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

    Neuronpedia

    INDEX
    Explanations

    problem issue Challenges

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     කොට
    0.52
    FirstFit
    0.51
    利润
    0.50
    asambhavam
    0.50
    𒋗
    0.49
    ﺔ
    0.49
     അമേരിക്ക
    0.48
    ᱙
    0.48
     Конгре
    0.47
    ୧
    0.47
    POSITIVE LOGITS
     (
    0.58
     }
    0.50
    0.48
    č
    0.48
     (*)
    0.47
     p
    0.47
     is
    0.45
     was
    0.45
    C
    0.44
      
    0.44
    Act Density 0.000%

    No Known Activations