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

    Neuronpedia

    INDEX
    Explanations

    safeguards

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     جال
    0.52
     Interesting
    0.45
    Interesting
    0.45
    Compact
    0.42
     padassa
    0.42
     Trends
    0.42
     Spatial
    0.41
     Leider
    0.41
     Weird
    0.41
    Networking
    0.40
    POSITIVE LOGITS
     safeguards
    1.46
     oversight
    1.34
     scrutiny
    1.19
     accountability
    1.05
     protections
    1.04
     safeguard
    1.04
     mechanisms
    1.02
     assurances
    0.99
     safeguarding
    0.98
     rigorous
    0.97
    Act Density 0.383%

    No Known Activations