INDEX

Explanations

units and abbreviations

This neuron activates on numeric measurements paired with scientific units or symbols (e.g. “143 GM,” “132 cGy,” “20.625 s,” “8.6 ± 8.0 f Mol/mg protein,” “1.5 degrees,” 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

Даже

-1.41

Any

-1.34

Those

-1.23

chenta

-1.23

Other

-1.20

If

-1.20

After

-1.20

There

-1.19

 really

-1.19

When

-1.17

POSITIVE LOGITS

yczny

1.27

 zeggen

1.25

iiii

1.25

LEMMA

1.23

kring

1.19



1.18

while

1.18

вил

1.17

müş

1.17

넴

1.16

Activations Density 0.088%