© 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. 47310
    Prev
    Next
    INDEX
    Explanations

    Alto-Shaam

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

    The neuron is picking out tokens that belong to source code (operators like “!”, program identifiers, comment‐only keywords) rather than ordinary prose.

    oai_token-act-pair · o4-miniTriggered 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
    Not in Any Lists

    No Comments

    Negative Logits
     Schlä
    -0.86
    čník
    -0.83
    boek
    -0.79
    oplayer
    -0.79
    BuildContext
    -0.76
     necesarias
    -0.76
    tagName
    -0.75
    -0.75
    gir
    -0.73
    REGISTRATION
    -0.73
    POSITIVE LOGITS
    承
    0.74
     pengurus
    0.73
     Incon
    0.73
    wantErr
    0.70
     lasagna
    0.70
     incarnate
    0.70
     oregano
    0.69
    ヱ
    0.69
    емного
    0.68
     做法
    0.67
    Activations Density 0.000%

    No Known Activations