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

    Neuronpedia

    INDEX
    Explanations

    prepositions

    np_max-act · gemini-2.0-flash

    This neuron detects words that express personal interest, intent, or motivational stance (e.g. “interested,” “forced,” “stating,” “suggested,” “driven”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     want
    -0.07
    Metro
    -0.06
     fa
    -0.06
    xp
    -0.06
    BUG
    -0.06
    -insert
    -0.06
    Northern
    -0.06
    “It
    -0.06
    ennial
    -0.05
    wanted
    -0.05
    POSITIVE LOGITS
     ALSO
    0.07
     TODAY
    0.06
    	AM
    0.06
     contraction
    0.06
     chalk
    0.06
     coordinating
    0.06
    _owner
    0.06
    ，每
    0.06
    _UNITS
    0.06
    (parameter
    0.06
    Act Density 0.261%

    No Known Activations