INDEX

Explanations

because / punctuation

The neuron fires strongly on relatively uncommon, content-bearing words (i.e. non-stopwords) rather than on frequent function words.

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

but

1.04

 lakini

1.04

 pero

1.02

but

0.90

 אבל

0.90

 before

0.89

BUT

0.87

 mutta

0.84

但在

0.83

 then

0.82

POSITIVE LOGITS

：

1.11

 Bởi

1.10

 কেননা

1.08

因为

1.05

：

1.05

 เนื่องจาก

1.04

 gdyż

1.02

 ponieważ

1.01

 زیرا

0.99

 क्योंकि

0.98

Activations Density 1.078%