© 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. 11386
    Prev
    Next
    INDEX
    Explanations

    structured lists or outlines

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

    This neuron strongly activates on formatting markers and metadata headings (e.g. bold markers like “**” and quoted section titles).

    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
    <start_of_image>
    0.90
    </b>
    0.73
    Еще
    0.67
    мын
    0.66
    كين
    0.65
    ស្ស
    0.64
    هاي
    0.64
    EF
    0.62
    ANDO
    0.62
    ટ
    0.62
    POSITIVE LOGITS
    <em>
    0.85
     继续访问
    0.70
    <strong>
    0.68
    lük
    0.66
    </em>
    0.64
    ità
    0.58
     gating
    0.58
    :</
    0.57
    ridate
    0.57
     annihil
    0.57
    Activations Density 0.029%

    No Known Activations