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

    Neuronpedia

    INDEX
    Explanations

    You are / we have

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     lohnt
    0.61
     admittedly
    0.61
     bolstering
    0.55
     역사
    0.54
     capace
    0.53
    Oste
    0.53
    Histor
    0.53
     historians
    0.53
     छुटकारा
    0.53
     unquestionably
    0.52
    POSITIVE LOGITS
     dont
    0.95
     have
    0.94
     telah
    0.94
     shall
    0.93
     didnt
    0.91
     אשר
    0.90
     đã
    0.87
     want
    0.87
     provided
    0.87
     will
    0.87
    Act Density 0.218%

    No Known Activations