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

    Neuronpedia

    INDEX
    Explanations

    reinforcement learning

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    unie
    0.71
    uni
    0.65
     аса
    0.64
    loge
    0.63
    友
    0.62
    LOG
    0.61
     downwards
    0.61
     summation
    0.60
     unis
    0.60
    sohn
    0.60
    POSITIVE LOGITS
     رہ
    0.74
     про
    0.73
    parse
    0.64
     permanent
    0.64
     penetration
    0.63
     perme
    0.63
    ناک
    0.61
    pered
    0.60
    manu
    0.60
    פא
    0.60
    Act Density 0.132%

    No Known Activations