© 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. 65604
    Prev
    Next
    INDEX
    Explanations

    numeric identifiers and punctuation

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

    The neuron fires on numeric tokens—digits or number‐laden strings (e.g. years, measurements, financial figures).

    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
     puisi
    -1.01
     Giochi
    -0.91
     amma
    -0.91
     دخترانه
    -0.90
     brilla
    -0.89
    Кстати
    -0.89
    kning
    -0.88
     testimon
    -0.87
     baj
    -0.85
     cristian
    -0.83
    POSITIVE LOGITS
    AfterViewInit
    0.95
     entregó
    0.93
    によく
    0.93
     utilizó
    0.92
     reclamar
    0.88
     esfor
    0.88
     鸡蛋
    0.87
     ofreció
    0.87
    oluted
    0.87
     approximate
    0.87
    Activations Density 0.066%

    No Known Activations