

-1.41

……。

-1.40

iyaki

-1.39

 will

-1.38

tischen

-1.36

POSITIVE LOGITS

žití

1.53

1.47

 calcetines

1.40

ες

1.39

 Menschheit

1.38

itsubishi

1.36

 honnête

1.33

 bertanya

1.31

 écout

1.31

ẩ

1.30

Activations Density 0.010%