© 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. 15971
    Prev
    Next
    INDEX
    Explanations

    Sagrada Familia, SageMaker, HOPE

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

    The neuron chiefly fires on tokens containing an uppercase S followed shortly by a “g” (e.g. “Sag,” “Seg,” “Sig” substrings in proper names).

    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
     Camin
    0.55
     موقف
    0.54
    ensioni
    0.54
    шит
    0.53
    elh
    0.53
    拍照
    0.52
     #+#
    0.50
    ाम्
    0.50
     finais
    0.50
     iniz
    0.50
    POSITIVE LOGITS
    ittarius
    0.75
     resetting
    0.58
    Seg
    0.57
    πως
    0.54
    seg
    0.53
     Sig
    0.53
    coating
    0.52
    昦
    0.51
     rapport
    0.50
    rapport
    0.50
    Activations Density 0.143%

    No Known Activations