© 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. 40014
    Prev
    Next
    INDEX
    Explanations

    html and php file extensions

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

    The neuron fires strongly on URL path segments and related tokens (e.g. “article,” “story,” “about_alerts,” file extensions like “html,” underscored names), i.e. bits of web‐link URLs.

    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
    s
    -1.60
    ––––
    -1.58
     The
    -1.56
     ><
    -1.52
     inklusive
    -1.48
    ???:
    -1.46
    一阵
    -1.45
    急忙
    -1.45
    盯着
    -1.44
     ▸
    -1.44
    POSITIVE LOGITS
     really
    2.06
     usually
    2.03
     often
    1.99
     actually
    1.91
     improve
    1.82
     always
    1.82
     about
    1.82
     also
    1.81
     nicely
    1.75
    僞
    1.74
    Activations Density 0.069%

    No Known Activations