© 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. 73066
    Prev
    Next
    INDEX
    Explanations

    introduction or context setting

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

    The neuron picks up Chinese prepositional markers—especially the word “基于” (meaning “based on”) and similar function words.

    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
    aş
    -2.30
    、“
    -2.28
    や
    -2.28
    a
    -2.28
     =
    -2.27
    atifs
    -2.17
    ")
    -2.16
    ，“
    -2.11
     detalladas
    -2.11
     decoradas
    -2.09
    POSITIVE LOGITS
    他
    2.55
    replyreport
    2.39
    峩
    2.38
    ים
    2.33
    Ꮉ
    2.25
     交換
    2.19
    ࿙
    2.14
     Kenntnis
    2.13
    >
    2.11
    籏
    2.11
    Activations Density 0.001%

    No Known Activations