© 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. 27585
    Prev
    Next
    INDEX
    Explanations

    scientific citations or technical descriptions

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

    citation years in academic or technical documents.

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

    This neuron spots structural citation markers—digits, hyphens, and apostrophes—commonly found in reference dates, numbering, and citation formatting.

    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
     graceful
    -0.95
    หลัง
    -0.93
     spazi
    -0.91
     موفق
    -0.85
    复杂
    -0.84
    เพราะ
    -0.83
     because
    -0.82
    mya
    -0.82
    hiya
    -0.82
     through
    -0.82
    POSITIVE LOGITS
    results
    0.92
    又有
    0.90
     puerto
    0.87
     useSelector
    0.86
     REQU
    0.84
     coupé
    0.84
    studies
    0.81
    stander
    0.79
    讣
    0.78
     FormsModule
    0.77
    Activations Density 0.001%

    No Known Activations