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

    Neuronpedia

    INDEX
    Explanations

    terminology related to safety and security in coding and technology

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    safe or unsafe

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    MLLoader
    -0.56
    EClass
    -0.52
    MenuInflater
    -0.52
     HasFactory
    -0.51
    AnchorStyles
    -0.51
    ToMany
    -0.49
    AnchorTagHelper
    -0.49
    ClientRect
    -0.49
    stateProvider
    -0.47
    EndGlobalSection
    -0.47
    POSITIVE LOGITS
    Safe
    2.50
     Safe
    2.47
    safe
    2.36
    SAFE
    2.19
     SAFE
    2.13
     safe
    2.02
    afe
    1.87
    saf
    1.50
     saf
    1.39
     safest
    1.27
    Act Density 0.007%

    No Known Activations