© 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. 82591
    Prev
    Next
    INDEX
    Explanations

    patent numbers and URLs

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

    the neuron detects tokens that signify technical resource references or import paths—such as question marks and slashes in URLs and keywords like “using” in code.

    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
     gjerne
    -1.02
    DLL
    -0.83
    isierte
    -0.82
     betyder
    -0.81
     Moira
    -0.80
    Lua
    -0.79
    |':'
    -0.78
     Iteration
    -0.76
    fär
    -0.75
     tän
    -0.74
    POSITIVE LOGITS
    hole
    0.77
    oogle
    0.77
    erweise
    0.76
    Poloha
    0.75
    նակ
    0.74
    bole
    0.73
    陋
    0.71
    ᆫ
    0.70
     Durchschnitt
    0.69
     aspetti
    0.68
    Activations Density 0.001%

    No Known Activations