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

    Neuronpedia

    INDEX
    Explanations

    describing actions or states of being

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     becoming
    0.73
     become
    0.66
    成为
    0.63
     becomes
    0.58
     incluir
    0.55
     diventare
    0.55
    变成
    0.55
     joining
    0.55
     became
    0.54
     Joining
    0.54
    POSITIVE LOGITS
     bezig
    1.01
     đang
    0.98
     sedang
    0.96
     مشغول
    0.95
     pondering
    0.94
    preparing
    0.94
     frantically
    0.93
    正在
    0.93
    Trying
    0.93
    Preparing
    0.89
    Act Density 0.389%

    No Known Activations