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

    Neuronpedia

    INDEX
    Explanations

    Asking and telling

    np_max-act · gemini-2.0-flash

    This neuron activates on first-person self-references that include “I” immediately followed by some form of “ask” (e.g., “I asked,” “I had asked”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     spat
    -0.06
     mbox
    -0.06
    Titulo
    -0.06
    (dt
    -0.06
     drž
    -0.06
    kání
    -0.06
     PLUGIN
    -0.06
    ữa
    -0.06
    än
    -0.06
    Seleccione
    -0.06
    POSITIVE LOGITS
    规定
    0.07
    ’ve
    0.06
     Correspond
    0.06
    ець
    0.06
     Oasis
    0.06
    _parm
    0.06
    (parseInt
    0.06
    earable
    0.06
    _ERR
    0.06
     normalized
    0.06
    Act Density 0.018%

    No Known Activations