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

    Neuronpedia

    INDEX
    Explanations

    concise alternatives

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ᾖ
    0.49
    ionine
    0.46
     definisi
    0.45
    uksen
    0.43
    ouses
    0.43
    ouk
    0.42
    モ
    0.42
     bhavanti
    0.42
     mirada
    0.42
    enas
    0.41
    POSITIVE LOGITS
     परस्मैपदी
    0.49
    ၃
    0.47
     Surprisingly
    0.46
    膻
    0.46
     ایل
    0.44
     Writes
    0.44
     Strange
    0.44
    $"
    0.43
    தல்
    0.43
     पीट
    0.43
    Act Density 0.002%

    No Known Activations