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

    Neuronpedia

    INDEX
    Explanations

    aprend

    np_max-act · gemini-2.0-flash

    The neuron selectively responds to Romance‐language terms built on the root “aprend-,” i.e. words referring to the concept of learning.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Fi
    -0.07
    uali
    -0.07
     �
    -0.07
    олі
    -0.06
     Goal
    -0.06
    cul
    -0.06
     rl
    -0.06
    oul
    -0.06
     Circle
    -0.06
    贸
    -0.06
    POSITIVE LOGITS
     aprend
    0.08
    	AND
    0.07
     upp
    0.07
    -only
    0.07
     από
    0.07
     امر
    0.07
    _apps
    0.07
     Apprent
    0.07
    	add
    0.07
    από
    0.07
    Act Density 0.019%

    No Known Activations