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

    Neuronpedia

    INDEX
    Explanations

    -offset

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     streamed
    -0.07
     WITH
    -0.07
     Sam
    -0.07
     shaken
    -0.07
     washing
    -0.07
    _done
    -0.07
    coming
    -0.07
    cents
    -0.06
    uzey
    -0.06
    ictory
    -0.06
    POSITIVE LOGITS
    -offset
    0.12
    915
    0.06
    _offset
    0.06
    ----------------------------------------------------------------------
    0.06
    -middle
    0.06
    _offsets
    0.06
    +t
    0.06
    cheap
    0.06
    phan
    0.06
     Blink
    0.06
    Act Density 0.000%

    No Known Activations