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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 92866
    Prev
    Next
    INDEX
    Explanations

    `.com` URLs and code snippets

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

    The neuron strongly activates on segments of technical identifiers—e.g. individual components and separators in file‐include paths, URIs/namespace URLs, and XML schema URLs.

    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
    デモ
    -0.73
    燮
    -0.65
    Ве
    -0.65
    Пере
    -0.64
    Сте
    -0.64
     гриб
    -0.63
     чита
    -0.63
    Чита
    -0.62
     Yas
    -0.62
    微微
    -0.61
    POSITIVE LOGITS
    uks
    0.73
     blauw
    0.71
    pocket
    0.70
     pocket
    0.69
     Roper
    0.68
     Nakamura
    0.65
    شوف
    0.65
    vga
    0.65
    upy
    0.64
     semicolon
    0.64
    Activations Density 0.088%

    No Known Activations