© 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. 34-GEMMASCOPE-RES-131K
    7. 116357
    Prev
    Next
    INDEX
    Explanations

    digits and international words

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

    The neuron selectively fires on numeric tokens—especially digits representing years, dates, or literal numbers in code.

    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
     value
    -0.68
    CRP
    -0.66
    heim
    -0.66
    oshiba
    -0.65
    gut
    -0.64
     nation
    -0.63
    応え
    -0.63
    🐠
    -0.63
     mode
    -0.62
    mode
    -0.62
    POSITIVE LOGITS
    Objec
    0.74
    Sensi
    0.73
    Adapt
    0.68
    תי
    0.67
     préparation
    0.66
    同上
    0.66
    Переваги
    0.65
     segregation
    0.65
    Dek
    0.65
    handleClick
    0.65
    Activations Density 0.124%

    No Known Activations