© 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. 2365
    Prev
    Next
    INDEX
    Explanations

    This neuron fires on mentions of religious or sacred buildings (e.g., temple, church, cathedral).

    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
     intrauterine
    0.56
    ंडर
    0.54
     घरात
    0.54
    嵓
    0.54
     indeterminate
    0.52
     |_{
    0.52
     اتح
    0.52
     thế
    0.52
    ին
    0.52
    ຜະລິດຕະພັນ
    0.51
    POSITIVE LOGITS
     வளாக
    1.09
     ખાતે
    1.06
    gebäude
    1.06
    门口
    1.02
     परिसर
    1.01
    での
    0.99
    址
    0.95
     walls
    0.94
     gates
    0.93
     doors
    0.92
    Activations Density 0.803%

    No Known Activations