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

    Neuronpedia

    INDEX
    Explanations

    keep out of reach

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    LG
    0.40
    涉及
    0.39
    胯
    0.38
    ಳಿ
    0.37
     agave
    0.36
    trp
    0.36
    ুদ্ধে
    0.36
    TRAS
    0.36
     లక్ష
    0.36
     Alerts
    0.36
    POSITIVE LOGITS
     inaccessible
    0.61
     away
    0.59
     curioso
    0.56
    好奇
    0.55
     curious
    0.53
     पहुंच
    0.52
     accessible
    0.51
     unreachable
    0.50
     curiosity
    0.50
     دسترس
    0.47
    Act Density 0.004%

    No Known Activations