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

    Neuronpedia

    INDEX
    Explanations

    first person

    np_max-act · gemini-2.0-flash

    The neuron signals on first‐person references, i.e. pronouns and words indicating the speaker’s own viewpoint (I, me, my, we).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Veget
    -0.06
    _solve
    -0.06
     Kag
    -0.06
    Verts
    -0.06
    oultry
    -0.06
    ckill
    -0.06
    Sarah
    -0.06
    closing
    -0.06
    reation
    -0.06
    /per
    -0.06
    POSITIVE LOGITS
     широк
    0.07
    0.06
     girişim
    0.06
     trú
    0.06
    istant
    0.06
     grounds
    0.06
    ?:
    0.06
     '/',↵
    0.06
    	val
    0.06
    _resize
    0.06
    Act Density 0.053%

    No Known Activations