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

    Neuronpedia

    INDEX
    Explanations

    track names and add commands

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Po
    0.58
     phenomen
    0.57
    -
    0.55
     Lie
    0.55
     s
    0.55
     Six
    0.55
     Les
    0.55
     grav
    0.54
     Eden
    0.53
     legacy
    0.51
    POSITIVE LOGITS
     获取
    0.88
     用于
    0.88
    筛选
    0.87
    ྛ
    0.85
     添加
    0.84
    删除
    0.84
    过滤
    0.83
    没有
    0.83
     或者
    0.83
    ྞ
    0.83
    Act Density 1.009%

    No Known Activations