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

    Neuronpedia

    INDEX
    Explanations

    will be held

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     progressed
    0.48
     undertook
    0.47
    進行
    0.46
    いつも
    0.45
     never
    0.44
     никогда
    0.43
     ঘটছে
    0.43
     acabó
    0.43
    Initially
    0.42
     initially
    0.42
    POSITIVE LOGITS
     held
    0.45
     Utility
    0.43
     Held
    0.40
     controllability
    0.39
     motivos
    0.39
     ModelAndView
    0.38
     utility
    0.38
     الار
    0.38
     Weird
    0.37
     disponibile
    0.37
    Act Density 0.003%

    No Known Activations