INDEX

Explanations

operations and processes

This neuron activates on numeric tokens and values (e.g. counts, measurements, or other numbers) in the text.

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

-2.03

了过来

-1.80

-1.66

等我

-1.63

などは

-1.59

了过去

-1.54

'':

-1.52

 there

-1.48

良かったです

-1.46

she

-1.43

POSITIVE LOGITS

</em>

2.34

㌧

1.59

摃

1.58

 forder

1.52

 těchto

1.52

푆

1.45

klü

1.45

匐

1.45

 Costs

1.42

 dalších

1.42

Activations Density 0.271%