© 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. 22-GEMMASCOPE-RES-131K
    7. 15666
    Prev
    Next
    INDEX
    Explanations

    citations like [@B1-nanomaterials]

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

    Roman numerals and bibliographic citation markers in academic papers.

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

    The neuron detects uppercase Roman numerals used as section or outline headings.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     głów
    -0.83
    therefore
    -0.75
     WASTE
    -0.74
     aquele
    -0.72
     décoration
    -0.71
    opa
    -0.70
    talks
    -0.68
     yana
    -0.68
     nguoi
    -0.68
    idespread
    -0.67
    POSITIVE LOGITS
    mayın
    0.92
    igem
    0.90
     EOF
    0.87
     translateY
    0.87
    älä
    0.86
    CGContext
    0.82
     ComboBox
    0.82
    iaire
    0.82
    льна
    0.79
    ulaires
    0.79
    Activations Density 0.047%

    No Known Activations