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

    Neuronpedia

    INDEX
    Explanations

    json objects and lists

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ()['
    0.45
    (['
    0.44
    ()["
    0.42
    ѣ
    0.39
    ',['
    0.38
    存档备份
    0.38
    гии
    0.38
    })}{\
    0.38
    ":["
    0.38
     生地
    0.38
    POSITIVE LOGITS
     {
    0.71
    {
    0.59
     ={
    0.57
    ={
    0.55
     निम्नलिखित
    0.52
     {//
    0.51
     다음과
    0.51
    如下
    0.47
     ){
    0.47
    ',{
    0.46
    Act Density 0.132%

    No Known Activations