INDEX

Explanations

structured data and source code

The neuron fires on standalone numeric tokens (i.e. sequences of digits).

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

 blusa

-1.60

 with

-1.45

if

-1.43

🕗

-1.41

 bahagia

-1.40

 italiano

-1.39

 bicicleta

-1.35

for

-1.34

 senang

-1.28

mbps

-1.27

POSITIVE LOGITS

</b>

1.82

1.55

</strong>

1.51

1.47

1.45

1.41

</em>

1.41

…

1.37

…

1.36

Activations Density 0.002%