© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 48180
    Prev
    Next
    INDEX
    Explanations

    numbers and code snippets

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

    The neuron primarily detects multi‐digit numeric tokens (e.g., years, street numbers, phone numbers).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    Accounting
    -0.76
     انگی
    -0.74
    Pager
    -0.74
    oane
    -0.73
    aliyet
    -0.73
     Laurel
    -0.72
    ThemeOverlay
    -0.72
    сен
    -0.72
    similarity
    -0.72
     الموقع
    -0.71
    POSITIVE LOGITS
     NW
    1.13
    NW
    0.88
     DC
    0.83
     客厅
    0.79
    DC
    0.75
    実感
    0.73
     predicting
    0.71
     Georgetown
    0.71
     unaware
    0.70
    wah
    0.70
    Activations Density 0.028%

    No Known Activations