© 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. 22-GEMMASCOPE-RES-131K
    7. 127709
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on code identifiers—especially variable names (including underscores or digits) in declarations and assignments.

    oai_token-act-pair · o4-mini
    Triggered by @jyhe0408

    code tokens and punctuation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
     bedenken
    -1.09
     into
    -1.04
     rằng
    -0.96
    rmdir
    -0.93
     significantly
    -0.92
     kleinere
    -0.91
     punctu
    -0.89
    //
    -0.88
     majority
    -0.88
    akteur
    -0.87
    POSITIVE LOGITS
     at
    1.19
     DbSet
    1.11
     permitan
    1.05
     durab
    0.95
    WillAppear
    0.91
    Kval
    0.91
     SIMBAD
    0.90
    ฅ
    0.90
    军团
    0.90
    éton
    0.90
    Activations Density 0.011%

    No Known Activations