© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 118768
    Prev
    Next
    INDEX
    Explanations

    mobile navigation menu

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

    The neuron is detecting HTML/XML markup tokens—i.e. angle‐bracket delimiters (like “<” and “</”) and tag names.

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
    地说
    -0.90
     about
    -0.84
    限り
    -0.81
    Бележки
    -0.80
    akaian
    -0.79
     no
    -0.79
    mpeg
    -0.77
    ingroup
    -0.77
     talento
    -0.75
    各大
    -0.75
    POSITIVE LOGITS
    MOBILE
    0.86
    )}-\
    0.85
     opar
    0.82
     Khe
    0.82
     mobilen
    0.81
     illi
    0.79
    Mobile
    0.77
    XtraBars
    0.76
    χαν
    0.75
     mobile
    0.75
    Activations Density 0.002%

    No Known Activations