© 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. 80753
    Prev
    Next
    INDEX
    Explanations

    `<refentry>` structure

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

    The neuron strongly activates on the “<” token that begins XML/HTML-style tags, i.e. the opening of markup elements.

    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
    抬头
    -0.89
    将在
    -0.84
    みよう
    -0.82
    はずです
    -0.80
     direttore
    -0.80
    ICHE
    -0.80
     viaggio
    -0.79
    ூர்
    -0.78
    ELIN
    -0.78
    اویر
    -0.77
    POSITIVE LOGITS
    omg
    0.88
    投手
    0.86
     coordinate
    0.82
     coordinación
    0.81
     coordination
    0.81
    0.80
     Pensions
    0.80
     Coppola
    0.79
    ThroughAttribute
    0.79
     Coordination
    0.78
    Activations Density 0.015%

    No Known Activations