© 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. 10-GEMMASCOPE-RES-131K
    7. 58964
    Prev
    Next
    INDEX
    Explanations

    code definitions

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

    The neuron fires on C++ member‐function declaration keywords—especially trailing “const” qualifiers and overloaded “operator” definitions.

    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
    蘗
    -1.75
    what
    -1.68
    子は
    -1.65
     utilize
    -1.56
    </h1>
    -1.55
    私たちは
    -1.54
    ၟ
    -1.52
     purposely
    -1.52
     proactively
    -1.52
    
    -1.52
    POSITIVE LOGITS
    sandalia
    1.77
    The
    1.63
    “
    1.51
    ￬
    1.50
     verticale
    1.48
    magaz
    1.44
    "";
    1.44
     itinéraires
    1.42
    ↵
    1.41
     vocale
    1.41
    Activations Density 0.001%

    No Known Activations