© 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. 123485
    Prev
    Next
    INDEX
    Explanations

    describing typical circumstances

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

    The neuron fires on discourse‐level frequency adverbs (e.g. “usually,” “typically”) used to introduce general statements.

    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
    でも
    -3.11
    OUTUBE
    -2.94
    镠
    -2.77
     kommende
    -2.75
    蟊
    -2.70
    嫑
    -2.64
    푘
    -2.59
     vijf
    -2.56
    虓
    -2.50
    푙
    -2.47
    POSITIVE LOGITS
    of
    3.25
    a
    2.91
    what
    2.86
    '
    2.64
    ot
    2.61
     seems
    2.61
     brings
    2.59
    at
    2.59
     takes
    2.56
     allow
    2.55
    Activations Density 0.029%

    No Known Activations