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

    Neuronpedia

    INDEX
    Explanations

    g-man, tar, gas, rice, buckets, char

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ש
    0.84
    ີ
    0.77
    潙
    0.76
    م
    0.75
     veineux
    0.71
    ሳሪያ
    0.70
    zonych
    0.69
    როს
    0.69
    isées
    0.68
    жие
    0.68
    POSITIVE LOGITS
    0
    0.96
     was
    0.95
     for
    0.95
     with
    0.93
     that
    0.91
    for
    0.91
    0.90
     {
    0.88
     at
    0.82
    was
    0.81
    Act Density 0.000%

    No Known Activations