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

    Neuronpedia

    INDEX
    Explanations

    contributes to harm

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ஒப்பு
    0.47
     onus
    0.46
     상관
    0.46
    isição
    0.44
    අ
    0.44
     Verfü
    0.43
     actualizaciones
    0.43
    𖥔
    0.43
     처
    0.42
    讣
    0.42
    POSITIVE LOGITS
    یدہ
    0.52
     prevent
    0.48
     robot
    0.47
     busy
    0.44
    し
    0.44
     stop
    0.44
     Chinese
    0.44
    戲
    0.44
     market
    0.43
    管理的
    0.43
    Act Density 0.002%

    No Known Activations