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

    Neuronpedia

    INDEX
    Explanations

    no-no

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     caregivers
    0.40
    အစိတ်အပိုင်း
    0.39
     SM
    0.38
    ?/
    0.37
     gär
    0.37
     పూ
    0.35
    CRE
    0.35
    .?
    0.35
     антенна
    0.35
    Env
    0.35
    POSITIVE LOGITS
     exchange
    0.74
     Exchange
    0.72
    Exchange
    0.68
    exchange
    0.65
    交換
    0.58
     এক্সচে
    0.56
    交换
    0.55
     обмен
    0.55
     échange
    0.55
     EXCHANGE
    0.54
    Act Density 0.000%

    No Known Activations