INDEX

Explanations

errors or mistakes

This neuron detects long runs of repeated punctuation characters (like sequences of dots, dashes, or equals signs) used as visual separators.

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

狽

-0.84

 ControllerBase

-0.78

aruhi

-0.77

ZIONI

-0.77

 screening

-0.72

 assurance

-0.72

ppa

-0.70

دارة

-0.70

}_{

-0.68

 نہیں

-0.68

POSITIVE LOGITS

 error

1.38

 Error

1.27

 mistake

1.24

 оши

1.16

 Mistake

1.05

Error

0.99

 ошибка

0.98

err

0.97

 ERROR

0.97

error

0.96

Activations Density 0.001%