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

    Neuronpedia

    INDEX
    Explanations

    definitions using relative clauses

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     другие
    1.00
    Он
    0.90
    После
    0.88
     других
    0.84
    Как
    0.84
     други
    0.84
    это
    0.83
    ವಾರು
    0.83
     respectivas
    0.81
    Sebagai
    0.81
    POSITIVE LOGITS
     that
    2.73
     whose
    2.60
     which
    2.56
     που
    2.33
    that
    2.31
    which
    2.27
     الذي
    2.25
     که
    2.22
     التي
    2.17
    whose
    2.10
    Act Density 0.219%

    No Known Activations