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

    Neuronpedia

    INDEX
    Explanations

    identifying as or not as

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     некоторых
    0.44
    简直
    0.42
     основных
    0.42
     alguma
    0.41
     alguna
    0.41
     algún
    0.41
     যেসব
    0.41
     algum
    0.41
    なんか
    0.40
     somehow
    0.40
    POSITIVE LOGITS
     ultimately
    0.61
     fundamentally
    0.54
     unequivocally
    0.50
     firstly
    0.50
     categorically
    0.50
     emphatically
    0.49
     uiteindelijk
    0.49
     neither
    0.48
     اولا
    0.46
    Ultimately
    0.46
    Act Density 0.046%

    No Known Activations