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

    Neuronpedia

    INDEX
    Explanations

    unpacking data structures

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     perturbative
    0.75
    ievement
    0.74
    ppled
    0.73
     丨
    0.73
     worsens
    0.71
    vadipine
    0.70
     reduzir
    0.70
     impeach
    0.70
     giường
    0.70
     endogenous
    0.70
    POSITIVE LOGITS
    و
    0.86
    де
    0.76
    ол
    0.73
    ’
    0.70
    彩
    0.69
     alma
    0.67
    cine
    0.66
    c
    0.66
    А
    0.66
    ка
    0.66
    Act Density 0.014%

    No Known Activations