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

    Neuronpedia

    INDEX
    Explanations

    code parameters and keywords

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Th
    0.57
    Level
    0.56
    Tomato
    0.56
     徐
    0.53
     शिव
    0.53
    Hyd
    0.53
    Dev
    0.53
    Cher
    0.52
     Stephen
    0.52
    DR
    0.51
    POSITIVE LOGITS
     title
    0.74
    title
    0.71
    appName
    0.70
    标题
    0.70
     titel
    0.69
    名称
    0.67
     BrowserModule
    0.66
     Title
    0.64
     vijj
    0.64
     hostname
    0.61
    Act Density 0.133%

    No Known Activations