© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 111453
    Prev
    Next
    INDEX
    Explanations

    advertisement navigation

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

    The neuron is detecting URL components—especially the “youtu.be” and “youtube.com/watch?v=” patterns—by firing strongly on domain tokens (like “be”), slashes, and URL query 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_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
    '
    -1.72
     .
    -1.66
    还有
    -1.65
     
    -1.61
     就
    -1.52
     So
    -1.52
    ほか
    -1.45
     $
    -1.43
    着一
    -1.43
     For
    -1.41
    POSITIVE LOGITS
    älfte
    1.77
     symboles
    1.63
    降临
    1.61
    癡
    1.59
    bundet
    1.58
    ızlık
    1.58
    鬭
    1.58
     êtres
    1.57
    śmy
    1.55
     内衣
    1.53
    Activations Density 0.004%

    No Known Activations