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

    Neuronpedia

    INDEX
    Explanations

    numerical ratios

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     কতকগুলি
    0.42
    fraction
    0.37
     комплекс
    0.37
     sheriff
    0.37
     Fractional
    0.36
    sher
    0.36
    णा
    0.36
    filterable
    0.36
    coeff
    0.36
     suatu
    0.36
    POSITIVE LOGITS
     split
    0.68
     Split
    0.61
     splitting
    0.58
    ／
    0.55
     evenly
    0.54
    split
    0.52
     ratio
    0.52
     Splits
    0.52
     splits
    0.51
    Split
    0.50
    Act Density 0.010%

    No Known Activations