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

    Neuronpedia

    INDEX
    Explanations

    install this, stage compiles

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ph
    0.46
    اط
    0.46
    tag
    0.45
    collect
    0.45
    ما
    0.45
    k
    0.44
    яз
    0.44
    ì
    0.44
    ಶಸ್
    0.44
    PH
    0.44
    POSITIVE LOGITS
    FirstFit
    0.54
     texted
    0.53
    >-->
    0.50
     воздух
    0.50
     extruder
    0.50
     remarked
    0.49
     evade
    0.49
     kleuren
    0.49
    ᴺ
    0.49
     dors
    0.49
    Act Density 0.000%

    No Known Activations