© 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. 5933
    Prev
    Next
    INDEX
    Explanations

    meta tags and title

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

    The neuron detects SEO and metadata–related terms and labels (e.g. keywords, meta description, Open Graph tags, image metadata, downloads).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    technical web development and SEO-related terms, particularly metadata fields and configuration settings.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    technical terms referring to metadata fields, annotations, and web page SEO/Open Graph configuration.

    oai_token-act-pair · gpt-5Triggered 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
    halen
    0.58
    ξης
    0.56
     nút
    0.55
    basan
    0.53
     Factories
    0.49
     hul
    0.49
     Mills
    0.49
    vora
    0.48
     균
    0.48
     ذریع
    0.47
    POSITIVE LOGITS
    keywords
    0.92
    author
    0.86
     keywords
    0.85
    Keywords
    0.84
     Keywords
    0.83
    robots
    0.80
     author
    0.76
     автор
    0.75
     Robots
    0.74
    twitter
    0.73
    Activations Density 0.023%

    No Known Activations