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

    Neuronpedia

    INDEX
    Explanations

    vintage, horse, Labs

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    directly
    0.44
    Brewer
    0.42
    ophan
    0.39
    હા
    0.39
    Down
    0.39
    Па
    0.39
    ractable
    0.39
    Scal
    0.38
    棹
    0.38
    Phys
    0.37
    POSITIVE LOGITS
     एडवांस
    0.46
     परेड
    0.44
     stewards
    0.43
     Gäste
    0.41
     heavy
    0.40
     Advance
    0.40
    了一些
    0.40
     Bxa
    0.40
     giraffe
    0.39
     assaulted
    0.39
    Act Density 0.001%

    No Known Activations