© 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. 34-GEMMASCOPE-RES-131K
    7. 63957
    Prev
    Next
    INDEX
    Explanations

    felicitous amelioration accoutrements

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

    The neuron strongly activates on subword pieces from longer, less common multisyllabic words (often of Latin or foreign origin).

    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
    どうして
    -1.05
    我會
    -1.00
    我覺得
    -0.93
    ほんとに
    -0.93
    dacia
    -0.91
    为啥
    -0.90
    マジで
    -0.89
     및
    -0.87
     видели
    -0.86
    </h2>
    -0.85
    POSITIVE LOGITS
    麩
    0.93
     prepping
    0.90
    茄子
    0.90
    垢
    0.89
    菊
    0.88
     mengha
    0.87
     alb
    0.87
    Yearly
    0.87
    getPackageName
    0.87
    kün
    0.85
    Activations Density 0.059%

    No Known Activations