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

    Neuronpedia

    INDEX
    Explanations

    key presses

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    opp
    0.49
     opa
    0.49
    ებლ
    0.48
    oy
    0.46
     ops
    0.45
    mark
    0.44
    jö
    0.44
    oyin
    0.44
     oma
    0.44
    mess
    0.44
    POSITIVE LOGITS
     to
    0.54
     Horizon
    0.50
    امج
    0.49
    缓冲
    0.49
     Buffer
    0.48
    范
    0.46
    伟
    0.46
     CGI
    0.45
     horizon
    0.43
     Riemann
    0.43
    Act Density 0.000%

    No Known Activations