© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 34784
    Prev
    Next
    INDEX
    Explanations

    programming directives and technical terms

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

    The neuron is triggering on code‐style punctuation and file-name extensions (e.g. the “.” before .h or .c, semicolons, and similar syntax markers in programming contexts).

    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
    ändern
    -0.99
    :+:
    -0.99
    __).
    -0.97
     wordt
    -0.96
    ฿
    -0.93
    issão
    -0.88
    примеча
    -0.88
     siis
    -0.86
    Ū
    -0.86
    umptive
    -0.86
    POSITIVE LOGITS
     dichas
    1.14
    Finalize
    0.98
    Oferta
    0.96
     aquellas
    0.92
     زنان
    0.90
     cali
    0.90
    Addresses
    0.89
     frattempo
    0.89
    Calidad
    0.89
     地址
    0.88
    Activations Density 0.000%

    No Known Activations