INDEX

Explanations

code file formats

The neuron responds to numbered labels or numeric references (e.g. section numbers, legal citations, figure/table numbers).

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

my

-1.69

-1.68

you

-1.50

-1.16

ność

-1.12

ૉ

-1.09

If

-1.09

stoffen

-1.05

领先

-1.04

You

-1.02

POSITIVE LOGITS

 ویتامین

1.42

はじめまして

1.35

こんばん

1.34

 pidana

1.32

饼

1.30

ञ्

1.22

excelente

1.20

 labdar

1.20

 again

1.20

 تعدى

1.19

Activations Density 0.000%