© 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. 22-GEMMASCOPE-RES-131K
    7. 14231
    Prev
    Next
    INDEX
    Explanations

    ever used or owned

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

    The neuron fires on superlative praise expressions—especially “best” (often in “best … ever”)—in positive product or service reviews.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    <blockquote>
    -0.98
    Appellant
    -0.98
    -0.94
    UserDefaults
    -0.93
     gewähr
    -0.90
    FIGURE
    -0.90
    ଶ
    -0.89
     判断
    -0.88
     चुनें
    -0.88
    ๋า
    -0.86
    POSITIVE LOGITS
    Semoga
    1.23
    今まで
    1.18
     experience
    1.17
    ادگی
    1.16
     since
    1.16
     ever
    1.09
     навіть
    1.09
    gemeente
    1.08
     in
    1.07
     even
    1.06
    Activations Density 0.006%

    No Known Activations