INDEX

Explanations

names and specific terms

This neuron detects the character sequence “Sant” within tokens, as found in names like Santana, Santaner, and Santini.

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

</i>

-2.69

And

-2.59

-2.56

-2.45

いい

-2.39

'})

-2.38

 This

-2.36

’

-2.33

 Should

-2.17

 Where

-2.11

POSITIVE LOGITS

Ꭳ

3.64



3.13

逦

3.00

礽

2.88

 incomparable

2.81

 perché

2.66

因为他

2.63

 perchè

2.58

缦

2.55

檍

2.53

Activations Density 0.007%