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

    Neuronpedia

    INDEX
    Explanations

    differential

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Handling
    -0.07
    DOMAIN
    -0.07
    hodob
    -0.07
    MAP
    -0.06
    galement
    -0.06
     spouses
    -0.06
    521
    -0.06
    都会
    -0.06
     Jon
    -0.06
     Lucas
    -0.06
    POSITIVE LOGITS
     стак
    0.07
     onlar
    0.07
     isteyen
    0.06
    ":"+
    0.06
     honored
    0.06
     оно
    0.06
     alley
    0.06
    /Delete
    0.06
    .Win
    0.06
     lor
    0.06
    Act Density 0.005%

    No Known Activations