© 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. 34-GEMMASCOPE-RES-131K
    7. 97796
    Prev
    Next
    INDEX
    Explanations

    original poster (OP)

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

    The neuron fires most strongly on high‐frequency function words (especially articles and prepositions like “the,” “to,” “by,” 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
     officials
    -0.84
    ające
    -0.79
     '';
    -0.77
    Officials
    -0.75
     Officials
    -0.73
    队
    -0.73
     ervoor
    -0.73
     Payable
    -0.72
    executive
    -0.69
    ߔ
    -0.69
    POSITIVE LOGITS
     OP
    5.66
    OP
    4.28
     op
    2.69
    op
    2.31
     Op
    2.17
    楼主
    2.14
    Op
    2.00
     original
    1.75
     ops
    1.55
    original
    1.52
    Activations Density 0.037%

    No Known Activations