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

    Neuronpedia

    INDEX
    Explanations

    combine

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     preseason
    0.65
     selected
    0.61
     progetto
    0.61
    স্তি
    0.61
     regulation
    0.59
    入門
    0.57
     summar
    0.56
    ikit
    0.56
     project
    0.54
    プロジェクト
    0.54
    POSITIVE LOGITS
     combine
    1.04
     Combine
    1.02
    combine
    0.96
    Combine
    0.91
     combines
    0.88
     combiner
    0.71
     combining
    0.68
     combinar
    0.64
     combin
    0.62
     Combining
    0.58
    Act Density 0.009%

    No Known Activations