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

    Neuronpedia

    INDEX
    Explanations

    explicitly prohibit

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ។
    0.42
     wynik
    0.39
     lakini
    0.38
    看似
    0.38
     nhưng
    0.38
     ngunit
    0.38
     resembling
    0.38
     Still
    0.37
     morphology
    0.37
    但不
    0.37
    POSITIVE LOGITS
    致力于
    0.57
     encourages
    0.52
     justement
    0.49
     encour
    0.47
     recognizes
    0.47
     помочь
    0.47
     помога
    0.46
    经常
    0.46
    ช่วย
    0.45
     تاکید
    0.45
    Act Density 0.280%

    No Known Activations