© 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. 78437
    Prev
    Next
    INDEX
    Explanations

    qualification competition

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

    The neuron strongly activates on code‐file and header‐name tokens—especially “.hh” filenames in #include directives, version numbers in source‐control diffs, and other file‐path or file‐extension identifiers.

    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
     svg
    -0.73
    validates
    -0.69
    smtp
    -0.68
    odynamics
    -0.65
    Download
    -0.64
     szö
    -0.64
    قي
    -0.61
    lave
    -0.61
    ENDIF
    -0.60
     Conte
    -0.59
    POSITIVE LOGITS
    ATEGORIES
    0.70
    eurovision
    0.69
    SAFETY
    0.68
    MONITOR
    0.68
    hotel
    0.66
     古い
    0.65
     beauties
    0.65
    OTHY
    0.65
    ITU
    0.64
     unearthed
    0.64
    Activations Density 0.068%

    No Known Activations