INDEX

Explanations

comma followed by especially

The neuron strongly activates on proper-noun or named-entity mentions (e.g. species, place names, product/feature names, organization or role titles).

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 মতোই

0.55

れば

0.54

 rồi

0.52

んだろう

0.49

 никакого

0.47

={

0.47

(!

0.47

={

0.47

=",

0.47

 沒有

0.46

POSITIVE LOGITS

 especially

4.97

 особенно

4.84

 particularly

4.72

 especialmente

4.72

especially

4.69

 특히

4.69

特に

4.66

 Especially

4.56

특히

4.56

Especially

4.50

Activations Density 1.137%