© 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. 129229
    Prev
    Next
    INDEX
    Explanations

    holy Spirit, Ghost, or Family

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

    The neuron fires strongly on the capitalized adjective “Holy” when it appears in religious names or titles (e.g. “Holy Spirit,” “Holy Family,” “Holy Cross”).

    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
    توضی
    -1.75
     southeast
    -1.74
     业
    -1.67
     verschiedenen
    -1.63
     hideous
    -1.63
    ſelves
    -1.62
    نکته
    -1.61
    مشخص
    -1.59
    囁
    -1.58
     bekämp
    -1.57
    POSITIVE LOGITS
     ʻ
    1.90
    Ги
    1.68
    }$
    1.61
     Attending
    1.59
    捯
    1.58
    麁
    1.57
    
    1.52
     sufic
    1.50
     marx
    1.49
     vender
    1.48
    Activations Density 0.006%

    No Known Activations