INDEX

Explanations

university admissions applications and policies

The neuron detects occurrences of the word “admission” or “admissions,” especially in titles and formal contexts.

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

一旁

-2.14

uſe

-2.03

ーズン

-1.95

一个

-1.91

を取る

-1.88

 مختلفة

-1.86

 帅气

-1.80

 Fällen

-1.80

かっこいい

-1.79

躄

-1.79

POSITIVE LOGITS

</strong>

2.48

狆

2.39

瑧

2.39

</h3>

2.36

葙

2.31

2.30

再度

2.25

庒

2.09

 transferring

2.08

 cautiously

2.06

Activations Density 0.003%