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

    Neuronpedia

    INDEX
    Explanations

    comma

    np_max-act · gemini-2.0-flash

    The neuron strongly activates on numeric tokens (e.g. years, channel numbers, digits) or tokens containing digits.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     FINAL
    -0.06
    -call
    -0.06
    opez
    -0.06
     из
    -0.06
     '<
    -0.06
     Nat
    -0.06
    あげ
    -0.06
     band
    -0.06
    ôi
    -0.06
    png
    -0.06
    POSITIVE LOGITS
    油
    0.07
     Alberto
    0.07
    crypto
    0.06
    AttributeValue
    0.06
     illustr
    0.06
    ubyte
    0.06
    상
    0.06
     latina
    0.06
     Sabha
    0.06
    _minimum
    0.06
    Act Density 0.071%

    No Known Activations