Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    defining classes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    いくつ
    0.89
    Also
    0.82
    сным
    0.80
     extrémités
    0.77
    FAQs
    0.77
    солю
    0.76
    insuff
    0.76
    absorbed
    0.75
    ≥
    0.75
    orylation
    0.74
    POSITIVE LOGITS
     class
    1.11
     Simple
    0.96
     simple
    0.90
     factory
    0.90
     custom
    0.86
     my
    0.85
    Simple
    0.85
     process
    0.83
     Custom
    0.83
     sınıf
    0.80
    Act Density 0.104%

    No Known Activations