© 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. 34-GEMMASCOPE-RES-131K
    7. 30082
    Prev
    Next
    INDEX
    Explanations

    code punctuation and structure

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

    The neuron strongly activates on code‐style package or module path tokens—i.e. dotted identifiers (like “com.facebook.react.bridge” or “android.os.Bundle”) in import/package lines.

    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_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
     Rückkehr
    -0.75
     BATTLE
    -0.75
    Poker
    -0.74
    BIZ
    -0.73
    toDate
    -0.73
     lah
    -0.71
     BIBLE
    -0.71
    グリップ
    -0.67
     מלח
    -0.67
     kız
    -0.67
    POSITIVE LOGITS
    ofila
    0.78
    หวัด
    0.76
    angelo
    0.72
    恤
    0.66
     paraffin
    0.66
    ivotopis
    0.65
    كز
    0.65
     電車
    0.63
    六十
    0.63
    mleri
    0.63
    Activations Density 0.053%

    No Known Activations