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

    Neuronpedia

    INDEX
    Explanations

    say consult

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     accessing
    0.68
     utilizing
    0.63
     engaging
    0.63
     interacted
    0.61
     engages
    0.59
     connecting
    0.59
     accessed
    0.59
     contacting
    0.55
     Engaging
    0.55
     interacting
    0.55
    POSITIVE LOGITS
    Kons
    1.12
    kons
    1.09
    Cons
    1.07
     kons
    1.06
    CONS
    1.06
     Kons
    1.05
     consul
    1.01
    cons
    0.99
     Consul
    0.95
    consult
    0.93
    Act Density 0.014%

    No Known Activations