INDEX

Explanations

account information or accounts

The neuron is triggered by any occurrence of the word “account” (in any capitalization or as part of an identifier).

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

 ของ

-2.11

 kulinar

-2.09

 неболь

-2.06

盌

-1.96

翃

-1.95

ちょっとした

-1.92

-1.88

 continuar

-1.86

–

-1.80

たまた

-1.78

POSITIVE LOGITS

镭

2.59

驀

2.20

激烈

2.20

<bos>

2.11

 сестра

2.03

 beina

1.95

十二月

1.94

文化的

1.93

.‏

1.90

酝

1.88

Activations Density 0.005%