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

    Neuronpedia

    INDEX
    Explanations

    Weak signals

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     fulfilled
    -0.09
     roses
    -0.08
     Ful
    -0.08
    polygon
    -0.08
    ldr
    -0.07
     hashing
    -0.07
     Ravens
    -0.07
    Bow
    -0.07
     Bever
    -0.07
     bishops
    -0.07
    POSITIVE LOGITS
     faint
    0.13
    Detection
    0.13
     Detection
    0.12
     tini
    0.12
     detectable
    0.12
     Detect
    0.11
    _detection
    0.11
    弱
    0.11
     detection
    0.11
    Detect
    0.11
    Act Density 0.018%

    No Known Activations