© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 101366
    Prev
    Next
    INDEX
    Explanations

    followed by or alongside

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

    The neuron fires strongly on two-word linking phrases that introduce or connect elements—especially verb+“by” collocations (e.g. “followed by,” “hosted by”) and coordinating pairs like “as well as” or “along with.”

    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
     who
    -1.55
     your
    -1.52
     in
    -1.38
     is
    -1.30
     of
    -1.27
     will
    -1.20
    see
    -1.19
     can
    -1.18
     would
    -1.17
     its
    -1.13
    POSITIVE LOGITS
    бовь
    1.29
    énégal
    1.27
     exportación
    1.25
    ครื่อง
    1.22
     Unterscheidung
    1.20
    notFound
    1.20
     folgte
    1.19
    ❓
    1.18
    mittag
    1.16
    了两
    1.16
    Activations Density 0.038%

    No Known Activations