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

    Neuronpedia

    INDEX
    Explanations

    :

    np_max-act · gemini-2.0-flash

    The neuron identifies numeric timestamp fragments (e.g., the hour:minute:second parts with optional fractional seconds) in log lines.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    _hor
    -0.08
    _texts
    -0.07
    	ob
    -0.06
    idal
    -0.06
    ักท
    -0.06
    	object
    -0.06
    mmm
    -0.06
    agine
    -0.06
    	msg
    -0.06
    uhn
    -0.06
    POSITIVE LOGITS
    自己的
    0.07
    Priority
    0.07
     sher
    0.07
     ResultSet
    0.06
     whom
    0.06
     залиш
    0.06
    ск
    0.06
     creditor
    0.06
     criminal
    0.06
    三年
    0.06
    Act Density 0.005%

    No Known Activations