INDEX

Explanations

reporting past events

This neuron detects mentions of the author or outlet reporting or writing (e.g. “I wrote,” “we reported,” “we posted”).

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

 дополнительных

-0.96

 further

-0.91

not

-0.90

本日の

-0.88

autant

-0.86

よくある

-0.85

んだから

-0.84

.")

-0.84

modificar

-0.83

lamations

-0.83

POSITIVE LOGITS

 HERE

1.73

 blog

1.55

HERE

1.38

 here

1.38

 предыду

1.27

 aqui

1.25

 blogs

1.25

 aquí

1.23

 здесь

1.20

 هنا

1.20

Activations Density 0.123%