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

    Neuronpedia

    INDEX
    Explanations

    still useful

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    s
    0.57
    0.50
     percent
    0.44
     percentage
    0.44
     rules
    0.43
      
    0.43
     code
    0.42
     council
    0.42
     pinc
    0.42
     Jo
    0.41
    POSITIVE LOGITS
    癶
    0.61
     ฝึก
    0.57
     dissati
    0.56
     βοη
    0.56
    0.56
    ងឺ
    0.55
    <unused1869>
    0.55
    <unused1825>
    0.54
     sistemat
    0.54
     学習
    0.54
    Act Density 0.002%

    No Known Activations