© 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. 24-GEMMASCOPE-2-RES-16K
    7. 3536
    Prev
    Next
    INDEX
    Explanations

    exclamations ending in punctuation

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

    The neuron fires on link‐style text—especially web addresses (e.g. “vill.edu”), “click here,” “posted in…,” and other URL or hyperlink patterns.

    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_24_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
    ine
    0.61
     अत्याधुनिक
    0.59
    u
    0.58
    GO
    0.57
    cled
    0.56
    ｏ
    0.55
    dolphin
    0.55
    ómetro
    0.55
    feitos
    0.54
    venidos
    0.54
    POSITIVE LOGITS
    🤭
    0.60
    </tr>
    0.57
    ។
    0.56
    😑
    0.55
     caranya
    0.55
     handleChange
    0.54
    ።
    0.54
     една
    0.52
     sommige
    0.52
    childNodes
    0.50
    Activations Density 0.011%

    No Known Activations