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

    Neuronpedia

    INDEX
    Explanations

    abusive or offensive language

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    エ
    0.39
    මන්
    0.38
    مش
    0.38
    جز
    0.38
    ಂಜ
    0.38
    الم
    0.37
     ponemos
    0.37
    います
    0.37
     Essas
    0.36
    ሁሉም
    0.36
    POSITIVE LOGITS
     argues
    0.45
    itrile
    0.43
    RowFilter
    0.42
     argued
    0.42
    vée
    0.42
     argument
    0.41
    rot
    0.40
    ករណ៍
    0.40
     grandchild
    0.40
    ell
    0.40
    Act Density 0.011%

    No Known Activations