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

    Neuronpedia

    INDEX
    Explanations

    tikz code blocks

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    <start_of_image>
    1.07
     dominion
    0.82
     VPN
    0.79
     Negative
    0.78
     Genetic
    0.78
     Mileage
    0.77
     austenite
    0.76
     Shang
    0.75
     Prom
    0.74
    »;
    0.73
    POSITIVE LOGITS
    ]{
    2.10
    ){
    1.69
    )]{
    1.54
     ){
    1.48
    }]{
    1.40
    "){
    1.34
    ]){
    1.34
    (){
    1.32
    ()){
    1.31
    '){
    1.28
    Act Density 0.001%

    No Known Activations