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

    Neuronpedia

    INDEX
    Explanations

    soul

    np_max-act · gemini-2.0-flash

    This neuron activates on the word “soul,” detecting uses of that token.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    _AdjustorThunk
    -0.07
    onders
    -0.07
    .BackgroundImageLayout
    -0.07
     Araştır
    -0.06
     kaufen
    -0.06
     Caucasian
    -0.06
    pec
    -0.06
     방법
    -0.06
    ucus
    -0.06
     abolition
    -0.06
    POSITIVE LOGITS
     маг
    0.07
     resend
    0.07
    /max
    0.06
     hack
    0.06
     Comple
    0.06
    maint
    0.06
     schl
    0.06
    Cancellation
    0.06
     Disconnect
    0.06
     Km
    0.06
    Act Density 0.002%

    No Known Activations