INDEX

Explanations

want to express a desire

The neuron is picking up on words that express desire, intent, or volitional action (e.g. want, like, wish, make, hold, show).

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

 нами

-0.94

 themselves

-0.91

 vérit

-0.88

 ourselves

-0.86

咱们

-0.84

 occuper

-0.83

 لنا

-0.80

zczegól

-0.79

+-+-

-0.79

 yourselves

-0.79

POSITIVE LOGITS

 people

5.28

people

3.27

 others

3.08

People

2.95

 everyone

2.88

人们

2.84

ppl

2.73

 People

2.66

 مردم

2.64

 peoples

2.63

Activations Density 0.120%