INDEX

Explanations

followed by a word

This neuron detects mentions of a programming context, especially the phrase “in C#.”

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

 only

-1.63

 this

-1.57

out

-1.37

 numai

-1.32

one

-1.29

 только

-1.29

 using

-1.23

زای

-1.23

 things

-1.22

Гц

-1.22

POSITIVE LOGITS

会有

1.48

 !!!!!

1.43

 XXVII

1.41

harapkan

1.41

Θ

1.39

 XXVIII

1.38

 XXIX

1.36

XLV

1.33

 XXIV

1.32

 XXXI

1.32

Activations Density 0.052%