INDEX

Explanations

Electronic identification and other aspects

The neuron flags numeric index arguments passed to child‐retrieval calls (e.g. the integers inside the parentheses of methods like getChildAt or item).

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

for

-1.76

by

-1.72

but

-1.71

one

-1.45

issors

-1.45

 notable

-1.44

to

-1.41

торая

-1.34

 lebt

-1.33

 rationality

-1.33

POSITIVE LOGITS

 been

1.62

Ⴀ

1.59

 forêts

1.54

砹

1.54

肏

1.52

 kilku

1.51

⸢

1.48

⣏

1.47

 keinerlei

1.45

觧

1.45

Activations Density 0.002%