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

    Neuronpedia

    INDEX
    Explanations

    start

    np_max-act · gemini-2.0-flash

    module and component identifiers related to electronics.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron is specifically activating on numeric and decimal values (e.g. floating‐point numbers in specifications).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     bức
    -0.07
     şik
    -0.07
    LI
    -0.06
    /pi
    -0.06
     Ni
    -0.06
     pathlib
    -0.06
    Parsing
    -0.06
     Hz
    -0.06
    _corr
    -0.06
    _struct
    -0.06
    POSITIVE LOGITS
     dakika
    0.07
     biodiversity
    0.07
     пн
    0.06
     bookstore
    0.06
    insi
    0.06
    AMERA
    0.06
     discard
    0.06
     现
    0.06
     NEED
    0.06
     само
    0.06
    Act Density 0.001%

    No Known Activations