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

    Neuronpedia

    INDEX
    Explanations

    sequence of actions and processing

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    获取
    0.64
    Capture
    0.58
    Fetching
    0.57
     capturing
    0.55
     fetching
    0.54
    Grab
    0.54
    Retrieve
    0.53
    拿起
    0.52
     retrieve
    0.52
     seizing
    0.52
    POSITIVE LOGITS
    放到
    0.75
     processing
    0.73
    加工
    0.72
     processed
    0.71
     Processing
    0.68
     PROCESSING
    0.68
     almacen
    0.65
     加工
    0.64
    将其
    0.64
     następnie
    0.63
    Act Density 0.033%

    No Known Activations