© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 38890
    Prev
    Next
    INDEX
    Explanations

    names that form titles or categories

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

    This neuron detects occurrences of the name “Paul” (especially as part of proper names like “John Paul II” or “Jean-Paul”).

    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
     rife
    -1.73
     sebagian
    -1.63
     worthwhile
    -1.47
     them
    -1.44
    ありません
    -1.41
     لئے
    -1.40
    ."
    -1.39
     it
    -1.38
    这些人
    -1.37
    היו
    -1.34
    POSITIVE LOGITS
    𓋼
    1.70
    ೢ
    1.67
    鹩
    1.63
    튿
    1.48
    IDAD
    1.47
    阄
    1.47
     Motorrad
    1.44
    Discografia
    1.43
    垸
    1.43
     Toilette
    1.42
    Activations Density 0.002%

    No Known Activations