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

    Neuronpedia

    INDEX
    Explanations

    setting environment variables

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    杝
    0.47
     outdated
    0.44
     proliferating
    0.44
    consistent
    0.44
     exams
    0.43
    感動
    0.43
     конца
    0.43
     ascendent
    0.42
     insistent
    0.42
    ன
    0.42
    POSITIVE LOGITS
    Ling
    0.49
    y
    0.48
    Wei
    0.46
    ères
    0.46
    éraux
    0.46
    ované
    0.45
     рез
    0.44
    ร้าย
    0.44
    Nes
    0.44
    nings
    0.43
    Act Density 0.001%

    No Known Activations