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

    Neuronpedia

    INDEX
    Explanations

    subject followed by would

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ಇದ್ದ
    0.43
     đang
    0.41
     đó
    0.41
    "}}>
    0.41
     seem
    0.40
    解决了
    0.39
    雖然
    0.38
    虽然
    0.38
     следующем
    0.38
     हालांकि
    0.37
    POSITIVE LOGITS
     would
    0.97
     Would
    0.97
    would
    0.94
    Would
    0.93
    acakt
    0.89
     devait
    0.88
     skulle
    0.86
     würde
    0.83
     হইত
    0.81
     seraient
    0.81
    Act Density 0.006%

    No Known Activations