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

    Neuronpedia

    INDEX
    Explanations

    fighter

    np_max-act · gemini-2.0-flash

    The neuron responds to Italian-language words (e.g. tema, poetica, opera, satira).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Achie
    -0.07
    Incorrect
    -0.07
    /List
    -0.06
     obedient
    -0.06
     impro
    -0.06
     розгля
    -0.06
    計算
    -0.06
    ตลอด
    -0.06
    ιο
    -0.06
     remed
    -0.06
    POSITIVE LOGITS
    NETWORK
    0.07
    باب
    0.06
    /packages
    0.06
    bet
    0.06
    	http
    0.06
    phere
    0.06
     мир
    0.06
     testosterone
    0.06
    .norm
    0.06
    core
    0.06
    Act Density 0.628%

    No Known Activations