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

    Neuronpedia

    INDEX
    Explanations

    PFLAG

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Compressed
    0.42
    Eff
    0.41
     אנ
    0.38
    אנ
    0.38
    ]?
    0.38
     flip
    0.37
     sfor
    0.37
     만들어
    0.37
    ولات
    0.36
     flipping
    0.36
    POSITIVE LOGITS
    дача
    0.45
    𝓂
    0.44
    𝚔
    0.44
    ática
    0.43
     contrato
    0.41
    ева
    0.41
    ueux
    0.40
     usual
    0.39
     कुमारी
    0.39
     अपनी
    0.39
    Act Density 0.001%

    No Known Activations