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

    Neuronpedia

    INDEX
    Explanations

    troubleshooting or expressing opinions

    np_max-act · gemini-2.0-flash

    The neuron primarily detects first-person pronouns and their contractions (e.g. “I,” “I’ve,” “I’m,” “we,” “we’re”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     ksi
    -0.07
    .AreEqual
    -0.06
    _invoke
    -0.06
    ={↵
    -0.06
     pains
    -0.06
    ครบ
    -0.06
     matrices
    -0.06
     Sm
    -0.06
    	rs
    -0.06
    rp
    -0.06
    POSITIVE LOGITS
    iedo
    0.07
    uat
    0.07
    forgot
    0.07
     Islamic
    0.07
     와
    0.06
     الأم
    0.06
    (Pointer
    0.06
     зна
    0.06
    essoa
    0.06
    illaume
    0.06
    Act Density 0.100%

    No Known Activations