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

    Neuronpedia

    INDEX
    Explanations

    Hidden objects/story elements

    np_max-act · gemini-2.0-flash

    This neuron activates on numeric tokens (floating-point numbers and digits).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    能源
    -0.07
     NIGHT
    -0.06
    	z
    -0.06
    (Call
    -0.06
    왔다
    -0.06
    _SSL
    -0.06
     bacteria
    -0.06
    Њ
    -0.06
    handling
    -0.06
    942
    -0.06
    POSITIVE LOGITS
     accompagn
    0.07
     indian
    0.07
     erot
    0.06
    .AspNet
    0.06
    /octet
    0.06
    =input
    0.06
     oblivious
    0.06
    Indian
    0.06
    ्रद
    0.06
    pol
    0.06
    Act Density 0.104%

    No Known Activations