INDEX

Explanations

code import statements

The neuron fires on dotted library or API namespace segments in code (e.g. module, computed, spreadsheet, utils, layout, etc.).

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

When

-1.45

೬

-1.45

時間

-1.41

不同

-1.41

Another

-1.34

、

-1.29

one

-1.29

-1.27

 when

-1.27

Numerous

-1.25

POSITIVE LOGITS

thanks

1.66

 chré

1.60

 firſt

1.58

despite

1.56

ㆆ

1.54

🧌

1.53

鹛

1.53

since

1.53

quelles

1.48

being

1.46

Activations Density 0.028%