© 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. 8191
    Prev
    Next
    INDEX
    Explanations

    words ending in up or s

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

    The neuron strongly responds to multi‐word or hyphenated technical/compound nouns (e.g. “coupon‐clearinghouse,” “file system hierarchies,” “trade‐offs,” “degrees of freedom”).

    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
     we
    -1.63
    críbete
    -1.53
    年底
    -1.47
    打算
    -1.45
    }=\
    -1.41
     to
    -1.41
     paſſ
    -1.38
     affitto
    -1.36
     nhưng
    -1.34
    ]=\
    -1.34
    POSITIVE LOGITS
    punkten
    1.45
     what
    1.37
     another
    1.33
     Как
    1.31
    脍
    1.24
    嚮
    1.23
     What
    1.23
    额头
    1.23
    壓力
    1.22
    
    1.22
    Activations Density 0.537%

    No Known Activations