© 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. 10-GEMMASCOPE-RES-131K
    7. 60248
    Prev
    Next
    INDEX
    Explanations

    academic research papers

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

    The neuron fires most strongly on abstract‐style meta‐phrases that introduce the paper’s aim or approach (e.g. “In this paper…,” “In this study we…,” “Here we show/ investigate/ review…”).

    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
    ſe
    -2.64
     Eſ
    -2.55
     ſi
    -2.42
     eſt
    -2.39
     Inſ
    -2.23
    ſo
    -2.20
    ſi
    -2.09
     enfans
    -2.09
     ſon
    -2.08
     パソコン
    -2.08
    POSITIVE LOGITS
     مباراة
    2.22
     Selasa
    1.79
    אָ
    1.75
     Кубка
    1.75
    According
    1.74
     препара
    1.71
     препарат
    1.71
     сливо
    1.70
     салат
    1.70
     сообщили
    1.68
    Activations Density 7.314%

    No Known Activations