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

    Neuronpedia

    INDEX
    Explanations

    simulations, neural networks, ads

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     multipl
    0.38
     Peut
    0.37
     certains
    0.37
     بعض
    0.36
     Plusieurs
    0.36
     certaines
    0.34
     algunas
    0.34
     two
    0.34
     plusieurs
    0.33
     algunos
    0.33
    POSITIVE LOGITS
    而非
    0.46
    iquement
    0.42
    norod
    0.39
    protein
    0.38
     ovviamente
    0.37
     పూర్తిగా
    0.37
    valho
    0.37
    boxylate
    0.37
     wyłącznie
    0.36
    只有
    0.36
    Act Density 0.139%

    No Known Activations