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

    Neuronpedia

    INDEX
    Explanations

    constant

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     reverse
    -0.09
     teaming
    -0.09
     ples
    -0.08
    reverse
    -0.08
     dinosaur
    -0.08
    .reverse
    -0.08
     Alfonso
    -0.07
    بيا
    -0.07
     algebra
    -0.07
     were
    -0.07
    POSITIVE LOGITS
     incess
    0.11
     relentless
    0.11
    不停
    0.10
     khiến
    0.10
     endlessly
    0.09
     endless
    0.09
     constamment
    0.09
     constantly
    0.09
     everywhere
    0.09
     cravings
    0.09
    Act Density 0.026%

    No Known Activations