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

    Neuronpedia

    INDEX
    Explanations

    there or they

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     übrigens
    0.45
     oczywiście
    0.44
    ôté
    0.44
     deluge
    0.42
    حه
    0.42
    也将
    0.42
     appunto
    0.41
     طبعا
    0.41
    athlon
    0.40
     sights
    0.40
    POSITIVE LOGITS
    there
    1.16
    There
    1.13
     there
    1.05
     There
    0.94
    THERE
    0.86
     هناك
    0.82
     THERE
    0.82
    They
    0.81
     theres
    0.75
     they
    0.73
    Act Density 0.012%

    No Known Activations