© 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. 24955
    Prev
    Next
    INDEX
    Explanations

    programming concepts and data structures

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

    The neuron strongly activates on code‐related tokens—programming keywords, identifiers, punctuation (like brackets or braces), and numeric literals—marking out code snippets.

    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_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
     LMS
    -0.81
     حاول
    -0.77
    ன்று
    -0.76
    会计
    -0.74
    btnNew
    -0.73
    หมาย
    -0.71
     neuve
    -0.71
    Ethernet
    -0.70
    ovem
    -0.70
    🏏
    -0.69
    POSITIVE LOGITS
    hunt
    0.81
    õi
    0.80
    tsu
    0.79
    ssr
    0.78
     marketed
    0.76
    こうして
    0.76
     Specialized
    0.75
    それだけ
    0.75
     ք
    0.74
    quias
    0.74
    Activations Density 0.085%

    No Known Activations