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

    Neuronpedia

    INDEX
    Explanations

    Consequences

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     overlying
    0.59
     idiosyncratic
    0.57
     idiosync
    0.57
     predominance
    0.57
    过度
    0.56
     overcomes
    0.55
     unstructured
    0.54
     disturbances
    0.54
     perturbations
    0.54
    hegan
    0.54
    POSITIVE LOGITS
     ostr
    1.02
     consecuencias
    1.00
     consequences
    0.96
     conséquences
    0.96
     punishable
    0.95
     repercussions
    0.93
    exposure
    0.92
    锒
    0.89
    Exposure
    0.89
     Consequences
    0.89
    Act Density 0.226%

    No Known Activations