INDEX

Explanations

p<number>

The neuron is tuned to detect floating‐point numeric tokens (numbers containing a decimal point).

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 than

0.75

 Auckland

0.75

 graduated

0.72

 positions

0.71

 comput

0.67

 computed

0.64

 bending

0.64

 Knowles

0.63

 contradicts

0.63

 sings

0.63

POSITIVE LOGITS

mdat

0.79

SalesRep

0.76

Բ

0.75

RefreshToken

0.71

ᱨ

0.71

jęcia

0.70

authenticate

0.68

ⴹ

0.68

accessToken

0.68

ładow

0.67

Activations Density 0.000%