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

    Neuronpedia

    INDEX
    Explanations

    importance and possibility

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    geom
    0.66
    Its
    0.65
    早速
    0.64
    ITS
    0.64
    就有
    0.63
    他就
    0.62
     Conductivity
    0.61
    நுட்ப
    0.61
     concat
    0.61
    xyz
    0.61
    POSITIVE LOGITS
     impossible
    1.61
     possible
    1.61
     important
    1.49
     understandable
    1.38
     clear
    1.38
     conceivable
    1.35
     easy
    1.35
    possible
    1.33
     unclear
    1.31
     tempting
    1.29
    Act Density 0.464%

    No Known Activations