© 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. 11678
    Prev
    Next
    INDEX
    Explanations

    ghormeh sabzi

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

    This neuron detects Arabic names—particularly the romanized personal names of early Muslim scholars and transmitters (with their characteristic apostrophes and “al-” prefixes).

    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
     gor
    0.53
    గానే
    0.53
     大学
    0.52
    ホール
    0.52
    ickers
    0.51
     总
    0.50
    ención
    0.49
     beach
    0.49
    疫
    0.49
    ኛውም
    0.49
    POSITIVE LOGITS
    ضب
    0.59
     مسئله
    0.57
    বাঁ
    0.56
    Stadt
    0.55
    bellar
    0.55
     растений
    0.55
    aryan
    0.55
    Ла
    0.54
    霈
    0.54
    лон
    0.54
    Activations Density 0.040%

    No Known Activations