© 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. 68387
    Prev
    Next
    INDEX
    Explanations

    code snippets with punctuation

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

    The neuron fires on literal “non-identifier” tokens in code—things like quoted strings, numeric literals, commas/brackets, and other punctuation symbols.

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

    No Comments

    Negative Logits
    Even
    -1.73
     If
    -1.66
    诸多
    -1.63
     it
    -1.60
     ALWAYS
    -1.53
    Every
    -1.49
    ra
    -1.48
    记得
    -1.44
    1
    -1.42
     This
    -1.41
    POSITIVE LOGITS
     such
    1.97
    tapis
    1.74
     something
    1.70
    ніка
    1.64
    – 
    1.55
    OUTUBE
    1.51
     nuov
    1.48
    镲
    1.47
    idlertid
    1.46
    tifkan
    1.46
    Activations Density 0.023%

    No Known Activations