© 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. 62778
    Prev
    Next
    INDEX
    Explanations

    CWE, extern, struct, class

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

    The neuron is detecting programming‐language type and generic declarations (e.g. class, struct, enum, protocol, and their angle‐bracketed type constraints).

    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
     klientów
    -0.85
    centaje
    -0.84
     bás
    -0.82
     景点
    -0.81
    tualmente
    -0.81
    giène
    -0.81
     willingness
    -0.80
    憶
    -0.79
    <0xD5>
    -0.79
     rigorous
    -0.79
    POSITIVE LOGITS
    ={({
    0.86
    int
    0.86
     DevComponents
    0.83
     achats
    0.77
     Zy
    0.76
     CCP
    0.76
    ле
    0.75
     ure
    0.74
    Tor
    0.72
    Foreign
    0.72
    Activations Density 0.003%

    No Known Activations