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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 8470
    Prev
    Next
    INDEX
    Explanations

    health and well-being

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

    The neuron activates on terms related to health, wellness, and well-being (e.g. health, harmful, overall, wellbeing, efficacy, healthy).

    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
     ሽ
    0.63
    নার্থ
    0.59
     defamation
    0.55
     اعتراض
    0.55
    шить
    0.52
     ineffective
    0.52
     ዓ
    0.51
    γκ
    0.50
     impeachment
    0.50
     sarcasm
    0.50
    POSITIVE LOGITS
     health
    3.06
    健康
    2.98
     건강
    2.98
    health
    2.92
     HEALTH
    2.81
    HEALTH
    2.72
    Health
    2.70
     Health
    2.69
     स्वास्थ्य
    2.69
     здоровье
    2.69
    Activations Density 0.641%

    No Known Activations