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

    Neuronpedia

    INDEX
    Explanations

    hyperparameters

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     (
    0.55
     that
    0.41
     άνθρω
    0.38
    iejsc
    0.37
     Muay
    0.36
    ’
    0.36
    ierzchn
    0.36
     владе
    0.36
    leza
    0.36
    that
    0.35
    POSITIVE LOGITS
    ة
    0.49
     esempio
    0.43
     unei
    0.43
     の
    0.43
     egy
    0.41
     eine
    0.41
    q
    0.41
    した
    0.41
     হয়েছিল
    0.41
    ിൽ
    0.41
    Act Density 2.519%

    No Known Activations