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

    Neuronpedia

    INDEX
    Explanations

    sensory taste descriptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ی
    0.53
    ?
    0.52
    ä
    0.51
    nement
    0.50
    gebaut
    0.49
    ру
    0.49
    sterreich
    0.47
    ъ
    0.47
     commande
    0.46
    ς
    0.46
    POSITIVE LOGITS
    ↵↵
    0.67
     as
    0.54
    య
    0.51
    ع
    0.49
    ↵
    0.48
    al
    0.46
    h
    0.46
    d
    0.45
    g
    0.45
     exons
    0.45
    Act Density 0.009%

    No Known Activations