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

    Neuronpedia

    INDEX
    Explanations

    German sentences describing complex verb constructs, especially those involving modal verbs like "können," "müssen," and "hätte."

    oai_token-act-pair · gemini-2.0-flash

    German language

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     الحره
    -0.69
    gitte
    -0.52
    saraba
    -0.50
     adjour
    -0.50
    oa̍t
    -0.49
     đốc
    -0.48
     betweenstory
    -0.48
    war
    -0.47
     gainera
    -0.47
    anthene
    -0.46
    POSITIVE LOGITS
    complexContent
    0.73
     dürfen
    0.62
     devront
    0.62
     mußten
    0.58
     sollen
    0.58
    cektir
    0.58
    XmlAccessType
    0.57
     wollt
    0.55
     wollten
    0.55
    Datuak
    0.53
    Act Density 1.597%

    No Known Activations