INDEX

Explanations

obsessive focus and obsession

The neuron detects occurrences of the root “obsess” in any morphological form (e.g., obsessive, obsession, obsessed).

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

."

-1.97

of

-1.67

䜣

-1.58

 nossas

-1.55

 ajuda

-1.54

ﮨ

-1.54

matitis

-1.53

 costado

-1.46

,"

-1.42

:'

-1.41

POSITIVE LOGITS

 Untuk

1.46

 focus

1.41

ńst

1.40

or

1.36

 prevent

1.34

唎

1.34

 avoid

1.33

1.30

 obsession

1.27

 همه

1.27

Activations Density 0.030%