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

    Neuronpedia

    INDEX
    Explanations

    refer

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Newton
    -0.07
     Nova
    -0.07
    Math
    -0.07
    xd
    -0.07
     Light
    -0.07
    -distance
    -0.07
    ngth
    -0.07
    -0.07
    -work
    -0.07
    ONT
    -0.06
    POSITIVE LOGITS
     refer
    0.13
     referred
    0.12
     Refer
    0.12
     refers
    0.11
     referring
    0.10
    Refer
    0.10
    ref
    0.09
     PREF
    0.09
    ir
    0.08
     представляет
    0.08
    Act Density 0.020%

    No Known Activations