© 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. 22-GEMMASCOPE-RES-131K
    7. 33878
    Prev
    Next
    INDEX
    Explanations

    month names, dates, numbers

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

    The neuron fires on standalone topic‐ or category‐labels (the little “tags” or headings like php, mysql, Ionic, Video, Slashdot, etc.).

    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
    مقدمه
    -1.06
    Ẓ
    -1.05
    Ḫ
    -1.03
    částí
    -1.00
    Kesimpulan
    -0.99
    werden
    -0.98
    لغة
    -0.98
    hető
    -0.97
    zehn
    -0.96
    توضیحات
    -0.96
    POSITIVE LOGITS
     is
    1.73
     always
    1.37
    keteers
    1.21
     included
    1.19
     should
    1.17
     maybe
    1.11
     shouldn
    1.10
     will
    1.09
    しくは
    1.07
     too
    1.06
    Activations Density 0.085%

    No Known Activations