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

    Neuronpedia

    INDEX
    Explanations

    open weights model

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     compos
    0.44
     rate
    0.43
     compounded
    0.41
     calculations
    0.41
     Rate
    0.41
     compound
    0.41
    בית
    0.39
     compounds
    0.39
    캅
    0.38
     entanglement
    0.38
    POSITIVE LOGITS
    клон
    0.41
     izvr
    0.40
    有所
    0.40
    yllic
    0.39
    amacare
    0.38
     Rentals
    0.38
    geoning
    0.38
    getRuntime
    0.38
    ስ
    0.38
    wan
    0.37
    Act Density 0.000%

    No Known Activations