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

    Neuronpedia

    INDEX
    Explanations

    ball bench

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     SUN
    -0.08
     soap
    -0.07
    antom
    -0.07
     Sun
    -0.07
     Download
    -0.07
     technicians
    -0.06
    илися
    -0.06
     FM
    -0.06
    華
    -0.06
    ete
    -0.06
    POSITIVE LOGITS
    ονται
    0.07
    _EPS
    0.06
    .loggedIn
    0.06
    mutation
    0.06
     erf
    0.06
     cartesian
    0.06
    ).
↵
    0.05
     lstm
    0.05
     iken
    0.05
    look
    0.05
    Act Density 0.010%

    No Known Activations