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

    Neuronpedia

    INDEX
    Explanations

    able

    np_max-act · gemini-2.0-flash

    the neuron detects tokens that are parts of contractions (pieces containing an apostrophe).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
    again
    -0.07
    Enumeration
    -0.06
    .crop
    -0.06
    .borderWidth
    -0.06
     compte
    -0.06
    ัวหน
    -0.06
     booming
    -0.06
     vrát
    -0.06
    .origin
    -0.06
     parted
    -0.06
    POSITIVE LOGITS
    -component
    0.07
     війсь
    0.07
     />
    0.06
    (mapStateToProps
    0.06
    _PREFIX
    0.06
    正在
    0.06
    aviest
    0.06
     vouchers
    0.06
    opsis
    0.06
    arah
    0.06
    Act Density 0.183%

    No Known Activations