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

    Neuronpedia

    INDEX
    Explanations

    very clear boundary setting

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ό
    0.75
     velké
    0.71
     cited
    0.70
     motiva
    0.70
     reciting
    0.70
     Ν
    0.68
     આરોપી
    0.66
    ॉ
    0.66
    ithi
    0.66
     hablan
    0.66
    POSITIVE LOGITS
    leneck
    0.69
    ^{-
    0.63
    ⠉
    0.63
    அந்த
    0.59
    장을
    0.57
    Forgotten
    0.56
    stressed
    0.56
    ^{
    0.55
    iyorum
    0.55
    }^{-
    0.55
    Act Density 0.092%

    No Known Activations