© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 47653
    Prev
    Next
    INDEX
    Explanations

    .m file paths

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

    This neuron detects capitalized code identifiers (e.g. class or component names in CamelCase or all‐caps acronyms).

    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
     Irán
    -1.92
     Souhaitez
    -1.75
    쨩
    -1.73
    橚
    -1.70
     WHEN
    -1.68
     Voyez
    -1.68
     Pourquoi
    -1.67
     Primero
    -1.65
     Prenez
    -1.63
     Tercera
    -1.61
    POSITIVE LOGITS
     from
    2.41
     which
    1.91
    </i>
    1.89
     it
    1.89
     for
    1.80
     with
    1.75
    d
    1.67
     in
    1.61
     two
    1.60
     面包
    1.58
    Activations Density 0.001%

    No Known Activations