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

    Neuronpedia

    INDEX
    Explanations

    convex optimization

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     all
    0.41
     new
    0.37
     one
    0.35
     money
    0.34
    0.33
     really
    0.33
     free
    0.32
     jail
    0.32
     list
    0.32
     bucket
    0.31
    POSITIVE LOGITS
     및
    0.57
     एवं
    0.57
     également
    0.55
     ಅಥವಾ
    0.54
     ಮತ್ತು
    0.54
     También
    0.53
     तथा
    0.53
     आणि
    0.52
     также
    0.52
    ￱
    0.51
    Act Density 0.402%

    No Known Activations