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

    Neuronpedia

    INDEX
    Explanations

    why not

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Temperatura
    0.71
    ight
    0.69
    ua
    0.64
    Ṕ
    0.64
     Temperatur
    0.63
     Bly
    0.59
     Wordsworth
    0.59
    ATH
    0.58
     Prothorax
    0.58
     Photography
    0.57
    POSITIVE LOGITS
    ان
    0.85
    zellen
    0.71
     actos
    0.67
    an
    0.65
     unwind
    0.65
     carcin
    0.64
     improvised
    0.64
    ères
    0.63
     amortization
    0.61
     rolled
    0.61
    Act Density 0.005%

    No Known Activations