© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 9365
    Prev
    Next
    INDEX
    Explanations

    empathy and compassion

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron responds to words naming acts or qualities of compassion and care (e.g. mercy, grace, compassion, caring).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     excitement
    0.58
     दमदार
    0.58
     बवाल
    0.58
    嚣
    0.56
    मिथ
    0.55
    alles
    0.54
    ɤ
    0.54
    战斗
    0.54
    ワク
    0.54
     rythme
    0.54
    POSITIVE LOGITS
     compassion
    2.39
     compassionate
    2.36
     kindness
    2.05
     empathetic
    1.83
     Kindness
    1.81
     empathy
    1.79
     empath
    1.74
     pity
    1.72
    compass
    1.70
    Compass
    1.70
    Activations Density 0.649%

    No Known Activations