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

    Neuronpedia

    INDEX
    Explanations

    lo-fi or saddle

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     rodeo
    0.64
     Emma
    0.52
     autism
    0.51
     antagonistic
    0.50
     Rodeo
    0.49
     clown
    0.48
    ید
    0.47
     ASU
    0.47
    פ
    0.47
     emanc
    0.47
    POSITIVE LOGITS
    abhas
    0.55
    你看
    0.52
     లేదా
    0.51
     بين
    0.50
    órd
    0.50
    鵝
    0.50
    _{\
    0.49
    ewel
    0.49
    ě
    0.49
     bölün
    0.47
    Act Density 0.000%

    No Known Activations