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

    Neuronpedia

    INDEX
    Explanations

    for

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    [block
    -0.07
     detections
    -0.07
     relationship
    -0.07
    Steps
    -0.07
    /location
    -0.06
     Differences
    -0.06
     (*
    -0.06
     differences
    -0.06
     customers
    -0.06
     terr
    -0.06
    POSITIVE LOGITS
    誉
    0.07
     zorun
    0.07
     esposa
    0.06
    _f
    0.06
    ')
↵
    0.06
     Gerry
    0.06
    dot
    0.06
    нів
    0.06
    essoa
    0.06
    適
    0.06
    Act Density 0.052%

    No Known Activations