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

    Neuronpedia

    INDEX
    Explanations

    fillers in [ ] brackets

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     palate
    0.34
     tom
    0.34
     cooks
    0.33
    pads
    0.33
     haloes
    0.32
     these
    0.32
     underlie
    0.32
     diesen
    0.31
     tunn
    0.31
     chén
    0.31
    POSITIVE LOGITS
    here
    0.55
    звание
    0.54
     এখানে
    0.53
     Aquí
    0.53
     اینجا
    0.53
     mention
    0.52
     هنا
    0.52
     Here
    0.52
     HERE
    0.52
     название
    0.52
    Act Density 0.094%

    No Known Activations