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

    Neuronpedia

    INDEX
    Explanations

    when followed by actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     domination
    0.38
     konnten
    0.38
    能夠
    0.38
     acknowledgment
    0.37
     সাফল্য
    0.37
     mampu
    0.36
     achieves
    0.36
     បាន
    0.35
     Having
    0.35
     capables
    0.35
    POSITIVE LOGITS
     dealing
    0.79
     preparing
    0.73
     discussing
    0.69
     designing
    0.69
    dealing
    0.64
     Dealing
    0.61
     evaluating
    0.61
    Dealing
    0.60
     deciding
    0.56
     التعامل
    0.56
    Act Density 0.010%

    No Known Activations