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

    Neuronpedia

    INDEX
    Explanations

    code variable initialization

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    いない
    0.43
    eau
    0.43
    いました
    0.43
    evening
    0.42
    லாம்
    0.41
    aap
    0.41
    ல்களில்
    0.40
    نات
    0.38
    ᑐ
    0.38
    运气
    0.38
    POSITIVE LOGITS
     j
    0.59
     i
    0.59
    ye
    0.54
    xt
    0.54
    xa
    0.54
    ği
    0.51
    ota
    0.47
    hi
    0.47
    ̇
    0.46
    ğin
    0.46
    Act Density 0.034%

    No Known Activations