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

    Neuronpedia

    INDEX
    Explanations

    you might can should

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Lik
    0.39
    ksam
    0.39
    番
    0.36
    RENGTH
    0.36
    เดิม
    0.36
    ）。
    0.35
    >.
    0.35
    Oriental
    0.35
    ）.
    0.34
    Kelas
    0.34
    POSITIVE LOGITS
    必须要
    0.54
     fortunately
    0.48
     necesitas
    0.46
    必须
    0.44
    しなければ
    0.44
     phải
    0.43
     thankfully
    0.43
    giore
    0.43
     occasionally
    0.42
     இருந்தால்
    0.42
    Act Density 0.017%

    No Known Activations