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

    Neuronpedia

    INDEX
    Explanations

    key-value pairs

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    uggish
    0.45
     threaten
    0.44
    issants
    0.43
    स्थ
    0.43
    oros
    0.42
     მაშინ
    0.41
     recipients
    0.40
    สห
    0.40
    が行
    0.40
    ственности
    0.40
    POSITIVE LOGITS
     KV
    0.45
    t
    0.44
    AMP
    0.43
    val
    0.41
     VA
    0.40
     pressed
    0.38
     UVA
    0.38
    ovaniyu
    0.38
    iy
    0.38
     Harv
    0.38
    Act Density 0.010%

    No Known Activations