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

    Neuronpedia

    INDEX
    Explanations

    hypothesis formation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ILIO
    0.50
    ar
    0.49
    wa
    0.48
     turi
    0.47
    ih
    0.46
    avk
    0.46
    arco
    0.46
    rests
    0.45
    sized
    0.45
    hij
    0.45
    POSITIVE LOGITS
     Ending
    0.45
     hypothesized
    0.44
     iridescent
    0.43
     endings
    0.43
    یند
    0.41
     Eğer
    0.41
    结尾
    0.41
    াস
    0.41
     وأ
    0.40
    结论
    0.40
    Act Density 0.007%

    No Known Activations