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

    Neuronpedia

    INDEX
    Explanations

    leaks and rumors of releases

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     announced
    0.63
    公布
    0.59
    umumkan
    0.59
     annoncé
    0.59
     announcement
    0.57
     anunció
    0.57
     Announced
    0.57
    announced
    0.56
     announce
    0.55
     발표
    0.55
    POSITIVE LOGITS
     leakage
    0.64
     prototypes
    0.63
    Leak
    0.61
     camouflage
    0.59
     Leak
    0.59
     prototype
    0.57
     leak
    0.57
    camouflage
    0.54
    leak
    0.54
     leaks
    0.53
    Act Density 0.007%

    No Known Activations