© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 70230
    Prev
    Next
    INDEX
    Explanations

    center alignment

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

    The neuron fires on numeric literals (digits) used in code—especially numbers specifying layout or style values (e.g. CSS flex properties, widths, margins, alignment settings).

    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
    DateFormatter
    -0.86
    getReal
    -0.79
    lierung
    -0.75
    ᾗ
    -0.74
    équences
    -0.74
     鱼
    -0.73
    leton
    -0.72
    ingar
    -0.72
    睡觉
    -0.71
     mengikut
    -0.71
    POSITIVE LOGITS
     center
    3.13
     centre
    2.61
    center
    2.53
     Center
    2.34
     centers
    2.27
    Center
    2.19
     CENTER
    2.05
     centres
    1.99
     цент
    1.90
     Centre
    1.85
    Activations Density 0.003%

    No Known Activations