INDEX

Explanations

gastrectomy, pyre, antipyretic, transgender

This neuron detects the letters "re" in words, particularly in medical or technical terminology.

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

-3.19

-2.94

-2.88

-2.70

-2.64

-2.61

-2.52

-2.41

-2.38

ೣ

-2.28

POSITIVE LOGITS

 tendremos

3.02

逦

2.86

</h3>

2.61

re

2.56

 Opiniones

2.41

ญี่ป

2.34

 Introducción

2.28

çok

2.27

 podríamos

2.27

 Instead

2.23

Activations Density 0.008%