INDEX

Explanations

process

The neuron strongly activates on occurrences of the word “process.”

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

隳

-3.33

-3.27

’

-2.53

</strong>

-2.52

 améli

-2.45

ity

-2.41

 hugely

-2.31

ところです

-2.28



-2.17

 ſaid

-2.14

POSITIVE LOGITS

cbd

2.63

2.58

})

2.45

También

2.41

◳

2.39

 всѣ

2.39

 保護

2.36

鋮

2.36

</h5>

2.34

งิน

2.34

Activations Density 0.058%