INDEX

Explanations

explanations and improvements

The neuron triggers on standalone decimal numbers (floating-point values) in technical output.

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

())

0.73

());

0.69

()),

0.61

したので

0.60

ULAR

0.56

()">

0.55

.)

0.55

 रु

0.55

()){

0.54

POSITIVE LOGITS

 Particularly

0.63

 市場

0.60

ocysts

0.58

 אף

0.58

 ビジネス

0.57

 Também

0.57

 לא

0.56

 Obviously

0.55

 risposta

0.55

ﻻ

0.55

Activations Density 0.003%