© 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. 118789
    Prev
    Next
    INDEX
    Explanations

    start of for loops

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

    This neuron fires strongly on loop headers—especially `for` loops and their surrounding parentheses and initialization 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_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
    ضاء
    -0.97
     Sonja
    -0.90
    derna
    -0.89
    нулся
    -0.87
    Karakteristik
    -0.87
     Alamy
    -0.86
    žně
    -0.84
    もあります
    -0.83
     marchio
    -0.83
    動き
    -0.82
    POSITIVE LOGITS
     I
    0.98
     Of
    0.88
    …”
    0.85
     kasi
    0.81
     Danh
    0.77
     i
    0.75
    ffective
    0.75
    mau
    0.73
     tých
    0.73
     “
    0.73
    Activations Density 0.010%

    No Known Activations