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

    Neuronpedia

    INDEX
    Explanations

    bad followed by negative descriptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     مزید
    0.41
     ಸರ್
    0.41
     Sil
    0.41
    ベージュ
    0.39
    もう
    0.38
     エ
    0.38
     Sip
    0.38
     ochrony
    0.37
     سر
    0.36
     производителя
    0.36
    POSITIVE LOGITS
     bad
    0.66
    坏
    0.66
     BAD
    0.60
    Bad
    0.59
     খারাপ
    0.57
     stereotypes
    0.57
     बद
    0.57
    minton
    0.55
    nesses
    0.54
     kötü
    0.53
    Act Density 0.043%

    No Known Activations