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

    Neuronpedia

    INDEX
    Explanations

    unlocking

    np_max-act · gemini-2.0-flash

    This neuron primarily activates on words and phrases that describe opening containers or locks (e.g., “open,” “unlock,” “jar opener”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    (zip
    -0.07
    >,</
    -0.06
    xA
    -0.06
    laws
    -0.06
    ica
    -0.06
    BACK
    -0.06
    .setFill
    -0.06
    Iter
    -0.06
     Sergio
    -0.06
     返回
    -0.06
    POSITIVE LOGITS
     otevř
    0.07
     obese
    0.07
     دیگری
    0.07
     �
    0.07
    Enabled
    0.06
    iao
    0.06
     hafif
    0.06
    vrir
    0.06
     Eylül
    0.06
    
    0.06
    Act Density 0.174%

    No Known Activations