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

    Neuronpedia

    INDEX
    Explanations

    #

    np_max-act · gemini-2.0-flash

    The neuron lights up on software version numbers and other numeric tokens with decimal points (e.g. “3.1”, “0.2607”, “--build-arg”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    녀
    -0.09
    ENC
    -0.07
    pressed
    -0.07
    Hidden
    -0.07
     standing
    -0.07
    .addSubview
    -0.07
     тверд
    -0.07
     싶
    -0.06
    \Controllers
    -0.06
     rhythm
    -0.06
    POSITIVE LOGITS
    PLY
    0.07
    entreprise
    0.06
     Belgium
    0.06
     splice
    0.06
    statements
    0.06
    .");
↵
    0.06
     Portsmouth
    0.06
     Shelby
    0.06
    .DEFAULT
    0.06
    -pagination
    0.06
    Act Density 0.006%

    No Known Activations