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

    Neuronpedia

    INDEX
    Explanations

    company/product codes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     introspection
    0.92
     reformulation
    0.85
     cáncer
    0.84
     comprobar
    0.83
     reinvigor
    0.82
     mendapat
    0.81
    Ɖ
    0.81
     receptions
    0.80
     autobi
    0.80
     uphe
    0.80
    POSITIVE LOGITS
    _
    0.83
    деа
    0.75
    ）。
    0.68
     freaking
    0.66
    aka
    0.66
    '.
    0.65
    и
    0.65
    willing
    0.64
    。
    0.63
    ity
    0.62
    Act Density 0.013%

    No Known Activations