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

    Neuronpedia

    INDEX
    Explanations

    offset and shifting

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Strictly
    0.39
     নিয়ন্ত্র
    0.38
    扩散
    0.38
     inscribed
    0.38
     createTask
    0.38
    采集
    0.37
     Plea
    0.36
     Falling
    0.36
    篤
    0.36
     Gentle
    0.35
    POSITIVE LOGITS
     offset
    0.94
    Offset
    0.89
    offset
    0.89
     Offset
    0.81
     offsets
    0.81
    OFFSET
    0.79
     offsetting
    0.79
     OFFSET
    0.69
    GetOffset
    0.68
    offsets
    0.68
    Act Density 0.122%

    No Known Activations