© 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. 78307
    Prev
    Next
    INDEX
    Explanations

    XML tag close or structure definition

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

    The neuron responds primarily to XML/HTML structural tokens—especially the `<key>` tag and similar element markers.

    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
    tanooga
    -0.88
    Ques
    -0.85
    yst
    -0.72
     statute
    -0.71
    también
    -0.70
     horizont
    -0.70
    cartão
    -0.68
    mafia
    -0.68
    nima
    -0.67
     horizontal
    -0.67
    POSITIVE LOGITS
    相談
    0.83
    山口
    0.66
     相談
    0.66
    rieta
    0.66
    REPLY
    0.64
    reply
    0.64
    vindo
    0.64
    bah
    0.63
    ♫
    0.62
    super
    0.61
    Activations Density 0.084%

    No Known Activations