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

    Neuronpedia

    INDEX
    Explanations

    prepositions

    np_max-act · gemini-2.0-flash

    The neuron primarily activates on numeric tokens representing decimal or fractional values (e.g. tokens beginning with “.”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Numbers
    -0.07
     choked
    -0.07
    zd
    -0.06
    -show
    -0.06
     مش
    -0.06
     Lawyer
    -0.06
    Mu
    -0.06
    arnings
    -0.06
     MB
    -0.06
     toes
    -0.06
    POSITIVE LOGITS
     систему
    0.08
     "'
    0.06
     встанов
    0.06
     ICU
    0.06
    rig
    0.06
    ュー
    0.06
    _friends
    0.06
     Coh
    0.06
     Expedition
    0.06
    ::::::::::::::::::::::::::::::::
    0.06
    Act Density 0.088%

    No Known Activations