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

    Neuronpedia

    INDEX
    Explanations

    cloud resource names

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    膻
    0.46
     leech
    0.40
     groupBox
    0.39
     knife
    0.39
    菜
    0.38
     ganglion
    0.38
     filet
    0.38
     shear
    0.38
     getter
    0.37
     mites
    0.37
    POSITIVE LOGITS
    }-${
    0.47
    dev
    0.46
    Unique
    0.44
    idev
    0.44
    app
    0.43
    prod
    0.43
    unique
    0.43
    your
    0.42
    deployment
    0.42
    PREFIX
    0.41
    Act Density 0.009%

    No Known Activations