© 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. 22-GEMMASCOPE-RES-131K
    7. 84112
    Prev
    Next
    INDEX
    Explanations

    address entries

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

    This neuron fires on tokens that serve as labels or field/identifier names marking definitions or sections (e.g. “address,” “code,” “report,” “On,” the footnote marker “ddag,” 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
    werf
    -1.11
    chea
    -0.93
    erver
    -0.89
    arita
    -0.89
    urige
    -0.86
    !!!!!!!!!
    -0.86
     страница
    -0.86
     beset
    -0.85
     PRT
    -0.84
     !!!!!
    -0.84
    POSITIVE LOGITS
     with
    1.03
     on
    0.92
     higher
    0.92
     experience
    0.90
     through
    0.90
     skilled
    0.86
     comparatively
    0.84
     usually
    0.83
     sadly
    0.83
    hankelijk
    0.83
    Activations Density 0.001%

    No Known Activations