© 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. 10-GEMMASCOPE-RES-131K
    7. 68326
    Prev
    Next
    INDEX
    Explanations

    In Vivo, DIY Day, Q:

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

    The neuron is highly sensitive to non-English or foreign tokens (especially those with diacritics or unusual letter sequences) indicating multilingual 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_10/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
    ;
    -1.98
     expansive
    -1.93
    ’
    -1.88
     inspiradoras
    -1.88
    にご
    -1.84
    维基
    -1.82
     daba
    -1.81
    ‌های
    -1.80
     inappropriés
    -1.75
    栒
    -1.75
    POSITIVE LOGITS
     Commonly
    1.91
     Sometime
    1.90
     suốt
    1.89
     разных
    1.84
     usher
    1.83
     Except
    1.83
     Exploring
    1.78
    簠
    1.78
     Examining
    1.78
     Ideally
    1.77
    Activations Density 0.002%

    No Known Activations