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

    Neuronpedia

    INDEX
    Explanations

    people and user groups

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Corollary
    0.65
    ி
    0.65
    en
    0.64
     המ
    0.63
     donc
    0.61
     soin
    0.61
     khu
    0.61
     Adapun
    0.61
     guise
    0.61
     אף
    0.60
    POSITIVE LOGITS
    sthe
    0.82
    nf
    0.79
    IMUM
    0.77
    nte
    0.75
    na
    0.72
    行う
    0.71
    tenance
    0.70
    م
    0.70
    رخ
    0.70
    tg
    0.69
    Act Density 0.206%

    No Known Activations