INDEX

Explanations

place names and specific names

The neuron activates on floating‐point numerals (i.e. numbers containing a decimal point).

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

</strong>

-3.58

↵

-2.84

紧接着

-2.67

毉

-2.34

 victories

-2.34

-2.33

 varieties

-2.27

幸好

-2.27

踯

-2.22



-2.20

POSITIVE LOGITS

);

2.59

</td>

2.44

and

2.38

2.25

我們

2.23

Kondisi

2.22

</b>

2.17

2.16

If

2.13

Activations Density 0.082%