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

    Neuronpedia

    INDEX
    Explanations

    elliptic curves, matrices, and groups

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     eigenvalues
    0.44
     Value
    0.43
     வரவே
    0.41
    ग्राहक
    0.41
     оттен
    0.40
     Orders
    0.40
    la
    0.40
     запол
    0.40
     vän
    0.40
     Worte
    0.39
    POSITIVE LOGITS
    之为
    0.50
    四
    0.48
    㡀
    0.48
    டக
    0.47
    은
    0.44
    和
    0.43
    ਆਂ
    0.43
    Iam
    0.42
    హం
    0.42
    清新
    0.41
    Act Density 0.001%

    No Known Activations