© 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. 13592
    Prev
    Next
    INDEX
    Explanations

    a)

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

    This neuron detects single-letter list or section markers (e.g. “a,” “b,” “c,” “i”) used as item or footnote labels.

    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
    ről
    0.77
    jeni
    0.77
    یر
    0.71
    ról
    0.70
    ből
    0.70
    Ỏ
    0.70
    orit
    0.69
    ek
    0.69
    elon
    0.68
     volver
    0.66
    POSITIVE LOGITS
     Кстати
    0.59
    хов
    0.59
    CMR
    0.56
    ճ
    0.55
    โร
    0.55
    𝑛
    0.53
    分かり
    0.52
    কম
    0.52
    SMP
    0.52
    CAG
    0.52
    Activations Density 0.442%

    No Known Activations