INDEX

Explanations

mm followed by separator

The neuron activates on any token containing the substring “mm,” e.g. in identifiers or package names like mm_…, com.tencent.mm, 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

↵

-3.25

“

-3.06

<em>

-2.73

…

-2.58

 yaklaşık

-2.44

と

-2.41

-2.39

 There

-2.38

But

-2.36

感じで

-2.36

POSITIVE LOGITS

 processus

2.39

 extré

2.34

 使っ

2.25

鬭

2.22

 pattes

2.20

蚣

2.19

馓

2.17

 composants

2.16

ाउनलोड

2.16

皞

2.14

Activations Density 0.007%