INDEX

Explanations

closing punctuation or symbols

The neuron mainly activates on numeric tokens—years, measurements, and other numbers.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 Honestly

1.38

 Thankfully

1.38

 Luckily

1.33

 😊

1.25

 Ironically

1.20

 Didn

1.18

 Wouldn

1.16

 Definitely

1.16

 Personally

1.13

 Interestingly

1.13

POSITIVE LOGITS

및

0.63

0.61

،

0.60

，

0.57

 expts

0.57

、

0.56

分野

0.56

->

0.55

，

0.54

--->

0.54

Activations Density 0.187%