© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 67083
    Prev
    Next
    INDEX
    Explanations

    academic citation markers

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

    This neuron detects numeric reference markers and metadata tokens (e.g. citation keys like “[@CR2]” or date stamps) in scholarly text.

    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
    yā
    -1.08
     ilustração
    -1.00
    に到着
    -0.89
     favoring
    -0.88
    年は
    -0.86
    рений
    -0.86
    詛
    -0.85
    ⾮
    -0.83
    нале
    -0.83
     добра
    -0.82
    POSITIVE LOGITS
    人都
    1.01
    Possible
    0.99
     "\
    0.96
     ходу
    0.95
     возможно
    0.94
    ecin
    0.94
    roned
    0.93
     возмо
    0.92
    っしゃ
    0.89
     людям
    0.88
    Activations Density 0.010%

    No Known Activations