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

    Neuronpedia

    INDEX
    Explanations

    identification words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ogado
    0.44
     pooling
    0.37
    itektur
    0.37
    myCollision
    0.37
    ঘুম
    0.36
    迫
    0.36
     গা
    0.36
    समझ
    0.36
    ববার
    0.36
    霆
    0.36
    POSITIVE LOGITS
     gossip
    0.41
    iken
    0.40
     RTS
    0.40
     recognised
    0.39
     erkennen
    0.39
     kent
    0.38
     PPS
    0.38
     CCS
    0.38
    键
    0.37
     getSize
    0.36
    Act Density 0.000%

    No Known Activations