INDEX

Explanations

checking conditions

The neuron is triggered by the “.” token when used in object-oriented property/method access (e.g. object.method or object.property).

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

 simply

-1.64

 eenvou

-1.59

 only

-1.55

 merely

-1.51

simply

-1.45

 instead

-1.42

⎘

-1.41

and

-1.38

 either

-1.38

 mostly

-1.38

POSITIVE LOGITS

This

1.57

 после

1.43

time

1.43

 или

1.41

If

1.38

ра

1.36

What

1.36



1.33

Does

1.33

It

1.31

Activations Density 0.005%