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

    Neuronpedia

    INDEX
    Explanations

    making requests or actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ثلاثة
    0.47
    reet
    0.44
    şan
    0.40
     veter
    0.39
     اثنين
    0.39
    quete
    0.38
     absch
    0.38
    mesinin
    0.38
    ussen
    0.37
     이날
    0.37
    POSITIVE LOGITS
    敏感
    0.38
     Bowls
    0.38
    痕
    0.38
     podrían
    0.37
     medallions
    0.37
    oha
    0.37
     haunted
    0.37
     grids
    0.36
    ہیں
    0.35
     dodgy
    0.35
    Act Density 0.002%

    No Known Activations