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

    Neuronpedia

    INDEX
    Explanations

    only estimates

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     interestingly
    0.90
    Interestingly
    0.84
    Finalmente
    0.80
     interesting
    0.78
     Interestingly
    0.77
     Finalmente
    0.77
     interesantes
    0.73
     inconspicuous
    0.72
     curiously
    0.72
    UpdateChoice
    0.71
    POSITIVE LOGITS
     only
    1.74
    only
    1.64
     лишь
    1.58
    只是
    1.54
     только
    1.53
     Only
    1.46
    仅
    1.46
    あくまで
    1.45
     hanya
    1.44
     лише
    1.44
    Act Density 0.375%

    No Known Activations