© 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. 34-GEMMASCOPE-RES-131K
    7. 116448
    Prev
    Next
    INDEX
    Explanations

    research report

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

    This neuron activates on metadata field names and numeric identifiers in structured markup or code (e.g. attribute/attributes in assembly info, image/synonyms/species in taxoboxes, “q” loci, HTML meta translations, copyright numbers, 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_34/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
    Sunlight
    -0.73
     SAMPLE
    -0.69
    cmd
    -0.67
    zul
    -0.66
    _{-\
    -0.66
     running
    -0.66
    fifty
    -0.64
     hundert
    -0.63
    cadas
    -0.63
    tz
    -0.62
    POSITIVE LOGITS
    læg
    0.68
    選挙
    0.68
     Trey
    0.67
     لیے
    0.66
     vont
    0.66
    printList
    0.65
     rescued
    0.65
     کارت
    0.65
    StackNavigator
    0.65
     Tbilisi
    0.64
    Activations Density 0.084%

    No Known Activations