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

    Neuronpedia

    INDEX
    Explanations

    we need or want

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     telecoms
    0.44
     мене
    0.40
    ご了承ください
    0.40
     बैठकर
    0.40
     télé
    0.39
     вашей
    0.39
     षटके
    0.39
     Needless
    0.39
     svým
    0.39
     будь
    0.39
    POSITIVE LOGITS
    我们需要
    0.52
    aremos
    0.51
     necesitamos
    0.50
     want
    0.50
     need
    0.49
    utiliser
    0.48
    この
    0.47
     gonna
    0.47
     можем
    0.46
     хотим
    0.46
    Act Density 0.095%

    No Known Activations