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

    Neuronpedia

    INDEX
    Explanations

    recommend

    np_max-act · gemini-2.0-flash

    This neuron activates on words and phrasing that refer to making or giving recommendations (e.g. “recommendation,” “recomendar,” etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ('.')↵
    -0.07
     Dry
    -0.06
     swapped
    -0.06
    apyrus
    -0.06
     функци
    -0.06
    нім
    -0.06
     дан
    -0.06
    Paper
    -0.06
    :',↵
    -0.06
    ोश
    -0.06
    POSITIVE LOGITS
     voy
    0.06
    傷
    0.06
    gıç
    0.06
     recommendation
    0.06
     supplemental
    0.06
    Eq
    0.06
    ได
    0.06
    ักษณ
    0.06
     фото
    0.06
     بسبب
    0.06
    Act Density 0.012%

    No Known Activations