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

    Neuronpedia

    INDEX
    Explanations

    extracting digit from number

    np_max-act · gemini-2.0-flash

    This neuron activates on numeric tokens that are decimal (floating-point) numbers.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Hentai
    -0.08
     aup
    -0.06
    _CAT
    -0.06
    .ny
    -0.06
     detta
    -0.06
    (-(
    -0.06
     yếu
    -0.06
     enact
    -0.06
     cherish
    -0.06
     Watt
    -0.06
    POSITIVE LOGITS
     indo
    0.06
    �
    0.06
    ereum
    0.06
    subscribe
    0.06
     virtues
    0.06
     ()=>{↵
    0.06
     requestOptions
    0.06
     음
    0.06
     buffers
    0.06
    (){↵↵
    0.06
    Act Density 0.002%

    No Known Activations