© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 84785
    Prev
    Next
    INDEX
    Explanations

    URLs with forum or blog

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

    This neuron detects references to online discussion forums, especially URLs or link text containing “forum” or “forums” and associated thread/index numbers.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     but
    -2.31
     traditionally
    -1.98
     also
    -1.98
     reputable
    -1.82
     çeşitli
    -1.65
    但也
    -1.64
     and
    -1.62
     commodities
    -1.60
     in
    -1.59
     dubious
    -1.55
    POSITIVE LOGITS
     itſelf
    1.81
    »;
    1.80
    🌊
    1.75
     rajut
    1.72
    !");
    1.71
     ★★★
    1.68
    hehehe
    1.66
    鹀
    1.62
     Sénat
    1.61
     HOLIDAY
    1.61
    Activations Density 0.028%

    No Known Activations