© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 12791
    Prev
    Next
    INDEX
    Explanations

    potential carcinogen

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

    The neuron strongly activates on clinical‐study language—especially biomedical outcome statements (e.g. “result in fewer deaths from colorectal cancer”).

    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
     lipase
    0.78
     ওষুধের
    0.70
     obat
    0.68
    药物
    0.66
     ওষুধ
    0.66
     antibiot
    0.66
     intravenously
    0.66
    藥
    0.65
     medicamento
    0.64
     dextrose
    0.64
    POSITIVE LOGITS
     carcinogenesis
    1.11
     Screening
    1.02
    Cancer
    0.96
     cancer
    0.95
     screening
    0.95
     स्क्रीनिंग
    0.95
     Preventive
    0.95
     Cancer
    0.94
     carcinogenic
    0.93
     Cerv
    0.93
    Activations Density 0.213%

    No Known Activations