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

    Neuronpedia

    INDEX
    Explanations

    math operations and variables before operators

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    筈
    0.57
     বর্তমান
    0.55
     ciudad
    0.54
     actuel
    0.54
     wordt
    0.54
     một
    0.53
     morreu
    0.53
     mengatakan
    0.53
     podría
    0.52
     werd
    0.52
    POSITIVE LOGITS
     ballistic
    0.61
    etter
    0.56
     PDEs
    0.54
     biofuels
    0.54
    wapV
    0.54
     बल्
    0.52
     Algebras
    0.52
     warehousing
    0.52
    etrot
    0.49
     radiometric
    0.49
    Act Density 0.207%

    No Known Activations