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

    Neuronpedia

    INDEX
    Explanations

    separate rooms and beds

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     cancers
    0.39
     abortions
    0.39
     abrog
    0.39
     algorithms
    0.38
     intramolecular
    0.38
     pathogenesis
    0.36
     crimes
    0.35
     militias
    0.35
     trastornos
    0.34
    jectives
    0.34
    POSITIVE LOGITS
     room
    0.62
     Room
    0.62
     bed
    0.61
    房间
    0.59
     Bedroom
    0.59
    房間
    0.59
     bedroom
    0.58
    Room
    0.58
    ベッド
    0.57
     ROOM
    0.56
    Act Density 0.033%

    No Known Activations