INDEX

Explanations

Many people or groups

This neuron detects mentions of proportions or quantified groups—phrases expressing statistics about people (e.g. “a quarter of them,” “most of these young people,” “percent of people”).

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

 semua

1.04

all

1.00

 mostly

1.00

 அனைத்தும்

0.98

 tất

0.89

 Mostly

0.86

 kaikki

0.86

 Semua

0.86

 অনেকটা

0.86

mostly

0.84

POSITIVE LOGITS

 perished

0.83

 outright

0.81

 ইতোমধ্যে

0.78

 died

0.77

 ইতিমধ্যে

0.73

 ইতিমধ্যেই

0.71

 ইতো

0.71

 affected

0.70

 ineligible

0.68

(>

0.68

Activations Density 0.500%