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

    Neuronpedia

    INDEX
    Explanations

    Descriptions

    np_max-act · gemini-2.0-flash

    This neuron activates on simile constructions—especially comparisons introduced by “like a” (e.g. “like a game of Simon,” “like a universal remote control,” etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    だろう
    -0.07
    (arguments
    -0.07
    Administrator
    -0.06
     duplication
    -0.06
    يم
    -0.06
    -filled
    -0.06
     беременности
    -0.06
    альная
    -0.06
     terrifying
    -0.06
    Jones
    -0.06
    POSITIVE LOGITS
    -aligned
    0.07
    Likes
    0.07
     hedge
    0.07
     gle
    0.06
    World
    0.06
    ़क
    0.06
    Shares
    0.06
    Chapter
    0.06
    yyy
    0.06
     ≤
    0.06
    Act Density 0.042%

    No Known Activations