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

    Neuronpedia

    INDEX
    Explanations

    omit

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     TREE
    -0.07
    (suffix
    -0.06
    >r
    -0.06
    ze
    -0.06
     Jah
    -0.06
    stru
    -0.06
     Mane
    -0.06
     Graz
    -0.06
     Fra
    -0.06
    PY
    -0.06
    POSITIVE LOGITS
     omitted
    0.11
     omit
    0.10
     omission
    0.08
     withholding
    0.07
     repeated
    0.07
     municip
    0.06
     emission
    0.06
     Obs
    0.06
     formatter
    0.06
    /Input
    0.06
    Act Density 0.004%

    No Known Activations