© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 3760
    Prev
    Next
    INDEX
    Explanations

    Almanjayar Health Centre

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

    citation or reference markers in text, such as slashes, brackets, or "In:" indicators.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron fires on tokens that serve as section or metadata markers—things like “almanac” in a title line, double‐slash separators (“//”), colons after “In”, and similar tokens used to mark headers, citations, or other structural divisions.

    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
     vermelhas
    -1.68
     femininas
    -1.64
     semelhantes
    -1.59
    ’
    -1.57
    นี่
    -1.55
     essenciais
    -1.49
     какво
    -1.47
    となりました
    -1.45
     adicionais
    -1.45
     nová
    -1.43
    POSITIVE LOGITS
    </h2>
    1.72
     of
    1.69
    </em>
    1.66
    </td>
    1.66
    もの
    1.55
    '],
    1.51
    </h6>
    1.48
    pinMode
    1.44
    '].
    1.44
    男士
    1.42
    Activations Density 0.002%

    No Known Activations