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

    Neuronpedia

    INDEX
    Explanations

    constructor calling

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    本
    0.44
    ').':
    0.43
    用
    0.42
    }:
    0.42
     disconnect
    0.41
    ):
    0.41
     train
    0.41
    豆瓣
    0.40
     generic
    0.40
     allgeme
    0.40
    POSITIVE LOGITS
    pall
    0.44
    hares
    0.43
    ré
    0.43
    thiaz
    0.42
    elong
    0.42
    araja
    0.41
    rég
    0.40
    °
    0.40
    gebaut
    0.40
     (°
    0.39
    Act Density 0.001%

    No Known Activations