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

    Neuronpedia

    INDEX
    Explanations

    flawed data leaks

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     more
    0.56
     It
    0.55
     By
    0.48
     it
    0.47
     More
    0.47
     பெரிய
    0.47
     প্রচুর
    0.47
     normalmente
    0.47
     más
    0.47
     Professional
    0.47
    POSITIVE LOGITS
    Qaeda
    0.53
    usste
    0.53
    cuando
    0.53
     যেদিন
    0.50
    Cuando
    0.49
     عندما
    0.49
    chanted
    0.48
     unveiled
    0.48
    issaient
    0.48
     secretly
    0.47
    Act Density 0.001%

    No Known Activations