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

    Neuronpedia

    INDEX
    Explanations

    units of measurement

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    崞
    0.49
     अधू
    0.46
     میان
    0.45
    囵
    0.45
    सेप्शन
    0.44
    㜖
    0.44
    㹮
    0.44
    ",[],"
    0.44
    䟝
    0.44
    棫
    0.44
    POSITIVE LOGITS
    8
    0.61
     tiny
    0.57
    ↵
    0.55
    5
    0.54
    9
    0.54
     (<
    0.51
     micro
    0.50
     ,
    0.49
    4
    0.47
     (.
    0.47
    Act Density 0.128%

    No Known Activations