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

    Neuronpedia

    INDEX
    Explanations

    security filters or firewalls

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     performing
    0.65
     muscular
    0.63
     anticipation
    0.61
     utterances
    0.61
     platforms
    0.60
     filtr
    0.60
     infrastructure
    0.59
     connections
    0.59
     accelerating
    0.59
     instability
    0.58
    POSITIVE LOGITS
     zieht
    0.68
    ACA
    0.59
    ंना
    0.58
    вши
    0.57
     zehn
    0.56
    inis
    0.56
    in
    0.55
    ijnlijk
    0.53
    DAQ
    0.53
    áp
    0.52
    Act Density 0.001%

    No Known Activations