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

    Neuronpedia

    INDEX
    Explanations

    needs, thinks, means

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     noticing
    0.36
     preferring
    0.36
     sine
    0.36
     Tasks
    0.36
    ↵↵↵↵↵↵
    0.35
     interesado
    0.34
     Request
    0.34
     requesting
    0.34
     pueden
    0.33
     flavoring
    0.33
    POSITIVE LOGITS
     rightfully
    0.57
     ought
    0.55
     rightful
    0.48
     truly
    0.47
    事实上
    0.44
     deems
    0.42
     dama
    0.41
     rightly
    0.41
     gerekti
    0.41
    本来
    0.40
    Act Density 0.021%

    No Known Activations