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

    Neuronpedia

    INDEX
    Explanations

    however, contrast

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     surprising
    0.45
     quirky
    0.37
     compiles
    0.36
     çox
    0.36
     disks
    0.36
     wedges
    0.36
     nyingi
    0.36
     compile
    0.35
     inj
    0.35
     surprisingly
    0.35
    POSITIVE LOGITS
    忑
    0.41
    validate
    0.41
    validation
    0.39
    不必
    0.39
    ameni
    0.38
    swering
    0.38
     memastikan
    0.38
    ද්ග
    0.38
     Terima
    0.38
    和社会
    0.38
    Act Density 0.009%

    No Known Activations