© 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. 22-GEMMASCOPE-RES-131K
    7. 66359
    Prev
    Next
    INDEX
    Explanations

    Christ suffering

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

    The neuron activates on numeric tokens or digit sequences (e.g., numbers, IDs, version numbers) embedded in 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
    兩
    -1.60
    來
    -1.49
    利
    -1.13
    數
    -1.13
    ώ
    -1.08
    論
    -1.07
    度
    -1.05
    流
    -1.04
    金
    -0.99
     encontramos
    -0.98
    POSITIVE LOGITS
     می‌گو
    1.00
     versioni
    0.96
     Inſ
    0.94
    🧫
    0.92
     acuer
    0.90
     proche
    0.88
     głoś
    0.87
    couts
    0.87
     erzielt
    0.87
     descriptive
    0.86
    Activations Density 21.351%

    No Known Activations