INDEX

Explanations

college grades and grading

This neuron detects mentions of academic grades or grade requirements, especially when paired with specific grade values (letter grades or numeric scores).

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

efits

-1.50

🢒

-1.49

he

-1.48

리아

-1.48

hooter

-1.48

edrez

-1.47

-1.46

usitis

-1.45

廃止

-1.45

POSITIVE LOGITS

BeanFactory

1.43

 bahwa

1.36

𝙚

1.29

quần

1.27

了一句

1.25

を作る

1.25

 می‌تواند

1.24

蛲

1.23

高兴

1.21

on

1.20

Activations Density 0.007%