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

    Neuronpedia

    INDEX
    Explanations

    fairy tales

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ल
    0.61
    тся
    0.54
     tây
    0.54
    ную
    0.53
    л
    0.53
    १
    0.53
    й
    0.52
    ↵↵
    0.51
    もし
    0.51
    *
    0.51
    POSITIVE LOGITS
    )。
    0.77
    on
    0.66
    )：
    0.64
    )،
    0.63
    um
    0.63
     for
    0.61
    )，
    0.61
     fable
    0.60
    》。
    0.58
    insurer
    0.57
    Act Density 0.003%

    No Known Activations