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

    Neuronpedia

    INDEX
    Explanations

    way things are done

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     not
    0.42
     wondering
    0.41
    த்திற்க
    0.40
     không
    0.39
     more
    0.39
    partial
    0.38
     Pin
    0.38
     শুধু
    0.38
     wondered
    0.37
     ഉണ്ടായി
    0.37
    POSITIVE LOGITS
     방식
    0.61
     manier
    0.57
    对待
    0.57
     manière
    0.56
    方式
    0.54
    看待
    0.54
     manner
    0.52
     interacts
    0.50
     demean
    0.50
     طریقے
    0.50
    Act Density 0.165%

    No Known Activations