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

    Neuronpedia

    INDEX
    Explanations

    appeal

    np_max-act · gemini-2.0-flash

    The neuron detects occurrences of “app”–root tokens in appellate or appeal‐related legal terminology.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ocre
    -0.09
    Algorithm
    -0.07
    áce
    -0.07
    ्यप
    -0.07
    арі
    -0.07
    trap
    -0.06
    ��
    -0.06
    처럼
    -0.06
    ably
    -0.06
    parable
    -0.06
    POSITIVE LOGITS
     aime
    0.07
     attire
    0.06
    .undo
    0.06
     Cypress
    0.06
     Anim
    0.06
     UIAlert
    0.06
    /features
    0.06
     statist
    0.06
     defeats
    0.06
     bone
    0.06
    Act Density 0.003%

    No Known Activations