INDEX

Explanations

last updated information

The neuron detects date‐ or time‐stamp metadata (e.g. “Last updated” timestamps and their associated numbers).

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

But

-1.73

tiously

-1.60

tuce

-1.43



-1.41

いしい

-1.39

 ($('#

-1.38

</h2>

-1.36

zzz

-1.34

ztes

-1.28

不敢

-1.27

POSITIVE LOGITS

 их

1.57

reportWebVitals

1.38

Takk

1.38

Makanan

1.37

 działanie

1.37

 loro

1.36

Armin

1.35

Fourni

1.34

Beweg

1.34

Spesifikasi

1.34

Activations Density 0.018%