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

    Neuronpedia

    INDEX
    Explanations

    reason

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    ่าค
    -0.07
    �
    -0.07
     stresses
    -0.06
     kelim
    -0.06
     pew
    -0.06
     Cush
    -0.06
    apos
    -0.06
     mound
    -0.06
    astreet
    -0.06
     يع
    -0.06
    POSITIVE LOGITS
     Lars
    0.07
    _dyn
    0.07
     eye
    0.07
    foot
    0.06
     piss
    0.06
    	errors
    0.06
    Secret
    0.06
     filenames
    0.06
    Protection
    0.06
     complement
    0.06
    Act Density 0.009%

    No Known Activations