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

    Neuronpedia

    INDEX
    Explanations

    actions and their purpose

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    짙
    0.41
    міні
    0.39
    뭇
    0.39
     plume
    0.38
    טן
    0.38
    нце
    0.37
     corro
    0.37
     isle
    0.37
    ты
    0.36
     plumes
    0.36
    POSITIVE LOGITS
    and
    0.46
     strategia
    0.43
    group
    0.42
    among
    0.42
     Strategies
    0.41
     Schwarzschild
    0.40
     nonfiction
    0.40
    esor
    0.40
     Brand
    0.39
    Brand
    0.39
    Act Density 0.001%

    No Known Activations