© 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. 10-GEMMASCOPE-RES-131K
    7. 129577
    Prev
    Next
    INDEX
    Explanations

    constructing things

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

    The neuron fires on words denoting the act or result of building or defining something—especially “construct,” “constructed,” “construction,” 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_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
    歃
    -1.87
     collè
    -1.71
     vixion
    -1.70
    킁
    -1.68
    ೣ
    -1.66
     cinquième
    -1.65
    cketed
    -1.62
    ㈯
    -1.61
    ícias
    -1.58
    妧
    -1.58
    POSITIVE LOGITS
     This
    2.19
     If
    1.94
     Since
    1.91
     for
    1.85
     Because
    1.77
     Now
    1.73
     with
    1.72
     Nearly
    1.71
     as
    1.70
     When
    1.70
    Activations Density 0.064%

    No Known Activations