© 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. 44322
    Prev
    Next
    INDEX
    Explanations

    Alu repeats, FET, VL

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

    names and acronyms, particularly technical terms like "Hua", "Tul", "Palazzo", "FET", and "VLTI".

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron strongly activates on capitalized proper nouns (named entities).

    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
    e
    -3.22
    u
    -3.08
    x
    -3.05
    le
    -2.97
    d
    -2.92
    k
    -2.91
    t
    -2.88
    ic
    -2.83
    <i>
    -2.72
    ad
    -2.69
    POSITIVE LOGITS
     și
    2.94
    蠖
    2.59
    سجيلات
    2.52
    ي
    2.42
     juara
    2.39
     Türkei
    2.39
     Auffassung
    2.38
    oulders
    2.38
     trabajó
    2.38
     доволь
    2.36
    Activations Density 0.016%

    No Known Activations