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

    Neuronpedia

    INDEX
    Explanations

    Insight Timer

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     обеспечи
    0.92
     يس
    0.92
    yout
    0.90
     discarding
    0.89
    ৌ
    0.89
     deliverables
    0.88
     düzen
    0.88
     heuristics
    0.87
    IERC
    0.87
    APPENDIX
    0.86
    POSITIVE LOGITS
    لا
    0.92
    .+\
    0.91
    >(
    0.86
    必要
    0.85
    𝘀
    0.82
    )}\
    0.81
    ても
    0.81
    न
    0.80
     zumindest
    0.79
    க
    0.78
    Act Density 0.001%

    No Known Activations