INDEX

Explanations

list of types

The neuron mainly fires on numeric and monetary tokens (e.g. prices, dollar amounts, quantities).

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 division

0.63

 mistress

0.58

 fission

0.57

division

0.57

onuclear

0.55

 convexity

0.55

 дея

0.54

henes

0.54

 ratione

0.54

boundingRect

0.54

POSITIVE LOGITS

 شامل

0.81

 includes

0.79

 inclui

0.79

税込

0.75

Includes

0.75

 Includes

0.74

 incluye

0.74

includes

0.72

美元

0.71

 دلار

0.71

Activations Density 0.348%