© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 130735
    Prev
    Next
    INDEX
    Explanations

    mathematical equations and symbols

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

    This neuron doesn’t reliably activate on any text—it appears effectively “dead” (it doesn’t detect any pattern).

    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
     of
    -2.09
     who
    -1.89
    鼉
    -1.63
    孩子
    -1.62
    r
    -1.44
     dieses
    -1.39
    鷽
    -1.35
     우리
    -1.34
     适合
    -1.34
    футболка
    -1.33
    POSITIVE LOGITS
    9
    1.63
    鬘
    1.52
     Berücksich
    1.41
     that
    1.40
     to
    1.38
     neuesten
    1.38
    nought
    1.34
    formerly
    1.34
     mittler
    1.28
    ときは
    1.27
    Activations Density 0.009%

    No Known Activations