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

    Neuronpedia

    INDEX
    Explanations

    challenging, difficult

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ’
    1.26
    ა
    0.98
    ن
    0.96
    want
    0.96
    '
    0.93
    ité
    0.92
    mselves
    0.92
    いた
    0.91
    tenant
    0.91
    chten
    0.90
    POSITIVE LOGITS
     khăn
    1.07
    ის
    0.96
     unsuccessfully
    0.90
     चिल्ड्रन
    0.90
    Еще
    0.89
    騇
    0.88
    ००
    0.86
    फ्ते
    0.86
     Targets
    0.86
    꼐
    0.84
    Act Density 0.089%

    No Known Activations