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

    Neuronpedia

    INDEX
    Explanations

    convert to other formats

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    اً
    1.91
    었던
    1.87
    و
    1.80
    ᅯ
    1.80
    ا
    1.77
    ర్
    1.77
    ان
    1.74
    ════
    1.74
    ARON
    1.74
    ंस
    1.73
    POSITIVE LOGITS
     Converts
    2.41
     Cause
    2.36
     Convert
    2.06
     Cur
    2.05
     Cost
    2.02
     Calcul
    1.96
     Cancel
    1.91
     Crowd
    1.90
     Caus
    1.90
     Filter
    1.89
    Act Density 0.050%

    No Known Activations