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

    Neuronpedia

    INDEX
    Explanations

    Code

    np_max-act · gemini-2.0-flash

    The neuron flags code terms that denote transformation or conversion operations—e.g. “replace,” “convert,” “Conversion,” etc.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    马
    -0.07
     Housing
    -0.07
    LEMENT
    -0.06
    _disc
    -0.06
    agers
    -0.06
    馬
    -0.06
    dict
    -0.06
     egg
    -0.06
     guild
    -0.06
    Hash
    -0.06
    POSITIVE LOGITS
    "],"
    0.07
    SSL
    0.07
    _REQUIRE
    0.07
     shoved
    0.07
    .'"
    0.07
    YG
    0.06
    EU
    0.06
    ':{'
    0.06
    horia
    0.06
    171
    0.06
    Act Density 0.042%

    No Known Activations