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

    Neuronpedia

    INDEX
    Explanations

    arrows and directions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Gonz
    0.43
    **/
    0.43
    derick
    0.42
    пас
    0.41
    ضم
    0.41
    Climate
    0.41
    Hull
    0.40
    Madame
    0.39
    erk
    0.39
    Sport
    0.39
    POSITIVE LOGITS
     ↑
    0.63
    ↑
    0.59
    ↓
    0.56
     ↓
    0.54
    uparrow
    0.52
    ↓↓
    0.47
    上が
    0.45
    上げる
    0.45
     jumping
    0.45
    上の
    0.44
    Act Density 0.001%

    No Known Activations