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

    Neuronpedia

    INDEX
    Explanations

    multi-script words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    membered
    0.52
    ryl
    0.52
    reflective
    0.49
    rier
    0.45
    othes
    0.44
    rosis
    0.44
    engers
    0.43
     applicator
    0.43
     grasses
    0.43
    uv
    0.43
    POSITIVE LOGITS
    ס
    0.60
    ሳሪያ
    0.55
    র
    0.50
    噦
    0.48
    蹌
    0.47
    Ꭻ
    0.47
    篂
    0.45
     רי
    0.45
    簌
    0.45
     Noch
    0.45
    Act Density 0.001%

    No Known Activations