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

    Neuronpedia

    INDEX
    Explanations

    questions

    np_max-act · gemini-2.0-flash

    The neuron primarily activates on the word “to” when used as the infinitive marker in user requests (e.g. “code to …”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ований
    -0.07
    리아
    -0.07
     UX
    -0.07
    IED
    -0.06
    Rock
    -0.06
    大
    -0.06
     stud
    -0.06
    gui
    -0.06
    ีการ
    -0.06
    �
    -0.06
    POSITIVE LOGITS
     єв
    0.07
     emphasis
    0.06
    .ContentAlignment
    0.06
     Scalars
    0.06
    ("--
    0.06
     Trinity
    0.06
    .Linq
    0.06
    ypo
    0.06
     Cyril
    0.06
    .P
    0.06
    Act Density 0.021%

    No Known Activations