© 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. 98687
    Prev
    Next
    INDEX
    Explanations

    starting phrases for clauses

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

    The neuron fires on words and phrases that begin new sentences or sections—i.e. sentence-initial or heading tokens.

    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
     from
    -2.09
     and
    -1.84
    
    -1.68
     grudnia
    -1.66
     dovr
    -1.62
    ceptibility
    -1.56
     sierpnia
    -1.52
     AUSSI
    -1.52
    segno
    -1.49
     atque
    -1.48
    POSITIVE LOGITS
     -
    1.53
    …
    1.48
    -
    1.40
    ：……
    1.39
     apresent
    1.39
    지만
    1.38
     mantém
    1.30
    '
    1.30
     zrobić
    1.30
    ut
    1.30
    Activations Density 0.610%

    No Known Activations