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

    Neuronpedia

    INDEX
    Explanations

    don't just blindly rely

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    没有什么
    0.46
     purposefully
    0.41
     intentionally
    0.40
     precau
    0.39
    🤨
    0.38
     provoqu
    0.38
     expres
    0.37
     unmistak
    0.37
     volut
    0.37
     deliberately
    0.36
    POSITIVE LOGITS
     blindly
    1.21
     rely
    1.11
     relying
    1.07
     Rely
    0.99
     solely
    0.98
     reliance
    0.90
    reliance
    0.86
     relied
    0.85
     blind
    0.80
     relies
    0.80
    Act Density 0.118%

    No Known Activations