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

    Neuronpedia

    INDEX
    Explanations

    Dealing with problems/stress

    np_max-act · gemini-2.0-flash

    The neuron consistently activates on words expressing personal emotional states or distress (e.g. “feeling,” “overwhelmed,” “angry,” “struggling”), so it’s detecting tokens that convey someone’s emotions.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     kan
    -0.07
    碼
    -0.07
    (of
    -0.07
    /mail
    -0.06
    Apple
    -0.06
    .network
    -0.06
    Since
    -0.06
     rog
    -0.06
    zen
    -0.06
     astronomical
    -0.06
    POSITIVE LOGITS
     @
    0.08
     initialState
    0.07
     '↵↵
    0.07
     Proceedings
    0.06
    inject
    0.06
     UIImagePickerController
    0.06
                                    ↵
    0.06
     menggunakan
    0.06
                                            ↵
    0.06
    .getWorld
    0.06
    Act Density 0.044%

    No Known Activations