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

    Neuronpedia

    INDEX
    Explanations

    safety-focused refusals that empathetically redirect from harmful or inappropriate requests and offer supportive guidance and crisis resources instead of compliance.

    oai_token-act-pair · gpt-5Triggered by @irajjmoradi
    New Auto-Interp
    Negative Logits
     buffs
    0.92
     industrialists
    0.86
     galera
    0.84
     merchants
    0.81
     suka
    0.80
     big
    0.80
     amateurs
    0.80
     stal
    0.79
     popul
    0.78
     consumers
    0.77
    POSITIVE LOGITS
     hopelessness
    1.11
     psychotherapy
    1.08
     compassion
    0.97
     trauma
    0.97
     compassionate
    0.96
     counseling
    0.93
     PTSD
    0.93
     grieve
    0.93
     emotionally
    0.93
     healing
    0.92
    Act Density 2.784%

    No Known Activations