© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 45737
    Prev
    Next
    INDEX
    Explanations

    nam, Mai, or names ending in nam

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

    The neuron fires whenever it sees the subword “nam” in a token (e.g. as part of proper names or “Gangnam”), regardless of context.

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
    
    -2.61
     feiner
    -2.61
     schönes
    -2.55
    垕
    -2.53
    婳
    -2.36
     großes
    -2.34
    驁
    -2.31
    鹮
    -2.30
    nord
    -2.27
    mitsub
    -2.22
    POSITIVE LOGITS
    ar
    3.53
    e
    3.31
    ’
    3.30
    re
    3.00
    ra
    2.89
     Ironically
    2.89
    s
    2.88
    us
    2.80
     Interestingly
    2.77
     Virtually
    2.66
    Activations Density 0.003%

    No Known Activations